Seminars in Fetal & Neonatal MedicineAmsterdam • Boston • London • New York • Oxford • Paris • Philadelphia • San Diego • St. Louis

Aims and ScopeSeminars in Fetal & Neonatal Medicine (formerly Seminars in Neonatology) is a bi-monthly journal which publishes topic-based issues, including current ‘Hot Topics’ on the latest advances in fetal and neonatal medicine. The change in title relates to the growing interest amongst obstetricians, midwives and fetal medicine specialists.

The Journal commissions review-based content covering current clinical opinion on the care and treatment of the neonate and draws on the necessary specialist knowledge, including that of the respiratory physician, the infectious disease physician, the surgeon, as well as the paediatrician and obstetrician.

Each topic-based issue is edited by an authority in their field and contains 8–10 articles.

Current and forthcoming events can be viewed on the Internet at: http://www.elsevier.com/locate/siny

Seminars in Fetal & Neonatal Medicine provides: • coverage of major developments in neonatal care; • value to practising neonatologists, consultant and trainee paediatricians,

obstetricians, midwives and fetal medicine specialists wishing to extend their knowledge in this field;

• up-to-date information in an attractive and relevant format.

Editorial BoardEditor-in-ChiefProfessor M I LeveneUniversity of Leeds

School of Medicine

D Floor, Clarendon Wing

The General Infirmary at Leeds

Leeds LS2 9NS, UK

Associate EditorsM Blennow, Huddinge, Sweden K Maršál, Lund, SwedenL Cornette, Brugge, Belgium D Peebles, London, UKD J Field, Leicester, UK S Sinha, Middlesbrough, UKI Laing, Edinburgh, UK A M Weindling, Liverpool, UK

Advisory Board

F A Chervenak, USA P C NG, Hong KongS M Donn, USA J M Perlman, USAN Evans, Australia E Saliba, FranceV Fellman, Sweden M Vento, SpainN N Finer, USA L de Vries, The Netherlands

at SciVerse ScienceDirect

Seminars in Fetal & Neonatal Medicine 17 (2012) 119

Contents lists available

Seminars in Fetal & Neonatal Medicine

journal homepage: www.elsevier .com/locate/s iny

Editorial

The late and moderate preterm baby

This edition of Seminars in Fetal and Neonatal Medicineconsiders birth at moderate and late preterm gestations, definedthroughout as birth at 32þ0�33þ6 weeks and 34þ0�36þ6 weeksof gestation respectively. Moderate and late preterm births accountfor over 75% of all preterm births and around 6–7% of all births; yetresearch in this group has been relatively limited, with the focushaving been predominantly on extreme prematurity.

During recent years, epidemiological studies have reportedadverse outcomes in these more mature preterm babies and thishas led to an upsurge in interest in this group. Within this rapidlygrowing area of research, results of many new studies are confirm-ing previous findings. As might be expected, mortality, severeneonatal morbidity and profound neurodevelopmental disabilityare certainly much less common outcomes in the moderate andlate preterm group than in extremely preterm infants. However,the sheer numbers of babies born at these gestations mean thateven a modest increase in adverse outcomes is likely to have animpact that should not be underestimated. The extent to whichlong term health and developmental problems can be attributedsimply to prematurity is not clear and it is likely that a numberof other factors also contribute, to a greater or lesser degree. Thesemight include the effects of longstanding maternal illness,

1744-165X/$ – see front matter � 2012 Elsevier Ltd. All rights reserved.doi:10.1016/j.siny.2012.02.005

pregnancy-related complications, fetal compromise, timing andmode of delivery, neonatal or post-neonatal management andsocial risk as well as other, perhaps as yet unidentified risk factors.Exploration of these issues is in its infancy; there is muchwork stillto be done to optimise care for women who deliver just a fewweeks prematurely and their babies.

In the papers that follow, we are fortunate to be able to presentreviews from authors who have generated and collatedmuch of theavailable evidence in the field of moderate and late prematurity.They provide a comprehensive and current review of key topics,as well as pointing to areas that are ripe for future research, ofwhich there are many.We are extremely grateful for their contribu-tions and trust that you will enjoy and learn from this interestingand stimulating collection of articles.

Elaine M. Boyle*Department of Health Sciences, 22-28 Princess Road West,

University of Leicester, Leicester LE1 6TP, UK

* Tel.: þ44 116 252 5447.E-mail address: eb124@le.ac.uk

at SciVerse ScienceDirect

Seminars in Fetal & Neonatal Medicine 17 (2012) 120e125

Contents lists available

Seminars in Fetal & Neonatal Medicine

journal homepage: www.elsevier .com/locate/s iny

Epidemiology of late and moderate preterm birth

Carrie K. Shapiro-Mendoza*, Eve M. LackritzMaternal and Infant Health Branch, Division of Reproductive Health, Centers for Disease Control and Prevention, Mailstop K-23, 4770 Buford Highway, NE, Atlanta,GA 30341-3717, USA

Keywords:EpidemiologyObstetrics standardsPregnancy outcomePremature birthPrenatal careRisk factors

* Corresponding author. Tel.: þ1 770 488 6263; faxE-mail address: ayn9@cdc.gov (C.K. Shapiro-Mend

1744-165X/$ e see front matter Published by Elseviedoi:10.1016/j.siny.2012.01.007

s u m m a r y

Preterm birth affects 12.5% of all births in the USA. Infants of Black mothers are disproportionatelyaffected, with 1.5 times the risk of preterm birth and 3.4 times the risk of preterm-related mortality. Thepreterm birth rate has increased by 33% in the last 25 years, almost entirely due to the rise in late pretermbirths (34e36 weeks’ gestation). Recently attention has been given to uncovering the often subtlemorbidity and mortality risks associated with moderate (32e33 weeks’ gestation) and late pretermdelivery, including respiratory, infectious, and neurocognitive complications and infant mortality. Thissection summarizes the epidemiology of moderate and late preterm birth, case definitions, risk factors,recent trends, and the emerging body of knowledge of morbidity and mortality associated with moderateand late preterm birth.

Published by Elsevier Ltd.

1. Introduction

Preterm birth (<37 weeks gestation) affects approximately onein eight (12.5% in 2008) of all births in the USA each year.1 Pretermbirth is the most frequent cause of infant mortality, as well as theleading cause of long term neurologic disabilities in children,including cerebral palsy and developmental delays. It is estimatedthat preterm birth costs the US healthcare system more than $26billion each year.2 Black infants are disproportionately affected;infants of non-Hispanic Black women have >1.5 times the risk ofpreterm birth and 3.4 times the risk of preterm-related mortalitycompared with infants of White mothers.3 Reducing the highburden of preterm birth, and its associated morbidity, mortality,and racial disparities, has thus been identified as a public healthpriority, as reflected by the 2006 PREEMIE Act,4 the 2007 Instituteof Medicine Report on preterm birth,2 the 2006 Surgeon General’sconference,5 the 2007 US Department of Health and HumanServices’ public awareness campaign,6 and Healthy People 2010and 2020 objectives.7

The survival of preterm infants improved greatly in recentdecades, primarily due to advances in clinical managementincluding neonatal intensive care units (NICUs), paediatric venti-lators, and use of surfactant and antenatal steroids. Concomitantwith improvements in preterm survival, the preterm birth rateincreased by 33% from 1981 to 2006, almost exclusively due to

: þ1 770 488 6283.oza).

r Ltd.

a rise in late preterm births (34e36 weeks’ gestation).1,8 Currently,w72% of all preterm births are due to infants born late preterm and84% are due to late and moderate (32e33 weeks) pretermcombined (Fig. 1). Although the majority of preterm-related deathsoccur among very preterm infants (<32 weeks’ gestation),increased attention has recently been given to better under-standing the reasons for the high rate of late andmoderate pretermbirth, its causes, short and long term sequelae, and opportunitiesfor prevention. Research is uncovering significant, though oftensubtle, increased risks for late preterm infants comparedwith thoseborn at term (i.e. 39e41weeks’ gestation) for complications at birthand long term neurodevelopmental problems. Increased knowl-edge about the epidemiology of these moderate and late pretermbirths is critical for informing practices and guidelines related tothe prevention of preterm-related morbidity and mortality.

2. Preterm birth definitions and subcategories

Fetal growth and maturation occur along a continuumthroughout pregnancy. As such, case definitions based on discretecategories of gestational age may appear somewhat arbitrary.However, standard categorization of preterm infants based ongestational age is valuable for assessments of morbidity andmortality risk, comparisons across populations and researchstudies, generating health policy guidelines, and guiding patientcare.9 Generally, preterm birth is defined as birth of an infant at<37weeks’ completed gestation. However, there has traditionally beena lack of consensus on standard gestational age categories forinfants born in the period near to term, i.e. from 32 to <37 weeks’

Fig. 1. Distribution of preterm births by gestational age: United States, 2008. Source:Centers for Disease Control and Prevention. National Center for Health Statistics.VitalStats. http://www.cdc.gov/nchs/vitalstats.htm. [August 26, 2011].

C.K. Shapiro-Mendoza, E.M. Lackritz / Seminars in Fetal & Neonatal Medicine 17 (2012) 120e125 121

gestation. These infants have been grouped by different gestationalcategories and identified by different names, such as ‘near term’,‘late preterm’, ‘marginally preterm’, ‘moderate preterm’, and‘borderline preterm’. Unfortunately, these terms have notadequately captured the important implications that these infantsare still premature and vulnerable.9,10 Lack of recognition ofprematurity, and its important physical and neurocognitivesequelae, could lead to increased comfort with early electivedeliveries, less rigorous newborn assessment, early discharge, orinadequate monitoring. The 2005 National Institutes of Healthworkshop recommended that infants born at 34 0/7 through 36 6/7weeks’ gestation after the onset of the mother’s last menstrualperiod (LMP) be referred to as ‘late preterm’.10 We will use thisdefinition to define late preterm births in this review and defineinfants born at 32 0/7 to 33 6/7 weeks’ gestation as ‘moderatepreterm’.

3. Estimating gestational age

Ensuring accurate and standardized estimation and reporting ofgestational age may often be challenging. Gestational age isroutinely estimated according to the number of weeks’ gestationafter the onset of the mother’s LMP. However, this estimate may beunreliable and prone to error in maternal recall. The gold standardfor accurate determination of gestational age is first trimesterultrasound.11 Nevertheless, early ultrasound is currently not rec-ommended for routine gestational age dating and, in the USA, isoften not reimbursed by public and private payers unless there isuncertainty about dates. In addition, women who seek prenatalcare late may be less likely to have accurate dating. Inaccuratedetermination of gestational age can affect multiple downstreamfactors, including estimation of fetal maturity before electivedelivery, clinical and epidemiologic investigations, and programevaluation.

US birth certificate records are the only source of national datafor surveillance and epidemiologic studies about preterm birth.Typically, reported gestational age is based on the interval betweenthe first day of the mother’s LMP and the date of birth.1 Wheninformation on the date of LMP is missing or when the birth weightis incongruent with the gestational age, the clinical or obstetricestimate of gestation is used in place of the LMP gestational age

(w5% of all births). This reliance on LMP results in a highpercentage of deliveries in which gestational age is misclassified.Basso and Wilcox12 reported that preterm births are more likely tobe misclassified than term births. Qin et al.13 found that US pretermbirth rates were lower when gestational age was based on theclinical estimate or on an LMP/clinical estimate compositemeasurement thanwhen it was based on the LMP alone. When Qinet al.13 applied different data editing methods in an attempt tocorrect for misclassification, they found that the increase inpreterm delivery rates from 1990 through 2002 persisted, but wasattenuated.

4. Causes of late and moderate preterm birth

Preterm birth is not a single entity, but a common final outcomeof a heterogeneous collection of underlying maternal and fetalfactors. Approximately two-thirds of all singleton preterm birthsare spontaneous, often with no known cause, and approximatelyone-third are the result of medical intervention (i.e. medicallyindicated) to protect the health of the mother or infant.14 Compli-cations of pregnancy that lead to both spontaneous and indicatedpreterm birth are multiple, complex, and vary according to gesta-tional age. Although the rates of indicated deliveries remain similarthroughout pregnancy, the medical reasons for intervention varyaccording to the changing biological processes and complicationsthat occur at different periods of gestation. For example, placentalinsufficiency becomes increasingly important later in pregnancydue to the increased demands for oxygen and nutrients duringa crucial period of fetal growth. The risks of some maternalcomplications also increase later in pregnancy, such as hyperten-sion, pre-eclampsia, diabetes, and placenta praevia. Decisionsregarding obstetric interventionmust weigh the risks of continuingthe pregnancy in a suboptimal uterine environment comparedwiththe risks of early delivery. Thus, understanding the risks ofmoderate and late preterm birth is critical to informing optimalclinical decision-making.15,16

5. Epidemiology of preterm birth

Surveillance of preterm birth is essential for informing andevaluating clinical practices, research, programs, and policies aimedat reducing infant morbidity and mortality. Surveillance andepidemiologic analyses can measure the contribution of pretermbirth to infant morbidity and mortality, identify populations athighest risk, detect changes in obstetric practices, and guide thedevelopment, implementation, and evaluation of programs.17

Increases in the singleton preterm birth rate since 1990 havebeen almost entirely due to infants born late preterm (Fig. 2).During 1990e2006, the late preterm birth rate for singleton birthsincreased 20.9%, from 6.7% to 8.1%. This rate declined slightly in2007 and 2008 to 8.0% and 7.8%, respectively. Preliminary data from2009, the latest year available, suggest that this decline iscontinuing.18

Although the reasons for the increasing rates of moderatepreterm and late preterm birth during the last two decades are notwell understood, several theories have been postulated. Theseinclude improved risk assessment and timing formaternal and fetaldisorders, more elective inductions and caesarean sections toreduce adverse fetal outcomes, increasing maternal age (>35years), and increasing rates of multiple gestations.10,19,20 Growingconcerns have been raised about more invasive medical manage-ment without clear indication, driving late preterm and early termrates due to providers’ concerns about medical malpractice, patientrequests, or even convenience of the family or obstetrician. Theactual prevalence of these practices is unknown. One recent study

Fig. 2. Trends in singleton preterm birth rates: United States, final 1990, 1995, 2000, and 2005e2008. Source: Centers for Disease Control and Prevention. National Center for HealthStatistics. VitalStats. http://www.cdc.gov/nchs/vitalstats.htm. [August 26, 2011].

C.K. Shapiro-Mendoza, E.M. Lackritz / Seminars in Fetal & Neonatal Medicine 17 (2012) 120e125122

reviewed medical records and estimated that more than half of alllate preterm infants delivered by obstetric intervention could be‘non-evidence based iatrogenic preterm birth’.21 However, cautiousinterpretation of this study is warranted, as information on theindication for the delivery was often missing in the medicalrecord.22

Concern about the practice of elective induction or caesareandelivery without medical indication prompted the March of Dimesto launch a national campaign, ‘Healthy Babies are Worth theWait’,23 to raise awareness among patients and providers on theimportance of preventing non-indicated intervention. Also, severallarge health insurance groups have moved to limit elective induc-tions and caesarean sections without medical indication before 39weeks.24,25 Preliminary results of these programs appear prom-ising, demonstrating rapid declines in elective inductions andcaesarean section without indication prior to 39 weeks’ gestation,as well as cost savings due to NICU and hospital stays.25 Moreinformation is needed to determine the extent of changes that haveoccurred in clinical practice, or just changes in reporting of indi-cations. The reduction in US preterm birth rates in 2007e2009, thereversal of a long term trend in increasing rates, may indicate thatthese programs are having a positive impact (Fig. 2).

5.1. Risk factors for moderate and late preterm birth

Multiple maternal and infant characteristics are associated withspontaneous preterm births, such as multiple gestations, birthdefects, maternal age, and race and ethnicity. Although many ofthese variables are not modifiable, knowledge of these risk factorsis important for targeting interventions to those at highest risk.

5.1.1. Multiple gestationsTwins and higher order multiples have elevated rates of late and

moderate preterm birth compared with singletons. Between 1990and 2008, the rate of singleton infants born late preterm increasedby 14.7%, whereas among multiples the rate increased by 27.4%.26

Much of the increase in multiples is thought to be attributable todelayed childbirth and increased use of assisted reproductivetechnologies (ART).1 The proportion of multiples due to ART is quitehigh (w48%),27 but the overall contribution of ART to the nationalpreterm birth rate is more limited; only about 1% of all US births in

2006 were attributable to ART.27 ART is also associated withincreased risk of preterm birth among singleton births, but it is notknown whether ART or the underlying biological reasons forinfertility increase the risk of preterm birth.28

5.1.2. Congenital malformationsCongenital malformations are associated with increased

preterm birth rates. A US multi-state study from 1995 to 2000showed that infants born at 32e36 weeks’ gestation hadmore thana two-fold risk of having congenital malformations than their termcounterparts.29 The risk was five times higher for those born atearlier gestations.

5.1.3. Maternal ageA U-shaped distribution is observed between maternal age and

late preterm birth (Fig. 3). In 2008, late preterm birth rates werehighest among women of <20 and >35 years of age; women of20e34 years of age had the lowest preterm rates. A similar patternis observed for those born moderately preterm (Fig. 3). Increasedpreterm risk among older women may be due to an increasedprevalence of co-morbid conditions such as diabetes and hyper-tension, as well as higher multiple birth rates and use of ART.27

Among teens, increased preterm risk may be due to biologicimmaturity, lower socioeconomic status, and behavioral risk factorssuch as tobacco use.30 The increase in the preterm birth rate since1990 has been greater for older mothers and less so for teens.8

5.1.4. Race and ethnicityPreterm birth rates vary for different racial and ethnic groups. In

2008, late preterm birth rates were highest for infants of non-Hispanic black mothers (11.3%), followed by American Indian orAlaskan Natives (9.7%), Hispanics (8.8%), non-Hispanic whites(8.2%) and Asian or Pacific Islanders (7.9%) (Fig. 4). A similar patternis observed for moderate preterm births. The reasons for theseracial and ethnic disparities remain poorly understood. Possibleexplanations include differences in access to care and quality ofcare, social determinants of health including the effects ofpsychosocial stress, poverty, and the environment; prevalence ofco-morbidities; and genetic factors.3 Although preterm birth riskdeclines with increasing level of maternal education and income,Black women with a college education still have higher preterm

Fig. 3. Percent of live births born moderate and late preterm by maternal age: United States, 2008. Source: Centers for Disease Control and Prevention. National Center for HealthStatistics. VitalStats. http://www.cdc.gov/nchs/vitalstats.htm. [August 26, 2011].

C.K. Shapiro-Mendoza, E.M. Lackritz / Seminars in Fetal & Neonatal Medicine 17 (2012) 120e125 123

birth rates than White women with a college education.31 Despitesocial and economic disadvantages, Hispanic women tend to havea lower risk of preterm birth compared with Black women, oftentermed the ‘Hispanic paradox’.32 Reasons for disparities in latepreterm birth and preterm-related mortality for non-Hispanicblacks, American Indian or Alaskan Natives, adolescent, and oldermothers have not been fully elucidated and represent a critical areafor research.

5.2. Morbidity and mortality rates

Risk of infant morbidity and mortality declines dramaticallywith increasing gestational age up until term and then increasesagain post-term.33 Understanding which infants are at greatest riskfor morbidity and mortality can assist providers in anticipatingpotential complications and with successful management, earlyintervention, and follow-up.19 During the last decade, studies haveshown consistently that late preterm infants are at higher risk thanterm infants for a number of neonatal complications includingrespiratory distress requiring ventilation, transient tachypnea of

Fig. 4. Percent of live births born moderate and late preterm by maternal race andHispanic origin: United States, 2008. Source: Centers for Disease Control and Preven-tion. National Center for Health Statistics. VitalStats. http://www.cdc.gov/nchs/vitalstats.htm. [August 26, 2011].

the newborn, intraventricular hemorrhage, bacterial sepsis,apnoea, hypoglycemia, temperature instability, jaundice andhyperbilirubinaemia, feeding problems, neonatal intensive careadmission, and even death.34e36 Although moderate and latepreterm birth is associated with increased neonatal complications,it is important to note that the direction of causality remainsunclear. Preterm delivery may cause adverse infant outcomes, orthe medical indications leading to the need for the pretermdelivery, such as intrauterine growth restriction or fetal anomalies,may be the cause of the adverse outcomes. A growing body ofevidence suggests that infants born late andmoderate preterm alsohave higher risks of later childhood morbidity and disability,including cerebral palsy, poor school performance, early interven-tion services, special education needs, and asthma.37e41

Infant mortality rates are highest among the most preterminfants (Fig. 5). During 2006e2008 infants born <28 weeks’gestation had by far the highest infant mortality rate (378.2 per1000 live births), whereas infants born at 39e41 weeks had thelowest mortality rates (2.1 per 1000 live births). It is important tonote that mortality risk continues to decline up to 39e41 weeks’gestation, including those born late preterm (7.1 per 1000 livebirths) and moderate preterm (16.2 per 1000 live). Infants born at32e33 and 34e36 weeks’ gestation have, respectively, w8 and 3times the rate of infant mortality compared with their termcounterparts.

Several studies have attempted to examine neonatal compli-cations and mortality by underlying factors resulting in pretermbirth. Using 2001 US vital statistics data, Chen et al.15 comparedneonatal mortality risk by preterm birth subcategory (i.e. spon-taneous preterm birth with intact fetal membranes, pretermpremature rupture of membranes before labor onset, and indi-cated preterm birth). Compared with spontaneous pretermdeliveries, late and moderate preterm infants delivered formedical indications had twice the risk of neonatal death. Inanother study restricted to late preterm births, Reddy et al.16 re-ported similar findings. Deliveries with recorded obstetriccomplications (e.g. polyhydramnios, oligohydramnios, incompe-tent cervix, cord prolapse, fetal distress) had at least twice the riskof neonatal and infant mortality compared with deliveries with norecorded indication.

Fig. 5. Infant mortality rates (per 1000 live births) by gestational age: United States, 2006. Source: United States Department of Health and Human Services (US DHHS), Centers ofDisease Control and Prevention (CDC), National Center for Health Statistics (NCHS), Office of Analysis and Epidemiology (OAE), Division of Vital Statistics (DVS), Linked Birth/InfantDeath Records 2003-2006 on CDC WONDER On-line Database. Accessed at http://wonder.cdc.gov/lbd-current.html on Sep 27, 2011 03:40:20 PM.

Practice points

� Among infants born preterm, 72% are born latepreterm, i.e. at 34 0/7 through 36 6/7 completed weeks’gestation.

� Although mortality risk declines with increasing gesta-tional age, even infants born close to term are atincreased risk for mortality.

� Compared with term births, infants born late pretermexperience higher rates of infant morbidity andmortality, as well as higher risks of childhooddisabilities.

C.K. Shapiro-Mendoza, E.M. Lackritz / Seminars in Fetal & Neonatal Medicine 17 (2012) 120e125124

Another large retrospective cohort study using vital statisticsand hospital discharge records in Massachusetts found thatmorbidity risk during birth hospitalization among late preterminfants decreasedwith increasing gestational age.42 Morbidity rateswere similar among infants born between 39 and 41 weeks’gestation (ranging from 2.5% to 2.8%). However, morbidity ratesapproximately doubled for each additional gestational week earlierthan 38 weeks, from 5.9% with morbidities if born at 37 weeks’gestation to 51.7% morbidity at 34 weeks’ gestation. Importantly,this study also found that the risk of neonatal morbidity increasedamong infants born late preterm if the pregnancy was complicatedby an underlying medical condition, especially for antepartumhaemorrhage and hypertensive disorders of pregnancy.

Because fetal lung maturation is completed late in gestation,respiratory morbidities are a frequent concern for late andmoderate preterm infants.43,44 In a comprehensive review, Colinet al. found that infants born 32e36 weeks’ gestation had a greaterrisk of respiratory morbidities compared with term infants.43 Ratesof infant deaths due to respiratory distress syndrome have beenfound to increase with each week of decreasing gestational agebefore 37 weeks (0.06 infant deaths per 1000 live births at 36weeks, 0.11 at 35 weeks, 0.26 at 34 weeks, 0.44 at 33 weeks, and3.09 at 28e32 weeks).44

5.3. Late preterm birth and fetal death

Many argue that advances in obstetric practice have led to moreintensive monitoring during pregnancy, which has led to increasedobstetric interventions. It follows that this increase in medicalinterventions resulting in early delivery would lead to a decrease instillbirths. Some evidence exists to support this argument.45e47

Joseph et al.46 examined US vital records from 1988e1999 toevaluate the temporal associations between trends in late pretermbirths and fetal deaths. The increase in obstetric intervention at34e36 weeks’ gestation appeared to be temporally associated witha significant decline in late fetal death rates. Others have examinedsimilar data and have shown that increases in preterm birth

precipitated by obstetric intervention correlated with a reductionin perinatal mortality.47 Though more direct assessments areneeded, these associations suggest that the rise in moderate andlate preterm births, perhaps due to increased obstetric interven-tion, may be conferring some improvement in perinataloutcomes.14

In summary, late preterm birth accounts for the vast majorityof preterm births in the USA and for the rise in the nationalpreterm birth rate over the past two decades. Late and moderatepreterm infants are both physiologically and developmentallyimmature and have higher risks for morbidity and mortalitycompared with infants born at term. Attention to increasing ratesof late preterm births during the last decade have led to severalcampaigns aimed at reducing non-medically indicated pretermbirths.23,25 It will be important to document the health andeconomic impact of these initiatives in the years to come.Prospective studies will further strengthen our knowledge aboutdecisions regarding obstetric intervention. We must go beyondcrude measures such as race, ethnicity, and age, and explore thecomplex mediators of disparities to further advance identificationof effective prevention strategies.

Research directions

� Detailed prospectively collected data would increaseknowledge about decisions regarding obstetricintervention.

� Racial and ethnic disparities in late preterm birthmortality are complex and remain poorly understood.

C.K. Shapiro-Mendoza, E.M. Lackritz / Seminars in Fetal & Neonatal Medicine 17 (2012) 120e125 125

Conflict of interest statementNone declared. The findings and conclusions in this paper are

those of the authors and do not necessarily represent the officialposition of the Centers for Disease Control and Prevention.

Funding sourcesNone.

References

1. Martin JA, Hamilton BE, Sutton PD, Ventura SJ, Mathews TJ, Osterman MJ.Births: final data for 2008. Natl Vital Stat Rep 2010;59:1e72.

2. Behrman RE, Butler AS, Institute of medicine (U.S.). Committee on under-standing premature birth and assuring healthy outcomes. Preterm birth: causes,consequences, and prevention. Washington, DC: National Academies Press;2007.

3. MacDorman MF. Race and ethnic disparities in fetal mortality, preterm birth,and infant mortality in the United States: an overview. Semin Perinatol2011;35:200e8.

4. Prematurity Research Expansion and Education for Mothers who deliver InfantsEarly Act (PREEMIE Act); 2006. [Available from: http://frwebgate.access.gpo.gov/cgi-bin/getdoc.cgi?dbname¼109_cong_public_laws&docid¼f:publ450.109].

5. Surgeon general’s conference on the prevention of pretermbirth; 2008. [Availableat http://www.surgeongeneral.gov/library/preterm_birth/background.html].

6. ‘A Healthy Baby Begins with You’ Campaign. [Available at http://minorityhealth.hhs.gov/templates/browse.aspx?lvl¼2&lvlID¼117].

7. Healthy People. [Available at: http://www.healthypeople.gov/2010/default.htm].

8. Martin JA, Osterman MJ, Sutton PD. Are preterm births on the decline in theUnited States? Recent data from the National vital statistics system. NCHS DataBrief 2010:1e8.

9. Engle WA. A recommendation for the definition of “late preterm” (near-term)and the birth weightegestational age classification system. Semin Perinatol2006;30:2e7.

10. Raju TN, Higgins RD, Stark AR, Leveno KJ. Optimizing care and outcome forlate-preterm (near-term) infants: a summary of the workshop sponsored bythe National Institute of Child Health and Human Development. Pediatrics2006;118:1207e14.

11. Kramer MS, McLean FH, Boyd ME, Usher RH. The validity of gestational ageestimation by menstrual dating in term, preterm, and postterm gestations.JAMA 1988;260:3306e8.

12. Basso O, Wilcox A. Mortality risk among preterm babies: immaturity versusunderlying pathology. Epidemiology 2010;21:521e7.

13. Qin C, Dietz PM, England LJ, Martin JA, Callaghan WM. Effects of different data-editing methods on trends in race-specific preterm delivery rates, UnitedStates, 1990e2002. Paediatr Perinat Epidemiol 2007;21(Suppl. 2):41e9.

14. Ananth CV, Joseph KS, Oyelese Y, Demissie K, Vintzileos AM. Trends in pretermbirth and perinatal mortality among singletons: United States, 1989 through2000. Obstet Gynecol 2005;105:1084e91.

15. Chen A, Feresu SA, Barsoom MJ. Heterogeneity of preterm birth subtypes inrelation to neonatal death. Obstet Gynecol 2009;114:516e22.

16. Reddy UM, Ko CW, Raju TN, Willinger M. Delivery indications at late-pretermgestations and infant mortality rates in the United States. Pediatrics2009;124:234e40.

17. German RR, Lee LM, Horan JM, Milstein RL, Pertowski CA, Waller MN. Updatedguidelines for evaluating public health surveillance systems: recommendationsfrom the Guidelines Working Group. MMWR Recomm Rep 2001;50:1e35. quizCE1e7.

18. Hamilton BE, Martin JA, Ventura SJ. Births: preliminary data for 2009. Natl VitalStat Rep 2010;59:1e19.

19. Engle WA, Tomashek KM, Wallman C. “Late-preterm” infants: a population atrisk. Pediatrics 2007;120:1390e401.

20. Davidoff MJ, Dias T, Damus K, et al. Changes in the gestational age distributionamong U.S. singleton births: impact on rates of late preterm birth, 1992 to2002. Semin Perinatol 2006;30:8e15.

21. Bannerman CG, Fuchs KM, Young OM, Hoffman MK. Non-spontaneous latepreterm birth: etiology and outcomes. Am J Obstet Gynecol 2011;205:456.e1e6.

22. Iams JD. Late preterm birth: more and better data needed. Am J Obstet Gynecol2011;205:395.

23. Healthy Babies are Worth the Wait; 2011. [Available at: http://www.marchofdimes.com/professionals/medicalresources_hbww.html].

24. Oshiro BT, Henry E, Wilson J, Branch DW, Varner MW. Decreasing electivedeliveries before 39 weeks of gestation in an integrated health care system.Obstet Gynecol 2009;113:804e11.

25. Clark SL, Frye DR, Meyers JA, et al. Reduction in elective delivery at <39 weeksof gestation: comparative effectiveness of 3 approaches to change and theimpact on neonatal intensive care admission and stillbirth. Am J Obstet Gynecol2010;203. 449 e1e449 e6.

26. VitalStats. [Available at: http://www.cdc.gov/nchs/vitalstats.htm].27. Sunderam S, Chang J, Flowers L, et al. Assisted reproductive technology

surveillance e United States, 2006. MMWR Surveill Summ 2009;58:1e25.28. Basso O, Baird DD. Infertility and preterm delivery, birthweight, and Caesarean

section: a study within the Danish National Birth Cohort. Hum Reprod2003;18:2478e84.

29. Honein M, Kirby R, Meyer R, et al. The association between major birth defectsand preterm birth. Matern Child Health J 2009;13:164e75.

30. Strobino DM, Ensminger ME, Kim YJ, Nanda J. Mechanisms for maternal agedifferences in birth weight. Am J Epidemiol 1995;142:504e14.

31. McGrady GA, Sung JF, Rowley DL, Hogue CJ. Preterm delivery and low birthweight among first-born infants of black and white college graduates. Am JEpidemiol 1992;136:266e76.

32. Brown HL, Chireau MV, Jallah Y, Howard D. The “Hispanic paradox”: an inves-tigation of racial disparity in pregnancy outcomes at a tertiary care medicalcenter. Am J Obstet Gynecol 2007;197. 197 e1e197 e7; discussion e7e9.

33. Kirby RS, Wingate MS. Late preterm birth and neonatal outcome: is 37 weeks’gestation a threshold level or a road marker on the highway of perinatal risk?Birth 2010;37:169e71.

34. Wang ML, Dorer DJ, Fleming MP, Catlin EA. Clinical outcomes of near-terminfants. Pediatrics 2004;114:372e6.

35. Escobar GJ, Greene JD, Hulac P, et al. Rehospitalisation after birth hospital-isation: patterns among infants of all gestations. Arch Dis Child2005;90:125e31.

36. Hibbard JU, Wilkins I, Sun L, et al. Respiratory morbidity in late preterm births.JAMA 2010;304:419e25.

37. Abe K, Shapiro-Mendoza CK, Hall LR, Satten GA. Late preterm birth and risk ofdeveloping asthma. J Pediatr 2010;157:74e8.

38. Chyi LJ, Lee HC, Hintz SR, Gould JB, Sutcliffe TL. School outcomes of late preterminfants: special needs and challenges for infants born at 32 to 36 weeksgestation. J Pediatr 2008;153:25e31.

39. Gurka MJ, LoCasale-Crouch J, Blackman JA. Long-term cognition, achievement,socioemotional, and behavioral development of healthy late-preterm infants.Arch Pediatr Adolesc Med 2010;164:525e32.

40. Petrini JR, Dias T, McCormick MC, Massolo ML, Green NS, Escobar GJ. Increasedrisk of adverse neurological development for late preterm infants. J Pediatr2009;154:169e76.

41. Romeo DM, Di Stefano A, Conversano M, et al. Neurodevelopmental outcome at12 and 18 months in late preterm infants. Eur J Paediatr Neurol 2010;14:503e7.

42. Shapiro-Mendoza CK, Tomashek KM, Kotelchuck M, et al. Effect of late-pretermbirth and maternal medical conditions on newborn morbidity risk. Pediatrics2008;121:e223e32.

43. Colin AA, McEvoy C, Castile RG. Respiratory morbidity and lung function inpreterm infants of 32 to 36 weeks’ gestational age. Pediatrics2010;126:115e28.

44. Joseph KS, Nette F, Scott H, Vincer MJ. Prenatal corticosteroid prophylaxis forwomen delivering at late preterm gestation. Pediatrics 2009;124:e835e43.

45. Ananth CV, Liu S, Joseph KS, Kramer MS. A comparison of foetal and infantmortality in the United States and Canada. Int J Epidemiol 2009;38:480e9.

46. Joseph KS, Demissie K, Kramer MS. Obstetric intervention, stillbirth, andpreterm birth. Semin Perinatol 2002;26:250e9.

47. Lisonkova S, Hutcheon JA, Joseph KS. Temporal trends in neonatal outcomesfollowing iatrogenic preterm delivery. BMC Pregn Childbirth 2011;11:39.

at SciVerse ScienceDirect

Seminars in Fetal & Neonatal Medicine 17 (2012) 126e131

Contents lists available

Seminars in Fetal & Neonatal Medicine

journal homepage: www.elsevier .com/locate/s iny

Developmental physiology of late and moderate prematurity

Tonse N.K. Raju a,b,*

aCenter for Developmental Biology and Perinatal Medicine, National Institutes of Health, Bethesda, Maryland, USAb Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, Maryland, USA

Keywords:ApneaHypoglycemiaHypothermiaInfectionPretermRespiratory distress

* 6100 Executive Blvd, Room 4B03, Bethesda, MD 2E-mail address: rajut@mail.nih.gov.

1744-165X/$ e see front matter Published by Elseviedoi:10.1016/j.siny.2012.01.010

s u m m a r y

This is a brief review of the developmental physiology of selected organ and functional systems inmoderate and late preterm infants. This outline provides a discussion of the physiological underpinningsfor some of the clinical conditions seen in this group of infants, including hypothermia, hypoglycemia,respiratory distress syndrome, transient tachypnea, severe respiratory failure, apnea, feeding difficulties,jaundice, and increased susceptibility to infections.

Published by Elsevier Ltd.

1. Introduction

In a 2005 workshop organized by the National Institute ofHealth and Human Development (NICHD), this author madea proposal to replace the vague and imprecise phrase ‘near term’

with ‘late preterm’, to reflect the physiological and developmentalimmaturity of this group of infants. The expert panel agreed, andsuggested a definition for late preterm as births between 340/7 and366/7 weeks of gestation. As discussed in the workshop summary,1

therewere compelling reasons for the choice of the new phrase andits lower boundaries. Themost important reasonwas to underscorethat infants born even by a fewweeks prior to termwere immature,with higher morbidity and mortality risks compared to those bornat term gestations.2

The NICHD expert panel could not have predicted the degree ofimpact of introducing the new phrase ‘late preterm’ into the lexiconof perinatal medicine. Several professional societies and organiza-tions in the USA3,4 and in other countries adopted the new term.The Centers for Disease Control and Prevention began tracking latepreterm births rates.5

The phrase also shifted the paradigm, and stimulated a hugewave of research interest. More than 300 papers have been pub-lished since 2006 on late preterm, including observational andintervention studies, systematic reviews, editorials, and committeeopinions. The obstetric and pediatric communities began animportant dialogue about this vulnerable group of infants. Thesedevelopments seem to have made an impact. As discussed else-where in this issue, the US preterm and late preterm birth rateshave begun to decline significantly.5

0952, USA.

r Ltd.

This review will focus on the developmental physiology of theorgan systems in moderate and late preterm infants that are theunderpinning of their morbidity and mortality risk.

2. Generic aspects of developmental maturation

Maturation is a continuous process with no specific endpoints or‘goals’ to be achieved. By contrast, maturational milestones aredeveloped by us as useful signposts to assess maturation, and tomake clinical decisions. Themilestones are also useful for statisticaland definitional purposes, and for developing managementguidelines.

The duration of gestation (or ‘time’) is only one of the manyfactors that influence fetal and neonatal developmental maturationand trajectory. Others include intrauterine environment, maternalhealth and disease status, maternal medication use, diet, nutrition,and lifestyle (e.g. exercise, stress, smoking, drug abuse etc.), thenumber of fetuses, fetal sex, and fetal health/diseases.

Maturation is non-linear, in that it is programmed to meet theneeds of the organism for an independent extrauterine existence atvarious time points. Thus, different organs tend to mature atdifferent trajectories relative to each other. For instance, soon afterbirth, most late preterm infants can breathe on their own withoutassisted respiratory support, reflecting their mature breathingapparatus. However, most of them will have difficulty in initiatingand maintaining breastfeeding, reflecting their immature brain-stem, sucking and swallowing mechanisms.

3. Developmental basis for clinical problems

The following presentation is organized in an approximatechronological order of clinical disorders in moderate and latepreterm infants.

Box 1. Physiological basis for hypothermia.

� Heat loss soon after birth: w135e155 W/m2

B Channels of heat loss� Evaporative: along with evaporative water loss

(w125 g/m2/h ) at room temperature of w26 �Cand 40% humidity. Heat loss through this route:w60e80 W/m2

� Radiation: direct loss of heat through electro-magnetic radiation in the infrared spectrum(into the cooler walls of the nursery or theincubator). This can be up to 50 W/m2

� Convection: through the air; a negligible routeof heat loss up tow25W/m2 especially when theincubator walls are cold

� Conductive heat loss: direct contact througha cooler object, such as the mattress e this isa negligible route of heat loss.

� Hypothermia and physiological immaturityB Deficient subcutaneous fat and non-keratinized, thinskin: increases the speed and severity of hypo-thermia with exposure to cool, extrauterineenvironment.

� High ratio of surface area:body massB Common to all newborn infants, but especially worsefor preterm infants, thus a large surface area isexposed to environmental cooling and hypothermia.

� Inability to shiverB Neither term nor preterm infants can mount a shiv-ering response (generating heat through themuscles) to environmental cooling.

� Deficiency of brown adipose tissue (BAT)B Oxidation of BAT is the main source of heatproduction in term infants

B BAT development begins around the 20th week ofgestation, reaching a peak at term, accounting for upto 1% of body weight

B Lack of sufficient BATmass leads to decreased abilityfor generating heat to offset hypothermia fromenvironmental cooling.

� Immature response of the temperature sensors in theposterior hypothalamusB Insufficient release of thermogenic hormones (T4and norepinephrine)

B Brainstem serotonin deficiency can lead to homeo-static deficits owing to a reduced sympatheticresponse to mild cold stress.

T.N.K. Raju / Seminars in Fetal & Neonatal Medicine 17 (2012) 126e131 127

3.1. Thermal instability/cold stress

Fetal thermal homeostasis is controlled and regulated by thematernal thermal homeostasis.6 Because the temperature of theamniotic fluid reflects mother’s core temperature, the fetus isconstantly nurtured in a warm intrauterine environment. In addi-tion, with an average fetal oxygen consumption of 8 ml/min/kg, itgenerates 3e4W/kg of heat. Since the fetus has to eliminate its heatthrough the mother, its temperature remains about 0.5e1.0 �Chigher than that of the mother. Through mechanisms not wellunderstood, fetal thermoregulatory pathways remain dormantuntil birth, and fetal temperature is thus passively ‘clamped’ to thatof the mother, changing along with her temperature. Even at term,the fetus has little ability to generate heat if challenged bymaternalhypothermia, or lose heat if challenged by maternal hyperthermia,as in pyrexia.

Under normal circumstances, the placenta eliminates 85% offetal heat intomaternal circulation. Thus, the fetus and the placentaare often considered ‘an engine and a radiator’, respectively. Theremaining 15% of fetal heat loss occurs through the conductancefrom the fetal skin to the amniotic fluid, and to the uterine wall.6

As the fetus emerges from a warm intrauterine environment tothe cold external world, hypothermia is inevitable.7,8 In response,the newborn has to mount a variety of physiological responsesimmediately to conserve heat loss, increase heat production andmaintain core temperate with the least amount of oxygenconsumption in the process.6,9,10 Most healthy term infants mountsuch responses, but moderate and late preterm infants may not, orcannot, mount such responses due to the deficient, immature ornon-existent thermoregulatory processes (Box 1).6e10

3.2. Respiratory morbidities (Box 2)

Studies using large datasets and systematic reviews11e13

confirm that compared with term infants, all forms of pulmonarydisorders occur at higher frequencies in moderate and late preterminfants. Developmental immaturity-related pulmonary conditionsare respiratory distress syndrome (RDS), transient tachypnea of thenewborn (TTN), pneumonia, hypoxic respiratory failure andpulmonary hypertension and apnea of prematurity.

Maturation of the pulmonary apparatus occurs throughprogressive stages. By about the 16th week of gestation, the 16generations of bronchioloar branching are completed. By 24 weeks’gestation, the primitive alveoli can engage in some gas exchange;but major surge in alveolar generation (alveolization) occurs byterm and continues through early childhood. Between 25 and 36weeks’ gestation, the alveolization process goes through a transient‘saccular phase’. The saccules participate in gas exchange, but theyare compact, thick and primitive. Due to such structural immatu-rity, they are susceptible for barotraumas. By a process of septation,the saccules mature into alveoli beginning from 32 weeks’ gesta-tion, with dramatic, weekly increments. By term, the alveolar countreaches about 80% of that of the adult alveolar count.14

A study published by the Consortium on Safe Labor11 reportedan incidence of RDS in late preterm infants to be 5.2%, compared to0.4% in infants born at 37e40 weeks’ gestation (computed fromtheir Table 3, Reference11). RDS thus remains the most commoncause of significant respiratory morbidity in late preterm infants.

RDS is the consequence of qualitative and/or quantitative defi-ciency of pulmonary surfactants superimposed upon an immaturecardiopulmonary system. Pulmonary surfactants are predomi-nantly phospholipids. In the mature fetus, about 80% of phospho-lipids is phosphatidylcholine, 5e10% is phosphatidylglycerol, withother minor lipids making up the remainder. However, pretermfetuses completely lack phosphatidylglycerol, which is an

important cause of deficient surfactant function leading to RDS inpreterm infants.

Around 32 weeks’ gestation, the surfactant pool size begins toincrease, with a significant surge around 35 weeks’ gestation,reaching a peak at term. Uterine contractions preceding labor anddelivery, and the associated surge of catecholamines and endoge-nous steroids, enhance pulmonary surfactant secretion from Type IIcells into the alveoli. Initiation of ventilation also contributes toadditional surfactant release. In moderate and late preterm infants,most of the above processes are either deficient or non-existent,thus setting up a stage for clinical RDS.12e14

TTN is diagnosed in about 4% of late preterm infants making itthe secondmost common pulmonary disorder.15e17 A lack of timelyclearance of the pulmonary fluid from the alveolar airspaces is thepathophysiological basis for TTN. The risk of TTN is high in infantsborn precipitously, without active labor, and in those delivered byelective cesarean section. These observations gave rise to the

Box 2. Physiological basis for respiratory disorders.

� Respiratory distress syndromeB Qualitative and/or quantitative deficiency in surfac-tant and its function

�Deficiency of phospatidylglycerol in pretermlungs, or its delayed maturation in diabeticpregnancy

B Elective caesarean delivery, depriving labor-associated catecholamine and steroid surge,causing decreased pulmonary surfactant release.

� Transient tachypnea of the newborn (TTN)B Elective caesarean delivery, depriving labour-associated catecholamine and steroid surge, pre-venting lung fluid clearance

B Delayed transition to postnatal life, due to dimin-ished transepithelial absorption of lung fluid fromdisrupted functions of the amiloride-sensitiveepithelial sodium channels.

� Hypoxic respiratory failureB Mild respiratory distress due to TTN is initiallytreatedwith oxygen using oxy-hoods, which can leadto nitrogenwashout, oxygen absorption and alveolaratelectasis

B Persistent mild-to-moderate hypoxia leading todevelopment of pulmonary hypertension, oftenrequiring treatment with inhaled nitric oxide and/orextracorporeal membrane oxygenation.

� PneumoniaB Fetal infection as a cause for preterm delivery, pre-disposing to sepsis and pulmonary infection

B Mechanical ventilation leading to injury to theimmature lungs, increasing the risk for pulmonaryinfections.

� Central, obstructive and mixed apneaB Increased chest wall and upper airway complianceleading to their collapse when the diaphragmcontracts, especially during rapid eye movementsleep, which is >;60% of total sleep time

B Biphasic ventilatory response to hypoxiaB Blunted ventilatory response to hypercapneaB Higher incidence of acute life-threatening events andsudden infant death syndrome in late preterm.

T.N.K. Raju / Seminars in Fetal & Neonatal Medicine 17 (2012) 126e131128

hypothesis that the ‘squeeze’ on the infant chest during vaginalbirth helps clear fetal lung fluid. However, a large body of evidencecontradicts this. The transition from fluid-filled fetal lung to air-filled postnatal breathing occurs via a multitude of coordinatedhormonal and biochemical events during fetal life, labor anddelivery, and into the hours and days after birth.15e17

Around near-term gestation, the rate of fetal lung fluidproduction begins slowdown. With the onset of labor, and uterinecontraction-induced hormonal surge, the lung fluid gets cleared viathe pulmonary interstitial lymphatics, blood vessels, upper airway,mediastinum, and the plural spaces. In addition, recent studiesshow that amiloride-sensitive epithelial sodium channels (ENaCs)in the alveolar cells also play a crucial role in lung fluid sodiumtransport, facilitating water clearance.16,17 Both Type I and Type IIcells in the alveoli are involved in this process, but since Type I cellsconstitute >95% of alveolar surface, their contribution to lung fluidclearance appears to be greater than previously considered.16

3.3. Hypoxic respiratory failure

Although TTN is considered a mild condition, sometimes thereare serious complications leading to ‘malignant TTN’ and severe

hypoxic respiratory failure. Such infants initially would havereceived high concentrations of inspired oxygen via an oxygen hoodor a nasal cannula, which can lead to pulmonary nitrogen washoutand alveolar atelectasis. This can be followed by severe hypoxia,metabolic and respiratory acidosis, and pulmonary hypertension.Such infants often require inhaled nitric oxide therapy or extracor-poreal membrane oxygenation.12 Keszler et al.17 reported that latepreterm infants deliveredbyelective cesarean andmanaged as ‘mildTTN’ in small hospitals were at great risk for such complications.

3.4. Other respiratory problems

Moderate and late preterm infants are prone for apnea ofprematurity. Similar to other aspects of developmentalphenomena, maturation of brainstem regions and control ofbreathing apparatus in the moderate and late preterm infants lie inbetween those of extremely preterm and term infants. Brainstemincreases in length and volume throughout gestation, with rapidincrease in first half of pregnancy, but progressive incrementsoccurring later.18e20

Obstructive and mixed apnea occur less frequently in moderateand late preterm infants compared to those born <28 weeks’ ofgestation. However, they occur at greater frequency and severitycompared to those born at term. Many underlying developmentalfeatures are at play in causing apnea of prematurity.18e20 The chestwall and the upper airways are highly compliant, which tend tocollapse paradoxically when the diaphragm contracts and gener-ates negative intrathoracic pressure. The contraction of the musclesof the lower jaw and the tongue tend to dilate the upper airwaysand prevent their collapse. However, it is the balance betweenthese forces that determines whether or not the upper airwaysremain open during breathing (Box 2).

3.5. Hypoglycemia

Low plasma glucose, usually defined as 2.0e2.5 mmol/l(<40e45 mg/dl), has been reported in 5e15% of ‘normal’newborn infants. The frequency and severity of this complication isprobably much higher in moderate and late preterm infants.

The energy demands of the fetus are met by themother throughthe placental transfer of glucose, amino acids, free fatty acids,ketone, and glycerol. Although normal fetus does not produceendogenous glucose, throughout gestation its blood glucoseconcentration remains >3 mmol/l (54 mg/dl).

Soon after birth, the blood glucose concentration falls precipi-tously due to the interruption of the placental glucose supply. Themagnitude and duration of such a fall depends upon the fetalconcentrations of glucose and plasma insulin, the infant’s ability tomobilize glucose from hepatic glycogen, to initiate gluconeogen-esis, and to feed adequately.21,22

In low birth weight and premature infants, gluconeogenesis isactive, accounting for 30e70% of endogenous glucose production.However, the risk of hypoglycemia remains high due to manyfactors (Box 3), including caregiver complacency of not monitoringthem, since many such infants ‘appear normal’.

3.6. Hepatic immaturity and jaundice

Jaundice is themost common reason for readmission after initialhospital discharge in late preterm infants. Although accurateepidemiological data are lacking, compared to term infants, the riskof bilirubin-induced neurological injury is more common inmoderate and late preterm infants.23

In all infants, hyperbilirubinemia is a consequence of increasedbilirubin production, diminished metabolism and elimination, or

Box 4. Developmental immaturity of bilirubin metabolism.

� Bilirubin loadB Higher rates of bilirubin production due to higherproportion of senescent RBCs

B Lack of feeding leading to increased enterohepaticload of bilirubin.

� Hepatic bilirubin metabolismB Limited bilirubin uptake in part due to deficiency oforganic anion transport proteins

B Immaturity of hepatic endoplasmic reticular enzyme,uridine diphosphoglucuronate glucuronosyltransfer-ase (UGT) that is required for conjugating the fat-soluble indirect bilirubin to water-soluble directbilirubin.

� Enterophepatic absorption of bilirubinB Conversion of the relatively unstable monoglucur-onide and diglucuronide bilirubin conjugates fromthe duodenum and colon by the enteric mucosalenzyme, b-glucuronidase causes about 25% of theexcreted bilirubin to be reloaded

B Breastfeeding and delayed ingestion of any milkleads to persistent activity of b-glucuronidaseactivity.

Box 3. Developmental physiological basis for hypoglycemia.

� Decreased glycogen storesB In moderate and late preterm infants, hepaticglycogen stores are lower relative to term infants

B In infants with intrauterine growth restriction,glycogen stores are depleted in fetal life, perhaps dueto reduced placental supply of energy needs.

� Insulin concentrations and insulin response tochanging glucose levelsB Immaturity of the glucose-regulated insulin responsein the b-islet cells of the pancreas, resulting inunregulated, continued insulin secretion despite fallsin postnatal glucose concentrations

B High circulating insulin concentrations secondary tomaternal diabetes.

� Increased energy demandsB Cold stressB HypoxiaB Sepsis.

� Diminished intakeB Difficulty in sucking and swallowing, and initiationand maintenance of breastfeeding.

T.N.K. Raju / Seminars in Fetal & Neonatal Medicine 17 (2012) 126e131 129

a combination of both. The catabolic breakdown of haemoglobinfrom the senescent red blood is the predominant source of bilirubinduring the neonatal period. Heme from hemoglobin is convertedinto biliverdin with the catalytic activity of heme oxygenase-1enzyme (HMOX-1); and biliverdin reductase converts the latter tobilirubin, which is taken up by the liver for conjugation. Theconjugated bilirubin is then released into the intestines for excre-tion through the bowel.

During theprocess of breakdownof 1moleof heme into 1mole ofbilirubin,1mole of carbonmonoxide is released. Thus, bymeasuringcarbonmonoxide (or thecarboxyhemoglobin) in theexhaledbreath,one can estimate the rate of bilirubin production. Studies utilizingsuch methods have estimated that compared to term infants, thebilirubin production rates are slightly higher in preterm infants,perhaps due to higher proportion of senescent red blood cells.

The placenta is the primary route of fetal bilirubin excretion,because of which the hepatic conjugating system remains dormantin fetal life and for a few days following birth. (In fact, conjugatedbilirubin is impermeable through the placenta.) Therefore, visiblejaundice is seen in >70% of infants during the first week of life. Inpreterm infants, the hepatic conjugation system remains immatureto a greater degree and for longer durations, which explains thereasons for early, more severe, and prolonged hyperbilirubinemia.In addition, feeding difficulties lead to a delay in the resolution ofthe enterohepatic re-circulation of bilirubin, resulting in furtherincrease in hepatic bilirubin load. Other factors for prematurity-related hyperbilirubinemia are listed in Box 4.

3.7. Gastrointestinal immaturity and feeding

Oro-buccal coordination and swallowing mechanisms areimmature in most moderate and late preterm infants. These infantshave considerable difficulty in establishing successful breastfeed-ing.18,24 Because of lack of prior experience, the problem is com-pounded in primiparous mothers. Mild hypotonia, cold stress, andgeneral lack of strength add to the difficulty of moderate and latepreterm infants to establish andmaintain adequate breastfeeding. Allpreterm infants have a higher frequency of gastro-oesophageal

reflux, further reducing food intake and affecting weight gain. Thischain of events may lead to dehydration and hypernatremia duringthefirst fewweeksof life. TheAcademyofBreastfeedingMedicinehasdeveloped a useful protocol to help breastfeed late preterm infants.24

3.8. Immunological maturation

The topic of fetal and neonatal immune maturation is broad andcomplex, and only a few aspects are highlighted here.25e38

It is traditionally taught that newborn infants are deficient inacquired or adaptive immune responses; however, immunematurational processes are more complex. Due to the non-exposure to antigens during intrauterine life, neonates lack‘immunological memory’, and hence cannot mount an adequate T-cell response soon after exposure to a pathogenic microbe.However, in a study of lymphocyte subpopulations, Walker et al.25

found that the pattern of T-cell responses was similar in pretermand term infants. Yet, with increasing prematurity, the absolutecounts of naïve helper T-lymphocytes were lower. With postnatalantigenic stimulation, the T- and B-lymphocyte populationspredictably increased to significant levels.

The ability to produce immunoglobulin in response to specificantigenic stimulation is acquired early in fetal life. The predomi-nant circulating immunoglobulin in the healthy fetus is placentallyderived, actively secreted IgG, with a concentration of about 5e10%higher than that of the mother. Colostrum and human milk are themajor sources of immunological protective agents. Thus, breastfedinfants quickly acquire other immunoglobulins, and a host of non-specific protective factors (Box 5),29,30 which promote the devel-opment of specific immune functions. However, due to the inabilityto feed at the breast, or to suck and swallow ineffectively, moderateand late preterm infants may not receive adequate human milk e

a great source of immunologically valuable diet, increasing theirrisks for sepsis, including necrotizing enterocolitis.

Despite a reasonably well-developed adaptive immune system,the innate immune system is relatively immature in the newborn.The polymorphonuclear neutrophils (PMNs) from term infantscannot mount an adequate chemotactic response (directed migra-tion towards the chemo-attracting antigen). This is even worse

Box 6. Developmental immaturity of the innate immuno-logical system.

� Immunoglobulin and complementsB Due to earlier than term birth, preterm infants receivelesser amounts of IgG via the placenta

B The complement system is decreased in preterminfants.

� Antimicrobial proteins and peptidesB The maturation of leukocyte-secreted antimicrobialproteins and peptides increases in a gestational age-dependent manner. Some of the critical ones are:bactericidal/permeability-increasing protein; a-defensins; and several phospholipases.

� Mannose binding lectinB Required for activating complement pathways, lowin infants <28 weeks, increasing their risk for severesepsis.

� Microbial-killing abilitiesB Diminished chemotaxis, lower neutrophil pool, defi-cient opsonization, and other factors due to imma-turity limit the ability for phagocytosis.

� Other recently discovered factorsB Functionally impaired dendritic cellsB Functional immaturity of toll-like receptorsB Reduced cytokine-producing capacity of neonatalmonocytes.

Box 5. Immunological advantages from breastfeeding andfeeding of human milk.

� ImmunoglobulinB Secretory IgA, IgM and IgGB IgE, secretory component of IgD.

� Hormones and other biochemical componentsB Epidermal growth factorB ProstaglandinB Thyroid-releasing hormoneB Thyroid-stimulating hormoneB Thyroxine and triiodothyronineB Adrenocortical hormoneB Melatonin with diurnal variationsmatching circadianrhythm

B SomatostatinB NeurotensinB ProlactinB Erythropoietin.

� CellularB Immunologically specific T- and B-lymphocytesB Neutrophils, macrophages and epithelial cellsB Commensal bacteria colonized on maternal skinB Other microbiota matching maternal dietary intake.

� Other non-specific factorsB ComplementB Chemotactic factorsB LactoferrinB TransferrinB Several variety of biniding proteinsB Enzymes: lipoproteinlipase, lysozyme, and leukocyteenzymes.

Practice points

� Preterm birth, even by a few weeks prior to term, isassociated with a high prevalence of clinical problemsdue to:B immaturity of organ systems (e.g. the lungs, brain;the gastrointestinal system).

B acquired problems (e.g. infections, hypothermia).B inadequate monitoring and/or follow-up plans (e.g.hypoxic respiratory failure; hypoglycemia; kernicterus).

� Developmental processes are non-linear, therefore:B different organ systems mature at specific rates andtrajectories that are specific to their functions (e.ginfants breathing normally may not feed well, due toimmature sucking/swallowing mechanisms).

� The degree of maturation can vary among infants bornat comparable gestations because ‘time’ is only one ofthe many factors influencing maturation. Others areB maternal health and diseaseB medication useB lifestyle factors (nutritional status; smoking; illicitdrug use; exercise)

B fetal sex; number of fetuses; fetal disordersB the dynamics of the intrauterine environment.

T.N.K. Raju / Seminars in Fetal & Neonatal Medicine 17 (2012) 126e131130

among the PMNs from preterm infants. PMNs, however, maintainadequate phagocytic activity, but because of the deficiency ofcomplement receptors on their surface, phagocytosis is not effec-tive, especially when the invadingmicrobial load is massive relativeto the PMN counts. Other aspects of the immunological basis forincreased susceptibility for neonatal infections in moderate andlate preterm are listed in Box 6.

The human intestinal system is also a major immunologicalorgan. Soon after birth, the intestinal surface is exposed to a varietyof microbial and dietary antigens. Since the newborn infant isimmunologically naïve, during vaginal birth and soon thereafter,they become colonized by multitudes of microbes, includingEscherichia coli and streptococci. By the end of the first week of life,anerobes such as Bacteroides spp., Bifidobacterium spp. and Clos-tridium spp. also colonize the infant gut.

The diversityofmicrobial ecology inpreterm infants is affected bymany factors, including chorioamnionitis, maternal intrapartumantibiotic exposure, and infant feeding type.38 Infants delivered bycaesarean have less diverse microbial ecology compared to thosedelivered vaginally.39 They tend to have fewer E. coli and Bifidobac-teria spp., but more Klebsiella spp. and Enterobacter spp. The immu-nological implications of diversity of infant intestinal microbialecologyneeds tobe studied to explore their possible effectson sepsis,necrotizing enterocolitis, infant atopy, allergy, and food intolerance.

4. Summary and conclusions

The foregoing discussion is intended to be a brief overview ofsome critical aspects of fetal maturation that operate as underpin-nings for common problems seen in moderate and late preterminfants. One should note that variation in the degree of maturationamong infants of similar gestational ages is very common. Since

development is non-linear, the extent of maturity among variousorgans in a given infant might also vary. An understanding of theseprinciples can help guide treatment strategies.

Conflict of interest statementNone declared.

Funding sourcesNone

Research directions

� Understanding the mechanisms that affect the rate andtrajectory of fetal/neonatal organ maturation anddevelopment.

� Assessing the role of intrauterine environment on fetalorgan maturation during the second half of pregnancy.

� Developing biomarkers to:B identify the status of maturation of specific organsystems

B predict outcomes of adverse clinical conditions.� Study impact of preterm birth on the pattern of intes-tinal microbiota, and their variations to understand howthey affect the current and future health and disease.

� Study how the extrauterine environment affects organmaturation and development in infants born preterm,and uncover the underlying process involved in sucheffects.

� Use the above knowledge to optimize care and follow-up of moderate and late preterm infants.

T.N.K. Raju / Seminars in Fetal & Neonatal Medicine 17 (2012) 126e131 131

References

1. Raju TNK, Higgins RD, Stark AR, Leveno KJ. Optimizing care and outcome forlate preterm (near-term) infants: a summary of the workshop sponsored by theNICHD. Pediatrics 2006;118:1207e14.

2. Wang ML, Dorer DJ, Fleming MP, Catlin E. Clinical outcomes of near-terminfants. Pediatrics 2004;114:372e6.

3. Engle WA, Tomashek KM, Wallman C. the Committee on Fetus and Newborn,American Academy of Pediatrics. Pediatrics 2007;120:1390e401.

4. ACOG Committee Opinion No. 404, 2008. Late-preterm infants. Committee onObstetric Practice. Obstet Gynecol 2008;111:1029e32.

5. Martin JA, Hamilton BE, Sutton PD, et al. Births: final data for 2008. NationalVital Statistics Reports; vol. 59, no. 1. Hyattsville, MD: National Center for HealthStatistics; 2010.

6. Schroder HJ, Power GG. Engine and radiator: fetal and placental interactions forheat dissipation. Exp Physiol 1997;82:403e14.

7. Laptook A, Jackson GL. Cold stress and hypoglycemia in late preterm (“nearterm”) infant: impact on nursery admission. Semin Perinatol 2006;30:24e7.

8. Fairchild KD, Sun CCJ, Gross GC, Okogbule-Wonodi AC, Chasm RM, Viscardi RM.NICU admission hypothermia, chorioamnionitis, and cytokines. J Perinat Med2011;39:731e6.

9. Cummings KJ, Li A, Nattie EE. Brainstem serotonin deficiency in the neonatalperiod: autonomic dysregulation during mild cold stress. J Physiol2011;589:2055e64.

10. Tews D, Wabitsch M. Renaissance of brown adipose tissue. Horm Res Paediatr2011;75:231e9.

11. Consortium on Safe Labor, Hibbard JU, Wilkins I, et al. Respiratory morbidity inlate preterm births. JAMA 2010;28(304):419e25.

12. Ramachandrappa A, Rosenber ES, Wagoner S, Jain L. Morbidity and mortality inlate preterm infants with severe hypoxic respiratory failure on extracorporealmembrane oxygenation. J Pediatr 2011;159:192e8.

13. Colin AA, McEvoy C, Castile RG. Respiratory morbidity and lung function inpreterm infants of 32 to 36 weeks’ gestational age. Pediatrics 2010;126:115e28.

14. Smith LJ, McKay KO, van Asperen PP, Selvadurai H, Fitzgerald DA. Normal devel-opment of the lung and premature birth. Pediatr Respir Rev 2010;11:135e42.

15. Jain L, Eaton DC. Physiology of fetal lung fluid clearance and the effect of labor.Semin Perinatol 2006;30:34e43.

16. Eaton DC, Helms MN, Koval M, Bao HF, Jain L. The contribution of epithelialsodium channels to alveolar function in health and disease. Annu Rev Physiol2009;71:403e23.

17. Keszler M, Carbone MT, Cox C, Schumacher RE. Severe respiratory failure afterelective repeat cesarean delivery: a potentially preventable condition leadingto extracorporeal membrane oxygenation. Pediatrics 1992;89:670e2.

18. Darnall RA, Ariagno RL, Kinney HC. The late preterm infant and the control ofbreathing, sleep, and brainstem development: a review. Clin Perinatol2006;33:883e914.

19. Barlow SM. Central pattern generation involved in oral and respiratory controlfor feeding in the term infant. Curr Opin Otolaryngol Head Neck Surg2009;17:187e93.

20. Barlow SM. Oral and respiratory control for preterm feeding. Curr Opin Oto-laryngol Head Neck Surg 2009;17:179e86.

21. Garg M, Devaskar SU. Glucose metabolism in late preterm infants. Clin Perinatol2006;33:853e70.

22. Committee on Fetus and Newborn, Adamkin DH. Postnatal glucose homeo-stasis in late-preterm and term infants. Pediatrics 2011;127:575e9.

23. Watchko JF. Hyperbilirubinemia and bilirubin toxicity in the late preterminfant. Clin Perinatol 2006;33:839e52.

24. Academy of Breastfeeding Medicine. ABM clinical protocol #10: breastfeedingthe late preterm infant (34(0/7) to 36(6/7) weeks gestation) (first revision June2011). Breastfeed Med 2011;6:151e6.

25. Walker JC, Smolders MA, Gemen EF, et al. Development of lymphocytesubpopulations in preterm infants. Scand J Immunol 2011;73:53e8.

26. Strunk T, Currie A, Richmond P, Simmer K, Burgner D. Innate immunity inhuman newborn infants: prematurity means more than immaturity. J MaternFetal Neonatal Med 2011;24:25e31.

27. Blumer N, Pfefferle PI, Renz H. Development of mucosal immune function inthe intrauterine and early postnatal environment. Curr Opin Gastroenterol2007;23:655e60.

28. Pérez A, Gurbindo MD, Resino S, Aguarón A, Muñoz-Fernández MA. NK cellincrease in neonates from the preterm to the full-term period of gestation.Neonatology 2007;92:158e63.

29. Calder PC, Krauss-Etschmann S, de Jong EC, et al. Early nutrition and immunitye progress and perspectives. Br J Nutr 2006;96:774e90.

30. KapurR,YoderM,PolinRA.The immunesystem. In:MartinRJ, FanaroffAA,WalshM,editors. Neonatal perinatal medicine. 9th ed. St Louis: Elsevier; 2011. p. 761e885.

31. Wagner CL, Taylor SN, Johnson D. Host factors in amniotic fluid and breast milkthat contribute to gut maturation. Clin Rev Allerg Immunol 2008;34:191e204.

32. van Nimwegen FA, Penders J, Stobberingh EE, et al. Mode and place of delivery,gastrointestinal microbiota, and their influence on asthma and atopy. J AllergyClin Immunol 2011;128:948e55. e1e3.

33. Murgas Torrazza R, Neu J. The developing intestinal microbiome and its rela-tionship to health and disease in the neonate. J Perinatol 2011;311:S29e34.

34. Dominguez-Bello MG, Costello EK, Contreras M, et al. Delivery mode shapes theacquisition and structure of the initial microbiota across multiple body habitatsin newborns. Proc Natl Acad Sci USA 2010;107:11971e5.

35. Sangild PT. Gut responses to enteral nutrition in preterm infants and animals.Exp Biol Med 2006;231:1695e711.

36. Newell SJ, Sarkar PK, Durbin GM, Booth IW, McNeish AS. Maturation of thelower oesophageal sphincter in the preterm baby. Gut 1988;29:167e72.

37. Neu J. Gastrointestinal development and meeting the nutritional needs ofpremature infants. Am J Clin Nutr 2007;85(Suppl.):629Se34S.

38. Mshvidadze M, Neu J, Shuster J, Theriaque D, Li N, Mai V. Intestinal microbialecology in premature infants assessed with non-culture based techniques.J Pediatr 2010;156:20e5.

39. Biasucci G, Benenati B, Morelli L, Bessi E, Boehm G. Cesarean delivery mayaffect the early biodiversity of intestinal bacteria. J Nutr 2008;138:1796Se800S.

at SciVerse ScienceDirect

Seminars in Fetal & Neonatal Medicine 17 (2012) 132e137

Contents lists available

Seminars in Fetal & Neonatal Medicine

journal homepage: www.elsevier .com/locate/s iny

Obstetric decision-making and the late and moderately preterm infant

Cynthia Gyamfi-Bannerman*

Division of MaternaleFetal Medicine, Columbia University Medical Center, 622 W 168th St, New York, NY, USA

Keywords:Late preterm birthPrematurityRespiratory morbidity

* Tel.: þ44 212 305 6293; fax: þ44 212 342 2717.E-mail address: cg2231@columbia.edu.

1744-165X/$ e see front matter � 2012 Published bydoi:10.1016/j.siny.2012.01.014

s u m m a r y

The decision of when to deliver a patient for medical or obstetric complication directly affects theneonatal outcome. When the fetus is in danger due to suspected utero-placental insufficiency, thedecision to deliver is thought to benefit the neonate. However, the opposite may be true when a normallydeveloping fetus needs to be delivered for a maternal indication such as a persistently bleeding placentapraevia. These decisions are made daily by obstetric providers. The following is a review of obstetricdecision-making for moderate and late preterm pregnancies.

� 2012 Published by Elsevier Ltd.

1. Introduction

In 2005 the phrase ‘late preterm’ was introduced by theNational Institute of Child Health and Human Development(NICHD) to identify infants born between 340/7 and 366/7 weeksof gestation as a high risk group with increased morbidities whencompared with term infants. This replaced the earlier phrase‘near term’, which implied that these infants behaved similarly toterm infants. Limiting this group to neonates born from 34 to 36weeks also helped focus research on this cohort, allowinginvestigators to better characterize their outcomes. Unfortu-nately, the same consensus has not been reached for moderatepreterm infants, so the literature regarding this group of infantshas various gestational age definitions starting at 30e32 weeks ofgestation and ending at 34e36 weeks.1,2 We now have ampledata showing that most morbidities related to prematurityare increased in the late preterm group when compared withinfants born at term, most markedly where respiratory morbid-ities are concerned.3e5 Intuitively, these morbidities are alsohigher in the moderate preterm group. Therefore, the decision fordelivery in these patients should be weighed against the knownmorbidities associated with prematurity in these groups. Thisarticle discusses both morbidity and common obstetric indica-tions for delivery among moderate preterm and late preterminfants.

Elsevier Ltd.

2. Common complications warranting delivery and theirmanagement in the late preterm period

2.1. Obstetric

2.1.1. Preterm labourPreterm labour is defined as contractions leading to cervical

change prior to 37weeks’ gestation. Though the etiology of pretermlabour is multifactorial, it is thought to result from one of thefollowing pathways: infection, decidual haemorrhage, uterineoverdistention, or hypothalamicepituitaryeadrenal (HPA) axis.6

Traditional treatments for active preterm labour include tocolysisto decrease contractions; antibiotics to prevent early onset group Bstreptococcus infection; and antenatal corticosteroids, generally inthe form of betamethasone or dexamethasone, to promote fetallung maturity as well as to decrease the rates of necrotizingenterocolitis and intraventricular haemorrhage.7

While the treatment of preterm labour is fairly consistentamong most medical centres in the USA, the decision as to thegestational age beyond which intervention is no longer warrantedis not clear. The most recent guidelines from the American Collegeof Obstetricians and Gynecologists (ACOG) suggest that the deci-sion of when to intervene with preterm labour should be based onthe neonatal intervention capacities at the hospital of the practisingobstetrician since it is known that the neonatal morbidity isinversely proportional to the gestational age at delivery.7 Becausethe role of tocolysis is primarily to allow for administration ofantenatal corticosteroids since the efficacy beyond 48e72 h isunclear, and since antenatal corticosteroids are traditionallyadministered up to 34 weeks’ gestation, many tertiary care centreswill not prevent delivery in an actively labouring patient before 34weeks’ gestation. In two reviews of indications for late preterm

C. Gyamfi-Bannerman / Seminars in Fetal & Neonatal Medicine 17 (2012) 132e137 133

birth from different regions in the USA, Holland et al. found that36.2% of late preterm birth was spontaneous, while Gyamfi-Bannerman et al. found this proportion to be higher at 53.1%.8,9

Both of these numbers reflect a large population of infants whowill be affected by practice patterns where they are born.

2.1.2. Preterm premature rupture of the membranes (PPROM)Similar to preterm labour with intact membranes, PPROM is

a spontaneous event with management options directly linked tothe gestational age. The American College of Obstetricians andGynecologists has advocated for delivery at or beyond 34 weeks’gestation if PPROM is to occur; however, they acknowledge thatthese recommendations are based on limited and inconsistent(Level B) data.10 Regardless, regional practice patterns and hospitalneonatal intensive care unit (NICU) capabilities have led to varia-tions in gestational age at delivery. A survey of practice patterns bymaternalefetal medicine (MFM) providers showed that nearly 50%delivered pregnancies complicated by PPROM at later than 34weeks’ gestation.11

The question of whether a delay in delivery is harmful forpatients with PPROM was recently addressed by a review in theCochrane database of the use of antibiotics with rupturedmembranes. They showed that administration of antibioticssignificantly decreased the rates of both neonatal infection (oddsratio: 0.67; 95% confidence interval: 0.52e0.85) and chorioamnio-nitis (0.66; 0.46e0.96) at up to 37 weeks’ gestation.12 Antibioticsalso made it less likely that a patient with ruptured membraneswould deliver by 48 h (0.71; 0.58e0.87). Patients who wereincluded in this review delivered at up to 37 weeks’ gestation. Adifferent Cochrane database review of expectantly managedPPROM to 37 weeks’ gestation showed that there were no differ-ences in the rates of neonatal sepsis (relative risk: 1.33; 95%confidence interval: 0.72e2.47), perinatal mortality (0.98;0.41e2.36), intrauterine deaths (0.26; 0.04e1.52) or neonataldeaths (1.59; 0.61e4.16) when comparing early delivery withexpectant management.13 Interestingly, early delivery increasedthe incidence of caesarean delivery (1.51; 1.08e2.10) and endo-metritis (2.32; 1.33e4.07), likely secondary to a higher caesareandelivery rate in that group. However, there was no statisticaldifference in the rates of chorioamnionitis (0.44; 0.17e1.14) sug-gesting safety, at least with short term delay. Ongoing prospectivetrials, including PPROMT and PPROMEXIL, will help to furtherelucidate whether prolonged expectant management to 37 weeks’gestation is appropriate for late preterm PPROM.14,15

2.1.3. Pre-eclampsia/gestational hypertensionThere is some debate and regional variation regarding the

appropriate gestational age for delivery for some of the hyperten-sive disorders of pregnancy. While most experts agree that patientswith clinically stable severe pre-eclampsia should be delivered by�34 weeks’ gestation, there remains discussion regarding theappropriate gestational age for delivery of mild pre-eclampsia andgestational hypertension. A recent clinical trial evaluating labourinduction versus expectant management for mild pre-eclampsiaand gestational hypertension at �36 weeks found that maternaloutcomes were improved with labour induction after 37 weeks.16

The authors did not find this trend with delivery between 36 and37 weeks, but they cited low numbers in that subgroup as a reasonfor further study at that gestational age. Habli et al. performeda secondary analysis of data from an NICHD clinical trial on calciumto prevent pre-eclampsia.17 Specifically, they tried to assesswhether the poorer neonatal outcomes related to gestationalhypertension and pre-eclampsia were secondary to the diseaseprocess or iatrogenic delivery resulting from these diagnoses ina cohort of both hypertensive and normotensive women delivering

at 35, 36, or 37 weeks’ gestation. The authors found a higher rate ofsmall for gestational age, NICU admission, and neonatal length ofstay in the hypertensive compared with the normotensive group.These differences did not vary by the severity of hypertension;rather, they seemed to be related to labour induction. The authorsconcluded that the need for intervention at these earlier gestationalages should be carefully evaluated. Recently, Barton et al. reviewedoutcomes specifically associated with gestational hypertension andlate preterm delivery using a retrospective database.18 They foundthat elective delivery from 34 to 36 weeks resulted in increasedneonatal morbidity without maternal benefit. The cited literaturesuggests that patients with mild pre-eclampsia or gestationalhypertension can be delivered at term (�37 weeks). The NICHD/Society for MaternaleFetal Medicine (SMFM) workshop hada similar conclusion recommending delivery at 37 weeks for mildpre-eclampsia and 37e38 weeks for gestational hypertension.19

A recent study randomized 756 patients to labour inductionversus expectant management at 36 weeks’ gestation.16 Theyincluded women with mild pre-eclampsia (defined as a diastolicblood pressure of >90 mmHg with proteinuria, or gestationalhypertension with blood pressures <170/110 mmHg). They foundimproved maternal outcomes in the subgroup of womenwho wereinduced at 37 weeks’ gestation. Women pregnant from 36 to 37weeks’ gestation benefited from expectant management. We canconclude from their findings that it is likely that women with mildpre-eclampsia or gestational hypertension would benefit fromdelivery at 37 weeks’ gestation rather than waiting until 39 weeks.However, it would be interesting to see whether the same benefitwould be found in the subgroup of women who had only gesta-tional hypertension.

2.1.4. OligohydramniosOligohydramnios, defined either as an amniotic fluid index (AFI)

of <5, an AFI <5th percentile for gestational age, or a maximumvertical pocket of <2 cm is a complication of pregnancy that leadsto concern both for patients and their providers. Since amnioticfluid decreases with advancing gestational age, this complication isoften noted in the late preterm period. Isolated oligohydramnios isnot thought to be related to adverse perinatal outcome.20 However,the most commonly followed practice for this complication isdelivery at term (>37 weeks’ gestation). Sverker et al.21 random-ized 54 women at 40 completed weeks’ gestation with isolatedoligohydramnios to labour induction versus expectant manage-ment. The primary neonatal outcomes were Apgar scores and cordpH. There were no differences between groups. There were also nodifferences between NICU admission rates. With regards tomaternal complications, there was no increase in complications inthe expectantly managed group, and the rate of caesarean deliverywas similar between groups. More recently, Melamed et al.reviewed 108 cases of isolated oligohydramnios diagnosed from240/7 to 366/7 weeks matched 3:1 (controls:cases) to controls withnormal fluid.19 They found that the majority of adverse perinataloutcomes in cases could be attributed to iatrogenic prematurityrelated to early delivery, rather than the oligohydramnios itself.These data suggest that isolated oligohydramnios, while certainlywarranting closer evaluation, should not be an indication fordelivery without other complicating factors before term.

2.1.5. Prior caesarean deliveryAlthough the timing of an elective repeat caesarean after a prior

low transverse cesarean is unambiguous at 39weeks’ gestation, it isless obvious when the prior scar is from a classical uterine incision.The rate of uterine rupture in women with a prior classical scar isthought to be between 4% and 9%.22 Chauhan et al. revieweduterine rupture and dehiscence rates in a cohort of 157 womenwho

C. Gyamfi-Bannerman / Seminars in Fetal & Neonatal Medicine 17 (2012) 132e137134

had had a prior classical cesarean.7 All were planned for repeatcaesarean. There was one uterine rupture in this series resulting inan IUFD, which was of a womanwho presented with bleeding at 29weeks’ gestation. Otherwise, there was a 9% dehiscence rate notedat the time of repeat caesarean, which did not alter maternal orneonatal outcomes compared with those women without dehis-cence. Of note, almost 50% of these patients presented in labour,before an amniocentesis could be performed to document fetallung maturity. Although the mean duration of labour was7.3 � 5.6 h, there were no episodes of uterine rupture in that group.The authors concluded that uterine rupture and dehiscence couldnot be accurately predicted or prevented. Other data not specific toclassical caesarean are supplied from a review of >40,000 deliv-eries in an Irish hospital.23 Of the 1355 women with one or moreprior caesareans attempting a trial of labour, there were six uterineruptures. There was one pre-labour rupture in a woman witha prior classical caesarean, but the authors did not provide infor-mation on how many patients had had a prior classical caesarean.There is very little other evidence to help guide timing of delivery inthese patients. Expert opinion regarding timing of delivery in thesecases still favours delivery prior to the onset of labour. This isconsistent with the recommendation from the NICHD/SMFMworkshop for delivery at 36e37 weeks’ gestation.24

2.1.6. Accreta/praeviaThe SMFM clinical opinion paper on placenta accreta states that

planned late preterm delivery (340/7 to 366/7 weeks’ gestation) is anacceptable management strategy since there are data showingincreased rates of haemorrhage once delivery occurs beyond 36weeks’ gestation.25 A decision analysis on the optimal timing ofdelivery for patients with placenta accreta suggested that 34weeks’gestation optimized both maternal and neonatal outcomes.26

Therefore, it is likely appropriate to deliver a patient suspected tohave placenta accreta if the pregnancy continues to the early termperiod.

The optimal gestational age at delivery for womenwith placentapraevia is based primarily on expert opinion and has been quotedas between 36 and 38 weeks’ gestation.27,28 Some authoritiesadvocate fetal lung maturity testing in a stable patient if delivery isplanned at �37 weeks.27 The gestational age of delivery weighsthe risk of prematurity versus the risk of maternal haemorrhage,which increases in the presence of contractions. Therefore,based on few data, it is likely appropriate to deliver women withplacenta praevia in the early term period if the pregnancy remainsongoing. However, the optimal timing of delivery in this group isunknown.

2.2. Maternal

Chronic maternal disease during pregnancy is common dueboth to the high prevalence of obesity and to an increase in womenwaiting for advanced maternal age to attempt their first pregnancy.Exacerbation of chronic disease resulting from older age andobesity can lead to preterm delivery.

Obesity is considered endemic in the USA.29 Aside from a higherprevalence of chronic hypertension and diabetes in obese women,they are also noted to be at increased risk of pre-eclampsia. Anepidemiologic study ofmore than 1.4millionwomen found that therisk of pre-eclampsia is doubled for every 5e7 kg/m2 increase inpre-pregnancy body mass index.30 As noted previously, pre-eclampsia is a common cause of preterm birth. Thus, obesity islinked to prematurity. Obese women also have a 10-fold increase inthe rate of chronic hypertension comparedwith non-obesewomen,further increasing the risk of pre-eclampsia.29 Since weight lossduring pregnancy is not recommended, this risk factor can be

modified postpartum, but ideally should be modified beforeattempting pregnancy.

Chronic hypertension is another common cause of prematurity.The prevalence of this disease is thought to bew7% inwomen aged<40 years.31 Pregnant women are given this diagnosis if they havehypertension outside of pregnancy, or if they first manifest elevatedblood pressures before 20 weeks’ gestation. Increased vascularresistance noted in hypertensive patients can lead to intrauterinegrowth restriction (IUGR, discussed below) which, in the setting ofabnormal testing, can lead to preterm delivery. Chronic hyperten-sive parturients are also more likely to develop pre-eclampsiacompared to their normotensive counterparts. A MaternaleFetalMedicine Units Network trial found that 25% of women withchronic hypertension develop superimposed pre-eclampsia,increasing the likelihood that preterm pregnancy interruptionwould be necessary.32 A secondary analysis of the aforementionedNetwork study showed that 33% of women with chronic hyper-tension delivered prior to 37 weeks with 18% delivering prior to 35weeks,33 making chronic hypertension a common obstetricproblem leading to prematurity.

Finally, connective tissue disorders, known to be more commonin females compared with males, are another common cause ofprematurity. Disorders such as lupus or antiphospholipid antibodysyndrome have in common immune-mediated complexes that canaffect various organ systems leading to end-organ damage andshortened life-spans. The 10- and 20-year survival rates for womenwith lupus are 75% and 50%, respectively.34 Pregnancy outcomewith lupus is related to the presence or absence of hypertensionand proteinuria. Evenwomenwith lupus in remission have a higherrisk of pre-eclampsia. This risk is increased if both hypertensionand renal insufficiency are present. The diagnosis of pre-eclampsiaversus a lupus flare can be difficult during pregnancy. Complementlevels may be helpful to distinguish the two disease processes, sothese could be obtained at the first prenatal visit.29 While lupuspatients with severe pre-eclampsia should be delivered, patientswith a lupus flare can bemanaged conservatively. Both late pretermand moderate preterm deliveries are not uncommon in this groupof women. Similarly, women with antiphospholipid antibodysyndrome are known to have an increase in adverse fetal outcomes.Treatment with aspirin and heparin has been the mainstay in thecohort and allows many of these women to go to term. However,fetal growth restriction and abnormal fetal testing are causes ofpreterm delivery for these women.29

2.3. Fetal

2.3.1. IUGRAlthough it is well documented that fetuses with IUGR have an

increase in both fetal and neonatal morbidity andmortality,35 thereare few prospective data that evaluate risks associatedwith delayeddelivery of IUGR at term. Most available data are related to delayingdelivery in preterm IUGR.36 Because of the increase in adverseoutcomes in this group, many authorities state that delivery forIUGR should occur at term.37 However, there is one recentprospective, randomized controlled trial of management of IUGRbeyond 360/7 weeks’ gestation.38 The authors randomized 321patients with IUGR with either normal or abnormal Dopplerultrasound to either induction or expectant monitoring. Theprimary outcome was short term neonatal morbidity and includeddeath before discharge from the hospital, 5 min Apgar of <7,umbilical artery pH of <7.05, or admission to the NICU; and thestudy was powered to assess for equivalence of this outcome. Theaverage gestational age in the induction group was 380/7 weeks,whereas the average gestational age in the expectant managementgroup was 393/7 weeks. There was no difference in their primary

C. Gyamfi-Bannerman / Seminars in Fetal & Neonatal Medicine 17 (2012) 132e137 135

outcome. Additionally, significantly more women in the expectantmonitoring group presented in spontaneous labour (46.0% vs 3.7%),but there was no difference in the caesarean delivery rate. The dataon longer term outcomes in this group is limited to delayed deliveryof preterm infants with IUGR, but these data also suggest nodifference in the rates of death or disability at age �2 years.39 Themost recent ACOG bulletin on IUGR (January, 2000; Number 12)states that delivery should be considered once there is completecessation of growth or a non-reassuring fetal assessment; a specificgestational age for delivery is not supplied. Many cases of suspectedIUGR are from constitutionally small parents. Based on the aboveinformation, the optimal timing of delivery for women with IUGRand otherwise reassuring testing remains unknown, with at leastone study suggesting that delivery after 37 weeks does not increasethe rate of short termmorbidity. Recommendations from the EuniceKennedy Shriver National Institute of Child Health and HumanDevelopment and the Society for MaternaleFetal Medicine work-shop titled ‘Timing of Indicated Late Preterm and Early Term Births’suggest that delivery occur between 38 and 39 weeks in cases ofIUGR without other comorbidities and otherwise reassuring fetaltesting.24

2.3.2. Multiple gestationIt is known that twin gestations deliver earlier than singletons;

the average gestational age of delivery is thought to be in the latepreterm period. However, there are very few indications for electivedelivery of dichorionic, diamniotic twins in this gestational agewindow outside of coexisting maternal or obstetric complicationsindicating delivery. The consensus of the NICHD/SMFM workshopparticipants regarding elective late preterm delivery for dichorionictwins is in the situation of an intrauterine fetal demise (IUFD) ofa co-twin.24

However, the appropriate gestational age for delivery ofuncomplicated monochorionic, diamniotic (mono/di) twins ishighly debated with some authorities arguing that it should occurin the moderate preterm period. The recommendations vary from32 to 37 weeks’ gestation.40e42 Barigye et al.40 identified a group of151 uncomplicated mono/dichorionic twin gestations and evalu-ated the risk of fetal demise. They found that the risk of IUFD at�32weeks in these seemingly normal twins was 1 in 23. Furthermore,4.6% of pregnancies had an IUFD within 2 weeks of a normal scan.They suggested that a policy of elective preterm delivery at �32weeks would obviate this risk. On the other end of the spectrumSmith et al.41 reviewed 236 ongoing mono/dichorionic twin preg-nancies and found that the likelihood of two live births in the‘uncomplicated’ group was 99.5% at �32 weeks. They recom-mended against elective preterm delivery of uncomplicated mono/dichorionic twins. Consensus and expert opinion suggest thatdelivery for uncomplicatedmono/dichorionic twins should occur inthe late preterm period between 34 and 37 weeks’ gestation.24

2.3.3. AnomaliesDefining an appropriate gestational age for delivery in the

setting of fetal anomalies is nearly impossible due to the largenumbers and varied complications related to the numerousanomalies. It has been suggested that there are >100 differentcategories of anomalies; however, most of these do not haveoutcomes that will be improved by delivery prior to term.43 Onenotable exception to this is an anomaly that had led to fetalhydrops. Obstructive lesions and fetal heart failure can often lead tohydrops, which confers a poor prognosis for the fetus. Mostauthorities believe that hydrops presenting beyond 34 weeks’gestation is an indication for late preterm delivery.43 Anotherexception is in the setting of fetal arrhythmias. Specifically, isolatedfetal tachyarrhythmias can lead to hydrops from heart failure. The

first line treatment for this type of anomaly involves treatment ofthe mother with medications such as digoxin, procainamide, fle-cainide, or amiodarone. Because of the potential for both maternaland fetal toxicity from some of these medications, late pretermdelivery may be indicated for fetuses refractory to treatment orwhen the initial diagnosis occurs near term.44

3. Common complications and their management in theModerately preterm (MP) period

3.1. Intrauterine growth restriction (IUGR)

Whereas intrauterine growth restriction without comorbiditiescan be delivered at term, there are certain cases of pathologicalIUGR that will require preterm delivery regardless of the gesta-tional age. Timing of delivery in pregnancies complicated by IUGRwas evaluated by the Growth Restriction Intervention Trial (GRIT).They randomized women with growth-restricted fetuses from 24to 36 weeks’ gestation to immediate delivery, specifically after 48 hfor steroids, versus delayed delivery, specifically, abnormal testingor other factor requiring delivery. They found equal proportions ofstillbirth in both groups.45 The 2-year follow-up to this studyshowed that gestational age at delivery was the primary determi-nant of neonatal outcome. Infants born at <31 weeks had lowerdevelopmental scores than infants born at>31 weeks, regardless ofwhether they were in the immediate or delayed delivery group.39

Results from the GRIT study group suggest that gestational age isthe primary determinant of neonatal outcome.

While Doppler waveforms were recorded by the GRIT studygroup, they were not used in the randomization scheme or as partof the protocol. This is in contrast to a retrospective cohort study byChalubinski et al. where they identified a cohort of IUGR fetusesover a 10-year period at a single institution and evaluated perinataloutcomes based on Doppler findings within 7 days of delivery.46

They divided identified pregnancies into three groups: (1)abnormal umbilical artery (UA) pulsatility index (PI) or absent UAend-diastolic flow, and normal median cerebral artery (MCA) PI; (2)abnormal UA PI, or absent or reversed UA end-diastolic flow, andabnormal MCA PI with normal ductus venosus (DV) PI; and (3)absent or reversed UA end-diastolic flow, and abnormal MCA PI,and abnormal DV PI (mean >2 SD, a-wave present or absent orreversed end-diastolic flow). They found rates of neonatal demiseof 0/17 (0%), 2/44 (4.5%), and 7/30 (23.3%) in infants from Dopplergroups 1, 2, and 3, respectively (P ¼ 0.019). Gestational age alsoplayed a role, suggesting that information on Doppler indicesshould be used in delivery planning after 28 weeks’ gestation.Therefore, moderate preterm infants with IUGR and comorbidities,specifically severe Doppler abnormalities such as those in group 3,are likely to be delivered by obstetricians to prevent demise.

3.2. Severe pre-eclampsia/HELLP (hemolysis, elevated liverenzymes, low platelet count)

Severe pre-eclampsia complicates 0.3% of pregnancies at <34weeks’ gestation.47 The decision of whether to expectantly managethese patients has direct implications on neonatal outcomes.Traditionally, severe pre-eclampsia is defined by a blood pressure of>160/110 mmHg in the presence of proteinuria. Clinicians weighthe decision for delivery versus expectant management on thematernal and fetal status at the time of initial diagnosis. Conditionsconsidered contraindications to conservative management arepulmonary edema, abruption, significant renal dysfunction, HELLPsyndrome, non-reassuring fetal status and eclampsia.48

Studies that have evaluated conservative management of severepre-eclampsia have used several gestational age parameters to

C. Gyamfi-Bannerman / Seminars in Fetal & Neonatal Medicine 17 (2012) 132e137136

define when this management should end by delivery. Mostnotably, data from expectant management to both 32 and 34weeks’ gestation suggest improved neonatal outcomes from thesestrategies.49,50 Therefore, women with severe pre-eclampsia arelikely to be delivered in both the moderate and late pretermperiods, adding to the numbers of indicated preterm births.Importantly, definitive treatment of severe pre-eclampsia andHELLP syndrome remains by delivery. Hence, a clinician faced withsevere pre-eclampsia not amenable to expectant management willrightly deliver the mother.

3.3. Spontaneous preterm labour or PPROM (preterm prematurerupture of the membranes)

Finally, whereas management of spontaneous preterm labour orPPROM in the late preterm periodmay be varied, treatment of theseconditions prior to 34 weeks is fairly straightforward. ACOGadvocates for tocolysis, antenatal corticosteroid administration,and antibiotic to prevent group B streptococcal sepsis in womenpresenting in preterm labour between 24 and 34 weeks’ gestation.7

For women with PPROM, in addition to the interventions used forpreterm labour, the management includes broader spectrum anti-biotics to increase the latency from ruptured membranes todelivery.10 For both conditions, moderate preterm delivery wouldbe indicated if chorioamnionitis were diagnosed.

4. Conclusion

There are many clinical presentations, both spontaneous andindicated, that lead to moderate and late preterm delivery. Whilethe optimal treatment and prevention of spontaneous pretermbirth is an ongoing conundrum that has had little influence on theproportion of patients that deliver preterm, there remain manymedically indicated preterm births in both gestational agewindows that have coincided with a decrease in the number ofstillbirths.51,52 Thus, obstetric practice continues to directly impactthe numbers of preterm infants delivered annually, which confersa significant health care burden. Further evaluation of these prac-tices may help to decrease the rates of prematurity.

Practice points

� There are many medical indications that result inpreterm delivery.

� Some common indications for late preterm birth,specifically mild pre-eclampsia and gestational hyper-tension, may not be justified.

� Indicatedmoderate preterm birth is much less commonand generally is related to pathologic fetal growth orsevere pre-eclampsia/HELLP.

� Obstetric practice has a direct impact on the rates ofprematurity.

Conflict of interest statement

None declared.

Funding sources

None.

References

1. Altman M, Vanpee M, Cnattingius S, Norman M. Moderately preterm infantsand determinants of length of hospital stay. Arch Dis Child Fetal Neonatal Ed2009;94:F414e8.

2. Wooldridge J, Hall WA. Posthospitalization breastfeeding patterns of moder-ately preterm infants. J Perinat Neonatal Nurs 2003;17:50e64.

3. Hibbard JU, Wilkins I, Sun L, et al. Respiratory morbidity in late preterm births.JAMA 2010;304:419e25.

4. McIntire DD, Leveno KJ. Neonatal mortality and morbidity rates in late pretermbirths compared with births at term. Obstet Gynecol 2008;111:35e41.

5. Yoder BA, Gordon MC, Barth Jr WH. Late-preterm birth: does the changingobstetric paradigm alter the epidemiology of respiratory complications? ObstetGynecol 2008;111:814e22.

6. Lockwood CJ, Kuczynski E. Markers of risk for preterm delivery. J Perinat Med1999;27:5e20.

7. Anonymous. ACOG practice bulletin: management of preterm labor 2003.8. Holland MG, Refuerzo JS, Ramin SM, Saade GR, Blackwell SC. Late preterm

birth: how often is it avoidable? Am J Obstet Gynecol 2009;201:404.e1e4.9. Gyamfi-Bannerman C, Fuchs KM, Young OM, Hoffman MK. Nonspontaneous

late preterm birth: etiology and outcomes. Am J Obstet Gynecol2011;205:456.e1e6.

10. American College of Obstetricians and Gynecologists. ACOG practice bulletin:premature rupture of membranes 2007. Number 80.

11. Ramsey PS, Nuthalapaty FS, Lu G, Ramin S, Nuthalapaty ES, Ramin KD.Contemporary management of preterm premature rupture of membranes(PPROM): a survey of maternalefetal medicine providers. Am J Obstet Gynecol2004;191:1497e502.

12. Kenyon S, Boulvain M, Neilson JP. Antibiotics for preterm rupture ofmembranes. Cochrane Database Syst Rev 2010:CD001058.

13. Buchanan SL, Crowther CA, Levett KM, Middleton P, Morris J. Planned earlybirth versus expectant management for women with preterm prelabourrupture of membranes prior to 37 weeks’ gestation for improving pregnancyoutcome. Cochrane Database Syst Rev 2010:CD004735.

14. Morris JM, Roberts CL, Crowther CA, Buchanan SL, Henderson-Smart DJ,Salkeld G. Protocol for the immediate delivery versus expectant care of womenwith preterm prelabour rupture of the membranes close to term (PPROMT)Trial [ISRCTN44485060]. BMC Pregn Childbirth 2006;6:9.

15. van der Ham DP, Nijhuis JG, Mol BW, et al. Induction of labour versus expectantmanagement inwomenwith pretermprelabour rupture ofmembranes between34 and 37 weeks (the PPROMEXIL-trial). BMC Pregn Childbirth 2007;7:11.

16. Koopmans CM, Bijlenga D, Groen H, et al. Induction of labour versus expectantmonitoring for gestational hypertension or mild pre-eclampsia after 36 weeks’gestation (HYPITAT): a multicentre, open-label randomised controlled trial.Lancet 2009;374:979e88.

17. Habli M, Levine RJ, Qian C, Sibai B. Neonatal outcomes in pregnancies withpreeclampsia or gestational hypertension and in normotensive pregnanciesthat delivered at 35, 36, or 37 weeks of gestation. Am J Obstet Gynecol2007;197:406.e1e7.

18. Barton JR, Barton LA, Istwan NB, et al. Elective delivery at 34(/) to 36(/) weeks’gestation and its impact on neonatal outcomes in women with stable mildgestational hypertension. Am J Obstet Gynecol 2011;204:44.e1e5.

19. Melamed N, Pardo J, Milstein R, Chen R, Hod M, Yogev Y. Perinatal outcome inpregnancies complicated by isolated oligohydramnios diagnosed before 37weeks of gestation. Am J Obstet Gynecol 2011;205:241.e1e6.

20. Zhang J, Troendle J,Meikle S,KlebanoffMA,RayburnWF. Isolatedoligohydramniosis not associated with adverse perinatal outcomes. BJOG 2004;111:220e5.

21. Sverker E, Andersson A, Johansson A, Kublicas M. Oligohydramnios inuncomplicated pregnancies beyond 40 completed weeks. Fetal Diagn Ther2005;20:182e5.

22. Cunningham FG, Hauth JC, Leveno KJ, Gilstrap III LC, Bloom SL, Wenstrom KD,editors. Williams’ obstetrics. 22nd ed. New York: McGraw-Hill; 2005.

23. Meehan FP, Magani IM. True rupture of the caesarean section scar (a 15 yearreview, 1972e1987). Eur J Obstet Gynecol Reprod Biol 1989;30:129e35.

24. Spong CY, Mercer BM, D’Alton M, Kilpatrick S, Blackwell S, Saade G. Timing ofindicated late-preterm and early-term birth. Obstet Gynecol 2011;118:323e33.

25. Belfort MA. Placenta accreta. Am J Obstet Gynecol 2010;203:430e9.26. Robinson BK, Grobman WA. Effectiveness of timing strategies for delivery of

individuals with placenta previa and accreta. Obstet Gynecol 2010;116:835e42.27. Oyelese Y, Smulian JC. Placenta previa, placenta accreta, and vasa previa. Obstet

Gynecol 2006;107:927e41.28. Hull AD, Resnik R. Placenta previa, placenta accreta, abruptio placentae, and vasa

previa. In: Creasy RK, Resnik R, Iams JD, editors. Maternalefetal medicine: prin-ciples and practice. 6th ed. Philadelphia: Saunders/Elsevier; 2009. p. 725e37.

29. Cunningham FG, Leveno KJ, Bloom SL, Hauth JC, Rouse DJ, Spong CY, editors.Williams’ obstetrics. 23rd ed. New York: McGraw-Hill; 2010.

30. O’Brien TE, Ray JG, Chan WS. Maternal body mass index and the risk ofpreeclampsia: a systematic overview. Epidemiology 2003;14:368e74.

31. Hajjar I, Kotchen TA. Trends in prevalence, awareness, treatment, and controlof hypertension in the United States, 1988e2000. JAMA 2003;290:199e206.

32. Caritis S, Sibai B, Hauth J, et al. Low-dose aspirin to prevent preeclampsia inwomen at high risk. National Institute of Child Health and Human Develop-ment Network of MaternaleFetal Medicine Units. N Engl J Med1998;338:701e5.

C. Gyamfi-Bannerman / Seminars in Fetal & Neonatal Medicine 17 (2012) 132e137 137

33. Sibai BM, Lindheimer M, Hauth J, et al. Risk factors for preeclampsia, abruptioplacentae, and adverse neonatal outcomes among women with chronichypertension. National Institute of Child Health and Human DevelopmentNetwork of MaternaleFetal Medicine Units. N Engl J Med 1998;339:667e71.

34. Jacobsen S, Petersen J, Ullman S, et al. Mortality and causes of death of 513Danish patients with systemic lupus erythematosus. Scand J Rheumatol1999;28:75e80.

35. Williams RL, Creasy RK, Cunningham GC, Hawes WE, Norris FD, Tashiro M.Fetal growth and perinatal viability in California. Obstet Gynecol1982;59:624e32.

36. Group TGS. A randomised trial of timed delivery for the compromised pretermfetus: short term outcomes and Bayesian interpretation. Br J Obstet Gynaecol2003;110:27e32.

37. Resnik R, Creasy RK. Intrauterine growth restriction. In: Creasy RK, Resnik R,Iams JD, editors. Maternalefetal medicine: principles and practice. 6th ed. Phil-adelphia: Saunders/Elsevier; 2009. p. 635e50.

38. Boers KE, Vijgen SM, Bijlenga D, et al. Induction versus expectant monitoringfor intrauterine growth restriction at term: randomised equivalence trial(DIGITAT). BMJ 2010;341:c7087.

39. Thornton JG, Hornbuckle J, Vail A, Spiegelhalter DJ, Levene M. Infant wellbeingat 2 years of age in the Growth Restriction Intervention Trial (GRIT): multi-centred randomised controlled trial. Lancet 2004;364:513e20.

40. Barigye O, Pasquini L, Galea P, Chambers H, Chappell L, Fisk NM. High risk ofunexpected late fetal death in monochorionic twins despite intensive ultra-sound surveillance: a cohort study. PLoS Med 2005;2:e172.

41. Smith NA, Wilkins-Haug L, Santolaya-Forgas J, et al. Contemporary manage-ment of monochorionic diamniotic twins: outcomes and delivery recommen-dations revisited. Am J Obstet Gynecol 2010;203:133.e1e6.

42. Lee YM, Wylie BJ, Simpson LL, D’Alton ME. Twin chorionicity and the risk ofstillbirth. Obstet Gynecol 2008;111:301e8.

43. Craigo SD. Indicated preterm birth for fetal anomalies. Semin Perinatol2011;35:270e6.

44. Tachyarrhythmias. In: Bianchi DW, Crombleholme T, D’Alton ME, Malone FD,editors. Fetology. New York: McGraw-Hill; 2010. p. 313e9.

45. GRIT Study Group. A randomised trial of timed delivery for the compromisedpreterm fetus: short term outcomes and Bayesian interpretation. BJOG2003;110:27e32.

46. Chalubinski KM, Repa A, Stammler-Safar M, Ott J. The impact of dopplersonography on intrauterine management and neonatal outcome in pretermfetuses with intrauterine growth retardation. Ultrasound Obstet Gynecol 2011May 5. doi:10.1002/uog.9039 [Epub ahead of print].

47. Catov JM, Ness RB, Kip KE, Olsen J. Risk of early or severe pre-eclampsia relatedto pre-existing conditions. Int J Epidemiol 2007;36:412e9.

48. Sibai BM. Evaluation and management of severe preeclampsia before 34 weeks’gestation. Am J Obstet Gynecol 2011;205:191e8.

49. Odendaal HJ, Pattinson RC, Bam R, Grove D, Kotze TJ. Aggressive or expectantmanagement for patients with severe preeclampsia between 28e34 weeks’gestation: a randomized controlled trial. Obstet Gynecol 1990;76:1070e5.

50. Sibai BM, Mercer BM, Schiff E, Friedman SA. Aggressive versus expectantmanagement of severe preeclampsia at 28 to 32 weeks’ gestation: a random-ized controlled trial. Am J Obstet Gynecol 1994;171:818e22.

51. Ananth CV, Vintzileos AM. Epidemiology of preterm birth and its clinicalsubtypes. J Matern Fetal Neonatal Med 2006;19:773e82.

52. Ananth CV, Gyamfi C, Jain L. Characterizing risk profiles of infants who aredelivered at late preterm gestations: does it matter? Am J Obstet Gynecol2008;199:329e31.

at SciVerse ScienceDirect

Seminars in Fetal & Neonatal Medicine 17 (2012) 138e142

Contents lists available

Seminars in Fetal & Neonatal Medicine

journal homepage: www.elsevier .com/locate/s iny

Obstetric management of moderate and late preterm labour

P.C. McParland*

University Hospitals of Leicester, Kensington Building, Leicester Royal Infirmary, Infirmary Square, Leicester LE1 5WW, UK

Keywords:FibronectinPreterm labourPreterm rupture of membranesProgesterone

* Tel./fax: þ44 116 2585923.E-mail address: pcm7@le.ac.uk.

1744-165X/$ e see front matter � 2012 Elsevier Ltd.doi:10.1016/j.siny.2012.01.013

s u m m a r y

Moderate and late preterm births account for the majority of preterm babies. The common perceptionthat birth at 32e36 weeks’ gestation carries few risks is now being challenged, as these babies haveincreased risk of neonatal mortality and morbidity. However, spontaneous labour at this gestationfrequently has no specific, easily identifiable precursor, although preterm birth per se has a number ofepidemiological and clinical associations. Prediction and prevention of preterm birth is currently largelyaimed at identifying women at high risk such as those with previous preterm birth, and targetingintervention at this group. Both cervical length assessment and fibronectin testing permit some modi-fication of the likelihood of preterm birth in this group. Progesterone treatment for the prevention ofpreterm birth is currently being researched widely, and appears a potentially promising strategy. Babiesborn at 32e36 weeks’ gestation need careful monitoring in labour, with modification of intervention inlabour due to their prematurity.

� 2012 Elsevier Ltd. All rights reserved.

1. Introduction

Preterm birth has been identified internationally as a researchpriority in reproductive medicine. It accounts for 6e10% of allbirths, and accounts for the majority of the perinatal deaths ofnormally formed babies.1 Most of those mortalities occur in babiesborn at <32 weeks of gestation, and it is this group towards whichmost of the research and clinical effort is targeted. However theoverall rate of preterm birth is steadily increasing.2

Preterm was defined by the World Health Organization in 1950as birth at <37 weeks’ gestation.2 There have been a number ofrefinements, including definitions of subgroups of preterm birth,with late preterm birth now widely accepted as birth at 34e36weeks’ gestation. Moderate preterm birth has been defined asbirth at 32e33 weeks, very preterm birth at <32 weeks andextremely preterm birth at <28 weeks’ gestation.2

Moderate and late preterm births account for the majority ofpreterm births. Late preterm births account for 60e70% of allpreterm births, and a further 20% are moderate preterm births.3

Many of these births will be ‘iatrogenic’ or clinically indicatedbirths for maternal or fetal reasons (these account for w20% of latepreterm births). Mortality and serious morbidity are consideredrare, hence increased willingness to deliver at these gestations formaternal and fetal indications. A recent study in the USA has

All rights reserved.

estimated that 1 in 15 of late preterm babies was delivered for ‘soft’or elective precursors,4 and was thus avoidable. There is ongoingdebate regarding whether reduction in late preterm births is eitherpossible or desirable.5,6

It must be remembered that infant mortality among latepreterm babies is three times higher than that of term babies (7.7per 1000 live births for late preterm births compared with 2.5 per1000 live births at term). Early and late neonatal complications areincreased, including respiratory distress, jaundice, sepsis, poorfeeding and hypoglycaemia, with higher subsequent readmissionrates and increased risk of cerebral palsy.7

This review focuses on the obstetric management of sponta-neous moderate and late preterm labour, which may be precededby preterm prelabour rupture of the membranes.

2. Aetiology of moderate and late preterm birth

Preterm labour is a multifactorial condition. For any one indi-vidual, a single cause is rarely identified, although in populationstudies and clinical investigations several associations are found.Preterm labour may be considered to represent early activation ofthe normal physiological pathway of labour, or to be the result ofa pathological insult. Its cause may be considered on two levels: thebiochemical and endocrine pathways associated with the onset ofnormal labour and those triggered by a pathological insult; and theepidemiology and clinical associations of preterm birth. Only a fewof these associations and processes can be linked to moderate andlate preterm birth specifically, compared with preterm birth per se.

P.C. McParland / Seminars in Fetal & Neonatal Medicine 17 (2012) 138e142 139

2.1. Pathophysiology of preterm labour

The pathophysiological mechanism of normal term labour is notwell understood, let alone that of preterm labour. The process ofparturition has been described as a continuum, with five recog-nised phases: implantation, uterine quiescence, activation, stimu-lation and involution.8 The activation and stimulation phases thatcomprise labour have been studied in most detail.

There are three principal theories underlying the onset oflabour, at any gestation: progesterone withdrawal, oxytocin initi-ation and decidual activation. Progesterone withdrawal is recog-nised as a mechanism in other mammals, as evidenced by a fall inthe circulating levels of progesterone. This phenomenon is not seenin the human; rather a theory has been generated of functionalprogesterone withdrawal, with modulation of progesteronereceptors in the tissues of the myometrium and fetal membranes.9

Although oxytocin is required to stimulate contractions in labour, inthe human there is no evidence of an increase in oxytocin levelsrelated to the onset of labour, although modulation of oxytocinreceptors is possible. However, the last pathway, decidual activa-tion, mediated by inflammation, infection, bleeding or uteropla-cental ischaemia appears a more important pathway.3,10

Preterm labour is the endpoint of a number of multifactorialprocesses, not a single well-defined condition. Those processes maylead to preterm labour at different gestations. Considering clinicallyrecognised pathways leading to preterm labour, it is possibleto identify infection, decidual haemorrhage, maternal/fetalhypothalamicepituitaryeadrenal activation (stress), uterine disten-sion, and cervical insufficiency. The first and last of these causes arespecifically associated with early preterm birth; moderate/latepreterm birth may be the consequence of any of the others.8

Intrauterine infection is the single most clear underlying aeti-ology that has been identified in association with the initiation ofpreterm labour, preceding 25e40% of preterm births.3 However, itis identified in themajority of births at extreme preterm gestations,and only in 10% of births at late preterm gestations.3 This suggeststhat other mechanisms, perhaps the early activation of physiolog-ical labour, predominate as the trigger for later preterm births.

2.2. Clinical associations with preterm labour

Many epidemiological and environmental factors have beenidentified associated with increased risk of preterm birth. Theseinclude socioeconomic and psychosocial factors, substance misuse,nutritional factors and infection. Many of these risk factors overlap,and may represent associations rather than direct cause.11 Theprincipal clinical risk factors for moderate and late preterm labourare identified below.

2.2.1. Multiple pregnancyA rise in multiple pregnancies has been associated with later

childbearing and assisted reproductive technology. Fifty-eightpercent of all twins are born preterm, the majority of whom areborn after 32 weeks’ gestation.12 The mechanism of preterm birthin multiple pregnancy is more commonly physiological, such asstretch-mediated, with lower incidence of infection than insingleton preterm births.13

2.2.2. Preterm prelabour rupture of the membranesPreterm prelabour rupture of the membranes (PPROM)

contributes to approximately one-third of preterm births. Aetiologyis multifactorial, and overlaps with preterm labour with intactmembranes. Up to 30% have intrauterine infection prior to ruptureof membranes,14 more so at lower gestations. Multivariate analysis

identifies cigarette smoking and vaginal bleeding as independentrisk factors for PPROM.15

2.2.3. Bacterial vaginosisBacterial vaginosis (BV) was first identified as associated with

subsequent late miscarriage and preterm birth in 1994.16 BV isa complex alteration of the vaginal flora, rather than a vaginalinfection. It is characterised by reduction in numbers of lactobacilli,increase in vaginal pH, and overgrowth of pathogens includingGardnerella and Bacteroides spp. Prevalence of BV in asymptomaticwomen is 12e25%. BV doubles the risk of preterm birth at <37weeks [odds ratio (OR): 2.19; 95% confidence interval (CI):1.54e3.12].17 However, BV diagnosed at <16 weeks’ gestation givesthe highest risk of preterm birth (OR: 7.55).18

2.2.4. LLETZ (large loop excision of the transformation zone)Excision cervical procedures for cervical intraepithelial

neoplasia increase the risk of preterm birth prior to 37 weeks.19e21

The greater the amount of tissue removed, and the more proce-dures undertaken, the greater the risk (2.8-fold increased risk ofdelivery at <37 weeks with one procedure, 9.9-fold increase withtwo procedures19). The mechanism is unclear, but may be bycompromising the cervical mucus barrier to ascending infectionfrom the vagina into the uterine cavity. At present this is an area ofongoing research, and the best management options for suchwomen in pregnancy are yet to be confirmed.

3. Prediction and prevention of preterm labour

Strategies to prevent preterm labour can be considered asstrategies targeted at the whole obstetric population, or thosetargeted at high risk women after risk assessment. They can also beconsidered by stratifying for asymptomatic or symptomaticwomen. Whole population strategies aimed at reducing the risk ofpreterm birth include smoking cessation programmes andscreening for asymptomatic bacteriuria.

Women may be identified at high risk of preterm birth based onrisk factor assessment. One of the strongest risk factors that may beidentified is having a prior preterm birth (recurrence risk w15%after one prior preterm birth, 30% after two prior preterm births).They can be further stratified, as the majority of women who havea recurrent preterm birth do so within 2 weeks’ gestation of theprevious preterm birth.22 Other clinical risk factors that may beidentified in the history include congenital uterine abnormality andprior cervical surgery.

3.1. Cervical length

Cervical length assessment for risk of preterm birth is carriedout at 20e24 weeks’ gestation by transvaginal ultrasound. It hasnot been demonstrated to be beneficial to screen the low riskobstetric population. To et al. scanned 47 123 women at 22e24weeks, and identified cervical length of <15 mm in 470 women.23

Of these women, 253 participated in their study, and were rando-mised to cerclage or expectant management. The proportion ofpreterm birth <33 weeks was similar in each group, with nodifference in perinatal morbidity or mortality.23

However, strategies targeting the high risk group of womenhave been demonstrated to be more beneficial. Meta-analysisstudying the strategy of cervical length scanning of women withprior preterm birth, and cervical cerclage if cervical length is<25 mm, demonstrates a reduction in preterm birth at <37, <35,<32, <28 and <24 weeks with cerclage compared to expectantmanagement [relative risk (RR): 0.70; 0.58e0.83 at <37 weeks].Composite perinatal mortality/morbidity was also improved with

P.C. McParland / Seminars in Fetal & Neonatal Medicine 17 (2012) 138e142140

cerclage (RR: 0.64; 95% CI: 0.45e0.91).24 Moreover the strategy ofcervical length scanning with selective cerclage gave comparableoutcomes to elective history-indicated cerclage (31% vs 32% risk ofdelivery at <37 weeks; RR: 0.97; 95% CI: 0.73e1.29).25

3.2. Cervical cerclage

Cervical cerclage has a role in prevention of preterm birth ina selected small group of women. Exact indications for its useremain controversial. It may be placed either electively based onobstetric history, ultrasound indicated after a short cervix is foundon ultrasound scan, or as a ‘rescue’ procedure when the cervix isalready dilated and membranes are visible at or beyond theexternal cervical os. Many of the studies on its efficacy are poorlydefined, and as such, the current clinical indications for the place-ment of a cervical cerclage are not clear. However, its primaryfunction is in overcoming cervical insufficiency, a condition mostlikely to lead to early preterm birth rather than moderate or latepreterm birth. It may act by providing physical support to a struc-turally weak cervix, or perhaps more likely in maintaining cervicallength and the mucus barrier to ascending infection.26

3.3. Biochemical markers

A large number of biochemical markers have been studied forthe prediction of preterm labour, the most widely used beingfibronectin. Originally identified as a putative diagnostic marker forfetal membrane rupture due to the high levels in amniotic fluid, itsrelease from the fetomaternal interface into the cervicovaginalsecretions prior to labour lead to its development as a marker forpreterm labour. The accuracy of fibronectin testing is limited by thelarge number of false-positive test results, false-negative resultsbeing uncommon. The overall sensitivity and specificity of this testfor predicting delivery at <37 weeks are 53% and 89% respectively.Among high risk women, a positive fibronectin test at 22 weeks hassensitivity of 78% and specificity of 78% for prediction of birth <37weeks.22 Perhaps greater clinical utility is found in fibronectintesting among women with symptoms of preterm labour, wherea positive fibronectin test predicts birth within 7 days with sensi-tivity and specificity of 77% and 87% respectively.27

3.4. Progesterone

Progesterone treatment is currently a major research focus inthe prevention of preterm birth. A study byMeis et al. administeredweekly injections of 17a-hydroxyprogesterone caproate to womenwith a history of prior preterm birth. A reduction in the pretermbirth rate <37 weeks (36.3% vs 54.9%; OR: 0.66; 95% CI: 0.54e0.81),<35weeks (20.6% vs 30.7%; 0.67; 0.48e0.93) and<32 weeks (11.4%vs 19.6%; 0.58; 0.37e0.91) was observed.28 Da Fonseca et al.administered 100mg progesterone as a daily vaginal suppository towomen at high risk of preterm birth, and observed a significantreduction in preterm delivery rates at <37 weeks (13.8% vs 28.5%;P ¼ 0.03) and <34 weeks (2.8% vs 18.6%; P ¼ 0.002).29 Subsequentstudies suggest that the effect of progesterone on prevention ofpreterm birth is greatest in women with an earlier prior pretermbirth (<34 weeks).30 Administration to women with a sono-graphically identified short cervix in the second trimester alsoreduces the risk of delivery at<34weeks’ gestation (19.2% vs 34.4%;RR: 0.56; CI: 0.36e0.86).31 Several large randomised controlledtrials under way internationally to confirm the fetal and neonatalsafety and benefit from such treatment (e.g. OPPTIMUM), andcurrent advice is to administer only in the context of randomisedclinical trials.

3.5. Antibiotics

The use of prophylactic antibiotics to prevent preterm birth,either in asymptomatic high risk women or in women withthreatened preterm labour, has not been shown to be beneficial. Asa substantial proportion of preterm births are infection-associated,the idea of giving antibiotics to women at high risk of preterm birth(based on prior history, prior to conception) appears attractive.However, trials have demonstrated that this strategy will does notdecrease the risk of preterm birth, conversely there may be a trendtowards an increased risk.32 Administration of metronidazole towomen with a prior preterm birth and a positive fibronectin test,but no clinical evidence of infection, also increased the risk ofpreterm birth.33

Routine prophylactic antibiotic administration in pretermlabour has not been demonstrated to be beneficial. The ORACLETrial randomised 6295 women in preterm labour with intactmembranes to (i) co-amoxiclav, (ii) erythromycin, (iii) both anti-biotics or (iv) placebo, and found no neonatal benefits.34 Of concernwas a higher incidence of cerebral palsy in children born aftertreatment with antibiotics in the 7-year follow-up after this trial.35

Antibiotics may also be used to treat bacterial vaginosis in anattempt to reduce the risk of preterm birth. There is little evidenceto routinely screen and treat the low risk obstetric population. Theprinciple role of antibacterial treatment for bacterial vaginosis is intreating women at high risk of preterm birth due to prior pretermbirth, and those identified prior to 20 weeks’ gestation.36

4. Clinical management in threatened preterm labour

4.1. Tocolysis

Tocolytic drugs are considered in the acute management whena clinical diagnosis of preterm labour is made. This diagnosis maybe aided by the use of fibronectin or insulin-like growth factorbinding protein (IGPBP-1) testing. Drugs used for tocolysis includeoxytocin antagonists (atosiban), calcium channel blockers (nifedi-pine), b-mimetics (e.g. terbutaline), and non-steroidal anti-inflammatory agents (e.g. indomethacin). Although these agentsreduce the number of births occurring within 7 days of adminis-tration, they do not reduce the overall risk of preterm birth.37

Moreover there is no clear evidence of improvement in neonataloutcomewith their use. Their use is only recommendedwhen shortdelay in delivery may gain clinical benefit (e.g. time for adminis-tration of corticosteroids or in-utero transfer). In clinical practicethis means that administration beyond 34weeks’ gestation is rarelyindicated.

4.2. Antenatal corticosteroids

Administration of corticosteroids prior to preterm birth enteredroutine obstetric practice following the systematic review ofCrowley et al. in 1990.38 The most recent Cochrane reviewdemonstrates that a single course of corticosteroids prior topreterm birth reduces the risk of neonatal death (RR: 0.69; CI:0.58e0.81), respiratory distress syndrome (0.66; 0.59e0.73),intraventricular haemorrhage (0.54; 0.43e0.69) and necrotisingenterocolitis (0.46; 0.29e0.74).39 The current Royal College ofObstetricians and Gynaecologists (RCOG) guideline recommendsadministration of a single course of either betamethasone ordexamethasone prior to preterm birth up to 34þ6 weeks’ gesta-tion.40 There is currently no evidence of benefit from the admin-istration of steroids at gestations beyond 34 weeks.39,41

Practice points

� Birth at 32e26 weeks’ gestation is associated withincreased neonatal morbidity and mortality.

� Themajority of womenwho labour spontaneously at 32e36 weeks’ gestation have no single identifiableunderlying cause.

� Prediction and prevention strategies are currently tar-geted at women at high risk for preterm birth.

� Additional care is required inmanaging labour at 32e36weeks’ gestation with modification of the interventionscarried out during labour.

Research directions

� The neonatal benefits and safety of antenatal proges-terone for prevention of preterm labour.

� The role of LLETZ in the aetiology of preterm labour.

P.C. McParland / Seminars in Fetal & Neonatal Medicine 17 (2012) 138e142 141

4.3. Timing of delivery for preterm prelabour rupture of themembranes

Management of prelabour rupture of the membranes after 37weeks’ gestation is evidence-based, and induction of labour eitherimmediately or after 24e48 h to reduce the risk of maternalinfection is established. Prior to 34 weeks’ gestation, conservativemanagement is recommended, in the absence of infection or otherfetal indication for delivery. Between 34 and 37 weeks’ gestationthe management is more controversial. RCOG advises that deliveryshould be considered at 34 weeks’ gestation.42 Not surprisinglyearly induction is associated with a higher risk of neonatal respi-ratory morbidity (oxygen requirement after 24 h, 7% vs 1.6%;P ¼ 0.05), and conservative management with a higher risk ofinfection (4.8% vs 09%; P ¼ 0.07).43 Naef et al. examined PPROM at34e37 weeks, and also found that expectant management wasassociated with a higher incidence of chorioamnionitis (16% vs 2%;P ¼ 0.007). They concluded that early delivery was considered safefor the neonate.44 A further study examining early induction oflabour versus expectant management at 32e36 weeks’ gestationwith PPROM found longer hospitalisation for the neonate withexpectant management.45 Two large ongoing randomisedcontrolled trials may clarify the evidence further.46,47

4.4. Intrapartum care

Good practice in the management of preterm labour wouldrequire the attendance of a neonatologist at any preterm birth, toassess the baby and the need for either admission to the neonatalunit, or need for further monitoring on the delivery suite or post-natal ward.

4.4.1. Intrapartum fetal monitoringContinuous fetal monitoring by cardiotocograph (CTG) would be

expected in all preterm labours where intervention for fetal indi-cations is considered appropriate. This would be expected toinclude all moderate and late preterm labours with structurallynormal babies. Further assessment of fetal wellbeing in labour mayrequire fetal blood sampling from the scalp; however, this is con-traindicated at <34 weeks’ gestation,48 and management in labourshould be based on clinical findings and interpretation of CTGalone.

4.4.2. Mode of deliveryThere is no benefit to delivery by caesarean section for the

cephalic presenting baby at late preterm gestations.49 The mode ofdelivery for the preterm breech has caused significant controversyover the years, with no consensus available from the evidence.50

The only attempted randomised controlled trial aiming to clarifythe best mode of delivery for the preterm breech baby failed torecruit more than a handful of women, with significant ethicalissues highlighted surrounding recruitment and randomisation.51

Retrospective studies of outcomes following breech presentationin preterm labour demonstrate mixed results with some showingimproved outcomes in those preterm breech babies born bycaesarean section, and others showing no benefit. In day-to-dayobstetric practice, the management of the preterm breech baby inlabour depends on the clinical skill and assessment of the obste-trician present during labour, with a vaginal breech deliveryremaining a reasonable option providing that fetal condition isgood and that a skilled obstetrician is in attendance for the birth.The risk of entrapment of the aftercoming fetal head is minimisedby maintaining intact fetal membranes for as long as possible,consideration of epidural analgesia, and delaying pushing until fullcervical dilatation is confirmed by an experienced practitioner.

Where assisted vaginal delivery is required for the cephalicpresenting baby, the instrument of choice will depend on gesta-tional age as well as the clinical skill and preference of the obste-trician. Vacuum delivery is considered to be contraindicated at<34weeks’ gestation due to risk of cerebral bleeding, and should beusedwith caution at<36weeks’ gestation.52 Forceps may thereforebe most appropriate when assisted delivery is indicated.

Conflict of interest statement

None declared.

Funding sources

None.

References

1. Green NS, Damus K, Simpson JL, et al. Research agenda for preterm birth:recommendations from the March of Dimes. Am J Obstet Gynecol2005;193:626e35.

2. Raju TNK. Epidemiology of late preterm (near-term) births. Clin Perinatol2006;33:751e63.

3. Goldenberg RL, Culhane JF, Iams JD, Romero R. Epidemiology and causes ofpreterm birth. Lancet 2008;371:75e84.

4. Laughon SK, Reddy UM, Sun L, Zhang J. Precursors for late preterm birth insingleton gestations. Obstet Gynecol 2010;116:1047e55.

5. Bannerman CG. Late preterm birth: can be reduced. Am J Obstet Gynecol2011;204:459e60.

6. Chauhan SP. Late preterm births: irreducible because E ¼ mc2. Am J ObstetGynecol 2011;204:459e60.

7. Engle WA, Tomashek KM, Wallman C. “Late preterm” infants: a population atrisk. Pediatrics 2007;120:1390e401.

8. Simmons LE, Rubens CE, Darmstadt GL, Gravett MG. Preventing preterm birthand neonatal mortality: exploring the epidemiology, causes, and interventions.Semin Perinatol 2010;34:408e15.

9. Zakar T, Hertelendy F. Progesterone withdrawal: key to parturition. Am J ObstetGynecol 2007;196:289e96.

10. Romero R, Espinoza J, Kusanovic JP, et al. The preterm parturition syndrome.BJOG 2006;113(Suppl. 3):17e42.

11. Murphy DJ. Epidemiology and environmental factors in preterm labour. BestPract Res Clin Obstet Gynaecol 2007;21:773e89.

12. Chauhan SP, Scardo JA, Hayes E, Abuhamad AZ, Berghella V. Twins: prevalence,problems and preterm births. Am J Obstet Gynecol 2010;203:305e15.

13. Stock S, Norman J. Preterm and term labour in multiple pregnancies. SeminFetal Neonatal Med 2010;15:336e41.

14. Romero R, Quintero R, Oyarzun E, et al. Intraamniotic infection and the onset oflabour in preterm premature rupture of the membranes. Am J Obstet Gynecol1988;159:661e6.

P.C. McParland / Seminars in Fetal & Neonatal Medicine 17 (2012) 138e142142

15. Harger JH, Hsing AW, Tuomala RE, et al. Risk factors for preterm prematurerupture of fetal membranes: a multicentre caseecontrol study. Am J ObstetGynecol 1990;163:130e7.

16. Hay PE, Lamont RF, Taylor-Robinson D, et al. Abnormal bacterial colonisation ofthe genital tract and subsequent preterm delivery and late miscarriage. BMJ1994;308:295e8.

17. Leitich H, Bodner-Adler B, Brunbauer M, et al. Bacterial vaginosis as a risk factorfor preterm delivery: a meta-analysis. Am J Obstet Gynecol 2003;189:139e47.

18. Guaschino S, De Seta F, Piccoli M, Maso G, Alberico S. Aetiology of pretermlabour: bacterial vaginosis. BJOG 2006;113(Suppl. 3):46e51.

19. Ortoft G, Henriksen TB, Hansen ES, Petersen LK. After conisation of the cervix,the perinatal mortality as a result of preterm delivery increases in a subsequentpregnancy. BJOG 2010;117:258e367.

20. Quenby S. Obstetric management of women after treatment for CIN. BJOG2010;117:243e4.

21. Van der Vijver A, Poppe W, Verguts J, Arbyn M. Pregnancy outcome aftercervical conisation: a retrospective cohort study in the Leuven UniversityHospital. BJOG 2010;117:268e73.

22. Spong CY. Prediction and prevention of recurrent spontaneous preterm birth.Obstet Gynecol 2007;110:405e15.

23. To MS, Alfirevic Z, Heath VC, et al. Cervical cerclage for prevention of pretermdelivery in women with short cervix: randomised controlled trial. Lancet2004;363:1849e53.

24. Berghella V, Rafael TJ, Szychowski JM, Rust AO, Owen J. Cerclage for shortcervix on ultrasonography in women with singleton gestations and previouspreterm birth. Obstet Gynecol 2011;117:663e71.

25. Berghella V, Mackeen AD. Cervical length screening with ultrasound-indicatedcerclage compared with history-indicated cerclage for prevention of pretermbirth. Obstet Gynecol 2011;118:148e55.

26. Royal College of Obstetricians and Gynaecologists. Cervical cerclage. Green-topguideline 60. London: RCOG; 2011.

27. Norman JE. Cervical function and prematurity. Best Pract Res Clin ObstetGynaecol 2007;21:791e806.

28. Meis PJ, Klebanoff M, Thom E, et al. Prevention of recurrent preterm delivery by17 alpha-hydroxyprogesterone caproate. N Engl J Med 2003;348:2379e85.

29. Da Fonseca EB, Bittar RE, Carvalho MH, et al. Prophylactic administration ofprogesterone by vaginal suppository to reduce the incidence of spontaneouspreterm birth in women at increased risk: a randomised placebo-controlleddouble-blind study. Am J Obstet Gynecol 2003;188:419e24.

30. Spong CY, Meis PJ, Thom EA, et al. Progesterone for prevention of recurrentpreterm birth: impact of gestational age at previous delivery. Am J ObstetGynecol 2005;193:1127e31.

31. Fonseca EB, Celik E, Parra M, Singh M, Nicolaides KH. Progesterone and pretermbirth among women with a short cervix. N Engl J Med 2008;357:462e9.

32. Andrews WW, Goldenberg RL, Hauth JC, et al. Interconceptional antibiotics toprevent spontaneous preterm birth: a randomized clinical trial. Am J ObstetGynecol 2006;194:617e23.

33. Shennan A, Crawshaw S, Briley A, et al. A randomised controlled trial ofmetronidazole for the prevention of preterm birth in women positive forcervicovaginal fetal fibronectin: the PREMET study. BJOG 2006;113:65e74.

34. KenyonSL, TaylorDJ, Tarnow-MordiW.Broad-spectrumantibiotics for spontaneouspreterm labour: the ORACLE II randomised trial. Lancet 2001;357:989e94.

35. Kenyon S, Pike K, Jones DR, et al. Childhood outcomes after prescription ofantibiotics to pregnant women with spontaneous preterm labour: 7 yearfollow up of the ORACLE II trial. Lancet 2008;372:1319e27.

36. McDonald HM, Brocklehurst P, Gordon A. Antibiotics for treating bacterialvaginosis in pregnancy. Cochrane Database Syst Rev 2007;1:CD000262.

37. Royal College of Obstetricians and Gynaecologists. Tocolysis for women inpreterm labour. Green-top Guideline 1b. London: RCOG; 2011.

38. Crowley P, Chalmers I, Keirse MJNC. The effects of corticosteroid administrationbefore preterm delivery: an overview of the evidence from controlled trials. BrJ Obstet Gynaecol 1990;97:11e25.

39. Roberts D, Dalziel SR. Antenatal corticosteroids for accelerating fetal lungmaturation for women at risk of preterm birth. Cochrane Database Syst Rev2006;3:CD004454.

40. Royal College of Obstetricians and Gynaecologists. Antenatal corticoste-roids to reduce neonatal morbidity. Green-top Guideline 7. London: RCOG;2010.

41. Porto AMF, Coutinho IC, Correia JB, Amorim MMR. Effectiveness of antenatalcorticosteroids in reducing respiratory disorders in late preterm infants:randomised clinical trial. BMJ 2011;342:d1696.

42. Royal College of Obstetricians and Gynaecologists. Preterm prelabour rupture ofmembranes. Green-top Guideline 44. London: RCOG; 2010.

43. Kayem G, Bernier-Dupreelle A, Goffinet F, Cabrol D, Haddad B. Active versusexpectant management for preterm prelabour rupture of membranes at 34e36weeks completed gestation: comparison of maternal and neonatal outcomes.Acta Obstet Gynecol Scand 2010;89:776e81.

44. Naef RW, Allbert JR, Ross EL, et al. Premature rupture of membranes at 34 to 37weeks gestation: aggressive versus conservative management. Am J ObstetGynecol 1998;178:126e30.

45. Mercer BM, Crocker LG, Boe NM, Sibai BM. Induction versus expectantmanagement in premature rupture of the membranes with mature amnioticfluid at 32 to 36 weeks: a randomised trial. Am J Obstet Gynecol1993;169:775e82.

46. Morris JM, Roberts CL, Crowther CA, et al. Protocol for the immediate deliveryversus expectant care of women with preterm prelabour rupture of themembranes close to term (PPROMT) Trial [ISRCTN44485060]. BMC PregnChildbirth 2006;6:9.

47. Van der Ham DP, Nijhuis JG, Mol BW, et al. Induction of labour versus expec-tant management in women with preterm prelabour rupture of membranesbetween 34 and 37 weeks (the PPROMEXIL-trial). BMC Pregn Childbirth2007;7:11.

48. National Institute for Health and Clinical Excellence. Intrapartum care. Care ofhealthy women and their babies during childbirth. NICE clinical guideline 55.London: NICE; 2007.

49. Dobak WJ, Gardner MO. Late preterm gestation: physiology of labor andimplications for delivery. Clin Perinatol 2006;33:765e76.

50. Royal College of Obstetricians and Gynaecologists. The management of breechpresentation. Green-Top guideline 20b. London: RCOG; 2006.

51. Penn ZJ, Steer PJ, Grant A. A multicentre randomised controlled trial comparingelective and selective caesarean section for the delivery of the preterm breechinfant. Br J Obstet Gynaecol 1996;103:684e9.

52. Royal College of Obstetricians and Gynaecologists. Operative vaginal delivery.Green-Top guideline 26. London: RCOG; 2011.

at SciVerse ScienceDirect

Seminars in Fetal & Neonatal Medicine 17 (2012) 143e145

Contents lists available

Seminars in Fetal & Neonatal Medicine

journal homepage: www.elsevier .com/locate/s iny

Impact of multiple births on late and moderate prematurity

Jerrie S. Refuerzo*

Division of MaternaleFetal Medicine, Department of Obstetrics and Gynecology, University of Texas Health Science Center at Houston, 6431 Fannin,Suite 3.270, Houston, TX 77030, USA

Keywords:Late preterm birthModerately preterm birthMultiple gestationsTripletsTwins

* Tel.: þ1 713 500 6416; fax: þ1 713 500 7860.E-mail address: Jerrie.S.Refuerzo@uth.tmc.edu.

1744-165X/$ e see front matter � 2012 Elsevier Ltd.doi:10.1016/j.siny.2012.01.012

s u m m a r y

Multiple gestations have an increased risk of pregnancy complications compared with singletons. Delay inchildbearing and assisted reproductive techniques have remained common reasons for the increase inmultiple gestations over the last few decades. Higher rates of both spontaneous and indicated pretermbirthin twins and triplets lead to a significant proportion of the moderate preterm birth and late preterm birthrates. Thearticle is a reviewof thecausesofpretermbirth andmorbidities associatedwith thesepregnancies.

� 2012 Elsevier Ltd. All rights reserved.

7

1. Introduction

Multiple gestations such as twins and triplet pregnancies area unique pregnancy population with inherent risks for moderatepreterm birth and late preterm birth. For decades, twins repre-sented one of the most rapidly growing conditions in pregnancy,increasing by 70% from 1980 to 2004.1 Since then, this twin birthrate seems to have remained constant at 32.2 per 1000 birthsaccording to the vital statistics in 2007.2 The triplet, quadruplet andother higher order multiples (HOM) rate has had a similar trend.From 1980 to 1990, there was a large increase in triplets and HOM.Since 1998, there has been a gradual decrease in this high riskobstetric group. Because of the perinatal risks associated with twinand triplet gestations, multiple gestations contribute a considerableamount to the moderate and late preterm birth rates.

The majority of twins deliver within the moderate and latepreterm birth period.3,4 Twins represent 3% of all births in the USA.5

However, twins and triplets account for w26% of all births in thelate preterm birth period.6 Among twins, the moderate pretermbirth rate is 14.5%, late preterm 49.8% and term birth rate 35.7%(Table 1).4 The mean gestational age at delivery for singletons is38.8 weeks. By contrast, twins have a mean age of delivery of 35.3weeks, triplets 32.2 weeks and quadruplets 29.9 weeks.3 Thus,multiple gestations represent a considerable proportion of preg-nancies delivering moderate or late preterm.

2. Risk factors for moderate and late preterm birth

Multiple gestations have inherent factors that place them athigher risk for a preterm birth compared with singletons. Amongboth twins and triplets, 30% of preterm births are due to

All rights reserved.

spontaneous preterm labour. Although twins and triplet preg-nancies have an increased rate of spontaneous preterm birth, theyalso have a 50% rate of indicated preterm birth due to fetal ormaternal pregnancy complications. The rate of maternal hyper-tensive conditions in twins ranges from 12% to 22% vs from 5% to 8%in singletons.8 The rate of pre-eclampsia is two to three timeshigher in twins in the late preterm period compared with single-tons.3,8 Low birth weight also occurs more frequently in twins andtriplets. Twenty-four percent of infants with low birth weight areborn tomultiple gestations, particularly in the late preterm period.8

Preterm premature rupture of membranes (PPROM) occurs inw10% of deliveries.7 The average gestational age at PPROM is w31weeks and only 22.4% achieve a latency period >7 days.9 Of thosethat have a prolonged latency period, iatrogenic delivery typicallyoccurs at 34 weeks at the start of the late preterm birth period.

3. Morbidities of twins with moderate and late preterm birth

Over the last few decades, the rate of preterm birth among twinpregnancies has risen from 48% to 60%, with the largest groupcomprised of late preterm births between 34 and 36 weeks ofpregnancy.1 The overall rate of preterm birth is w69.9%.9 Themoderate preterm birth rate was 14.5%, late preterm 49.8% andterm birth rate 35.7%.4 In singleton pregnancies, both neonatalmortality and morbidity including respiratory distress syndrome(RDS), sepsis, intraventricular hemorrhage (IVH), phototherapy andintubation in the delivery room in those born late preterm areincreased compared with those born after 37 weeks.10,11 Similarfindings also occur in twin gestations.

Nonetheless, the overall outcomes for twins born within the latepretermperiod are favorable. A secondary analysiswas conducted byRefuerzo et al.4 of a multicenter, randomized controlled trial ofmultiple gestations. The primary outcome was a neonatal outcome

Table 1Rate of moderate preterm birth (MPTB) and late preterm birth (LPTB) in twins andtriplet pregnancies.

Neonatal outcome MPTB LPTB Term

Twins4 14.5% 49.8% 35.7%Triplets18 35.5% 43.6% 20.9%

J.S. Refuerzo / Seminars in Fetal & Neonatal Medicine 17 (2012) 143e145144

composite consistingof oneormore of the following: neonatal death,RDS, sepsis, stage 2 or 3 necrotizing enterocolitis (NEC), broncho-pulmonary dysplasia, grade 3 or 4 IVH, periventricular leukomalacia,pneumonia or severe retinopathy of prematurity (ROP). Becauserespiratory complications represent a large proportion of morbidityin the moderate and late preterm birth period, they examineda respiratory composite consisting of one or more of the followingoutcomes per pregnancy: RDS, transient tachypnea of the newborn(TTN), need for mechanical ventilation (MV) or need for oxygen. Therate of the primary outcomewas progressively higher in those twinsborn moderate and late preterm compared with term, (moderate30.0% vs late 12.8% vs term 0.5%; P < 0.001). Compared with termneonates, the primary neonatal outcome composite was increasedfollowingmoderate [relative risk (RR): 58.5; 95% confidence interval(CI): 11.3e1693.0] and late (RR: 24.9; 95% CI: 4.8e732.2) pretermbirth. The rate of the secondary respiratory composite was alsoprogressively higher in moderate and late preterm twins comparedwith term (moderate 67.5% vs late 33.8% vs term8.1%; P< 0.001). Therespiratory outcome composite was increased following moderate(RR: 8.3; 95% CI: 5.1e13.6) and late (RR: 4.2; 95% CI: 2.5e6.9) pretermbirth compared with term neonates. Most of the differences inneonatal outcomes were respiratory morbidities. RDS contributedthe greatest effect to the primary outcome. Both moderate and latepreterm twins had significantly higher rates of RDS, neonatal sepsis,TTN, MV, and neonatal intensive care unit (NICU) admissionscompared with term. Additionally, the lengths of NICU stay, MV andsupplemental oxygen were significantly longer in those twins withmoderate and late preterm birth compared with term.

Other investigators have found similar findings in morbidity andmortality. In comparison with singletons, risks of neonatal morbidityandmortalityarehigher.3,4,6 The rateofneonatalmortality is increasedfive to seven times compared with singletons.5 Major morbidityincluding grade 3 or 4 IVH, NEC and ROP do not appear increased inmoderate or late preterm twins with compared with singletons.4,12 Ingeneral, the riskof cerebralpalsy is lowinthe latepretermbirthperiod.However, twinpregnancieshavea fourfoldhigher riskof cerebralpalsyin the late preterm birth period compared with singletons.13

4. Chorionicity

Although it is known that twin gestations have a higher rate ofpregnancy complications and neonatal morbidity compared withsingletons, these risks are further dichotomized based on chorio-nicity.14 Investigators have reviewed risks of stillbirth of twins withexpectant management compared with elective delivery accordingto chorionicity in the late preterm and early term period. Multiplestudies have shown that the rate of stillbirth increases after 37e39weeks in dichorionic twins.14e17 The fetal mortality rate for twinswas lowest at 36e37 weeks’ gestation.17 Thus, Newman et al.13

concluded that due to the increased mortality risk after 38e39weeks, elective delivery is reasonable at 38 weeks in a well-dated,uncomplicated dichorionic twin pregnancy.

Monochorionic twins have different risks compared withdichorionic twins. These types of twins have a higher rate of stillbirth(monochorionic 3.6% vs dichorionic 1.1%).14 Thismay be due to acuteepisodes of twin-to-twin transfusion syndrome (TTTS) or failure ofclose fetal surveillance.18 Thus, Newman et al.13 recommend offering

elective delivery at 34e37 weeks in uncomplicated monochorionictwins. Monoamniotic twins occur in only 1% of all monozygotic twinpregnancies but carry the highest risk of stillbirth.14 The etiology forstillbirth includes fetal growth restriction, TTTS and fetal anomalies,but the largest proportion of perinatal mortality in monoamniotictwins is due to umbilical cord entanglement resulting in cord acci-dent. Despite variations in management of these high risk cases, themean gestational age at delivery occurs in the moderate pretermbirth period at 32.9 weeks. The current recommended timing ofdelivery for this group is between 32 and 34 weeks.

5. Morbidity of triplets

Triplets have higher risks than twins and therefore have anearlier gestational age at delivery and morbidity. In a retrospectivestudy of 55 triplets conducted by Kaufman et al., the rate ofmoderate preterm birth was 35.5% and the rate of late preterm birthwas 43.6%.18 Those born moderate preterm had higher morbiditythan those born in the late preterm period. In those born moderatepreterm, survival was 100%, NICU admission 23.6%, need for MV33.1%, RDS 16.6% and there were no cases of chronic lung disease. Inlate preterm infants, survival was 100%, NICU admission 28.5%, needfor MV 6.9% and there were no cases of RDS or chronic lung disease.

A randomized controlled trial conducted between 2004 and2006 of 134 triplets receiving 17a-hydyroxyprogesterone caproatecompared with placebo showed similar rates of preterm birth.19

The mean gestational age at delivery occurred in the moderatepreterm period at 33.0 weeks (range: 31.6e34.3). The rate ofdelivery at <35 weeks due to spontaneous labor was 43% and was41% in indicated births.

Garite et al. conducted a prospective study of 2155 tripletsbetween 1997 and 2002.12 Themean gestational age at deliverywas31 � 3 weeks. The cesarean section rate was 95%. Concordantgrowth was demonstrated in all triplets up until 29e30 weeks, atwhich time birth weight significantly diverged primarily due tofetal growth restriction.

Investigators inNorway found similarfindings.20 In a population-based cohort study of 1007 triplets between 1988 and 2006, thetriplet ratewas 2.7 per 10,000 pregnancies. Themeangestational ageat delivery occurred in the moderate preterm birth period at32.1�3.3weeks. The cesarean section ratewas95%and theperinataldeath rate was 7.1%. The average birth weight was 1736 � 544 g.Thus, monitoring closely is important in triplet pregnancies.

6. Morbidities associated with assisted reproductivetechniques

Over the past few decades, the option to delay childbearing hascontributed to an increase in the rate of pregnancy in women aged�40 years.2 An increased rate in assisted reproductive techniques(ART) has paralleled this trend. The birth rate for women agedbetween 40 and 45 years was 9.9 births per 100 women in 2008, anall-time high over the past four decades. The birth rate for womenbetween 45 and 49 years was 0.7 births per 1000 women. This ratehas more than tripled since 1990.

Because of reduced fertility and potential to conceive spontane-ously in women aged �40 years, the rate of ART is higher in thisgroup of women including in-vitro fertilization (IVF). A known riskfactor to such ART is multiple gestation. Studies have comparedoutcomes of both twins and triplets conceived spontaneously tothose conceivedwith ART. Fitzsimmons et al.21 examined both twinsand triplets between 1985 and 1996. They compared 56 IVF twinswith 108 non-IVF twins and 16 IVF triplets with 16 non-IVF triplets.The gestational age at delivery was similar in the IVF pregnanciescompared with non-IVF pregnancies. The mean gestational age at

Research directions

� Assess inherent risk factors and clinical practices thatlead to indicated preterm birth in twins and triplets.

� Assessing interventions in the late preterm period suchas antenatal corticosteroids that may impact neonatalrespiratory outcomes.

J.S. Refuerzo / Seminars in Fetal & Neonatal Medicine 17 (2012) 143e145 145

delivery for the IVF twin was 35.4 � 3.3 weeks compared with34.6 � 4.5 weeks non-IVF twins. The mean gestational age was32.4 � 3.3 weeks in IVF triplets compared with 32.6 � 3.8 weeks inthenon-IVF triplets. Bycontrast, therewere significantly higher ratesof perinatal mortality in non-IVF twins (24%) compared with IVFtwins (2%). The triplet mortality rate was similar between groups(w3%). The rate of monochorionic, diamniotic twins was greater inthe non-IVF group (34 sets IVF vs 2 sets non-IVF). Monochorionictriplets occurred in one set of non-IVF gestation. There were nosignificant differences between birth weight, rate of spontaneouspreterm labor, PPROM, pre-eclampsia or gestational diabetesbetween groups. There were also no differences in neonatalmorbidity including NICU admission, days of MV, IVH, sepsis or NEC.

Nassar et al.22 conducted a case-controlled study of 56 IVF twinsvs 112 non-IVF twins between 1996 and 2000. Compared with Fitz-simmons et al., they found a higher rate of preterm birth among theIVF twins (51.5%) vs non-IVF twins (22.3%). There was a significantincrease in the rate of delivery at <37 weeks’ gestation (IVF twins76.3% vs non-IVF twins 46.1%). There was also an increase in the rateof moderate and late preterm birth in IVF twins (IVF twins 57.1% vsnon-IVF twins 35.7%). The gestational agewas lower in IVF twins (IVF35� 3weeks vs non-IVF 36� 3weeks). The rate of both spontaneousand medically indicated inductions was similar (spontaneous: IVF69.6% vs non-IVF 67.9%) and (induction: IVF 5.4%% vs non-IVF 6.3%).

In Norway, 365 triplets conceived by IVF had similar outcomesto 1007 non-IVF triplets.20 Themean gestational age at deliverywas31.8 � 3.3 weeks. The cesarean section rate was 93.7% and themortality rate was 8.5%. The mean birth weight was 1686 � 542 g.

7. Conclusion

Multiple gestations comprise w3% of all births. This rateincreased steadily from the 1990s up until 2006, sincewhen the ratehas remained constant. Because of the increase in both spontaneousand indicated preterm birth, the mean gestational age at delivery oftwins occurs in the late preterm period and triplets in the moderatepreterm birth period. Outcomes for late preterm twins are generallyfavorable. The majority of morbidities are respiratory and typicallytemporary. Chorionicity of multiple gestations plays an importantrole in morbidity. Morbidity for triplets is generally higher andprimarily related to earlier gestational age at delivery. Delay inchildbearing age has increased the rate of ART including IVF. Thishas contributed greatly to the rate of multiple gestations. Althoughthere are some conflicting reports on outcomes, women whoconceive with IVF have overall favorable outcomes due to deliveryrates within the moderate and late preterm birth periods. Insummary, multifetal pregnancies carry inherent risk factors leadingto delivery in the late and moderate preterm birth periods.

Practice points

� Inherent risk factors for multiple gestations increase therate of both spontaneous and indicated preterm birthcompared with singletons.

� Twins’ mean gestational age is within the late pretermbirth period and triplets within the moderate pretermbirth period.

� Although the neonatal morbidity and mortality of twinsand triplets is higher compared with singleton preg-nancies, the outcomes are overall very favourable.

� Assisted reproductive techniques such as IVF havecontributed to the increase inmultiple gestations. Thereremain conflicting results on neonatal outcomes in IVFpregnancies compared with non-IVF pregnancies.

Conflict of interest statement

None declared.

Funding sources

None.

References

1. Martin JA, Kung HC, Mathews TJ, Hoyert DL, Strobino DM, Guyer B, Sutton SR.Annual summary of vital statistics. Pediatrics 2006;2008(121):788e801.

2. Mathews TJ, Minino AM, Osterman MJ, Strobino DM, Guyer B. Annual summaryof vital statistics. Pediatrics 2008;2011(127):146e57.

3. Lee YM, Cleary-Goldman J, D’Alton ME. Multiple gestations and late preterm(near-term) deliveries. Semin Perinatol 2006;30:103e12.

4. Refuerzo JS, Momirova V, Peaceman AM, et al. Neonatal outcomes in twinpregnancies delivered moderately preterm, late preterm and term. Am JPerinatol 2010;27:537e42.

5. American College of Obstetricians and Gynecologists. Multiple gestation;complicated twin, triplet and higher order multifetal pregnancy. Practice Bulletin56. Washington, DC: ACOG; 2004.

6. Vachharajani AJ, Vachharajani NA, Dawson JG. Comparison of short-termoutcomes of late preterm singletons and multiple births: an institutionalexperience. Clin Pediatr 2009;49:922e5.

7. Chauhan SP, Scardo JA, Hayes E, Abuhamad AZ, Berghella V. Twins: prevalence,problems and preterm birth. Am J Obstet Gynecol 2010;203:305e15.

8. Lee YM, Cleary-Goldman J, D’Alton ME. The impact of multiple gestations onlate preter (near-term) births. Clin Perinatol 2006;33:777e92.

9. Trentacost SV, Jean-Pierre C, Baergen R, Chasen ST. Outcomes of pretermpremature rupture of membranes in twin pregnancies. J Matern Fetal NeonatalMed 2008;21:555e7.

10. RouseDJ, Caritis SN, PeacemanAM, et al. A trial of 17 alpha-hydroxyprogesteronecaproate to prevent prematurity in twins. N Engl J Med 2007;357:454e61.

11. McIntire DD, Leveno KJ. Neonatal mortality and morbidity rates in late pretermbirths compared with births at term. Obstet Gynecol 2008;111:35e41.

12. Newman RB, Unal ER. Multiple gestations: timing of indicated late preterm andearly-term births in uncomplicated dichorionic, monochorionic and mono-amniotic twins. Semin Perinatol 2011;35:277e85.

13. Garite TJ, Clark RH, Elliot JP, Thorp JA. Twins and triplets: the effect of pluralityand growth on neonatal outcome compared with singleton infants. Am J ObstetGynecol 2004;191:700e7.

14. Kahn B, Lumey LH, Zybert PA, Lorenz JM, Cleary-Goldman J, D’Alton ME,Robinson JN. Prospective risk of fetal death in singleton, twin and tripletgestations: implications for practice. Obstet Gynecol 2003;102:685e92.

15. Sairam S, Costeloe K, Thilaganathan B. Prospective risk of stillbirth in multiplegestation pregnancies: a population-based analysis. Obstet Gynecol 2002;100:638e41.

16. Luke B, Keith LG. The contribution of singletons, twins and triplets to low birthweight, infant mortality and handicap in the United States. J Reprod Med1992;37:661e6.

17. Lee YM, Wylie BJ, Simpson LL, D’Alton ME. Twin chorionicity and risk of still-birth. Obstet Gynecol 2008;111:301e8.

18. Kaufman GE, Malone FD, Harvey-Wilkes KB, Chelmow D, Penzias AS,D’Alton ME. Neonatal morbidity and mortality associated with triplet preg-nancy. Obstet Gynecol 1998;91:342e8.

19. Caritis SN, Rouse DJ, Peaceman AM, et al. Prevention of preterm birth in tripletsusing 17 alpha-hydroxyprogesterone caproate. Obstet Gynecol 2009;113:285e92.

20. Tandberg A, Bjorge T, Nygard O, Bordahl PE, Skjaerven R. Trends in incidenceand mortality for triplets in Norway 1967e2006: the influence of assistedreproductive technologies. BJOG 2010;117:667e75.

21. Fitzsimmons BP, Bebbington MW, Fluker MR. Perinatal and neonatal outcomesin multiple gestations: assisted reproduction versus spontaneous conception.Am J Obstet Gynecol 1998;179:1162e7.

22. Nassar AH, Usta IM, Rechdan JB, Harb TS, Adra AM, Abu-Musa AA. Pregnancyoutcome in spontaneous twins versus twins who were conceived throughin vitro fertilization. Am J Obstet Gynecol 2003;189:513e8.

at SciVerse ScienceDirect

Seminars in Fetal & Neonatal Medicine 17 (2012) 146e152

Contents lists available

Seminars in Fetal & Neonatal Medicine

journal homepage: www.elsevier .com/locate/s iny

Neonatal problems of late and moderate preterm infants

J.-B. Gouyon a,b,c,d,*, S. Iacobelli a,c, C. Ferdynus a,b, F. Bonsante a,c

aCentre d’Etudes Périnatales de l’Océan Indien, Centre d’Investigation Clinique et d’Epidémiologie Clinique (CIC-EC), La Réunion, FrancebCentre d’Epidémiologie des Populations (EA 4184), Université de Dijon, Dijon, Francec Service de Néonatologie et Réanimation Néonatale, GHSR, La Réunion, Franced Service de Néonatologie et Réanimation Néonatale, CHU de Dijon, France

Keywords:HypoglycaemiaHypothermiaJaundiceMortalityPretermRespiratory disease

Practice points

� Four-fifths of premature babiesweeks’ gestation.

* Corresponding author. Service de NéonatologieGHSR, BP350, 97448 Saint-Pierre Cedex, France. Tel.:

E-mail address: jean-bernard.gouyon@chr-reunion

1744-165X/$ e see front matter � 2012 Elsevier Ltd.doi:10.1016/j.siny.2012.01.015

s u m m a r y

Late and moderate preterm infants account for >80% of premature births. These newborns experienceconsiderable mortality and morbidity in comparison with full-term born infants. The purpose of thispaper is to summarise the most common morbidities of late and moderate preterm infants in theneonatal period, their incidence, severity, risk factors and need for admission to the different levels ofcare. The recent findings on preventive strategies and management priorities for clinical care of thesevulnerable babies are also reviewed.

� 2012 Elsevier Ltd. All rights reserved.

1. Introduction

In the previous decade, research on mortality and morbidity inpreterm births focused mainly on the highest risk births, thoseoccurring at<32 weeks’ gestation. However, the category of infantsborn at 32e36 weeks represents >80% of preterm births andincludes moderate preterm (320/7 and 336/7) and late preterminfants (340/7 to 366/7 weeks). Late preterm infants, who are oftenviewed as normal newborns by parents and care providers, expe-rience morbidity during the birth hospitalisation 3.5 times morefrequently than term infants and their neonatal mortality is 4.6times higher.1 They represent one-third of the neonatal intensivecare unit (NICU) admissions in the USA.1 Moderate preterm infantshave an intermediate rate of morbidity and mortality between latepreterm and very preterm infants but they receive little attention.

This review aims to describe neonatal problems in infants bornbetween 32 and 36 weeks’ gestation.

are born at 32e36

et Réanimation Néonatale,þ33 6 81 48 32 70..fr (J.-B. Gouyon).

All rights reserved.

2. Rates of neonatal morbidities

The relationship between gestational age and neonatalmorbidity is continuous between 32 and 36 weeks’ gestation1e3

without any threshold of gestational age.2 For instance, the rateof severe respiratory disorders (treated by mechanical ventilationand/or nasal continuous positive airway pressure, nCPAP)continuously declined with gestation from 19.8% at 34 weeks to0.28% at 39e41 weeks in a French epidemiological study ofsingleton live-born babies.3 Similarly, the risk of hypoglycaemia,hyperbilirubinaemia and poor prognosis as well as the length ofthe hospital stay has been inversely correlated with gestationalage.3e5

Clinicians should also be aware that the outcome of late preterminfants also depends on maternal complications during pregnancyand on obstetrical practice.1,3 Late preterm birth and, to a lesserextent, maternal conditions are each independent risk factors ofsevere newborn morbidity (particularly severe respiratory disor-ders) especially when late preterm infants have been exposed toantepartum haemorrhage, hypertensive disorders of pregnancy,diabetes, elective caesarean section (CS) and intrauterine growthrestriction (IUGR).2e4,6

Finally, the adaptation of professional practices to the risk levelrecorded in epidemiological studies is of prime importance. Forinstance, Reddy et al.7 have recently shown that the highestmortality rate in late preterm infants was observed when no indi-cation for delivery was recorded in the medical files of the USABirth Cohort, thus suggesting that the absence of medical indica-tions for delivery at 34e36 weeks was associated with an excessrisk of mortality.

Practice points

� A continuous relationship exists between gestationalage and neonatal morbidity/mortality between 32 and36 weeks of gestation.1

Practice points

� Each week gained until 39 weeks’ gestation reduces therisk of respiratory morbidity and improves the prog-nosis of the newborn.

J.-B. Gouyon et al. / Seminars in Fetal & Neonatal Medicine 17 (2012) 146e152 147

3. Hospitalisation and organisation of care

The incidence of hospitalisation of moderate and late preterminfants depends on gestational age, comorbidities and organisationof care. The latter probably explains the variation in rates of hos-pitalisation of late preterm infants in NICUs: 62% in a French study1;100% for all babies born at <35 weeks in hospitals of the KaiserPermanente Medical Care Program8; 36.5% in 19 other US hospi-tals.9 In all studies the rate of hospitalisation in NICU declinedwhengestational age increased. This point was well illustrated when thepopulation of the Burgundy study3 was extended by addingneonates of 32 and 33 weeks’ gestation and by one further year ofdata collection as shown in Fig. 1 (J-B. Gouyon et al., unpublisheddata).

The pragmatic choice of the appropriate level of care formoderate and late preterm infants closely relies on the universalmedical risks related to each gestational age and also to the tech-nical limitations of the local neonatal care units (well baby nursery,intermediate care, special care, intensive care). These infantsshould be managed according to precise medical guidelines1

including the admission criteria in the different neonatal careunits of the birth place or in another hospital as necessary. Theguidelines have also to consider the increased need of resuscitationat birth in moderate and late preterm infants. A comprehensiveunderstanding of the neonatal issues of infants born between 32and 36 weeks’ gestation by health care providers and administra-tors of hospitals and insurance companies is essential to determinethe resources needed to take care of those infants.

0%

10%

Weeks of gestion

20%

30%

40%

50%

60%

70%

32 33 34 35 36 37 38

39-41

Antenatal steroids

Nasal CPAP

Mechanical ventilation

Surfactant

Admission in NICU

Fig. 1. Gestational age and rates of respiratory treatments and admission in neonatalintensive care unit (NICU) in a population of 173,058 live-born infants (years2000e2009). CPAP, continuous positive airway pressure. Adapted and extended fromGouyon et al.3

Practice points

� Guidelines for care providers should include admissioncriteria in neonatal units.

� Give information about characteristics of MPI and LPI toadministrators.

4. Respiratory diseases

4.1. Rates of respiratory diseases

Twenty-four studies published between 2000 and 2009consistently revealed that infants born at 32e36 weeks experiencerespiratory distress syndrome (RDS), transient tachypnoea of thenewborn (TTN), pneumonia and pulmonary hypertension of thenewborn (PPHN) at higher rates than term infants.10 Subsequentstudies delivered similar conclusions. The rate of severe respiratorydisorders e treated by mechanical ventilation and/or nasal CPAP(nCPAP) e in a population-based study of 150,426 live-bornsingleton neonates was 8.31% in late preterm, 0.84% in early term(37e38 weeks) and 0.28% in late term infants (39e41 weeks).3 Therate of respiratory compromise in 19 US hospitals was 10.5% of19,334 late preterm and 1.13% of 165,993 term infants.9 In thisstudy, all forms of respiratory morbidity and overall respiratoryfailure decreased significantly with gestational age until 39 weekswith the exception of meconium aspiration syndrome whichincreased.9 Extended data from Burgundy (Fig. 2) show a similartrend when moderate and late preterm infants are consideredtogether.

4.2. Elective caesarean section and neonatal iatrogeny

Elective caesarean section is now regarded as a major contrib-utor to respiratory morbidities in both late preterm and early termneonates.11,12 Elective caesarean section has been shown to beiatrogenic in 25.5% of late preterm deliveries among 1251 singletonpregnancies with mild gestational hypertension without protein-uria.12 A large retrospective study has shown that late preterm andterm infants delivered via elective caesarean section had increasedrespiratory morbidity but also significantly higher rates ofmortality [adjusted risk ratio (aRR): 2.1; 95% confidence interval(CI): 1.1e4.1] and risk of special care admission (aRR: 1.4; 95% CI:1.2e1.6) compared with infants delivered via planned vaginaldelivery.13

1,60 2,52 1,61 0,65 0,58 0,26 0,16 0,16

45,87

26,05

17,65

8,22

3,651,11 0,48 0,21

0%

10%

20%

30%

40%

50%

32 33 34 35 36 37 38 39-41

Poor prognosis Severe respiratory disorders

Weeks of gestion

Fig. 2. Gestational age and rates of poor prognosis (death and/or severe neurologicalcondition) and severe respiratory disorders (treated by nasal continuous positiveairway pressure and/or mechanical ventilation) in a population of 173,058 live-borninfants (years 2000e2009). Adapted and extended from Gouyon et al.3

J.-B. Gouyon et al. / Seminars in Fetal & Neonatal Medicine 17 (2012) 146e152148

Obstetric management strategies probably play an importantrole in delaying deliveries and reducing late preterm birth rates, aswas recently suggested by the national USA results.14

A recent randomised trial has provided results indicating thatantenatal steroid administration reduces the risk of respiratorydistress in term infants born by elective caesarean section.15 Similarstudies should be addressed in late preterm infants born bycaesarean section.

Practice points

� Follow current guidelines on antenatal corticosteroidadministration up to 34 weeks’ gestation.

Practice points

� Overall, elective caesarean section increases the risk ofneonatal morbidities and mortality, special careadmission and separation from the mother.

4.3. Clinical characteristics of respiratory diseases

The respiratory disease in late preterm infants often begins asdelayed respiratory transition. The course of the respiratorydistress can be unpredictable at its outset andmany of these infantswho have no or a few respiratory symptoms at birth subsequentlydemonstrate a growing need for respiratory support leading ina few cases to PPHN.16 The aetiology of the respiratory disease isfrequently unclear as late preterm infants often present charac-teristics of RDS, TTN and PPHN in the course of the same respiratorydisease.

Because of their birth weight and a misleading clinical picture,the respiratory disease severity in late preterm infants isfrequently underestimated. The initial diagnosis is often TTN asthese infants usually have tachypnoea with few retraction symp-toms. Consequently, treatment and monitoring are ofteninappropriate.

Practice points

� Initial picture of severe respiratory diseases in LPI isoften misleading.

4.4. Prevention of respiratory diseases by antenatal corticosteroids

Antenatal corticosteroids enhance maturity of surfactantproduction and lung liquid clearance. National guidelines17,18

endorse the use of a single course of antenatal corticosteroids tothe mother if birth up to 34 weeks’ gestation is a risk. Thisstatement is supported by a Cochrane review19 showing that RDSwas significantly reduced in corticosteroid-treated infantsentering a trial from 33 to 34þ6 weeks’ gestation (RR: 0.53; 95% CI:0.31e0.91; 434 infants) but not from 35þ0 to 36þ6 weeks (RR:0.61; 95% CI: 0.11e3.26; 189 infants). These results were obtainedfrom data limited to two studies. A recent single centre rando-mised controlled trial specifically addressed the use of antenatalcorticosteroids at 34 to 36þ6 weeks and concluded that antenatalbetamethasone failed to reduce the risk of respiratory morbidity,even after adjustment for subgroups of gestational age.20

However, this study was markedly underpowered especially forpreterm infants born at 34þ0 to 34þ6 weeks. In addition, it is ofconcern that studies performed in Canada21 and the USA8 showedthat moderately preterm births benefit from antenatal steroid

administration but to a much lesser degree than do preterm birthsat <33 weeks. Therefore, clinicians should follow current guide-lines on antenatal corticosteroid administration up to 34 weeksand await new data on antenatal corticosteroids before consid-ering any changes in the practice of antenatal corticosteroidadministration at 34 weeks.

4.5. Treatment of respiratory diseases

Among singleton neonates with gestational age ranging from 32to 41 weeks the rates of antenatal corticosteroid exposure, nCPAP,mechanical ventilation and surfactant administration in a Frenchpopulation are shown in Fig. 1 for each week of gestation (J-B.Gouyon et al., unpublished data). The potential severity ofa moderate preterm birth is obvious as w30% of moderate preterminfants needed nCPAP alone, another 30% were mechanicallyventilated and 35% were administered surfactant. In late preterminfants with respiratory failure, 45% were intubated and receivedexogenous surfactant which is similar to the 41% recently reportedby Condò et al.22

Therefore, care providers have to anticipate that moderatepreterm birth is often associated with severe respiratory disorderswhereas this condition is less frequent in late preterm infants. Inevery case, a high level of neonatal care is required.

Because respiratory diseases in moderate and late preterminfants may be severe, inhaled nitric oxide (INO) has been assessedas a selective pulmonary vasodilator in preterm as well as in terminfants with hypoxaemic respiratory failure.

A pooled analysis of data from three clinical trials in late pretermand term infants with hypoxic respiratory failure requiringmechanical ventilation found that INO at a starting dose of 20 ppmwas acutely associated with improved oxygenation and a reducedmedian duration of mechanical ventilation.23

Whether or not infants have clear echocardiographic evidenceof PPHN does not appear to affect the outcome for the baby.24

Echocardiographic examination is recommended by the AmericanAcademy of Paediatrics (AAP),25 and allows the exclusion ofcongenital cardiac malformation. The beneficial effect of INO inmoderate preterm infants with hypoxic respiratory failure is notestablished. The most recent Cochrane review26 about INO inpreterm infants has included 14 studies in infants of �34 weeks’gestation and has concluded that preterm infants cannot benefitfrom INO given as early rescue treatment or as a routine treatmentor as a late treatment based on the risk of bronchopulmonarydysplasia. Unfortunately the design of the studies does not allowthe identification of specific outcomes for moderate preterminfants. Definitive results of a meta-analysis using the individualpatient data from all infants enrolled in these trials (MAPPiNOcollaboration) are expected.27 Therefore, it appears reasonable touse INO in cases of hypoxic respiratory failure in late preterminfants if they do not have a diaphragmatic hernia, the outcome forwhich could be worsened by INO.24 At the moment, INO cannot berecommended in a systematic approach of moderate preterminfants with severe respiratory failure. However, INO administra-tion in this group can be discussed on an individual basis as a rescuetherapy.

Practice points

� Severe respiratory morbidity is frequent in moderatepreterm infants and can also be observed in latepreterm infants.

� Efficacy of inhaled NO in hypoxic respiratory failure isdemonstrated in late preterm infants and can be dis-cussed in moderate preterm infants on an individualbasis.

Practice points

� Neonatal hypoglycaemia: 8% of late preterm infants;16% of moderate preterm infants.

� Prevention: favour early enteral nutrition.

J.-B. Gouyon et al. / Seminars in Fetal & Neonatal Medicine 17 (2012) 146e152 149

4.6. Hypothermia

Irrespective of the infant’s gestational age, neonatal hypo-thermia has been defined as body temperature below 36.5 �C.28

However, a normal body temperature is not sufficient to assertthat the thermal environment is optimal in premature babies, asthey might show clinical signs of cold stress, such as peripheralvasoconstriction, even though rectal temperature remains at 37 �C.

In moderate and late preterm infants, as in smaller babies,thermal instability may extend from birth to the early hours anddays of life. In the delivery room, the attention taken by health careprofessionals to prevent heat loss may determinewhether the babywill be admitted to the NICU or to the neonatal nursery rather thanremaining with the mother. Indeed, clinical symptoms induced bycold stress may alter early adaptation of the otherwise ‘well’ babyand can be misinterpreted as presenting signs of neonatal sepsis,this in turn resulting in triage to higher than necessary levels ofcare.29 Unfortunately, a large number of infants born at 32e36weeks have admission temperatures of 34.5e36.5 �C and thecourse of their hospital stay has been associated with hypothermiain up to 10% of cases.5,30,31

Interventions to minimise the extent of heat loss in the deliveryroom should be tailored according to the infant’s gestation andneeds for assistance following birth: a meta-analysis of threerandomised controlled trials reported that increasing gestationalage from 23 to 33 weeks was independently associated withincreasing admission temperature and that moderate preterminfants at 32e33 weeks may benefit from the use of occlusivewraps for preventing heat loss in the delivery room.32 It is alsosuggested that late preterm infants who necessitate no or minimalintervention after birth should be managed by the same protectiveintervention recommended for term babies, that is rapid drying ona warm surface such as the mother’s chest or abdomen (skin-to-skin contact). Early starting of breastfeeding will further provideinfants with warmth. When carefully performed skin-to-skincontact is not possible, wrapping and placement under an over-head heat source may be an alternative option.28,33 Of course, moreactive interventions for preventing heat loss and providing activewarming are needed for the sickest infants immediately after birthand during transport to the neonatal unit, considering that coldstress may worsen concomitant RDS and precipitate PPHN. In casesof hypothermia, the means and the rapidity chosen to rewarm thebaby depend on the degree of hypothermia and on the infantresponse to rewarming.1,28

Practice points

� Hypothermia and cold stress threaten moderate andlate preterm infants during the first days of life.

4.7. Hypoglycaemia

Neonatal hypoglycaemia may be considered a part of multiple‘adaptational’ pathologies particularly for moderate and latepreterm infants, in whom hypoglycaemia is frequently investigatedin association with RDS, hypothermia, feeding problems andhyperbilirubinaemia.4,29,31,34 The risk of hypoglycaemia approaches8% in different series of late preterm infants4,30,31 and increases upto 16% inmoderate preterm infants.35 Late preterm infants cared forin the nursery are exposed to hypoglycaemia with a RR of 12.4 (95%CI: 10.1e15.4) compared with term infants.36 In growth-restrictedlate preterm infants, hypoglycaemia is a leading cause for NICUadmission and prolonged hospital stay.

The chief causes of hypoglycaemia in moderate and late preterminfants are limited enteral intake, poor suckeswallow coordination,delayed or ineffective oral feeding, associated pathologies (coldstress, sepsis) and limited compensatory mechanisms.30,31

The management of this common condition remains difficult.Indeed, no specific value or range of plasma glucose potentiallyresulting in brain injury has been identified for infants born at32e36 weeks. With specific regard to moderate preterm infants,there is a paucity of data on the optimal prevention and manage-ment of neonatal hypoglycaemia. However, recent works haveextensively addressed the importance of correct management ofpostnatal glucose homeostasis in late preterm infants34 anda practical guide for the screening and subsequent treatment ofhypoglycaemia in late preterm infants has recently been providedby the AAP.37 Preventive strategies for neonatal hypoglycaemia aremainly based on encouraging and establishing early breastfeeding.Babies should be fedwithin 1 h after birth. After the first feed, bloodglucose levels should be systematically checked as early as 3 h oflife and subsequently tailored to the infant’s individual adaptationof glucose homeostasis. Finally, interventions for treatment of lowglucose concentrations may range from enteral re-feeding tointravenous glucose infusion, according to the infant’s postnatalage, glucose values, and associated clinical signs of hypoglycaemiaand concomitant pathologies. For instance, asymptomatic infantswith glucose concentrations <25 mg/dL (birth to 4 h of life) or<35 mg/dL (4e24 h of life) should be firstly re-fed and receiveintravenous glucose infusion only in case of failure to increase theirglycaemia after enteral feeding. Symptomatic babies shouldconversely be treated by intravenous infusion as plasma glucosevalues fall to <40 mg/dL.37

Even with the tool of these practical guides, care providers forlate preterm infants finally have the unique challenge to ensure safemanagement of postnatal glucose homeostasis together with thepreservation of mothereinfant bonding. This consideration mayalso apply for moderate preterm infants, especially when relatively‘healthy’.

� Use guidelines.37

4.8. Feeding problems

Some evidence suggests that both moderate and late preterminfants are more likely to present feeding intolerance than theirterm counterparts.4,5,36 Specific considerations are needed with

Practice points

� An increased risk of jaundice and hyperbilirubinaemia-induced neurological injury in moderate and latepreterm infants.

� Be aware of the bilirubin kinetic pattern and usenomograms.44

� Avoid early discharge.

J.-B. Gouyon et al. / Seminars in Fetal & Neonatal Medicine 17 (2012) 146e152150

respect to gestational age ranges. Nutritional requirements afterbirth change according to gestation and it may be difficult to ach-ieve the recommended protein and caloric intakes withoutparenteral nutrition support at lower gestations, especially duringthe transitional period.38

Some evidence suggests that moderate preterm infants hospi-talised in the NICU fail to achieve intrauterine growth rates.39 Earlyinitiation of enteral feeding, use of gavage, caloric supplementationand parenteral nutrition support should rapidly be considered fornutritional management of moderately preterm infants andaccording to the infant’s needs, as they currently are for smallerbabies. This approach should be used jointly with care policies thatpromote breastfeeding. For late preterm infants, feeding difficultiessuch as poor sucking, milk intolerance and poor weight gain aremore usually interrelated with all the ‘minor morbidities’, i.e.neonatal hypoglycaemia and hyperbilirubinaemia, often puttingthe baby in a vicious circle which prolongs hospitalisation andincreases the risk of late morbidity. So, avoidance of feedingproblems remains based on the global strategy described forpromoting early breastfeeding. Although some papers reportedthat exclusive breastfeeding represents an independent risk factorfor morbidity and subsequent need for hospital care or re-admission in late preterm infants, these studies do not allowconclusions on this point as they did not assess breastfeedingquality in the analysed populations.

Practice points

� Increased risk of extrauterine growth restriction inmoderate preterm infants.

� Prevention: early nutrition; favour breastfeeding.

Practice points

� Increased risk of early and late onset sepsis in moderateand late preterm infants.

5. Jaundice

Premature birth is known to place infants at risk of hyper-bilirubinaemia. Preterm infants are exposed because of exagger-ated bilirubin production, hepatic immaturity in the uptake andconjugation of bilirubin, and excessive re-uptake due to intestinalimmaturity and delayed enteral feeding.

Although this risk decreases with increasing gestational age dueto a progressive developmental maturity, jaundice rates remainelevated in moderate and late preterm infants when comparedwith term ones.2,31 Jaundice also appears to be one of the mostimportant causes of hospital readmission, accounting for almosthalf of cases in late preterm infants.40

Preterm infants are vulnerable to brain damage induced byhyperbilirubinaemia, with particular relevance for kernicteruscases with bilirubin levels <20 mg/dL, so-called ‘low bilirubinkernicterus’.41 Despite a clear reduction in the prevalence of lowbilirubin kernicterus among preterm infants in recent decades, anincreased risk persists in moderate preterm infants.42 Concerninglate preterm infants, a retrospective study based on the Pilot Ker-nicterus Registry showed that kernicterus was more frequent inlate preterm than in term infants. In this population of late preterminfants, the large for gestational age babies were over-represented;many infants had a suboptimal lactation experience and they werecared for as healthy term infants.43

Guidelines for hyperbilirubinaemia are missing for infants <35weeks or <2500 g. Attention should be paid to bilirubin kineticpatterns, by reporting several measures on specific nomograms forpreterm babies44 and clinicians should also determine all possiblejaundice causes and risk factors.

For preterm infants of 35e36 weeks and >2500 g, the AAP45

recommendations for term and near term infants may be applied.They suggest accurate estimation of the risk of developing severehyperbilirubinaemia before discharge (bilirubin level based on theinfant’s age in hours, peak and evolution), correct dischargeinstructions to parents, and close follow-up especially in exclu-sively breastfed infants.45 The formal recommendation of the sameinstitution to avoid early discharge in these infants (<48 h) has tobe respected.45

6. Infections

Moderate and late preterm infants are more likely to developsevere infections such as sepsis, meningitis and pneumonia thanterm infants. In a large retrospective population report, McIntireand Leveno46 found an increased risk of sepsis [odds ratio (OR):2.18] and of work-up for infection (OR: 1.73) in late preterm versusterm infants. The rate of sepsis inmoderate preterm infants has alsobeen shown to be almost double that in late preterm infants: ina twin population, Refuerzo et al.47 found 5% of sepsis in moderateversus 2.2% in the late preterm infants. Cohen-Wolkowiez et al.47

analysed the prevalence of infections in a very large cohort of latepreterm infants. Early onset sepsis was established in 0.44% of latepreterm infants. Gram-positive organisms caused the majority ofearly onset sepsis (66%) and Gram-negative organisms represented27%. Group B streptococcus and Escherichia coli accounted for mostepisodes. Only 1% of late preterm infants with early onset sepsisdied. Late onset sepsis was found in 0.63% of infants, most beingcaused by Gram-positive bacteria (59%), Gram-negative bacteria(30%) and yeast (7%). The most commonly observed pathogensresponsible for late onset sepsis were coagulase-negative staphy-lococcus, Staphylococcus aureus, and Escherichia sp. A total of 7% ofinfants with late onset sepsis died.47

7. Other problems

Some common problems of very preterm babies may rarelyaffect the moderate and late preterm infant.35 In two large retro-spective cohorts the prevalence of necrotising enterocolitis (NEC)was significantly increased in late preterm compared with termnewborns2,46 but the NEC rate shows large variations from onestudy to another: 0.7% in US babies at 33e34 weeks8; 4.8% inTurkish late preterm infants hospitalised in NICU.30

Intraventricular haemorrhage (IVH), often limited to grades 1 or2, is rare in population-based cohorts of late preterm infantsalthough it is more frequently observed than in term infants.2,46

J.-B. Gouyon et al. / Seminars in Fetal & Neonatal Medicine 17 (2012) 146e152 151

Higher rates of IVH in late preterm infants have been reported bysingle centre studies conducted in tertiary hospitals but a bias atinclusion was likely. It is worth noting that IUGR was the mainpredisposing factor of IVH in the late preterm infant.30

Chronic lung disease (CLD, i.e. oxygen dependency at 36 weeks’postconceptional age) decreases with gestational age in moderateand late preterm infants. In Californian hospitals the CLD rate was6.7% at 30e326/7 weeks and 1.0% at 33e346/7 weeks.8 Another studyfound an incidence of 3.7% among late preterm infants whenrespiratory failure was present at birth.22

Apnoea of prematurity decreases with advancing gestationalage. Recurrent apnoea may be a symptom of infection, anaemia,hypoxia, respiratory failure, intracerebral haemorrhage, drug side-effect, hypoglycaemia and other metabolic disorders. Caffeine hasbeen found to be an effective treatment of recurrent apnoea up to37 weeks.48 The routine use of caffeine cannot be recommended inbabies born from 32 weeks’ gestation but may be valuable in thoseinfants with recurrent apnoea and/or weaning from respiratorysupport.

Moderate and late preterm infants show a reduction in drugclearance and prolonged half-lives, exposing them to an increasedrisk of drug toxicity especially when drugs have a narrow thera-peutic index. Reduction in drug absorption and protein binding ofdrugs contribute to the immaturity of the metabolic system. Someclinical conditions (hypoxaemia, acidosis, low serum albuminconcentration, hypotension) may modify drug metabolism.

A point of concern is the consequence on prognosis of lateprematurity in an infant born with a severe congenital malforma-tion. Natarajan et al.49 showed the particular vulnerability of latepreterm infants with congenital cardiac defects. The associationbetween late prematurity and cardiac malformation showeda greater than additive effect on both morbidity and mortality,although this conclusion is not universal for all malformations. Forinstance, a Canadian study of a national cohort of congenital dia-phragmatic hernia (CDH) did not show an optimal gestational agefor prenatally diagnosed CDH.50 Therefore further studies areneeded to assess the role of gestation in the prognosis of infantsborn with severe malformations.

Practice points

� Amoderately increased risk of NEC, low grade IVH, CLDand apnoea in moderate and late preterm infants.

� Drug metabolism is modified in moderate and latepreterm infants.

� Birth between 32 and 36 weeks’ gestation may mark-edly worsen the prognosis of some congenitalmalformations.

8. Conclusion

The rate of morbidity and mortality declines continuously whengestational age increases up to 39 weeks. This finding rendersobsolete the classification of births by categories of gestational age<39 weeks (very preterm, moderate preterm, late preterm, earlyterm). Clinicians should bear in mind that infants born at 34e36weeks are not ‘near term’ infants and that infants born at 37e38weeks are not full term infants. There is a particular need toeducate health care providers and parents about the vulnerabilityof infants born between 34 and 36 weeks’ gestation.

Conflict of interest statement

None declared.

Funding sources

None.

References

*1. Engle WA. Morbidity and mortality in late preterm and early term newborns:a continuum. Clin Perinatol 2011;38(3):493e516.

2. Melamed N, Klinger G, Tenenbaum-Gavish K, et al. Short-term neonataloutcome in low-risk, spontaneous, singleton, late preterm deliveries. ObstetGynecol 2009;114(2 Pt 1):253e60.

3. Gouyon JB, Vintejoux A, Sagot P, et al. Neonatal outcome associated withsingleton birth at 34e41 weeks of gestation. Int J Epidemiol 2010;39:769e76.

4. Dimitriou G, Fouzas S, Georgakis V, et al. Determinants of morbidity in latepreterm infants. Early Hum Dev 2010;86:587e91.

5. Bird TM, Bronstein JM, Hall RW, Lowery CL, Nugent R, Mays GP. Late preterminfants: birth outcomes and health care utilization in the first year. Pediatrics2010;126:e311e9.

6. Pulver LS, Guest-Warnick G, Stoddard GJ, Byington CL, Young PC. Weight forgestational age affects the mortality of late preterm infants. Pediatrics2009;123:e1072e7.

7. Reddy UM, Ko CW, Raju TN, Willinger M. Delivery indications at late-pretermgestations and infant mortality rates in the United States. Pediatrics2009;124:234e40.

8. Escobar GJ, McCormick MC, Zupancic JA, et al. Unstudied infants: outcomes ofmoderately premature infants in the neonatal intensive care unit. Arch DisChild Fetal Neonatal Ed 2006;91:F238e44.

9. Hibbard JU, Wilkins I, Sun L, et al, Consortium on Safe Labor. Respiratorymorbidity in late preterm births. JAMA 2010;304:419e25.

10. Colin AA, McEvoy C, Castile RG. Respiratory morbidity and lung function inpreterm infants of 32 to 36 weeks’ gestational age. Pediatrics 2010;126:115e28.

11. Gouyon JB, Ribakovsky C, Ferdynus C, et al. Severe respiratory disorders interm neonates. Paediatr Perinat Epidemiol 2008;22:22e30.

12. Barton JR, Barton LA, Istwan NB, et al. Elective delivery at 340/7 to 366/7 weeks’gestation and its impact on neonatal outcomes in women with stable mildgestational hypertension. Am J Obstet Gynecol 2011;204:44.e1e5.

13. De Luca R, Boulvain M, Irion O, Berner M, Pfister RE. Incidence of earlyneonatal mortality and morbidity after late-preterm and term cesareandelivery. Pediatrics 2009;123:e1064e71.

*14. Mathews TJ, Miniño AM, Osterman MJ, Strobino DM, Guyer B. Annualsummary of vital statistics: 2008. Pediatrics 2011;127(1):146e57.

15. Stutchfield P, Whitaker R, Russel I. Antenatal betamethasone and incidence ofneonatal respiratory distress after elective cesarean section: pragmatic rand-omised trial. BMJ 2005;331:662.

16. Ramachandrappa A, Jain L. Elective cesarean section: its impact on neonatalrespiratory outcome. Clin Perinatol 2008;35:373e93. vii.

*17. ACOG Committee on Obstetric Practice. ACOG Committee Opinion No. 475:Antenatal corticosteroid therapy for fetal maturation. Obstet Gynecol2011;117(2 Pt 1):422e4.

18. Roberts D, Antenatal corticosteroids to reduce neonatal morbidity andmortality. Greentop guideline no. 7. London: Royal College of Obstetrics andGynaecology; 2010.

19. Roberts D, Dalziel S. Antenatal corticosteroids for accelerating fetal lungmaturation for women at risk of preterm birth. Cochrane Database Syst Rev2006;(3):CD004454.

20. Porto AM, Coutinho IC, Correia JB, Amorim MM. Effectiveness of antenatalcorticosteroids in reducing respiratory disorders in late preterm infants:randomised clinical trial. BMJ 2011;342:d1696.

21. Joseph KS, Nette F, Scott H, Vincer MJ. Prenatal corticosteroid prophylaxis forwomen delivering at late preterm gestation. Pediatrics 2009;124:e835e43.

22. Condò V, Colnaghi M, Vanzatiz M, et al. Respiratory failure in “late preterm”

infants: a retrospective cohort study. Pediatr Med Chir 2009;31:241e5.23. Golombek SG, Young JN. Efficacy of inhaled nitric oxide for hypoxic respiratory

failure in term and late preterm infants by baseline severity of illness:a pooled analysis of three clinical trials. Clin Ther 2010;32:939e48.

24. Finer NN, Barrington KJ. Nitric oxide for respiratory failure in infants born at ornear term. Cochrane Database Syst Rev 2006;(18):CD000399.

25. American Academy of Pediatrics. Committee on Fetus and Newborn. Use ofinhaled nitric oxide. Pediatrics 2000;106(2 Pt 1):344e5.

*26. Barrington KJ, Finer N. Inhaled nitric oxide for respiratory failure in preterminfants. Cochrane Database Syst Rev 2010;(12):CD000509.

27. Askie LM, Ballard RA, Cutter G, et al. Meta-Analysis of Preterm Patients oninhaled Nitric Oxide (MAPPiNO) Collaboration. BMC Pediatr 2010;23:10e5.

28. World Health Organization: Department of Reproductive Health and Research(RHR). Thermal protection of the newborn: a practical guide. Geneva: WHO.Available from,http://www.who.int/making_pregnancy_safer/MSM_97_2;1997.

*29. Laptook AR, Jackson GL. Cold stress and hypoglycemia in the late preterm(“near term”) infant: impact on nursery of admission. Semin Perinatol2006;30(1):24e7.

30. Kalyoncu O, Aygün C, Cetino�glu E, Küçüködük S. Neonatal morbidity andmortality of late-preterm babies. J Matern Fetal Neonatal Med 2010;23:607e12.

31. Wang ML, Dorer DJ, Fleming MP, Catlin EA. Clinical outcomes of near-terminfants. Pediatrics 2004;114:372e6.

J.-B. Gouyon et al. / Seminars in Fetal & Neonatal Medicine 17 (2012) 146e152152

32. Cramer K, Wiebe N, Hartling L, Crumley E, Vohra S. Heat loss prevention:a systematic review of occlusive skin wrap for premature neonates. J Perinatol2005;25:763e9.

*33. Laptook AR, Watkinson M. Temperature management in the delivery room.Semin Fetal Neonatal Med 2008;13(6):383e91.

34. Adamkin DH. Late preterm infants: severe hyperbilirubinemia and postnatalglucose homeostasis. J Perinatol 2009;29(Suppl. 2):S12e7.

35. Altman M, Vanpée M, Cnattingius S, Norman M. Neonatal morbidity inmoderately preterm infants: a Swedish national population-based study.J Pediatr 2011;158:239e44.

36. Dani C, Corsini I, Piergentili L, Bertini G, Pratesi S, Rubaltelli FF. Neonatalmobidity in late preterm and term infants in the nursery of a tertiary hospital.Acta Paediatr 2009;98:1841e3.

*37. Adamkin DH, Committee on Fetus and Newborn. Postnatal glucose homeo-stasis in late-preterm and term infants. Pediatrics 2011;127(3):575e9.

38. Rigo J, De Curtis M. Parenteral nutrition in premature infants. In: Guandalini S,editor. Textbook of pediatric gastroenterology and nutrition. New York: Taylor &Francis; 2004. p. 619e37.

39. Blackwell MT, Eichenwald EC, McAlmon C, et al. Interneonatal intensive careunit variation in growth rates and feeding practices in healthy moderatelypremature infants. J Perinatol 2005;25:478e85.

40. Maisels MJ, Kring E. Length of stay, jaundice, and hospital readmission. Pedi-atrics 1998;101:995e8.

41. Gartner LM, Snyder RN, Chabon RS, Bernstein J. Kernicterus: high incidence inpremature infants with low serum bilirubin concentrations. Pediatrics1970;45:906e17.

42. Watchko JF, Oski FA. Kernicterus in preterm newborns: past, present, andfuture. Pediatrics 1992;90:707e15.

43. Bhutani VK, Johnson L. Kernicterus in late preterm infants cared for as termhealthy infants. Semin Perinatol 2006;30:89e97.

44. Van Imhoff DE, Dijk PH, Hulzebos CV. BARTrial study group, NetherlandsNeonatal Research Network. Uniform treatment thresholds for hyper-bilirubinemia in preterm infants: background and synopsis of a nationalguideline. Early Hum Dev 2011;87:521e5.

*45. American Academy of Pediatrics Subcommittee on Hyperbilirubinemia.Management of hyperbilirubinemia in the newborn infant 35 or more weeksof gestation. Pediatrics 2004;114(1):297e316.

46. McIntire DD, Leveno KJ. Neonatal mortality and morbidity rates in latepreterm births compared with births at term. Obstet Gynecol2008;111:35e41.

47. Cohen-Wolkowiez M, Moran C, Benjamin DK, et al. Early andlate onset sepsis in late preterm infants. Pediatr Infect Dis J2009;28:1052e6.

*48. Henderson-Smart DJ, De Paoli AG. Methylxanthine treatment for apnoea inpreterm infants. Cochrane Database Syst Rev 2010;(12):CD000140.

49. Natarajan G, Anne SR, Aggarwal S. Outcomes of congenital heartdisease in late preterm infants: double jeopardy? Acta Paediatr2011;100:1104e7.

50. Safavi A, Lin Y, Skarsgard ED. Canadian Pediatric Surgery Network. Perinatalmanagement of congenital diaphragmatic hernia: when and how shouldbabies be delivered? Results from the Canadian Pediatric Surgery Network.J Pediatr Surg 2010;45:2334e9.

at SciVerse ScienceDirect

Seminars in Fetal & Neonatal Medicine 17 (2012) 153e158

Contents lists available

Seminars in Fetal & Neonatal Medicine

journal homepage: www.elsevier .com/locate/s iny

Neonatal management and safe discharge of late and moderate preterm infants

Robin K. Whyte*

Dalhousie University, IWK Health Centre G2216, 5980 University Avenue, Halifax, Nova Scotia, Canada B3J 6R8

Keywords:ApneaBody temperature regulationFeeding methodsLate preterm infantsLow birth weightModerate preterm infants

* Tel.: þ1 902 470 6466.E-mail address: Robin.whyte@dal.ca.

1744-165X/$ e see front matter � 2012 Elsevier Ltd.doi:10.1016/j.siny.2012.02.004

s u m m a r y

Late and moderate preterm infants form the majority of admissions for prematurity to special careneonatal nurseries. Although at risk for acute disorders of prematurity, they do not suffer the serious longterm risks and chronic illnesses of the extremely premature. The special challenges addressed here are oftransition and of thermal adaptation, nutritional compensation for postnatal growth restriction, theestablishment of early feeding, and the avoidance of post-discharge jaundice or apnea. These ‘healthy’premature infants provide challenges for discharge planning, in that opportunities may be available fordischarge well before the expected date of delivery, which should be pursued. Barriers to early dischargeare rigid conservative protocols and unwarranted investigations; facilitators of discharge are individu-alized care by nurses expert in cue-based feeding, early management of the thermal environment,support of family preferences and encouragement of motherebaby interactions. Safe discharge dependson recognizing these opportunities and applying strategies to address them.

� 2012 Elsevier Ltd. All rights reserved.

1. Introduction

The definitions of moderate and late preterm infants areaccepted as infants of 32 and 33 and of 34, 35 and 36 weeksrespectively. This review addresses management issues particularto prematurity but does not address in detail the disease-specificinterventions of intensive care. The management issues of prema-ture infants addressed here will be confined to discussing theselection of infants for admission to special or intensive neonatalcare units (hereafter called special care nurseries) and their prep-aration for discharge in industrialized countries with establishedmaternity and neonatal intensive care services.

The birth weights of this population overlap with infants of verylow and even extremely low birth weight (Table 1).1 Mostmorbidity of prematurity is related to gestational age, but somechallenges (such as the maintenance of thermal homeostasis or there-establishment of growth) are related to size at birth. Moderatepreterm infants form a continuum with very preterm infants (<32weeks’ gestation) and will therefore occasionally require dischargeplans usually applied to this group of smaller infants. Neonatalfollow-up programmes do not routinely enrol infants >32 weeks’gestation, and infants with chronic lung disease requiring homeoxygen are rarely included this group. Infants of >32 weeks’gestation are not ordinarily at risk of retinopathy of prematurity.2,3

All rights reserved.

Aspects of discharge related to more extreme prematurity will notbe addressed here.

2. Triage at birth

2.1. The delivery room

A resuscitation physician or team, with appropriate equipment,capable of assessing the newborn, of establishing an airway and ofachieving thermal neutrality must be present. If the birth is antic-ipated the team should have reviewed the maternal antenatalrecord and should be, in particular, knowledgeable as to the esti-mated gestational age and the confidence with which this has beenestimated, and to the use of antenatal steroids, the estimated fetalweight and the presence or absence of maternal drugs or infection.Resuscitation and its controversies will not be addressed here,other than to say that the risks of respiratory depression anddistress aremuch higher in this population than at term. There is noprerequisite for prophylactic surfactant, added oxygen or antibi-otics and a minimal-handling, expectant approach may be taken,with close attention to thermal homeostasis.

2.2. Post-delivery triage of the premature infant

A decision must be made to admit the baby either to a specialcare nursery or to a postnatal ward, which in practice includesa decision either to separate the baby from the mother or tomaintain the mothereinfant dyad. Admission to a special care

Table 1Distribution of birth weight by gestational age for singletons. Approximated from the data of Kramer et al.1

Gestationalage (weeks)

Median birthweight (g)

Cumulative percentage below birth weight limits

<1000 g <1250 g <1500 g <1750 g <2000 g <2250 g <2500 g

32 1862 1% 5% 16% 38% 65% 86% 96%33 2081 0% 1% 6% 19% 41% 67% 87%34 2313 0% 0% 2% 7% 21% 44% 69%35 2553 0% 0% 0% 2% 8% 22% 45%36 2795 0% 0% 0% 1% 3% 9% 23%

R.K. Whyte / Seminars in Fetal & Neonatal Medicine 17 (2012) 153e158154

nursery is generally required for infants who need incubator care,cardiorespiratory monitoring, assessment of adaptation ormanagement of the complications of prematurity. This triage isspecific to institutions and depends greatly on nursing skills andresources allocated to postnatal wards and the availability ofpediatric expertise. Early separation can be usefully delayed ifdelivery room staff has the resources to monitor premature infantswith their mothers and has the skills to manage the neonatalenvironment with skin-to-skin contact and early breast-nippling.4,5

The stable maintenance of an axillary temperature in the range36.7e37.3 �C, the absence of respiratory distress and the presenceof resources for ongoing surveillance are prerequisites fora prolongation of maternaleinfant contact in the delivery room;skin-to-skin contact requires continued surveillance of the infantsby nursing staff.6 Blood sugar screening in the delivery room is ofunclear benefit and the consequences of intervention unevaluated;screening for hypoglycemia is advocated for preterm infants at 1½to 2 h of age.7,8

The abilities and benefits to themother of continued shared carewith their infants must also be addressed. Exhausted mothers orthose recovering from anesthesia cannot be expected to participatein infant care, but innovative and individualized strategies (some-times involving other family members) may reduce separation andsupport the early establishment of breastfeeding. Limitation ofnursing resources and medical skills in the delivery room andpostnatal wards reduces the safety of individualized triagedecision-making; in these circumstances clear guidelines must beimposed which determine the criteria that will permit admission ofthe infant to the postnatal ward.

3. Admission to and early management in the special carenursery

Many special care nurseries have universal admission protocolsfor premature infants, which include incubator care and cardiore-spiratory monitoring. An early medical assessment, includingreview of the maternal perinatal history, birth events and physicalexamination of the newborn should determinewhether an infant isto be admitted for care limited to observation of adaptation or forspecific investigation and management of disorder of prematurity(such as respiratory distress). For many infants, where there are norisk factors for infection and no signs of delayed adaptation orcardiorespiratory disease, management can be confined to obser-vation, establishment of feeding, monitoring and care of thethermal environment.

The admission process must not result in additional morbidityor prolonged hospital stay. There is no place for ‘routine’ investi-gations such as blood sampling; in the presence of low diseaseprevalence (risk) such pursuits generate a high rate of false-positiveresults which frequently result in further investigation, interven-tion and prolongation of stay. Likewise there is no role for routineblood culture or recourse to antibiotics; these strategies must bereserved for infants in whom there is a defined risk of perinatally

acquired infection from the maternal history or clinical findings.The requirement for blood sugar measurements on admission andat selected intervals thereafter is inescapable,8 but these must belimited by protocol or to response to clinical suspicion of hypo-glycemia; repeated unsolicited estimations lead to false-positiveobservations, further confusion and unwarranted intervention.

4. Establishing thermoneutrality

All caregivers should have a good understanding of thecomponents of the thermal environment.9 In order to facilitateobservation, incubator placement may be a prerequisite foradmission in some units, inwhich case great caremust be exercisednot to overheat larger infants. Thermoneutrality can be assumed ifthe axillary temperature is between 36.7 and 37.3 �C.10

5. Early feeding

The importance of the establishment of very early nutrition hasrecently been recognised, with the goal of establishing, so far as theconstraints of fluid allowances permit, full nutritionwithin the first2e5 days of life.11,12 Whereas these discussions have focused moreon extremely low birth weight infants, the nutritional deficienciesofmoderate and late preterm infantsmust be similarly addressed, asthese infants have been born before the major part of fetal growthhas been completed. The special needs of the small-for-datespreterm have long been recognised, but all premature infantsincluding those born of appropriate size-for-dates suffer frompostnatal growth restriction attributable to the prolonged delay inre-attaining birth weight characteristic of the preterm infant.13

Feeding should be fully enteral where possible, often by gavage,and should be the expectation in infants of >1500 g birthweight.Mother’s milk will generally not be immediately available, andhuman milk banks are not universally accessible, in which case anindividualized decision must be taken, preferably with the mother,to feed with infant formula, using additional maternal colostrum ormilk as it comes available, or to feed by an intravenous route. Earlyenteral feeding and advancement are not risk factors for necro-tizing enterocolitis.14e16 Trophic feeding may be introduced whenfull enteral tolerance cannot be achieved, and it may accelerate theachievement of full feeding.16,17 Where intravenous feeding isunavoidable, amino acids and lipids should be introduced as soonas these are available in fully supportive quantities.12,18,19

6. Ongoing care and preparation for discharge

Safe discharge of late preterm infants has been addressed inreviews20,21 but including moderate premature infants requires theconsideration of an expanded profile of precedent conditions.Despite the greatly increased morbidity of the late preterm ascompared with the term infant20,22 there is still a considerableproportion of infants who appear sufficiently stable at birth to bemanaged in a low risk newborn nursery and discharged according

R.K. Whyte / Seminars in Fetal & Neonatal Medicine 17 (2012) 153e158 155

to protocols offered to healthy term infants. Discharges at 2e5 daysof age can be characterized as ‘low risk’ newborn discharges; inrecent years there has been an increase in discharges within 2 days(and sometimes within hours) of birth. These have been termed‘early’ discharges. To qualify for early discharge American23 andCanadian24 guidelines specify that an infant must be term (i.e. �37weeks’ gestation). British guidelines do not apply this limitation toearly discharge,25 preferring instead to stress the individualizeddecision-making for all mothereinfant dyads concerning discharge.

Infants who experienced some of the morbidities of prematu-rity, such as respiratory distress, cannot be discharged until theirrequirements for intensive care support have been addressed.

Moderate preterm infants, even if free of otherwise seriousmorbidity of prematurity, do not qualify for ‘low risk’ newborndischarges and will always have to be demonstrably physiologicallystable enough for the home environment. Guidelines specific to thisgroup of infants are not available, but may be adapted from thosemade for smaller infants.26,27 Infants born at 32e34 weeks tend togo home before due date, as they have had time to recover fromearly complications and opportunities for early feeding andparental planning can be addressed.28

7. Conditions required for discharge

Conditions for discharge include infant, community and familyfactors. Preterm infants must be breathing room air withoutrespiratory embarrassment or interruption, feeding adequately toestablish and maintain weight gain, and capable of maintainingthermal homeostasis without stress in the receiving environment.These achievements depend in part on physiological maturation,and they tend to mature together in individual infants.29 A furthercondition, which applies to late preterm infants who may qualifyfor ‘term’ discharge, is that the metabolic adaptations of earlypostnatal life are either complete or provided for.

8. Resolution of jaundice

All newborn infants should now be screened for hyper-bilirubinemia, but current guidelines address late preterm but notmoderate premature infants.30 Phototherapy is very effective atreducing high levels of bilirubin and has few adverse effects,31 but itdisturbs maternal access and impedes nursing care. Unless thereare clear indications to the contrary (such as established hemolysisor early jaundice) bilirubin measurements should be made whenphototherapy is likely to be effective and the predictive power ofthe test greatest (i.e. at >24 h of age).

The preterm infant has a slower rise in serum bilirubin than theterm infant, reaching its peak at 4e9 days.32 Moderate and latepremature infants are at greater risk for kernicterus33; moreconservative levels for phototherapy, currently undefined, shouldbe used for these infants.34 Moderate premature infants, with theirmore prolonged admissions, canmore readily receive phototherapy,but the late preterm infant is at risk of discharge before the peakbilirubin has been reached or even before the bilirubin has risen tomoderate risk values. Late preterm infants who are breastfeedingare at particular risk of kernicterus35 or extreme hyper-bilirubinemia,36 and this complication can be avoided by ensuringadequate post-discharge supervision for all infants who are notknown to be both free of jaundice and in a low risk category fromearlier screening.

9. Maintenance of thermal homeostasis

Infants must be adapted to the thermal environment of thehome to achieve a successful discharge. Several trials, combined in

a Cochrane review,37 have confirmed the feasibility of transfer ofmedically stable infants to a cot at 1600 g (as compared with1800 g) for infants of �32 weeks’ gestation. Furthermore, earlytransfer to a cot is associated with greater weight gain and earlierdischarge. The larger and more recent of these trials38,39 took placein nursery temperatures of 24 �C with humidities approaching 50%,with infants dressed in light clothing with single blankets. Possiblythe reported increased weight gain reflects a more optimal self-regulated thermal environment for these dressed infants than isprovided by an incubator. It is important not to exceed the thermalinsulation described in these and other studies40; overbundling isassociated with sudden infant death syndrome.41

Current guidelines for term infants cared for at home suggestthat the infant should be lightly clothed for sleep,40 and the roomtemperature should be kept comfortable for a lightly clothedadult.42

10. Establishment of feeding to discharge

Enteral feeding is often delayed by respiratory and otherillnesses. Late and moderate preterm babies may retain thepremature patterns of poor suck, swallow and breathing coordi-nation,43 as well as reflux or poor satiety associated with slowgastric emptying.44 Staged, timed introduction of scheduled feedsof predetermined volume has traditionally been achieved by pro-longed gavage feeding. There is now substantial evidence that earlymaturation of sucking may be accelerated by the early cue-basedintroduction of suckling feeding ‘on demand’ (where the infantcontrols the timing) and ‘ad libitum’ (where the infant controls theamount).45 The skilled recognition of and response to infant signsof hunger and satiety can result in successful infant-cued feeding ator before 32 weeks’ gestation and a shortened length of stay.46

Non-nutritive sucking may be used as either a tool for cue recog-nition or as an infant training technique and may also contribute tothe earlier establishment of feeding.47 For breastfeeding infants,a staged sequence from feeding expressed milk by gavage or bottlemay be safely combined with early breast nippling to achieve earlyinfant-cued and maternally responsive breast-nipple feeding.

11. Goals and content of feeding

The practice of staged increases in feeds in the first week to life toachieve a goal of 150ml/kg/day have changed little from the original2½ oz/lb/day from which this number was derived. Models of idealpostnatal weight gain are based on estimates of fetal growthcombined with growth rates of healthy term infants.11,48 From 32 to50 weeks postconceptional age (pca) these idealized weight gainscan be approximated by a daily weight gain of 30 g/day. Weight gainis close to linear, despite feedingbeing increasedonaperbodyweightbasis, because the energy and nutrient costs of weight gain increasewith postnatal age.11,49,50 Matching the fetal rate of weight gain maynot be enough, as this does not compensate for the prolonged post-natal delay inweight gain experienced by the more preterm infants,who may, despite establishing ‘adequate’ rates of weight gain, bedischarged below the tenth centile for weight at discharge.11e13 This‘postnatal growth restriction’ can only be addressed by strategiesdesigned to increase nutrient intake to the maximum tolerated.50

Recommended protein intakes for infants >1200 g are �3.6 g/kg/daywith a protein:energy ratio of 2.8.11 Strategies for achieving thesegoals include fortification of humanmilk with commercial fortifiers,theuseofhighnutrientpreterm formulas, and theadditionofproteinsources to the limitofmetabolic tolerance, asdefinedbyurea levels.51

Weight gain goals of�35g/daymaybeachievedwith such strategies.Someconcernhasbeenaddressed that rapidweight gain in infancy isassociated with adverse outcomes in adult life52; clarification of this

R.K. Whyte / Seminars in Fetal & Neonatal Medicine 17 (2012) 153e158156

dilemma must await further long term follow-up of early feedingtrials, but as previous enriched nutritional intervention waspredominantly focused on increased energy intake, resulting inexcess growth of fat mass,53 it may be that re-establishing lean bodymass with high protein:energy and phosphate:energy ratios willaddress this concern.

12. Feeding beyond discharge

Full enteral feeding is generally a prerequisite for discharge,although in specific circumstances infants may be discharged withgavage feeds.54 In most circumstances safe tolerance and retentionof feeds and at least a reasonable expectation of weight gain arerequirements for discharge home. High nutrient post-dischargeformulas are available,55,56 as are human milk fortifiers57; theiruse has not been fully evaluated or compared to a strategy of fullyliberalizing infant intake to demand.58

13. Respiratory stability

Respiratory stability depends on complete recovery fromrespiratory illness and the absence of apnea of prematurity. Wellpreterm and term infants have median pulse oximetry readings inexcess of 97% after the first days of life59,60 (partly because pulseoximeters may read errantly high in the range 97e100%). Whenadditional oxygen therapy has been withdrawn, oximetry readingsshould be �93% when the infant is asleep or quietly awake in roomair for a least one night prior to discharge.61

14. Apnea of prematurity

Apnea of prematurity increases in frequency and severity withdecreasing gestational62 and postconceptional age,63 and isa particular component of discharge planning for the moderatepreterm. Apnea may persist well beyond term,29 but thisphenomenon is far more common in very premature infants andrare for infants born at or after 32 weeks’ gestation.63,64 Therequired apnea/bradycardia-free interval that safely defines thisperiod has been estimated conservatively at 8 days64 where apneais defined as apnea for >20 s, bradycardia <60 bpm or oxygensaturation of <80%, not provoked by feeding. Infants who haveapnea of increasing frequency or a new episode beyond an 8-dayapnea-free interval are likely to be manifesting new illnesses,such as infection.64 There is a reassuring tendency for apnea for allinfants to resolve by 44 weeks’ pca.65 Apnea at home is a fright-ening event which usually leads to emergency readmission, but itrarely results in mortality and does not appear to be a predictor ofsudden infant death syndrome.65,66 Families should be so reassuredand instructed on the simple stimulatory techniques required toresolve such an event; their knowledge of access to emergencyservices should be reviewed on discharge. Caffeine has been usedon an ambulatory basis, but its discontinuation may require read-mission for monitoring.26,67

15. Resources for care: transition between hospital and homeresources

The resources that are provided by ongoing hospitalization of aninfant are facility-based, health profession-based and physician-based; discharge may take place when these are no longerrequired. Health profession support includes individualized expertpatient care from nursing and allied health professions. Thephysician, in most systems, provides medical expertise as inter-mittent care, and generally bears the responsibility for the decisionto discharge to the community. Expert continuous nursing care and

observation is the cornerstone of evaluation of readiness fordischarge. The community, on the other hand, generally includesa highly variable (but often superior) home environment, familycare which may be continuous, community health resources whichcan be intermittent, and physician resources which require accessand often transport.

A supportive home environment is an opportunity for earlierdischarge which should be seized. The convalescent preterm infantis at increased risk of infection, but more so of serious nosocomialbacterial infection in a hospital nursery than of a sibling-acquiredviral infection in a well-managed home. Continuous observationof the infant may be more effectively and reliably conducted (andreported) by a parent supported by an extended family than bya nurse sharing responsibilities with other sicker infants andworking intermittent shifts. Marked improvements in feeding mayfollow discharge.68

Merritt et al.25,69 have critically reviewed strategies for effectingsafe, early discharge of very low birth weight infants that can beapplied to moderate and late preterm infants. Parent involvement,empowerment and education70 can and should be used to reducethe length of hospital stay where the opportunity arises.

16. Anticipating discharge

Safe discharge of a healthy infant is the primary goal of neonatalcare, and plans should be clear and documented from admission.Parents should be expected to plan for discharge home from thetime of birth, and this date should be estimated with them soonafter admission; if there are uncertainties, these should be explicitand the estimation repeatedly adjusted. Planning for dischargefollowing premature birth is more difficult for parents than it is forinstitutions, as it may involve adjustment of parental leave, changesin domestic planning, recruitment of family support, earlier acqui-sition of baby needs, and changes or improvements to housing.Maternal medical needs, twins or other multiples, and parentalseparation or discordmay further impede timely discharge. A nursedischarge specialist or team can evaluate and address parentingskills and confidence, community resources, and access to healthcare and resources. These evaluations and arrangements are time-consuming and require excellent coordination and communica-tion skills, but may be expected to be cost-effective.69

Attention must be paid to the safe transport of the infant to thecommunity. Recommendations for car seat use for prematureinfants should be applied.26

17. Types of discharge

Hospital discharge may take the following forms:

1. Complete discharge home with support from local communitycare services (health visitors, family doctor appointments).

2. Provisional discharge home with ongoing daily managementby hospital personnel, which may include daily blood testingand management (for example, in hyperbilirubinemia oranemia) or monitoring (e.g. weight gain).

3. Suspended discharge, where a hospital admission (‘bed’) ismaintained for the infant who goes home ‘on a pass’, usuallywith a daily return for evaluation or care. The hospital staffretains the responsibility although not the direct administra-tion of infant care.

4. Rooming in or placing the baby under family care in accom-modationwithin the hospital, supervised by hospital staff, is animportant step in evaluating and managing the infant with thefamily outside the directly monitored environment of thenursery.

� Is the body composition of the preterm newborn at termdue date predictive of body habitus in later childhood?

� Canwe safely identify full vascularization of the pretermretina? If so, can we release our restraints on higherlevels of oxygen in respiratory therapy?

� How effective is replacement of breast milk withformula in the treatment of prolonged hyper-bilirubinemia in the breast-fed, late preterm infant?

� When should phototherapy be applied or discontinuedin the prevention of exchange transfusion or kernicte-rus in otherwise-healthy preterm infants?

� Are blood glucose levels in otherwise-healthy preterminfants predictive of anything, and should they betreated?

R.K. Whyte / Seminars in Fetal & Neonatal Medicine 17 (2012) 153e158 157

Institutions should recognize and develop flexibility indischarge arrangements with a view to making opportunities forsafe, earlier but sometimes graduated discharge available. Thefollowing is a list of requirements for safe discharge of themoderate or late preterm infant, modified from a plan for the verylow birth weight infant.26

1. Sustained or anticipated weight gain sufficient to attain ante-natal growth expectations.

2. Stable feeding and infant cue-driven feeding patterns.3. Stable temperature when lightly clothed at room temperature.4. A scheduled immunization plan.5. Nutritional supplementation, where indicated, to include iron,

vitamin D or fortification.6. Review of requirements for retinal fundoscopy (generally

limited to infants of �32 weeks’ gestation).7. Written copy of discharge summary for parents to share with

medical care services.8. Satisfactory evaluation of, or education of, parenting skills.9. Review of medications with parents.

10. Home environment evaluation and availability of communitysupport services.

11. Community-based home or office assessment within 48 h ofdischarge.

Practice points

� Moderate and late preterm infants are increasinglyvulnerable to disorders of prematurity with decreasinggestational age.

� An individualized approach to care (expecting differentbabies to respond in different ways) leads to safe,earlier discharge in selected infants.

� Otherwise-healthy preterm babies should be evaluatedat 1600 g for selected transfer from an incubator toa cot.

� Otherwise-healthy preterm babies should be evaluatedearly for early oral feeding and satiety cuing; selectedinfants may be nipple fed from 32 weeks on.

� Some stable preterm babies will do better at home andbe safer and more successfully fed than in a nurseryenvironment.

� Individual review ofmetabolic tolerance should be usedto maximize and optimize nutrient intakes to achieveweight gains sufficient to compensate for the postnatalmalnutrition of the premature.

� Discharge plans and expectations should be sharedwith parents and community resources from the day ofadmission.

� Discharge, and preparation for discharge, may takemany forms and should be flexible, with easy recourseto readmission or to suspended discharge schemes.

Research directions

� More on cue-based feeding: does recognition of earlysatiety signals lead to better weight gain or avoidanceof later onset obesity or metabolic syndrome?

� Does early servocontrol lead to earlier or latersuccessful response to transfer from incubator to cot?

� What are the long term neurodevelopmental andphysical health outcomes of early aggressive adaptivefeeding of the preterm newborn?

References

1. Kramer MS, Platt RW, Wen SW, et al. A new and improved population-basedCanadian reference for birth weight for gestational age. Pediatrics2001;108:E35.

2. Smith LE. Through the eyes of a child: understanding retinopathy throughROP. The Friedenwald lecture. Invest Ophthalmol Vis Sci 2008;49:5177e82.

3. Reynolds JD, Dobson V, Quinn GE, et al. Evidence-based screening criteria forretinopathy of prematurity: natural history data from the CRYO-ROP andLIGHT-ROP studies. Arch Ophthalmol 2002;120:1470e6.

4. Moore ER, Anderson GC, Bergman N. Early skin-to-skin contact for mothersand their healthy newborn infants. Cochrane Database Syst Rev2007;3:CD003519.

5. Bergman NJ, Linley LL, Fawcus SR. Randomized controlled trial of skin-to-skincontact from birth versus conventional incubator for physiological stabiliza-tion in 1200- to 2199-gram newborns. Acta Paediatr 2004;93:779e85.

6. Andres V, Garcia P, Rimet Y, Nicaise C, Simeoni U. Apparent life-threateningevents in presumably healthy newborns during early skin-to-skin contact.Pediatrics 2011;127:e1073e6.

7. Adamkin DH, Committee on Fetus and Newborn. Postnatal glucose homeo-stasis in late-preterm and term infants. Pediatrics 2011;127:575e9.

8. Aziz K, Fetus and Newborn Committee. Canadian Paediatric Society (CPS).Screening guidelines for newborns at risk for low blood glucose. Paediatr ChildHealth 2004;9:723e40.

9. Schwartz RH, Hey EN, Baum JD. Management of the newborn’s thermalenvironment. In: Sinclair JC, editor. Temperature regulation and energymetabolism in the newborn. New York: Grune & Stratton Ltd; 1978. p. 205e25.

10. Sauer PJ, Dane HJ, Visser HK. New standards for neutral thermal environmentof healthy very low birthweight infants in week one of life. Archs Dis Childh1984;59:18e22.

11. Ziegler EE. Meeting the nutritional needs of the low-birth-weight infant. AnnNutr Metab 2011;58(Suppl. 1):8e18.

12. Ehrenkranz RA. Early nutritional support and outcomes in ELBW infants. EarlyHum Dev 2010;86(Suppl. 1):21e5.

13. Ehrenkranz RA, Younes N, Lemons JA, et al. Longitudinal growth of hospital-ized very low birth weight infants. Pediatrics 1999;104:280e9.

14. Henderson G, Craig S, Brocklehurst P, McGuire W. Enteral feeding regimensand necrotising enterocolitis in preterm infants: a multicentre caseecontrolstudy. Archs Dis Childh Fetal Neonatal Ed 2009;94:F120e3.

*15. Morgan J, Young L, McGuire W. Delayed introduction of progressive enteralfeeds to prevent necrotising enterocolitis in very low birth weight infants.Cochrane Database Syst Rev 2011;3:CD001970.

*16. Morgan J, Young L, McGuire W. Slow advancement of enteral feed volumes toprevent necrotising enterocolitis in very low birth weight infants. CochraneDatabase Syst Rev 2011;3:CD001241.

17. Bombell S, McGuire W. Early trophic feeding for very low birth weight infants.Cochrane Database Syst Rev 2009;3:CD000504.

18. Van Goudoever JB, Colen T, Wattimena JL, Huijmans JG, Carnielli VP, Sauer PJ.Immediate commencement of amino acid supplementation in preterminfants: effect on serum amino acid concentrations and protein kinetics on thefirst day of life. J Pediatr 1995;127:458e65.

19. Rivera Jr A, Bell EF, Bier DM. Effect of intravenous amino acids on proteinmetabolism of preterm infants during the first three days of life. Pediatr Res1993;33:106e11.

20. Engle WA, Tomashek KM, Wallman C. Committee on Fetus and Newborn,American Academy of Pediatrics. “Late-preterm” infants: a population at risk.Pediatrics 2007;120:1390e401.

*21. Whyte RK, Fetus and Newborn Committee, Canadian Paediatric Society. Safedischarge of the late preterm infant. Paediatr Child Health 2010;15:655e60.

22. Pulver LS, Denney JM, Silver RM, Young PC. Morbidity and discharge timing oflate preterm newborns. Clin Pediatr 2010;49:1061e7.

23. American Academy of Pediatrics Committee on Fetus and Newborn. Hospitalstay for healthy term newborns. Pediatrics 2004;113:1434e6.

24. MacMillan D, Fetus and Newborn Committee, Canadian Paediatric Society,Maternal Fetal Medicine Committee, Society of Obstetricians and

R.K. Whyte / Seminars in Fetal & Neonatal Medicine 17 (2012) 153e158158

Gynaecologists of Canada, Clinical Practice Obstetrics Committee, Society ofObstetricians and Gynaecologists of Canada. Facilitating discharge homefollowing a normal term birth. Paediatr Child Health 1996;1:165e8.

25. National Institute for Health and Clinical Excellence. Routine post-natal care ofwomen and their babies. Clinical Guideline 37. London: NICE; 2006.

*26. Merritt TA, Pillers D, Prows SL. Early NICU discharge of very low birth weightinfants: a critical review and analysis. Semin Neonatol 2003;8:95e115.

27. American Academy of Pediatrics Committee on Fetus and Newborn. Hospitaldischarge of the high-risk neonate. Pediatrics 2008;122:1119e26.

28. Rawlings JS, Scott JS. Postconceptional age of surviving preterm low-birth-weight infants at hospital discharge. Arch Pediatr Adolesc Med 1996;150:260e2.

29. Eichenwald EC, Aina A, Stark AR. Apnea frequently persists beyond termgestation in infants delivered at 24 to 28 weeks. Pediatrics 1997;100:354e9.

30. American Academy of Pediatrics Subcommittee on Hyperbilirubinemia.Management of hyperbilirubinemia in the newborn infant 35 or more weeksof gestation. Pediatrics 2004;114:297e316.

31. Maisels MJ, McDonagh AF. Phototherapy for neonatal jaundice. N Engl J Med2008;358:920e8.

32. Sarici SU, Serdar MA, Korkmaz A, et al. Incidence, course, and prediction ofhyperbilirubinemia in near-term and term newborns. Pediatrics2004;113:775e80.

*33. Watchko JF, Maisels MJ. Jaundice in low birthweight infants: pathobiology andoutcome. Archs Dis Childh Fetal Neonatal Ed 2003;88:F455e8.

34. Maisels MJ, Watchko JF. Treatment of jaundice in low birthweight infants.Archs Dis Childh Fetal Neonatal Ed 2003;88:F459e63.

35. Bhutani VK, Johnson L. Kernicterus in late preterm infants cared for as termhealthy infants. Semin Perinatol 2006;30:89e97.

36. Newman TB, Xiong B, Gonzales VM, Escobar GJ. Prediction and prevention ofextreme neonatal hyperbilirubinemia in a mature health maintenance orga-nization. Arch Pediatr Adolesc Med 2000;154:1140e7.

*37. New K, Flenady V, Davies MW. Transfer of preterm infants from incubator toopen cot at lower versus higher body weight. Cochrane Database Syst Rev2011;9:CD004214.

38. New K, Flint A, Bogossian F, East C, Davies MW. Transferring preterm infantsfrom incubators to open cots at 1600 g: a multicentre randomised controlledtrial. Archs Dis Childh Fetal Neonatal Ed 2011 Aug 13 [Epub ahead of print].

39. Zecca E, Corsello M, Priolo F, Tiberi E, Barone G, Romagnoli C. Early weaningfrom incubator and early discharge of preterm infants: randomized clinicaltrial. Pediatrics 2010;126:e651e6.

40. Wigfield RE, Fleming PJ, Azaz YE, et al. How much wrapping do babies need atnight? Archs Dis Childh 1993;69:181e6.

*41. Fleming PJ, Gilbert R, Azaz Y, et al. Interaction between bedding and sleepingposition in the sudden infant death syndrome: a population based caseecontrolstudy. BMJ 1990;301:85e9.

42. American Academy of Pediatrics Task Force on Sudden Infant DeathSyndrome. The changing concept of sudden infant death syndrome: diag-nostic coding shifts, controversies regarding the sleeping environment, andnew variables to consider in reducing risk. Pediatrics 2005;116:1245e55.

43. Mizuno K, Ueda A. The maturation and coordination of sucking, swallowing,and respiration in preterm infants. J Pediatr 2003;142:36e40.

44. Berseth CL. Gastrointestinal motility in the neonate. Clin Perinatol1996;23:179e90.

*45. Breton S, Steinwender S. Timing introduction and transition to oral feeding inpreterm infants: current trends and practice. Newborn Infant Nurs Rev2008;8:153e9.

46. Puckett B, Grover VK, Holt T, Sankaran K. Cue-based feeding for preterminfants: a prospective trial. Am J Perinatol 2008;25:623e8.

47. Pinelli J, Symington A. Non-nutritive sucking for promoting physiologicstability and nutrition in preterm infants. Cochrane Database Syst Rev2005;4:CD001071.

48. Sparks JW. Human intrauterine growth and nutrient accretion. Semin Perinatol1984;8:74e93.

49. Whyte RK, Bayley HS, Sinclair JC. Energy intake and the nature of growth inlow birth weight infants. Can J Physiol Pharmacol 1985;63:565e70.

50. Agostoni C, Buonocore G, Carnielli VP, et al. Enteral nutrient supply forpreterm infants: commentary from the European Society of PaediatricGastroenterology, Hepatology and Nutrition Committee on Nutrition. J PediatrGastroenterol Nutr 2010;50:85e91.

51. Arslanoglu S, Moro GE, Ziegler EE. Adjustable fortification of human milkfed to preterm infants: does it make a difference? J Perinatol 2006;26:614e21.

52. Thureen PJ. The neonatologist’s dilemma: catch-up growth or beneficialundernutrition in very low birth weight infants-what are optimal growthrates? J Pediatr Gastroenterol Nutr 2007;45(Suppl. 3):S152e4.

53. van Goudoever JB, Sulkers EJ, Lafeber HN, Sauer PJ. Short-term growth andsubstrate use in very-low-birth-weight infants fed formulas with differentenergy contents. Am J Clin Nutr 2000;71:816e21.

54. Collins CT, Makrides M, McPhee AJ. Early discharge with home support ofgavage feeding for stable preterm infants who have not established full oralfeeds. Cochrane Database Syst Rev 2003;4:CD003743.

55. Amesz EM, Schaafsma A, Cranendonk A, Lafeber HN. Optimal growth andlower fat mass in preterm infants fed a protein-enriched postdischargeformula. J Pediatr Gastroenterol Nutr 2010;50:200e7.

56. Schanler RJ, Shulman RJ, Lau C. Feeding strategies for premature infants:beneficial outcomes of feeding fortified human milk versus preterm formula.Pediatrics 1999;103:1150e7.

57. Zachariassen G, Faerk J, Grytter C, et al. Nutrient enrichment of mother’s milkand growth of very preterm infants after hospital discharge. Pediatrics2011;127:e995e1003.

58. Henderson G, Fahey T, McGuire W. Nutrient-enriched formula versus standardterm formula for preterm infants following hospital discharge. CochraneDatabase Syst Rev 2007;4:CD004696.

59. Brockmann PE, Poets A, Urschitz MS, Sokollik C, Poets CF. Reference values forpulse oximetry recordings in healthy term neonates during their first 5 days oflife. Archs Dis Childh Fetal Neonatal Ed 2011;96:F335e8.

60. Richard D, Poets CF, Neale S, Stebbens VA, Alexander JR, Southall DP. Arterialoxygen saturation in preterm neonates without respiratory failure. J Pediatr1993;123:963e8.

61. Poets CF. When do infants need additional inspired oxygen? A review of thecurrent literature. Pediatr Pulmonol 1998;26:424e8.

62. Henderson-Smart DJ. The effect of gestational age on the incidence andduration of recurrent apnoea in newborn babies. Aust Paediatr J 1981;17:273e6.

*63. Ramanathan R, Corwin MJ, Hunt CE, et al. Cardiorespiratory events recordedon home monitors: comparison of healthy infants with those at increased riskfor SIDS. JAMA 2001;285:2199e207.

64. Darnall RA, Kattwinkel J, Nattie C, Robinson M. Margin of safety for dischargeafter apnea in preterm infants. Pediatrics 1997;100:795e801.

65. Hunt CE, Corwin MJ, Lister G, et al. Longitudinal assessment of hemoglobinoxygen saturation in healthy infants during the first 6 months of age.Collaborative Home Infant Monitoring Evaluation (CHIME) Study Group.J Pediatr 1999;135:580e6.

66. Esani N, Hodgman JE, Ehsani N, Hoppenbrouwers T. Apparent life-threateningevents and sudden infant death syndrome: comparison of risk factors.J Pediatr 2008;152:365e70.

67. Ducrocq S, Biran-Mucignat V, Boelle PY, Lebas F, Baudon JJ, Gold F. Apnea ofprematurity: risk factors and ambulatory treatment with caffeine citrate. ArchPediatr 2006;13:1299e304.

68. Lucas A, King F, Bishop NB. Postdischarge formula consumption in infantsborn preterm. Archs Dis Childh 1992;67:691e2.

69. Merritt TA, Raddish M. A review of guidelines for the discharge of prematureinfants: opportunities for improving cost effectiveness. J Perinatol1998;18:S27e37.

70. Melnyk BM, Feinstein NF, Alpert-Gillis L, et al. Reducing premature infants’length of stay and improving parents’ mental health outcomes with theCreating Opportunities for Parent Empowerment (COPE) neonatal intensivecare unit program: a randomized, controlled trial. Pediatrics 2006;118:e1414e27.

at SciVerse ScienceDirect

Seminars in Fetal & Neonatal Medicine 17 (2012) 159e162

Contents lists available

Seminars in Fetal & Neonatal Medicine

journal homepage: www.elsevier .com/locate/s iny

Health outcomes in infancy and childhood of moderate and late preterm infants

Pooja Harijan, Elaine M. Boyle*

Department of Health Sciences, University of Leicester, 22e28 Princess Road West, Leicester LE1 6TP, UK

Keywords:HealthLate pretermModerate pretermOutcomesPreterm

* Corresponding author. Tel.: þ44 (0) 116 252 5447E-mail address: eb124@leicester.ac.uk (E.M. Boyle)

1744-165X/$ e see front matter � 2012 Elsevier Ltd.doi:10.1016/j.siny.2012.02.002

s u m m a r y

There has been a long-held belief that outcomes for babies born at moderate and late preterm gestationsdo not differ substantially from those of infants born at full term. This has recently been challenged bystudies highlighting an increased risk of adverse neonatal outcomes, and of poorer cognitive, behaviouraland educational outcomes in this population. Data about the effects of birth at moderate and latepreterm gestations on later health outcomes are limited, but emerging evidence suggests that ongoingphysical health may also be worse in those born just a few weeks before full term. This reviewsummarises the available evidence, considers the factors influencing health outcomes and discusses theimplications for the planning and provision of children’s health care services.

� 2012 Elsevier Ltd. All rights reserved.

1. Introduction

Historically, interest in health outcomes of preterm infants hasbeen centred onverypreterm infants, i.e. thoseborn at<32weeksofgestation. This is understandable, as this group includes thosebabiesfor whom severity of neonatal illness is likely to be greatest and inwhomadverse long-termoutcomesaremost likely tooccur.1e4Mostneonatologists, therefore, ‘routinely’ follow up infants born at <32weeks until around 2 years of age andmost research studies to datehave been concentrated on this population. By contrast, few centreshave a programmeof ongoing surveillance formoremature preterminfants, as their outcomes have been assumed to be similar to thoseof term-born infants. In theUK there is currentlyno routineneonataldata collection for this group. However, it has become apparentover recent years, from retrospective analyses of large cohorts inNorth America, that the risk of significant neonatal morbidity isgreater inmoderate (32e33weeks) and late (34e36weeks) preterminfants compared with infants born at �37 weeks of gestation.5e8

Published data on later health outcomes in children who havebeen born at 32e36 weeks of gestation are relatively few, butemerging evidence suggests that for some, health problems maypersist into later infancy and childhood.9

2. Hospitalisation during infancy and childhood

Increased likelihood of readmission of late preterm infants tohospital following discharge from neonatal care is well

..

All rights reserved.

documented. In a US study, 4.4% of 2153 infants born at 34 to 36þ6

weeks of gestation were readmitted within the first 15 days afterdischarge, compared with 2.0% of 30 261 infants born at >37weeks.10 Following early neonatal discharge in a UK cohort, read-mission in the first month occurred in 6.3% of infants born at 35e37weeks of gestation compared with 3.4% and 2.4% respectively ofthose born at 38e40 and >40 weeks.11 Jain and Cheng found thatemergency department attendances in the neonatal period werealso increased in late preterm infants, with the greatest proportionof these being in infants born at 36 weeks of gestation.12 Jaundicehas consistently been found to be the most frequent reason forhospital readmission in the first month of life.10,13,14 Other commonreasons are infection,14 feeding difficulties and dehydration.5 Animportant risk factor for early post-discharge morbidity appears tobe failure to successfully establish adequate breastfeeding,13,15 butearly hospital discharge has also been implicated.14

Hospitalisation beyond the neonatal period has been less wellinvestigated. Escobar found that late preterm infants were morelikely to require at least one hospital admission within the first sixmonths after birth.5 McLaurin et al.16 showed that health care-related costs were higher for late preterm infants than for termborn infants, with the greatest costs being related to hospitalinpatient admissions; regardless of the timing of discharge fromneonatal care, infants born late pretermwere almost twice as likelyto have received inpatient hospital care during the first year.Secondary analysis of data from the Millennium Cohort Study(MCS), a UK nationally representative prospective cohort study,examined parental reports of their children’s hospital admissions.Compared with infants born at full term (39e41 weeks of gesta-tion), the odds of having three or more admissions within the firstninemonths of lifewas higher for bothmoderate preterm [adjusted

P. Harijan, E.M. Boyle / Seminars in Fetal & Neonatal Medicine 17 (2012) 159e162160

odds ratio (aOR): 7.8; 95% CI: 2.9e20.7] and late preterm (aOR: 5.1;95% CI: 3.0e8.8) infants. The most common reasons for hospital-isation were respiratory disorders such as chest infection andwheezing, and gastrointestinal problems including infection andgastro-oesophageal reflux. The odds of being admitted to hospitalthree or more times between nine months and 5 years of age werealso higher in children born moderate and late preterm than inthose born at full term (aOR: 3.0; 95% CI: 1.4e6.2; and 1.9; 95% CI:1.3e2.7 respectively).9

3. Respiratory morbidities

Acute respiratory disorders immediately after birth such asrespiratory distress syndrome, transient tachypnoea of thenewborn and pneumonia are more common, and incur a greaterrisk of respiratory failure, in late preterm than in full term infants.17

This increased respiratory morbidity is thought to be related tofunctional immaturity of the lung structure at 34e37 weeks ofgestation, which predisposes to delayed intrapulmonary fluidabsorption, surfactant insufficiency, and ultimately to inefficientgas exchange.18

The risk of ongoing respiratory morbidity after the earlyneonatal period is unclear. Although respiratory outcomes havebeen investigated in only a small number of studies, some suggesta continued risk of respiratory problems into infancy and child-hood. It has been suggested that preterm birth, even when therehas been no significant neonatal respiratory disease, may haveadverse effects on lung growth and development with subsequentreduced pulmonary function.19

Kotecha et al.20 conducted an analysis of data from childrenparticipating in the Avon Longitudinal Study of Parents and Chil-dren (ALSPAC; n ¼ 14 049) who had spirometry measurementstaken at ages 8e9 years (n ¼ 6705) and/or 14e17 years (n ¼ 4508).This study included 691 children who had been born between 33and 36 weeks of gestation. The researchers demonstrateddecreased lung function at school age in children born at 33e34weeks of gestation. At 8e9 years of age, all spirometry measuresin the 33e34 week gestation group were significantly lower than inthose born at �37 weeks and spirometry decrements were similarto those observed in children who had been born between 25 and32 weeks of gestation. At 14e17 years of age, some differencesbetween the groups had resolved, but forced expiratory volume(FEV1)/forced vital capacity (FVC) and forced expiratory flow at25e75% FVC (FEF 25e75%) remained significantly lower in thepreterm group than that in term controls. By contrast, children bornat 35e36 weeks had values similar to those of children born at �37weeks. As might be expected, mechanical ventilation in theneonatal period was more common in the lower gestational agegroups; this may contribute to the later differences observed.Although there was considerable loss to follow-up from this study,non-attenders were from families of lower socioeconomic statusand in which maternal smoking was more prevalent, raising thepossibility that the study results might provide an under-representation of the true differences.

It has been suggested that there is increased susceptibility torespiratory infection in infants born at moderate and late pretermgestations. In particular, a number of studies have investigated therisk of respiratory syncytial virus (RSV) bronchiolitis.21 There issome evidence for this being associated with risk of wheezing andother respiratory events in the year following RSV infection, toa greater extent than in term infants.22,23 Law et al. conducteda multicentre prospective cohort study of infants born at 33e35weeks of gestation to identify risk factors for RSV infectionrequiring hospitalisation in this group. The most important of thesewere day-care attendance, birth during winter months and

preschool age sibling(s).24 A risk assessment tool for RSV infectionin late preterm infants was devised25 and its predictive validity wasfound to be good.26 A Canadian health economics study suggestedthat the use of prophylaxis may be cost-effective in the subgroup ofmoderate to late preterm infants identified as being at high risk forRSV infection.27

Asthma is a major cause of respiratory morbidity in children.There is conflicting evidence regarding the association betweenmoderate and late preterm birth and a diagnosis of asthma in earlychildhood. A retrospective cohort study included 582 late preterminfants born in 2007 and found evidence of more diagnoses ofpersistent asthma (aOR: 1.68; 95% CI: 1.01e2.80), inhaled cortico-steroid use (aOR: 1.66; 95% CI: 1.20e2.29), and numbers of acuterespiratory visits (incidence rate ratio 1.44; 95% CI: 1.24e1.67) inlate preterm infants compared with those born at 37e38 weeks ofgestation.28 Vogt et al. examined inhaled corticosteroid use in6e19-year-olds and found an increased use of medication in thoseborn at 33e34 weeks’ (aOR: 1.35; 95% CI: 1.25e1.46) and 35e36weeks’ gestation (aOR: 1.24; 95% CI: 1.19e1.30) compared withthose who were born at more mature gestations.29 In the Millen-nium Cohort Study group, Boyle et al.9 found an increased inci-dence of asthma and wheeze at 3 and 5 years in children bornmoderate or late preterm, and medication for asthmawas the mostprescribed medication at 5 years (aOR: 2.2; 95% CI: 1.6e3.1). Bycontrast, Abe et al.30 found no statistically significant associationbetween late preterm birth and the risk of developing asthma inchildhood.

4. Feeding and growth

Adequate feeding in infancy is crucial for healthy weight gain.Ongoing feeding problems are not infrequently seen in verypreterm infants and are often attributed to immaturity of suck andswallowing coordination. There is, however, very limited evidencein more mature preterm infants, and the number in this groupaffected by feeding difficulties is not known. A prospective cohortstudy used parent questionnaires to collect information aboutfeeding behaviour during the first year of life in infants who wereborn preterm, from 25 to 36 weeks of gestation.31 The studyexamined outcomes such as appetite, oromotor dysfunction andfeeding avoidance, comparing 571 late preterm infants with 319infants born before 34 weeks. Although the researchers had ex-pected to find evidence of more mature feeding patterns andfewer difficulties in the late preterm infants, parents in bothgroups reported similar rates of feeding dysfunction. Feedingdifficulties improved during the first 12 months of life in bothgroups and episodes of hospitalisation for feeding problems didnot differ between the two groups. The study was limited by thelack of comparison with term-born infants, but is supportive of thefindings of an earlier small study in 20 infants of 32e37 weeks’gestation, who showed delayed feeding development comparedwhen compared with 10 full term infants at 11e17 months ofage.32

Whereas very preterm infants are known to have patterns oflater growth and weight gain that are poorer than those of term-born infants, there is a paucity of data for those born between 32and 36 weeks. A Brazilian population-based study found signifi-cantly increased risks of poor weight and height attainment in latepreterm infants at 12 and 24 months of age (aOR: 3.36; 95% CI:1.56e7.23; and aOR: 2.30; 95% CI: 1.40e3.77, respectively)compared with those born at term.33 Data from the MillenniumCohort Study9 showed lower height attainment at both 3 and 5years in late preterm infants when compared with infants born at39e41 weeks.

Research directions

Contemporary prospective longitudinal studies are neededto:

� clarify long term outcomes of moderate and latepreterm infants;

� identify risk factors and predictors for poor healthoutcomes in children born moderate or late preterm;

� determine appropriate follow-up strategies for infantsborn at moderate and late preterm gestations.

Practice points

� Children born moderate or late preterm are more likelyto be admitted to hospital in infancy and childhood thanthose born at full term.

� Children born moderate or late preterm may be moresusceptible to respiratory disease than those born atfull term.

� Large numbers of moderate and late preterm infantsmean that the impact of this group on health careservices may be substantial.

P. Harijan, E.M. Boyle / Seminars in Fetal & Neonatal Medicine 17 (2012) 159e162 161

5. Longstanding illness

A large longitudinal study of all births in Norway between 1967and 1983 showed that those born late preterm had an increasedrisk of major disabilities that included epilepsy and visual andhearing impairments.34 As adults, they were more likely to expe-rience disability affecting their work capacity. Advances inobstetric, neonatal and paediatric care since the birth of thesechildren may mean that these findings are less relevant today.However, more recent data from the Millennium Cohort Studyshowed that longstanding illness at 3 and 5 years was morecommon in children bornmoderate or late preterm than those bornat full term and that their illness was more likely to limit theiractivities.9 Parents of the children born pretermweremore likely toreport their child’s health as being poor than those with term-bornchildren. There has been only one study assessing health-relatedquality of life in children born at 32e36 weeks of gestation.35

This study included 362 pre-school children and showed lowerhealth-related quality of life scores for lung problems in those whohad received continuous positive airway pressure in the neonatalperiod.

6. Mortality

A number of studies have consistently reported increasedmortality during the neonatal period in infants born moderate andlate preterm,36e39 but until recently there have been no publisheddata specifically addressing the relationship between gestationalage at birth and later mortality. A national cohort study in Swedenof 674 820 singletons, born between 1973 and 1979, recently re-ported a gradient of increasing mortality with decreasing gesta-tion.40 The study included 22 590 late preterm infants, and birthbetween 34 and 36 weeks was associated with increased mortalityin early childhood (age 1e5 years) and young adulthood (aged18e36 years). The adjusted hazard ratio for all-cause mortality inthe late preterm group compared with the 37e42 week gestationgroup was 1.31 (95% CI: 1.13e1.5). There was a non-significantassociation between late preterm birth and mortality in latechildhood and adolescence. The strongest associations withmortality in adulthood were congenital anomalies, respiratory,endocrine and cardiovascular disorders; no association was foundwith mortality from neurological disorders, cancer or injury.Sensitivity analyses showed that the association was not explainedby congenital anomalies.

7. Influences on health outcomes in moderate and latepreterm infants

Despite a recent increase in interest in moderate and latepreterm birth, considerably less is known about the long termrisks associated with birth at more mature preterm gestationscompared with those of very preterm birth. Nevertheless,although sparse, much of the available information to date pointsto worse outcomes in this group than in individuals born at term.Also poorly understood are the factors leading to birth at 32e36weeks of gestation and those influencing long term health ininfants, children and adults. Such influences may originate beforeor during pregnancy, in the perinatal or neonatal period or duringchildhood. In deciding to what extent outcomes are due toprematurity per se, many other factors must be considered. Infantswith congenital anomalies and multiple births commonly deliverspontaneously at moderate and late preterm gestations. In addi-tion, obstetric concerns about either maternal or fetal conditionoften lead to medically indicated deliveries during this period ofgestation. Obstetric decision-making with respect to indications

for delivery, timing of delivery and mode of delivery may thereforeinfluence outcomes. During the neonatal period place of care,discharge policies and duration of postnatal hospitalisation varyand the consequences of this variation have not been fullyexplored. Following discharge, parental education and familylifestyle are other factors that may potentially affect longer termhealth outcomes in these infants. Many of these are inextricablylinked and their relative contributions to infant and child healthoutcomes remain to be disentangled.

8. Implications for provision of children’s health care services

Reasons for preterm birth at 32e36 weeks of gestation are notalways clearly understood, but a recent small decrease in the ratesof late preterm birth suggests that strategies for reducing avoidablebirths at these gestations may be having some success. Despite this,moderate and late preterm births continue to represent >75% of allpreterm births. Emerging evidence about health outcomes in theseinfants appears to mirror reports of worse neurodevelopmentaland educational outcomes, although contemporary and prospec-tive data remain limited. Poor health in large numbers of infantsand children is likely to have a substantial impact on primary careand public health services. Many preterm infants born at �32weeks do not receive follow-up or surveillance for their health careor developmental needs. Now that these children are increasinglybeing highlighted as a group at greater risk than has previouslybeen appreciated, it is important to ensure that their ongoinghealth care needs are being adequately met. Reports of adverseoutcomes have led to many calls for routine follow-up in this groupuntil at least 18 months of age.41e43 Such a strategy would haveimportant financial and logistic implications for the provision ofhealth care services for children. Further work is required toidentify those at greatest risk of poor health outcomes in order toallow appropriate and, if possible, targeted follow-up in childrenborn moderate or late preterm.

P. Harijan, E.M. Boyle / Seminars in Fetal & Neonatal Medicine 17 (2012) 159e162162

Conflict of interest statement

None declared.

Funding sources

None.

References

1. Hack M. Young adult outcomes of very-low-birth-weight children. Semin FetalNeonatal Med 2006;11:127e37.

2. Vohr BR, Wright LL, Dusick AM, et al. Neurodevelopmental and functionaloutcomes of extremely low birth weight infants in the National Institute ofChild Health and Human Development Neonatal Research Network,1993e1994. Pediatrics 2000;105:1216e26.

3. Wood NS, Costeloe K, Gibson AT, Hennessy EM, Marlow N, Wilkinson AR. TheEPICure study: growth and associated problems in children born at 25 weeks ofgestational age or less. Arch Dis Child Fetal Neonatal Ed 2003;88:F492e500.

4. Wood NS, Marlow N, Costeloe K, Gibson AT, Wilkinson AR. Neurologic anddevelopmental disability after extremely preterm birth. EPICure Study Group.N Engl J Med 2000;343:378e84.

5. Escobar GJ, Clark RH, Greene JD. Short-term outcomes of infants born at 35 and36 weeks gestation: we need to ask more questions. Semin Perinatol2006;30:28e33.

6. Khashu M, Narayanan M, Bhargava S, Osiovich H. Perinatal outcomes associ-ated with preterm birth at 33 to 36 weeks’ gestation: a population-basedcohort study. Pediatrics 2009;123:109e13.

7. Kitsommart R, Janes M, Mahajan V, et al. Outcomes of late-preterm infants:a retrospective, single-center, Canadian study. Clin Pediatr 2009;48:844e50.

8. Wang ML, Dorer DJ, Fleming MP, Catlin EA. Clinical outcomes of near-terminfants. Pediatrics 2004;114:372e6.

9. Boyle EM, Poulsen G, Field DJ, et al. Population-based cohort study of the effectsof gestational age at birth on health outcomes at three and five years. BMJ2012;344:e896.

10. Escobar GJ, Greene JD, Hulac P, et al. Rehospitalisation after birth hospitalisation:patterns among infants of all gestations. Archs Dis Childh 2005;90:125e31.

11. Oddie SJ, Hammal D, Richmond S, Parker L. Early discharge and readmission tohospital in the first month of life in the Northern Region of the UK during 1998:a case cohort study. Archs Dis Childh 2005;90:119e24.

12. Jain S, Cheng J. Emergency department visits and rehospitalizations in latepreterm infants. Clin Perinatol 2006;33:935e45. abstract xi.

13. Shapiro-Mendoza CK, Tomashek KM, Kotelchuck M, Barfield W, Weiss J,Evans S. Risk factors for neonatal morbidity and mortality among “healthy,”late preterm newborns. Semin Perinatol 2006;30:54e60.

14. Tomashek KM, Shapiro-Mendoza CK, Weiss J, et al. Early discharge among latepreterm and term newborns and risk of neonatal morbidity. Semin Perinatol2006;30:61e8.

15. Radtke JV. The paradox of breastfeeding-associated morbidity among latepreterm infants. J Obstet Gynecol Neonatal Nurs 2011;40:9e24.

16. McLaurin KK, Hall CB, Jackson EA, Owens OV, Mahadevia PJ. Persistence ofmorbidity and cost differences between late-preterm and term infants duringthe first year of life. Pediatrics 2009;123:653e9.

17. Hibbard JU, Wilkins I, Sun L, et al. Respiratory morbidity in late preterm births.JAMA 2010;304:419e25.

18. Dudell GG, Jain L. Hypoxic respiratory failure in the late preterm infant. ClinPerinatol 2006;33:803e30. abstract viiieix.

19. Colin AA, McEvoy C, Castile RG. Respiratory morbidity and lung function inpreterm infants of 32 to 36 weeks’ gestational age. Pediatrics2010;126:115e28.

20. Kotecha SJ, Watkins WJ, Paranjothy S, Dunstan FD, Henderson AJ, Kotecha S.Effect of late preterm birth on longitudinal lung spirometry in school agechildren and adolescents. Thorax 2012;67:54e61.

21. Resch B, Paes B. Are late preterm infants as susceptible to RSV infection as fullterm infants? Early Hum Dev 2011;87(Suppl. 1):S47e9.

22. Palmer L, Hall CB, Katkin JP, et al. Healthcare costs within a year of respiratorysyncytial virus among Medicaid infants. Pediatr Pulmonol 2010;45:772e81.

23. Shi N, Palmer L, Chu BC, et al. Association of RSV lower respiratory tractinfection and subsequent healthcare use and costs: a Medicaid claims analysisin early-preterm, late-preterm, and full-term infants. J Med Econ2011;14:335e40.

24. Law BJ, Langley JM, Allen U, et al. The Pediatric Investigators CollaborativeNetwork on Infections in Canada study of predictors of hospitalization forrespiratory syncytial virus infection for infants born at 33 through 35completed weeks of gestation. Pediatr Infect Dis J 2004;23:806e14.

25. Sampalis JS, Langley J, Carbonell-Estrany X, et al. Development and validationof a risk scoring tool to predict respiratory syncytial virus hospitalization inpremature infants born at 33 through 35 completed weeks of gestation. MedDecision Making 2008;28:471e80.

26. Paes B, Steele S, Janes M, Pinelli J. Risk-Scoring Tool for respiratory syncytialvirus prophylaxis in premature infants born at 33e35 completed weeks’gestational age in Canada. Curr Med Res Opin 2009;25:1585e91.

27. Lanctot KL, Masoud ST, Paes BA, et al. The cost-effectiveness of palivizumab forrespiratory syncytial virus prophylaxis in premature infants with a gestationalage of 32e35 weeks: a Canadian-based analysis. Curr Med Res Opin2008;24:3223e37.

28. Goyal NK, Fiks AG, Lorch SA. Association of late-preterm birth with asthma inyoung children: practice-based study. Pediatrics 2011;128:e830e8.

29. Vogt H, Lindstrom K, Braback L, Hjern A. Preterm birth and inhaled cortico-steroid use in 6- to 19-year-olds: a Swedish national cohort study. Pediatrics2011;127:1052e9.

30. Abe K, Shapiro-Mendoza CK, Hall LR, Satten GA. Late preterm birth and risk ofdeveloping asthma. J Pediatr 2010;157:74e8.

31. DeMauro SB, Patel PR, Medoff-Cooper B, Posencheg M, Abbasi S. Postdischargefeeding patterns in early- and late-preterm infants. Clin Pediatr2011;50:957e62.

32. Dodrill P, McMahon S, Ward E, Weir K, Donovan T, Riddle B. Long-term oralsensitivity and feeding skills of low-risk pre-term infants. Early Hum Dev2004;76:23e37.

33. Santos IS, Matijasevich A, Domingues MR, Barros AJ, Victora CG, Barros FC. Latepreterm birth is a risk factor for growth faltering in early childhood: a cohortstudy. BMC Pediatr 2009;9:71.

34. Moster D, Lie RT, Markestad T. Long-term medical and social consequences ofpreterm birth. N Engl J Med 2008;359:262e73.

35. Ketharanathan N, Lee W, de Mol AC. Health-related quality of life, emotionaland behavioral problems in mild to moderate prematures at (pre-)school age.Early Hum Dev 2011;87:705e9.

36. Kramer MS, Demissie K, Yang H, Platt RW, Sauve R, Liston R. The contributionof mild and moderate preterm birth to infant mortality. Fetal and Infant HealthStudy Group of the Canadian Perinatal Surveillance System. JAMA2000;284:843e9.

37. McIntire DD, Leveno KJ. Neonatal mortality and morbidity rates in late pretermbirths compared with births at term. Obstet Gynecol 2008;111:35e41.

38. Tomashek KM, Shapiro-Mendoza CK, Davidoff MJ, Petrini JR. Differences inmortality between late-preterm and term singleton infants in the UnitedStates, 1995e2002. J Pediatr 2007;151:450e6. 56 e1.

39. Young PC, Glasgow TS, Li X, Guest-Warnick G, Stoddard G. Mortality of late-preterm (near-term) newborns in Utah. Pediatrics 2007;119:e659e65.

40. Crump C, Sundquist K, Sundquist J, Winkleby MA. Gestational age at birth andmortality in young adulthood. JAMA 2011;306:1233e40.

41. Chyi LJ, Lee HC, Hintz SR, Gould JB, Sutcliffe TL. School outcomes of late preterminfants: special needs and challenges for infants born at 32 to 36 weeksgestation. J Pediatr 2008;153:25e31.

42. Petrini JR, Dias T, McCormick MC, Massolo ML, Green NS, Escobar GJ. Increasedrisk of adverse neurological development for late preterm infants. J Pediatr2009;154:169e76.

43. Talge NM, Holzman C, Wang J, Lucia V, Gardiner J, Breslau N. Late-preterm birthand its association with cognitive and socioemotional outcomes at 6 years ofage. Pediatrics 2010;126:1124e31.

at SciVerse ScienceDirect

Seminars in Fetal & Neonatal Medicine 17 (2012) 163e169

Contents lists available

Seminars in Fetal & Neonatal Medicine

journal homepage: www.elsevier .com/locate/s iny

School outcome, cognitive functioning, and behaviour problems in moderateand late preterm children and adults: A review

Marjanneke de Jong*, Marjolein Verhoeven, Anneloes L. van BaarDepartment of Child and Adolescent Studies, Utrecht University, The Netherlands

Keywords:Behaviour problemsCognitive functioningLate pretermModerate pretermPsychiatric disordersSchool problems

* Corresponding author. Address: P.O. Box 80.1Netherlands. Tel.: þ31 30 2534601.

E-mail address: M.deJong1@uu.nl (M. de Jong).

1744-165X/$ e see front matter � 2012 Elsevier Ltd.doi:10.1016/j.siny.2012.02.003

s u m m a r y

A large number of children (6 to 11% of all births) are born at a gestational age between 32 and 36 weeks.Little is known of long term outcomes for these moderate and late preterm children. In this review,results of 28 studies on school outcome, cognitive functioning, behaviour problems, and psychiatricdisorders are presented. Overall, more school problems, less advanced cognitive functioning, morebehaviour problems, and higher prevalence of psychiatric disorders were found in moderate and latepreterm born infants, children, and adults compared with full term peers. Suggestions for future researchare discussed.

� 2012 Elsevier Ltd. All rights reserved.

1. Introduction

Many infants are born too soon and therefore are at risk fordevelopmental problems. In 2007, 10.7% of all children in the USA(6.1% in The Netherlands1) were born at a gestational age between32 and 36 weeks, i.e. moderate (32e33 weeks) and late (34e36weeks) preterm.2 This entails 84% of all preterm births. Duringthe last two decades, the number of preterm births increased byalmost 25%, with a 4% increase in very preterm births (i.e. below 32weeks’ gestation) versus a 30% increase in moderate and latepreterm births.2 Hence, the number of children in this groupespecially has increased over the last 20 years. Although mortalityrates are much lower (7.5 times) in these children than in verypreterm children, these rates are almost 10 times higher than in fullterm children.3 With respect to neonatal complications, moderateand late preterm infants experience fewer illnesses than verypreterm infants, but they are at elevated risk for breathing andfeeding difficulties, hypoglycaemia, and hyperbilirubinaemiacompared with full term infants.4 Another risk factor results fromthe fact that the brain of moderate and late preterm children is stillimmature; at 34 weeks of gestation, the brain weighs only 65% ofthe weight at 40 weeks of gestation.5 Despite the high prevalenceand increasedmedical risk of moderate and late preterm birth, littleis known of long-term developmental outcomes of these children.In this review the available information on school outcome,cognitive functioning, behaviour problems, and psychiatric

40, 3508 TC Utrecht, The

All rights reserved.

disorders of infants, children, and adults born moderate or latepreterm is presented.

2. Method

2.1. Search strategy and study selection

The Scopus database was searched until 23 June 2011, using theterms: ‘late preterm’, ‘moderately preterm’, and ‘moderatepreterm’. In addition, reference lists of selected articles wereexamined to find additional studies.

Studies were included if: (i) these were published after 1January 2000; (ii) gestational age of participants was between 32and 36þ6 weeks; and (iii) these investigated school outcome,cognitive functioning, behaviour problems or psychiatric disorders.

3. Results

3.1. Included studies

The three search terms resulted in 485 hits. Based on title andabstract, 449 studies were excluded. Thirty-six articles were readfull text and 16 of these were excluded because outcome measureswere beyond the scope of this review (n ¼ 11) or because thestudies did not concern empirical data collection and analyses(n ¼ 5). In four of the included studies, gestational age of theparticipants differed slightly from our inclusion criterion (i.e.selected 31 and/or 37 weeks of gestation).6e9 Nevertheless, wedecided to include these studies as most participants fell in ourdefined range of gestational age. An additional eight papers were

M. de Jong et al. / Seminars in Fetal & Neonatal Medicine 17 (2012) 163e169164

included, based on the reference lists of the selected articles. Finallya total number of 28 papers were included in this review.

3.2. Characteristics of included studies

Characteristics and results of the included studies are presentedin Table 1. Regarding the participants, 17 studies focused onmoderate and late preterm children (32e36 weeks of gestationalage),6e8,10e23 10 focused on late preterm children (34e36 weeks ofgestational age),9,24e32 and one focused on children born between32e34 weeks of gestational age.33 Twenty-three studies investi-gated outcomes in infancy or childhood (0e15 years ofage),8e11,13,15e17,19e33 four studies reported outcomes in (young)adulthood (18e36 years of age),7,12,14,18 and one reported onoutcomes in both childhood and adulthood.6 School outcome wasinvestigated in seven studies,6,11,16,17,23,27,29 cognitive functioning in19,6,7,9,12e15,20e26,29e33 behaviour problems in six,8,13,16,23,25,29 andpsychiatric disorders in seven.7,10,12,18,19,21,28

3.3. School outcome

All seven studies found an increased risk for school problems formoderate and late preterm children compared with full termchildren. Compared with full term children, they were at 1.3e2.8-fold increased risk for attending special education11,16,23,29 andhad repeated grades between 1.3 and 2.2 times more often at 5e10years of age.23,27 Three studies focused on specific school abilities in5e10-year-olds and found lower scores for moderate/late pretermchildren on reading, spelling, and arithmetic compared with fullterm children.6,11,16 A higher risk for reading and spelling difficul-ties was also found in the preterm children at age 9e11 years.17

With regard to educational attainment in adults, it was foundthat individuals born moderate or late preterm had more learning-related disabilities at age 7 years, which was related to lowereducational attainment in adulthood.6

3.4. Cognitive functioning

Five studies investigated cognitive functioning in infancy (0e24months of age).9,13,15,20,24 One of these found lower developmentalscores for preterm children compared with full term children at 24months of age (not corrected for prematurity).24 In a longitudinalstudy by Romeo et al.,20 a delay in cognitive functioning was foundfor children born at 33e36 weeks compared with full term childrenat 12 and 18 months of age. These group differences were no longersignificant when age was corrected for prematurity. Two otherstudies that corrected age for prematurity also found no differencesin scores between moderate/late preterm and full term children at12 and 24 months of age.9,13 Hillemeier et al.15 performed a longi-tudinal study in which cognitive functioning was investigated at 9and 24 months of age (corrected for prematurity). They founda two-fold higher risk for scoring in the lowest 10% inmoderate/latepreterm children compared with full term children at 9 months ofage, but this difference was no longer seen at 24 months of age.

Different results emerge from 11 studies concerning cognitivefunctioning in moderate/late preterm children at 3e15 years ofage.6,21e23,25,26,29e33 Two studies on large samples of more than100,000 and 300,000 children found an increased risk of 1.3e1.9 formental retardation (i.e. IQ scores <70) in moderate/late pretermchildren compared with full term children.21,26 In line with this, thepreterm children were twice as likely to score <85 on an IQ testthan their full term peers at age 5, 6 and 10 years.22,25,33 In two ofthese studies (at ages 5 and 6 years), no differences were foundbetween mean scores of moderate/late preterm and full termchildren,25,33 which suggests a different distribution in scores with

more moderate/late preterm children showing developmentaldelay than in the population. Six studies did find differences inmean IQ scores between moderate/late preterm and full termchildren aged 3e10 years, with preterm children scoring belowtheir full term peers.6,22,23,30e32 Medical risk of the childrenregarding their need for neonatal intensive care unit treatmentseemed to differ between these six studies. Three of these studiesonly investigated moderate/late preterm children with a highmedical risk,22,31,32 one included only low risk children,23 andanother study did not report any neonatal characteristics regardinghigh or low risk.6 Baron et al.30 showed that medical risk might beimportant, as they found lower scores for late preterm childrenonly in those children that were at high risk, but not for low riskchildren. Gurka et al.29 also found no differences in mean IQ scoresbetween low risk late preterm and full term children in theirlongitudinal study following children from 4 to 15 years of age.

Results regarding cognitive functioning in adulthood aresomewhat inconsistent.7,12,14 An increased risk for mental retar-dation was found in adults born moderate/late preterm comparedwith adults born at full term in a large epidemiological study.7

Ekeus et al.14 also found slightly lower scores on a general intel-lectual performance test used for the military service in 18e19-year-old preterm born men, compared with full term. By contrast,Dalziel et al.12 found no differences betweenmoderate/late pretermand full term born adults at 30 years of age regarding their scoreson the Wechsler Abbreviated Scale of Intelligence.

3.5. Behaviour

3.5.1. Behaviour problems as reported by parents and teachersSix studies examined the prevalence of behaviour problems in

moderate/late preterm children as reported by parents andteachers.8,13,16,23,25,29 A longitudinal study showed that at ages 3, 5and 8 years, w20% of the preterm children scored in the clinicalrange of the Child Behavior Check List total problem scale asreported by their parents; twice as much as the expected 10%.8

Likewise, Darlow et al.13 found that parents reported slightly morebehaviour problems in 2-year-old preterm compared with full termchildren, as indicated by the presence of two or more out of fiveproblem behaviours (i.e. poor positive affect, frequent negativeaffect, poor attention, highly active, low self-confidence). Teachersof 6-year-old children reported more internalizing behaviour andattention problems, and slightly more externalizing behaviour inlate preterm children compared with full term children.25 Similarly,in the study byVan Baar et al.,23 bothmothers and teachers reportedmore internalizing behaviour and attention problems in moderate/late preterm children compared with full term children at 8 year ofage. Next to that,mothers reportedmore symptoms of hyperactivityin moderate/late preterm children than in full term children. At 7years of age, Huddy et al.16 also found that parents and teachersrated moderate/late preterm children more often as being hyper-active, but did not report higher levels of emotional symptoms,conduct problems, peer problems, or prosocial behaviour for thesechildren. A longitudinal study investigating children between 4 and15 years of age found no differences in parent-reported external-izing, internalizing, aggressive, and anxiety/depressive behaviourbetween a small group of late preterm (n¼ 53) and a large group offull term children (n ¼ 1254).29

3.5.2. Prevalence of psychiatric disordersSeven studies reported on the prevalence of psychiatric disor-

ders.7,10,12,18,19,21,28 Lindström et al.18 found a 30% higher risk forpsychiatric disorders inmoderate/late pretermadults comparedwithfull term adults, especially in the domain of organic/neuropsychiatricdisorders. Moster et al.7 found a 30e40% higher risk for

Table 1Characteristics and results of included studies.

Authors Gestational agerange (weeks)preterm

Subject characteristics No. of subjects School outcomes Cognitive fun ioning[mean (SD)/( nge)or confidenc nterval]

Behaviour problemsand/or psychiatricdisorders

Nomura et al.6 MPT/LPT: 33e37 7e8 and 27e33 years old;born in 1960e1964

MPT/LPT: 226FT: 1393

Reading: MPT/LPT < FT.Spelling: MPT/LPT < FT.Arithmetic: MPT/LPT < FT(age 7e8 years).This is associated withlower educationalattainment in adulthood.

MPT/LPT < F 89(11.8) vs 94 .6).

Moster et al.7 MPT/LPT: 31e36MPT: 31e33LPT: 34e36

20e36 years old;born in 1967e1983

MPT: 6363LPT: 31 169FT: 828 227

Mental retar tion:MPT: OR ¼ 2LPT: OR ¼ 1.

Schizophrenia:MPT: OR ¼ 1.4LPT: OR ¼ 1.3.Disorders ofpsychologicaldevelopment,behaviour, andemotion:MPT: OR ¼ 1.4LPT: OR ¼ 1.5.Autism: NS.

Gray et al.8 MPT/LPT: 31e37MPT: 31e34LPT: 35e37

3, 5 and 8 years old;born in 1985

MPT: 435LPT: 262

w20% of MPT/LPThad clinical scoreson CBCL.

Cheatham et al.9 LPT: 35e37 12 months; yearof birth unknown

LPT: 29FT: 20

LPT ¼ FT: 97 1e111)and 95 (77e 2) respectively.

Buchmayer et al.10 MPT/LPT: 32e36 0e10 years old;born in 1987e2002

MPT/LPT: 420FT: 6207

Autism: OR ¼ 1.55

Chyi et al.11 MPT/LPT: 32e36MPT: 32e33LPT: 34e36

5e10 years old;born in w1993e1994

MPT: 203LPT: 767FT: 13 671

Poor school outcomes:MPT: OR ¼ 1.0e2.0LPT: OR ¼ 1.0e1.3.Need for help:MPT: OR ¼ 1.9e2.8LPT: OR ¼ 1.1e1.4.Special education:MPT: OR ¼ 2.0e2.9LPT: OR ¼ 1.2e2.1.

Dalziel et al.12 MPT/LPT: 32e35 30 years old;born in 1969e1974

MPT/LPT: 126FT: 66

MPT/LPT ¼ F 103 (13.4)and 104 (12. respectively.

MPT/LPT ¼ FT fordepression, anxiety,schizophrenia andattention deficit disorder.

Darlow et al.13 MPT/LPT: 33e36 2 years old; born in 2001e2002;NICU admittance for MPT/LPT group

MPT/LPT: 112FT: 94

MPT/LPT ¼ F 90 (14.2)and 92 (11.0 espectively.

Showing two or moreout of five problematicbehaviours: MPT/LPT(26%) > FT (20%).

Ekeus et al.14 MPT/LPT: 33e36MPT: 33e34LPT: 35e36

18e19 years old; born in 1973e1976;males conscripted for military service

MPT: 1088LPT: 3981FT: 94 821

MPT (4.8) an LPT (4.9) < FT (5.1).

Hillemeier et al.15 MPT/LPT: 33e36 9 and 24 months; born in 2001 MPT/LPT: 1224(592 singletons,632 multiple births)FT: 7173 (6570singletons, 603multiple births)

Scoring in lo st 10%:9 months OR 2.4(singletons),(multiple bir s);24 months O ¼ 0.9(singletons),(multiple bir s), NS.

(continued on next page)

M.de

Jonget

al./Sem

inarsin

Fetal&Neonatal

Medicine

17(2012)

163e169

165

ctrae i

T;(11

da.16.

(811

T:2)

T:) r

d

we¼3.6thR1.3th

Table 1 (continued )

Authors Gestational agerange (weeks)preterm

Subject characteristics No. of subjects School outcomes Cognitive functioning[mean (SD)/(range)or confiden interval]

Behaviour problemsand/or psychiatricdisorders

Huddy et al.16 MPT/LPT: 32e35 7 years old; born in 1990 MPT/LPT: 117 School problems:MPT/LPT (w30%) >population (10e20%).Special education:MPT/LPT (4%) >population (1.7%).Need help at school:w25% of MPT/LPT.

Abnormal hyperactivityscores: MPT/LPT (parents:18%, teachers: 18.8%) >population norm (10%)

Kirkegaard et al.17 MPT/LPT: 33e36 9e11 years old;born in 1990e1992

MPT/LPT: 169FT: 3081

Spelling difficulties:OR ¼ 1.6.Reading difficulties:OR ¼ 1.2.

Lindström et al.18 MPT/LPT: 33e36 23e29 years old;born in 1973e1979

MPT/LPT: 2037FT: 450 165

Psychiatric disorders:OR ¼ 1.3Neuropsychiatricdisorders (e.g. ADHD):OR ¼ 2.1

Lindström et al.19 MPT/LPT: 33e36MPT: 33e34LPT: 35e36

6e19 years old;born in 1987e2000

MPT/LPT: 56 650FT: 813 606

ADHD:MPT: OR ¼ 1.4LPT: OR ¼ 1.3

Romeo et al.20 MPT/LPT: 33e36 12 and 18 months old;born in 2005e2006; lowrisk group (e.g. no serious illness)

MPT/LPT: 61FT: 60

MPT/LPT < T [at 12 months92 (9.3) vs 0 (8.7) and at18 months 8 (9.9) vs 98 (8.1)],only when e was notcorrected f prematurity.

Schendel and Bhasin21 MPT/LPT: 33e36 3e10 years old; born in 1981e1993 MPT/LPT: 26 319FT: 241 888

Mental ret ation: OR ¼ 1.9. Autism: MPT/LPT ¼ FT

Schermann and Sedin22 MPT/LPT: 32e36 10 years old; born in 1986e1989;NICU admittance for preterm group

MPT/LPT: 82FT: 72

MPT/LPT < T; 101 (12.9)vs 109 (14 .

Van Baar et al.23 MPT/LPT: 32e36MPT: 32e33LPT: 34e36

8 years old; born in 1996e1998;low risk group (e.g. no NICU admittance)

MPT/LPT: 377MPT: 62LPT: 315FT: 182

Special education:MPT/LPT (7.7%) > FT (2.8%)MPT (9.7%) ¼ LPT (7.3%).Grade retention:MPT/LPT (19%) > FT (8.8%)MPT (30%) > LPT (17%).

MPT/LPT < T; 105(14) vs 108 15).

Internalizing problems:MPT/LPT > FT.Externalizing problems:MPT/LPT ¼ FT.Attention problems:MPT/LPT > FT.Behaviour problems:MPT < LPT.

Woythaler et al.24 LPT: 34e36 24 months old; born in 2001 LPT: 1200FT: 6300

LPT < FT, 1 (10.3) vs 128 (10.4).Percentage 70: LPT(21.2%) > F (16.4%).

Talge et al.25 LPT: 34e36 6 years old; born in 1983e1985 LPT: 168FT: 168

LPT ¼ FT; 1 1 (SE ¼ 1.3) and102 (SE ¼ ) respectively.

Internalizing problems:LPT > FT.Externalizing problems:LPT > FT (slightly).Attention problems:LPT > FT.

Petrini et al.26 LPT: 34e36 0e5.5 years old; born in 2000e2004 LPT: 8341FT: 128 955

Developme tal delay/mentalretardation ompared: OR ¼ 1.4.

Morse et al.27 LPT: 34e36 0e5 years old; born in 1996e1997;low risk group (<3 days hospitalization)

LPT: 7152FT: 152 661

School-related problems(e.g. grade retention,special learning needs):OR ¼ 1.1e1.3.

M.de

Jonget

al./Sem

inarsin

Fetal&

NeonatalM

edicine17

(2012)163

e169

166

ce

F108agorard

F.4)F(

26<

T01.3

nc

Linnet

etal.28

LPT:

34e36

Allch

ildrenbo

rnin

1980

e19

94diagn

osed

withHKD

andco

ntrol

child

ren

LPT:

37cases,

544co

ntrols.

FT:45

6cases,

1236

5co

ntrols

HKD:OR¼

1.8

Gurkaet

al.29

LPT:

34e36

4.5e

15ye

arsold;bo

rnin

1991

;low

risk

grou

p(e.g.n

oseriou

sillness)

LPT:

53FT

:12

54Sp

ecialed

ucation

:OR¼

1.3

LPT¼

FTLP

T¼

FTon

CBCL

Baron

etal.30

LPT:

35e36

3ye

arsold;bo

rnin

2004

e20

06;

both

complic

ated

and

unco

mplic

ated

grou

ps

LPT:

118

(90co

mplic

ated

;28

unco

mplic

ated

)FT

:10

0

Com

plic

ated

LPT

(103

e10

9)<

FT(109

e11

5).

Unco

mplic

ated

LPT(104

e11

5)¼

FT.

Baron

etal.31

LPT:

34e36

3ye

arsold;bo

rnin

2004

e20

05;

complic

ated

grou

p(e.g.N

ICU

admittance)

LPT:

60FT

:35

LPT<

FT:10

5(15.9)

vs11

2(12.9).

Baron

etal.32

LPT:

34e36

3ye

arsold;bo

rnin

2004

e20

06;

complic

ated

grou

pLP

T:75

FT:40

LPT<

FT:10

6(15.0)

vs11

6(10.3).

Marretet

al.33

MPT

:32

e34

5ye

arsold;bo

rnin

1997

MPT

:78

8Scores

<85

:MPT

(w24

%)>

pop

ulation

(15%

).

MPT

,mod

eratepreterm

;LPT

,latepreterm

;FT,fullterm

;OR,o

ddsratio;

NS,non

-significant;CBCL,Child

Beh

aviorCheckList;A

DHD,atten

tion

defi

cit/hyp

eractivity

disorder;N

ICU,n

eonatal

intensive

care

unit;H

KD,h

yperkinetic

disorder.

M. de Jong et al. / Seminars in Fetal & Neonatal Medicine 17 (2012) 163e169 167

schizophrenia and 40e50% higher risk for psychiatric disorders inmoderate/late preterm adults compared with full term adults,including developmental, behavioural, and emotional disorders.

Three studies reported the prevalence of autism in largesamples.7,10,21 Buchmayer et al.10 found a 50% increased risk forautism in moderate and late preterm children compared with fulltermchildren,whereas theother two studies foundnodifferences inprevalence of autism.7,21 Two studies examining the prevalence ofattention deficit/hyperactivity disorder (ADHD) or hyperkineticdisorder (HKD) in large samples both reported a higher risk(30e80%) for these disorders inmoderate/late pretermchildren.19,28

On the other hand, Dalziel et al.12 found no negative effect ofmoderate/late preterm birth on the presence of attention deficitdisorder, as well as depression, anxiety, and schizophrenia inadulthood. It should, however, be noted that the sample used in thisstudy was relatively small (126 preterm and 66 full term adults)compared with the above-mentioned studies, which were allepidemiological7,18,19 or nested caseecontrol10,21,28 designs.

4. Discussion

The aim of this review has been to gain more insight into longterm developmental outcomes of moderate and late preterm chil-dren, specifically in school outcome, cognitive functioning, behav-iour problems, and psychiatric disorders. Based upon the 28 studiesincluded in this review, we conclude that moderate/late pretermchildren show more school problems, have lower IQ scores, andmore behaviour problems than their full term peers. For psychiatricdisorders it is concluded that especially ADHD is more frequentlyreported for these preterm children; concerning autism, inconsis-tent findings have been reported.

The first explanation for the developmental problems ofmoderate/late preterm children lies in their immature (brain)development at birth. At 34 weeks of gestation, the brain weighsonly 65% of the weight at 40 weeks of gestation,5 so a lot of the‘hardware’ brain tissue still has to grow during this period. Inaddition, other organs of the preterm infant, such as the lungs andthe heart, have to adapt to extrauterine life at an earlier stage ofdevelopment, which may stress brain development.

Correction for prematurity is important, as it provides at leastthe same amount of time for moderate or late preterm children todevelop as could be used by full term children. This correction forprematurity explained the differences in results of several studiesdone in infancy from 0 to 2 years of age.9,13,15,20,24 Lower scores ondevelopmental tests were found for moderate/late pretermchildren compared with full term children only when age was notcorrected for prematurity. However, when age was corrected andboth groups of children could be considered equally mature,differences in developmental tests were no longer visible. Thecontrasting finding of Hillemeier et al.,15 that at 9 months of agemoderate/late preterm children performed worse than full termchildren even after age correction for prematurity whereas nogroup differences appeared at 24months, may be very important. Itmight be that differences between the preterm and full term bornchildren are larger at younger ages (<12 months). This suggestsa different developmental pattern for young moderate or latepreterm infants. Further research is needed on such developmentaltrajectories of these infants over the first years.

Moderate and late preterm children were found to be at risk forschool problems and low scores on IQ tests. This might result fromthe regulation difficulties, or more specifically from attentiondifficulties involved in ADHD symptoms.23 In very pretermchildren, attention difficulties were repeatedly found,34,35 anddifferences in attention capacities were suggested as a mechanismpartially explaining cognitive and school problems.36 In this review

Practice points

� Follow-up of moderate and late preterm children isindicated.

� Design of intervention programmes based on earlyindications of insufficient attention capacities.

Research directions

� Developmental trajectories after hospital dischargeover the first years in moderate and late pretermchildren.

� Medical treatment, caretaking practice during thehospital stay and discharge routines in relation todevelopmental outcome.

� Development of attention capacities of moderate andlate preterm children.

� Differences within themoderate and late preterm groupand factors explaining these differences.

� Consequences of moderate and late preterm birth inadulthood.

M. de Jong et al. / Seminars in Fetal & Neonatal Medicine 17 (2012) 163e169168

it was found that ADHD symptoms and attention problems aremore common in moderate/late preterm children.19,23,28 The firstyears of life are very important in the development of attentioncapacities.37 Neurological maturation is one of the factors influ-encing the development of attention capacities. Important braindevelopment takes place between 32 and 37 weeks’ gestation andpreterm children cannot benefit from the neurobiological processes(e.g. related to maternal thyroid hormone exchange) during thisperiod in the womb. Attention capacities of moderate and latepreterm children should be studied in detail and compared withthe development of processes in attention of full term children.

Not all moderate/late preterm children actually show develop-mental problems. The finding that a higher percentage of thesechildren had an IQ score <85 compared with full term children,although mean scores on a standardized intelligence test did notdiffer between the preterm and full term children,25,33 indicatesdifferences within the moderate/late preterm group. Even withinthemoderate/late pretermgroup, gestational agemaybe important.In only two studies included in the reviewweredifferences betweenmoderate and late preterm born children discussed or analysed.11,23

Chyi et al.11 found that themoderate preterm children showedmoredifficulties than the late preterm children. Van Baar et al.23 foundthat moderate preterm children more frequently repeated a grade,but also found that late preterm children showed more behaviourproblems than moderate preterm children. Hence, further study inrelation to a further differentiation of gestational age, as well as inrelation to different developmental outcomes, is necessary.

Another factor of influence might be differences in neonatalcomplications that may have occurred. Some of the preterm chil-dren needed neonatal intensive care treatment and are consideredto be at high medical risk. Looking at the studies that investigatedspecific samples of high or low medical risk children, it seems thathigh risk moderate/late preterm children in particular score loweron standardized intelligence tests than full term children.22,23,29e32

In general not much attention has been paid to the importance ofspecific neonatal complications and the medical treatments inrelation to developmental outcome of moderate or late pretermchildren. Future research could focus on the association betweenmedical treatments (e.g. for hypoglycaemia or infections), care-taking habits during the hospital stay [e.g. caregiving or kangar-ooing by parents, Newborn Individualized Developmental Care andAssessment Program (NIDCAP) or individualised caretaking], orhospital discharge routines and developmental outcome.

Another factor in explaining differences within the pretermgroup might be parentechild interaction and the quality of care-giver stimulation. Regarding very preterm infants, it was found thatindividual differences in cognitive development at infant andtoddler age could be explained by maternal behaviours such as theamount of stimulation provided.38 Further research is needed onthe influence of parentechild interaction on the development ofmoderate and late preterm children.

Only a few studies investigated cognitive functioning in adultsborn moderate or late preterm, and comparison between these ishampered by their differences in selection criteria. As these studiesshowed inconsistent results, more research is needed before anyconclusions on outcome in adulthood can be drawn. A generalproblem with studies on adults born preterm is that they weretreated a long time ago (in these studies from 1967 to 1983).Information on the quality of their functioning may not be appro-priate any more for infants treated with the current medical tech-niques and nursing methods. Nevertheless such information mayprovide some insight into developmental trajectories and theseverity of problems in developmental outcome.

A strength of this review is that it presents an overview of longterm outcome of moderate and late preterm birth across the life

span on the domains of school, cognitive, and behavioral func-tioning. Furthermore, the focus was not solely on late preterm(34e36 weeks) infants but also included studies investigatingchildren born after 32e33 weeks of gestation. A limitation may bethat the focus was on recent findings, as only studies publishedafter 2000 were included. Second, the studies included varied indesign and sample size (varying from 29 to 56 650 moderatepreterm children). Finally, year of birth of the samples variedstrongly across the studies from 1960 to 2006. Because quality ofneonatal care has increased a great deal during the last severaldecades, year of birth might influence the outcome for moderateand late preterm children. However, this review shows no clearassociation between developmental outcome and year of birth.

A clinical implication of the information from this review is thatcareful follow-up monitoring of moderate and late preterm chil-dren is required to provide early intervention when needed and totry to reduce the amount of school and behavior problems. Selec-tive intervention, perhaps based on early indications of insufficientattention capacities, might be worthwhile.

In conclusion, moderate and late preterm children are at risk forschool problems, lower cognitive functioning, behaviour problems,and psychiatric disorders. Future research should focus on devel-opmental trajectories over the first years. Also the associationsbetween medical treatments, caretaking practice during thehospital stay in the neonatal period, hospital discharge routinesand developmental outcome need to be studied in greater detail.Research should focus on attention capacities of moderate and latepreterm children, as well as on differences within the group andfactors explaining these differences. Intervention programmesbased on improvement of attention skills could be worthwhile.Finally, consequences of moderate and late preterm birth inadulthood might be the subject of study. In short: moderate andlate preterm children need attention!

Conflict of interest statement

None declared.

Funding sources

None.

M. de Jong et al. / Seminars in Fetal & Neonatal Medicine 17 (2012) 163e169 169

References

1. Stichting Perinatale Registratie Nederland. Perinatale Zorg in Nederland 2007.Utrecht, Netherlands: Stichting Perinatale Registratie Nederland; 2009.

2. Martin JA, Hamilton BE, Sutton PD, et al. Births: final data for 2007. Natl VitalStat Rep 2010;58:1e85.

3. Mathews TJ, MacDorman MF. Infant mortality statistics from the 2005 periodlinked birth/infant death data set. Natl Vital Stat Rep 2008;57:1e32.

4. Shapiro-Mendoza CK, Tomashek KM, Kotelchuck M, et al. Effect of late-pretermbirth and maternal medical conditions on newborn morbidity risk. Pediatrics2008;121:e223e32.

5. Kinney HC. The near-term (late preterm) human brain and risk for periven-tricular leukomalacia: a review. Semin Perinatol 2006;30:81e8.

6. Nomura Y, Halperin JM, Newcorn JH, et al. The risk for impaired learning-related abilities in childhood and educational attainment among adults bornnear-term. J Pediatr Psychol 2009;34:406e18.

7. Moster D, Lie RT, Markestad T. Long-term medical and social consequences ofpreterm birth. N Engl J Med 2008;359:262e73.

8. Gray RF, Indurkhya A, McCormick MC. Prevalence, stability, and predictors ofclinically significant behavior problems in low birth weight children at 3, 5, and8 years of age. Pediatrics 2004;114:736e43.

9. Cheatham CL, Bauer PJ, Georgieff MK. Predicting individual differences in recallby infants born preterm and full term. Infancy 2006;10:17e42.

10. Buchmayer S, Johansson S, Johansson A, Hultman CM, Sparén P, Cnattingius S.Can association between preterm birth and autism be explained by maternal orneonatal morbidity? Pediatrics 2009;124:e817e25.

11. Chyi LJ, Lee HC, Hintz SR, Gould JB, Sutcliffe TL. School outcomes of late preterminfants: special needs and challenges for infants born at 32 to 36 weeksgestation. J Pediatr 2008;153:25e31.

12. Dalziel SR, Lim VK, Lambert A, et al. Psychological functioning and health-related quality of life in adulthood after preterm birth. Dev Med Child Neurol2007;49:597e602.

13. Darlow BA, Horwood LJ, Wynn-Williams MB, Mogridge N, Austin NC.Admissions of all gestations to a regional neonatal unit versus controls: 2-yearoutcome. J Paediatr Child Health 2009;45:187e93.

14. Ekeus C, Lindström K, Lindblad F, Rasmussen F, Hjern A. Preterm birth, socialdisadvantage, and cognitive competence in Swedish 18- to 19-year-old men.Pediatrics 2010;125:e67e73.

15. Hillemeier MM, Farkas G, Morgan PL, Martin MA, MacZuga SA. Disparities inthe prevalence of cognitive delay: how early do they appear? Paediatr PerinatEpidemiol 2009;23:186e98.

16. Huddy CLJ, Johnson A, Hope PL. Educational and behavioural problems inbabies of 32e35 weeks gestation. Arch Dis Child 2001;85. F23e8.

17. Kirkegaard I, Obel C, Hedegaard M, Henriksen TB. Gestational age and birthweight in relation to school performance of 10-year-old children: a follow-upstudy of children born after 32 completed weeks. Pediatrics 2006;118:1600e6.

18. Lindström K, Lindblad F, Hjern A. Psychiatric morbidity in adolescents andyoung adults born preterm: a Swedish national cohort study. Pediatrics2009;123:e47e53.

19. Lindström K, Lindblad F, Hjern A. Preterm birth and attention-deficit/hyperactivity disorder in schoolchildren. Pediatrics 2011;127:858e65.

20. Romeo DM, Di Stefano A, Conversano M, et al. Neurodevelopmental outcome at12 and 18 months in late preterm infants. Eur J Paediatr Neurol 2010;14:503e7.

21. Schendel D, Bhasin TK. Birth weight and gestational age characteristics ofchildren with autism, including a comparison with other developmentaldisabilities. Pediatrics 2008;121:1155e64.

22. Schermann L, Sedin G. Cognitive function at 10 years of age in children whohave required neonatal intensive care. Acta Paediatr 2004;93:1619e29.

23. van Baar AL, Vermaas J, Knots E, de Kleine MJK, Soons P. Functioning at schoolage of moderately preterm children born at 32 to 36 weeks’ gestational age.Pediatrics 2009;124:251e7.

24. Woythaler MA, McCormick MC, Smith VC. Late preterm infants have worse24-month neurodevelopmental outcomes than term infants. Pediatrics2011;127:e622e9.

25. Talge NM, Holzman C, Wang J, Lucia V, Gardiner J, Breslau N. Late-preterm birthand its association with cognitive and socioemotional outcomes at 6 years ofage. Pediatrics 2010;126:1124e31.

26. Petrini JR, Dias T, McCormick MC, Massolo ML, Green NS, Escobar GJ. Increasedrisk of adverse neurological development for late preterm infants. J Pediatr2009;154:169e176.e3.

27. Morse SB, Zheng H, Tang Y, Roth J. Early school-age outcomes of late preterminfants. Pediatrics 2009;123:e622e9.

28. Linnet KM, Wisborg K, Agerbo E, Secher NJ, Thomsen PH, Henriksen TB.Gestational age, birth weight, and the risk of hyperkinetic disorder. Arch DisChild 2006;91:655e60.

29. Gurka MJ, Locasale-Crouch J, Blackman JA. Long-term cognition, achievement,socioemotional, and behavioral development of healthy late-preterm infants.Arch Pediatr Adolesc Med 2010;164:525e32.

30. Baron IS, Erickson K, Ahronovich MD, Baker R, Litman FR. Cognitive deficit inpreschoolers born late-preterm. Early Hum Dev 2011;87:115e9.

31. Baron IS, Erickson K, Ahronovich MD, Coulehan K, Baker R, Litman FR.Visuospatial and verbal fluency relative deficits in ‘complicated’ late-pretermpreschool children. Early Hum Dev 2009;85:751e4.

32. Baron IS, Erickson K, Ahronovich MD, Litman FR, Brandt J. Spatial locationmemory discriminates children born at extremely low birth weight andlate-preterm at age three. Neuropsychology 2010;24:787e94.

33. Marret S, Ancel P, Marpeau L, et al. Neonatal and 5-year outcomes after birth at30e34 weeks of gestation. Obstet Gynecol 2007;110:72e80.

34. Rose SA, Feldman JF, Jankowski JJ. Attention and recognition memory in the 1styear of life: a longitudinal study of preterm and full-term infants. Dev Psychol2001;37:135e51.

35. Landry SH. The development of joint attention in premature low birth weightinfants: effects of early medical complications and maternal attention-directingbehaviors. In: Moore C, Dunham P, editors. Joint attention: its origins and role indevelopment. Hillsdale, NJ: Lawrence Erlbaum; 1995. p. 223e50.

36. van de Weijer-Bergsma E, Wijnroks L, Jongmans MJ. Attention development ininfants and preschool children born preterm: a review. Infant Behav Dev2008;31:333e51.

37. Ruff HA, Rothbart MK. Attention in early development. New York: OxfordUniversity Press; 1996.

38. Landry SH, Smith KE, Miller-Loncar CL, Swank PR. Predicting cognitive-language and social growth curves from early maternal behaviors in childrenat varying degrees of biological risk. Dev Psychol 1997;33:1040e53.

at SciVerse ScienceDirect

Seminars in Fetal & Neonatal Medicine 17 (2012) 170e178

Contents lists available

Seminars in Fetal & Neonatal Medicine

journal homepage: www.elsevier .com/locate/s iny

Economic costs associated with moderate and late preterm birth: Primary andsecondary evidence

Stavros Petrou*, Kamran KhanWarwick Clinical Trials Unit, Division of Health Sciences, Warwick Medical School, University of Warwick, Coventry CV4 7AL, UK

Keywords:CostEconomicModeratePreterm birthReview

* Corresponding author. Tel.: þ44 2476 151124.E-mail address: s.petrou@warwick.ac.uk (S. Petrou

1744-165X/$ e see front matter � 2012 Elsevier Ltd.doi:10.1016/j.siny.2012.02.001

s u m m a r y

Despite constituting the vast majority of preterm births, relatively little is known about the clinical andeconomic outcomes of children born either moderately or late preterm. This paper outlines the economicconsequences of moderate and late preterm birth for the health services, for other sectors of theeconomy, for families and carers and, more broadly, for society. The paper reviews both the peer-reviewed literature and additional sources for information on the economic consequences of moderateand late preterm birth. It then goes on to present the results of a decision-analytic modelling study thataimed to estimate the societal costs associated with moderate and late preterm birth throughout thechildhood years. Finally, the requirements for future methodological and applied research in this area arebriefly outlined.

� 2012 Elsevier Ltd. All rights reserved.

1. Introduction

Preterm births are generally considered to occur before 37weeks of completed gestation.1 Despite advancing knowledgeabout the causes of preterm birth and the introduction of manypublic health and medical interventions targeted at its prevention,the incidence of preterm birth has continued to increase in manyindustrialized countries.1 Factors that have been postulated ascontributing to its increasing incidence include increasing rates ofmultiple births, greater use of assisted reproduction, increasedobstetric intervention, such as induced labour and caesareansection, and developments in clinical practice, such as the use ofultrasonography to estimate gestational age.1e3 The mortality andmorbidity arising from preterm birth imposes an immense burdenon finite public sector resources. Babies born preterm are atincreased risk of adverse neonatal outcomes, including chroniclung disease,4 intraventricular haemorrhage,5 retinopathy ofprematurity,6 necrotizing enterocolitis7 and neonatal sepsis,8 andon average they requiremore intensive and longer stays of neonatalcare. Following this initial period of hospitalization, preterm infantsaremore likely to be readmitted to hospital and havemore frequentcontact with community health and social care professionals thaninfants born at term.9 In the longer term, preterm infants are also atincreased risk of impairment, disability and handicap, includingmotor and sensory impairment,10 learning difficulties,11 and social

).

All rights reserved.

and behavioural problems.12 The health and neurodevelopmentaloutcomes of children born extremely preterm (usually defined as<28 weeks’ gestation) or very preterm (usually defined as <32weeks’ gestation) are reasonably well documented; a number ofstudies report on the early outcomes and follow-up assessments ofseveral extremely and very preterm birth cohorts in countriespredominately in North America,Western Europe, and Australasia.7

Despite constituting the vast majority of all preterm births, less isknown about those born moderately preterm (between 32þ0 and33þ6 weeks’ gestation) or late preterm (between 34þ0 and 36þ6

weeks’ gestation). Although data on the consequences of moderateand late preterm birth are sparse, those that are available suggestthat infants born at between 32 and 36 weeks’ gestation inclusivemay be at substantial risk of adverse growth,13,14 neuro-psychological,13,15 educational16 and behavioural16 outcomes. Thispaper outlines the economic consequences of moderate and latepreterm birth for the health services, for other sectors of theeconomy, for families and carers and, more broadly, for society. Itbuilds on previous research by one of the authors (S.P.)17 with thedistinguishing feature that it focuses on moderate and late pretermbirth rather than preterm birth as a whole. We review both thepeer-reviewed literature and additional sources for information onthe economic consequences of moderate and late preterm birth.We then go on to present the results of a decision-analyticmodelling study that aimed to estimate the societal costs associ-ated with moderate and late preterm birth. Finally, we outline therequirements for future methodological and applied research inthis area.

S. Petrou, K. Khan / Seminars in Fetal & Neonatal Medicine 17 (2012) 170e178 171

2. Methods

We conducted a literature search of relevant economic studiesthrough the following databases: MEDLINE, CINAHL, EconLit,Science Citation Index (SCI), Social Science Citation Index, Index toScientific and Technical Proceedings (ISTP), British Library InsideInformation (BLII), EMBASE, Cochrane Library (CDSR), York Data-base of Abstracts of Reviews of Effectiveness (DARE), NationalHealth Service (NHS) Economic Evaluation Database (NEED) andthe Database of Consortium of University Research Libraries(COPAC). Our search strategy was piloted on MEDLINE and oursearch terms eventually included all minor and major topicscovered by MeSH terms for ‘preterm birth’, ‘prematurity’ and ‘lowbirthweight’ combined with ‘cost’, ‘economic’, ‘financial’ and‘burden’. We also examined the bibliographies of identified papersfor additional studies that estimated the economic costs associatedwith moderate or late preterm birth. The search strategy coveredthe period January 1980 to September 2011. Studies were excludedfrom the literature searches if they had been conducted in non-developed countries [defined, for the purposes of this review, ascountries outside of the Organisation for Economic Co-operationand Development (OECD)], if the abstract had not been publishedin the English language or if the focus was animal research. A two-stage screening process of studies was followed: first, a screen ofthe title and abstract using predetermined criteria; second, a screenof the full report to determine whether the study explicitlydescribed, measured and valued the economic implications of careprovided to moderate or late preterm infants. A total of 20 relevantstudies were identified by the searches: 10 reported the costsassociated with the initial period of hospitalisation; 13 reportedcosts incurred following the initial hospital discharge, three ofwhich also reported costs associated with the initial period ofhospitalisation. Methodological variations between studies pre-vented a pooling of economic data akin to the meta-analyses per-formed on the clinical effectiveness literature. Rather, the results ofthe studies are presented and discussed in a qualitative manner.

3. Results

3.1. Costs associated with the initial hospitalisation

A total of 10 studies were identified that reported the economiccosts associated with moderate or late preterm birth during theinfant’s initial hospitalisation.18e27 Overviews of the location,sample size and methodology of each of these studies are sum-marised in Table 1; overviews of the economic results of each ofthese studies are summarised in Table 2. For live-born infants,mean hospital costs associated with the initial hospitalisationvaried between $1,929 for a term infant (US$, 2003 prices)25 andV13,655 for a moderate or late preterm infant (2001 prices).24 Forsurviving infants, mean hospital costs associated with the initialhospitalisation varied between $1,334 for a term infant and £32,153for amoderate or late preterm infant (US$, 2008 prices).20 Variationin costs between studies can be explained by a number of factors,including the wide time-frame over which studies were conducted,with some older studies conducted before the widespread use ofeffective perinatal and neonatal practices; geographical diversity ofstudies across systems that vary in the way that perinatal andneonatal care is organised and delivered; methodological diversitywith some economic studies based on prospective cohorts andothers based on retrospective analyses of routine observationaldatasets; and variations in the underpinning costing methodolo-gies with some studies valuing hospital stays in terms of chargesrather than costs generated through rigorous accounting proce-dures.18,21,27 Nevertheless, regardless of date of study, location of

study or economic methodology, a consistent inverse relationshipis observed between gestational age at birth and hospital costsassociated with the initial hospitalisation. In studies estimating theeconomic costs of preterm birth across the gestational age spec-trum,mean hospital costs associatedwith the initial hospitalisationfor infants born either moderate or late preterm appear closer tothose for infants born at term rather than to infants born eitherextremely or very preterm; nevertheless, they are consistently atleast double those for infants born at term (Table 2).

In addition to estimates of economic costs, a number of studiesprovide information on resource use associated with moderate orlate preterm birth, most commonly in the form of the number ofdays of hospitalisation by level of intensity (Table 2). For live-borninfants, the mean duration of initial hospitalisation varied between1.9 days for a term infant19 and 29.9 days for a moderate preterminfant.24 For surviving infants, the mean duration of initial hospi-talisation varied between 1.2 days for a term infant26 and 22.6 daysfor a moderate preterm infant.23 A relatively small number ofstudies also report increased lengths of stay for the motherfollowing birth of a moderate or a late preterm infant.19,24

A number of recent studies have used population-based data toestimate the potential economic benefits of interventions that areeffective at delaying moderate or late premature delivery. Aninteresting example of this line of enquiry is provided by the studyby Phibbs and Schmitt.23 Using cohort data for all births in Cal-ifornia in 1998e2000, the authors estimated that the mean(median) economic savings (in terms of neonatal costs prevented)for a 1-week increase in gestational age at birth were US$15,972(US$12,387) for infants born at 32 gestational weeks and US$4,528(US$1,116) for infants born at 34 gestational weeks (2003 prices).Similarly, the mean (median) economic savings for a 2-weekincrease in gestational age at birth were estimated at US$35,583(US$30,652) for infants born at 32 gestational weeks and US$7,090(US$1,287) for infants born at 34 gestational weeks. Data of thistype can be incorporated into future economic evaluations ofpreventive interventions for moderate or late preterm birth.

3.2. Costs following the initial hospital discharge

A total of 13 studies were identified that reported the economiccosts associated with moderate or late preterm birth following theinfant’s initial hospitalisation,18,20,24,28e37 three of which also re-ported costs associated with the initial period of hospital-isation.18,20,24 Overviews of the location, sample size andmethodology of each of these studies are summarised in Table 3;overviews of the economic results of each of these studies aresummarised in Table 4. As with interpretation of studies thatfocused on the costs associated with the initial hospitalisation,comparability of results across studies is complicated by a numberof factors. In particular, the cohort dates of these studies variedbetween 1970 and 2006, over which time there were severaldevelopments in the way that perinatal and neonatal care wasorganised and delivered with concomitant variations in mortalityand morbidity outcomes; this, in turn, is likely to have led tovariation in demand for health and broader services following theinitial hospital discharge. In addition, the studies vary in terms oftheir epidemiological design for selecting the underpinning patientdata, their cost categories and costing methodologies, and theirperiod of follow-up.

All but two studies28,30 focused on hospital costs followinginitial discharge, thereby excluding costs to other sectors of theeconomy, to families and carers and, more broadly, to society.Studies that estimated costs across the gestational age spectrumrevealed an inverse relationship between gestational age at birthand costs following the initial hospital discharge. In general, mean

Table 1Studies reporting the costs of initial hospitalisation associated with moderate or late preterm birth (studies since 1980).

Reference Date of cohort Location Type ofcohort

Sample size Currency Price date Gestationalage(s)

Type ofcosts

Cost vscharges

Datasources

Cholletet al.18

1989e1991 USA Conveniencesample

Extreme PT: 986 US$ Notspecified

All Hospital;professional

CostsBilled charges

ClaimsdatabaseNormal PT: 946

Term: 44 041Gilbert

et al.191996 USA

(California)Geographic(state)

�27 weeks: 77128e31 weeks: 297832e36 weeks: 37 388(32e33 weeks: 5093)(34e36 weeks: 32 295)�37 weeks: 106 087

US$ Notspecified

25e38weeks

Hospital CCR (hospitalspecific)

Stateelevellinked vitalstatistics anddischargerecords

Korvenrantaet al.20

2000e2003 Finland National 23 weeks: 1724e25 weeks: 13526e27 weeks: 23128e29 weeks: 38530e31 weeks: 756�32 weeks: 228Term: 200 609

V 2008 All Hospital Costs Hospitaldatabases

Lo et al.21 2000e2008 USA Not stated Total 240 179 US$ Notspecified

All Hospital Charges Not specified

Luke et al.22 1991e1992 USA(Illinois)

Hospitalspecific

Singletons: 106Twins: 111

US$ Notspecified

All Hospital Not specified Hospital billsHospital chart

Phibbs andSchmitt23

1998e2000 USA(California)

Geographic(state)

�27 weeks: 282028e31 weeks: 586832e36 weeks: 92 058(32e33 weeks: 7518)(34e36 weeks: 84 540)�37 weeks 92 421

US$ Dec2003 CPI

All Hospital CCR (hospitalspecific

State-levellinked vitalstatistics anddischargerecords

Ringborget al.24

1998e2001 Sweden National �27 weeks: 41828e31 weeks: 113632e36 weeks: 6092(32e33 weeks: 1365)(34e36 weeks: 4727)�37 weeks: 16 852

V 2001 All Hospital Costs Dischargerecords

Schmittet al.25

2000 USA(California)

Geographic(state)

�32 weeks: 828233e36 weeks: 34 800Other: 394 432

US$ 2003 All Hospital CCR (hospitalspecific)

State-levellinked vitalstatistics anddischargerecords

St Johnet al.26

1989e1992 USA(Alabama)

Hospitalspecific

� 27 weeks: 18028e31 weeks: 31532e36 weeks: 266(32e33 weeks: 163)(34e36 weeks: 103)�37 weeks: 197

US$ Notspecified

24e32 vs33e42weeks

Hospital;professional

CCR (hospitalspecific)

Hospital chartreview andbillingdatabase

Xu et al.27 2003 USA(Michigan)

Geographic(state)

Total 111 264 US$ 2007 All Hospital Charges State-levellinked vitalstatistics anddischargerecords

PT, preterm; CCR, cost-to-charge ratio; CPI, consumer price index.

S. Petrou, K. Khan / Seminars in Fetal & Neonatal Medicine 17 (2012) 170e178172

costs for infants born either moderate or late preterm appear closerto those for infants born at term rather than to infants born eitherextremely or very preterm (Table 4). However, the differential incost following initial hospital discharge between infants bornbetween 32 and 36 weeks’ gestation and those born at term variedbetween a 1.3-fold difference30 and an almost 10-fold difference.20

Notably, only six studies have estimated economic costsassociated with moderate or late preterm birth beyond the first 2years of life,20,28,30,32,33,37 one of which did not include a termcomparison group.37 Korvenranta et al.20 estimated the economiccosts of preterm birth during the first 4 years of life using Finnishpopulation-based national registry data. Total hospitalisationcosts averaged V43,325 for children born either moderate or latepreterm compared to V4,580 for children born at term (2008prices). Duration of rehospitalisations following initial dischargeaveraged 5.5 days for children born either moderate or latepreterm compared with 2 days for children born at term. Simi-larly, the number of non-emergency outpatient visits averaged

17.2 for children born either moderate or late preterm comparedwith 2.5 for children born at term. Petrou et al.32,33 have alsoconducted a number of economic studies in this area based ondata extracted from the Oxford Record Linkage Study, a largecollection of linked, anonymised birth registrations, deathcertificates and statistical abstracts of NHS hospital inpatient andday-case admissions for part of southern England. A multivariatenegative binomial regression performed on the 5-year hospitalservice use profile of 239,694 infants born in Oxfordshire andWest Berkshire during 1970e1993 revealed that the total dura-tion of hospital admissions for infants born between 32 and 36gestational weeks was three times that for term infants, takinginto account duration of life.32 A subsequent multi-level multipleregression model revealed that the adjusted effect regardinghospital inpatient admissions, days and costs over the first 10years of life was 1.37, 1.41 and 1.86, respectively, for childrenbetween 32 and 36 gestational weeks when compared withchildren born at term.33

Table 2Studies reporting the costs of initial hospitalisation associated with moderate or late preterm birth: Resource use and cost results.

Reference Gestationalage(s)

Per patientinfant cost(mean)

Cost persurvivor(mean)

Cost per livebirth (mean)

Per patientmaternal cost(mean)

LOS infantdays (mean)

LOS survivorsdays (mean)

LOS maternaldays (mean)

Chollet et al.18 Extreme PT $55,424 N/A N/A $14,815 N/A N/A N/ANormal PT $15,363 $13,017Term $2,376 $7,850

Gilbert et al.19 �27 weeks28e31 weeks32e36 weeks(32e33 weeks)(34e36 weeks)�37 weeks

N/A $156,937$55,350$8,800($18,744)($7,232)$3,860

N/A $7,843$7,057$3,524($4,776) ($33,26)$2,535

76.031.64.7(11.2)(3.7)1.9

N/A 6.85.93.0(4.2)(2.9)2.0

Korvenranta et al.20 23 weeks24e25 weeks26e27 weeks28e29 weeks30e31 weeks�32 weeksTerm

N/A V147,398V120,179V88,188V56,588V35,147V32,153V1,334

N/A N/A N/A 136.7122.194.467.548.043.63.5

N/A

Lo et al.21 All a

Luke et al.22 25e27 weeks28e30 weeks31e34 weeks35e38 weeks39e42 weeks

$195,254$91,343$18,367$4,308$2,230

N/A N/A $84,892$81,971$23,759$8,532$7,573

71.239.011.83.83.1

N/A N/A

Phibbs and Schmitt23 �27 weeks28e31 weeks32e36 weeks(32e33 weeks)(34e36 weeks)�37 weeks

N/A $231,852$98,599$7,694($35,635)($5,238)$1,966

$207,004$99,261$7,928($36,091)($5,424)$2,027

N/A 79.449.85.9(22.6)(4.4)2.6

91.850.85.9(22.6)(4.4)2.6

N/A

Ringborg et al.24 �27 weeks28e31 weeks32e36 weeks(32e33 weeks)(34e36 weeks)�37 weeks

N/A N/A V61,886V27,074V13,655(V16,917)(V12,713)V5,543

V4,788V4,785V3,462(V4,243)(V3,340)V2,679

95.354.117.9(29.9)(14.6)7.0

N/A 9.79.66.9(8.4)(6.6)3.0

Schmitt et al.25 �32 weeks33e36 weeksOther

N/A N/A $66,813$7,081$1,929

N/A N/A N/A N/A

St John et al.26 �27 weeks28e31 weeks32e36 weeks(32e33 weeks)(34e36 weeks)�37 weeks

$100,717$41,567$13,338($18,385)($5,041)$1,351

$71,643$40,671$17,018($20,845)($10,961)$6,953

N/A N/A 80.540.313.8(18.9)(5.4)1.2

N/A

Xu et al.27 All b

LOS, length of stay; PT, preterm; N/A, not available.a 24 weeks: $98,162; 25 weeks: $83,070; 26 weeks: $90,936; 27 weeks: $79,463; 28 weeks: $59,928; 29 weeks: $52,998; 30 weeks: $40,568; 31 weeks: $30,853; 32 weeks:

$22,612; 33 weeks: $16,597; 34 weeks: $9,740; 35 weeks: $5,015; 36 weeks: $2,413; 37 weeks: $1,469; 38 weeks: $1,070; 39 weeks: $994; 40 weeks: $1,017; 41 weeks:$1,058; 42 weeks: $1,072.

b 20weeks: $11,397; 21weeks: $11,703; 22weeks: $25,367; 23weeks: $48,908; 24weeks: $99,477; 25weeks: $82,296; 26weeks: $103,980; 27weeks: $72,212; 28weeks:$56,933; 29 weeks: $45,598; 30 weeks: $34,642; 31 weeks: $29,679; 32 weeks: $24,623; 33 weeks: $21,887; 34 weeks: $18,617; 35 weeks: $15,864; 36 weeks: $12,305; �37weeks: $6,368.

S. Petrou, K. Khan / Seminars in Fetal & Neonatal Medicine 17 (2012) 170e178 173

3.3. Modelling economic costs throughout childhood

Building on previous research by one of the authors (S.P.),30 weconstructed and populated aMarkov model to estimate the societalcosts associated with preterm birth over the first 18 years of life.The model structure is presented in Figure 1. A hypothetical cohortof children, the size of which was set at 669,601 to reflect thenumber of live births in England and Wales in 2006, was assumedto enter the model on live birth. Subsequently, children wereassumed to proceed to hospital discharge or neonatal care, with theexception of a small proportion assumed to die in the delivery roomor during transfer. Admission to a neonatal unit was treated asa proxy for neonatal complications, with length of neonatal stayacting as an indicator for the severity of morbidity. Followingdischarge from hospital, children were assumed to enter a healthstate defined by the time period between discharge and 2 years of

age. Survivors at 2 years were then allocated to one of four statesthat described their overall level of disability: none, mild, moderateor severe. For each subsequent year of childhood, childrenwere assumed to remain in the same disability state, move toanother disability state, or to die. Model parameter inputs,including gestation-specific transitional probabilities and costs,were largely drawn from three cohorts that incorporated paralleleconomic studies: (i) the EPICure cohort of infants born in the UKand Republic of Ireland between March and December 1995 at <26weeks’ gestation; (ii) the 1991e1992 cohort of the Victorian InfantCollaborative Study Group (VICSG) of infants born in the Victoriastate of Australia at 26 or 27 weeks’ gestation; and (iii) the OxfordRecord Linkage Study for infants born in Oxfordshire or Berkshirebetween 1990 and 1993 at 28 weeks’ gestation or later.30 Costestimates were further disaggregated into cost categories. Notably,using evidence drawn from a parallel economic study,38 we also

Table 3Studies reporting the costs following initial hospitalisation discharge associated with moderate or late preterm birth (studies since 1980).

Reference Date of cohort Location Type of cohort Sample size Currency Price date Gestationalage(s)

Type of costs Cost vs charges Data sources Time frame

Bird et al.35 2001e2005 USA(Arkansas)

Geographic (state) 34e36 weeks: 519937e42 weeks: 50 907

US$ Not specified Late PTand term

Hospital Charges Claims First year of life

Chollet et al.18 1989e1991 USA Convenience sample Extreme PT: 986Normal PT: 946Term: 44 041

US$ Not specified All Hospital;professional

CostsBilled charges

Claims database First 2 years

Clementset al.28

1999e2000 USA(Massachusetts)

Geographic (state) �27 weeks: 25328e31 weeks: 69532e36 weeks: 6679(32e33: 997 weeks;34e36 weeks: 5682)�37 weeks: 69 274

US$ 2003 All Health andsocial services

Charges Claims database Dischargeto 3 years

Kirkby et al.29 2001e2004 USA Hospital specific 32 weeks: 122133 weeks: 150734 weeks: 2204

US$ Not specified 32e34weeks

Hospital Charges Claims database Birth to 2 weekspost initialdischarge

Korvenrantaet al.20

2000e2003 Finland National 23 weeks: 1724e25 weeks: 13526e27 weeks: 23128e29 weeks: 38530e31 weeks: 756�32 weeks: 228Term: 200 609

V 2008 All Hospital Costs Hospital databases First 4 yearsof life

Manghamet al.30

2006 UKEnglandand Wales

National �27 weeks: 318528e31 weeks: 557532e36 weeks: 39 222(32e33 weeks: 6410;34e36 weeks: 32 812)�37 weeks: 621 618

£ 2006 All Public Sector Costs Secondary data First 18 yearsof life

McLaurin et al.31 2004 USA National (insurancemembership specific)

LPT 1683Term 33 745

US$ Not specified 33e36weeks

Hospital Charges Claims Birth to 1 year

Petrou et al.32 1970e1993 UK(Oxfordshireand WestBerkshire)

Geographic (region) <28 weeks: 50028e31 weeks: 134632e36 weeks: 11 728>37 weeks: 226 120

£ 1998e1999 All GA Hospital Costs Linked vital statisticsand NHS recordsfinancial returns

First 5 years

Petrou33 1978e1988 UK(Oxfordshireand WestBerkshire)

Geographic (region) <28 weeks: 24128e31 weeks: 59632e36 weeks: 4485>37 weeks: 90 236

£ 1998e1999 All GA Hospital Costs Linked vital statisticsand NHS recordsfinancial returns

First 10 years

Profit et al.37 Not specified Mexico National 24e26 weeks: 920027e29 weeks: 16 50030e33 weeks: 58 300

US$ 2005 24e33weeks

Hospital Costs Secondary data Lifetime

Ringborg et al.24 1998e2001 Sweden National �27 weeks: 41828e31 weeks: 113632e36 weeks: 6092(32e33 weeks: 1365;34e36 weeks: 4727)�37 weeks: 16 852

V 2001 All Hospital Costs Discharge records First 12 months

Russell et al.34 2001 USA Hospital specific <28 weeks: 29 20028e36 weeks: 305 500

US$ Not specified All Hospital CCR (hospitalspecific)

Database First 12 months

Underwoodet al.36

1992e2000 USA(California)

Geographic (state) �27 weeks: 11 61928e31 weeks: 34 00532e36 weeks: 218 259(32e33 weeks: 1365;34e36 weeks: 4727)

US$ Not specified All Hospital CCR (hospitalspecific)

Stateelevel linkedvital statistics anddischarge records

Dischargeto 1 year

PT, preterm; LPT, late preterm; CCR, cost-to-charge ratio; GA, gestational age; NHS, National Health Service.

S.Petrou,K.K

han/Sem

inarsin

Fetal&Neonatal

Medicine

17(2012)

170e178

174

Table 4Studies reporting the costs following initial hospitalisation discharge associated with moderate or late preterm birth: Resource use and cost results.

Reference Gestationalage(s)

Total cost per patientincluding initialhospitalisation (mean)

Total cost per survivorincluding initialhospitalisation (mean)

Total cost per livebirth including initialhospitalisation (mean)

Inpatient costs perpatient (mean)

Outpatient cost perpatient (mean)

Birth hospitalisationcosts (mean)

Rehospitalisationcosts (mean)

RehospitalisationLOS (mean)

Bird et al.35 34e36 weeks N/A $4,541 N/A $3,027 $1,560 N/A N/A 0.6537e42 weeks $3,472 $2,183 $1,316 0.48

Chollet et al.18 Extreme PT $70,239 N/A N/A $63,910 $6,329 N/A N/A N/ANormal PT $28,380 $23,916 $4,463Term $10,226 $7,984 $2,243

Clements et al.28 �27 weeks N/A $7,439 N/A N/A N/A N/A N/A N/A28e31 weeks $4,64832e36 weeks $1,579(32e33 weeks)(34e36 weeks)

($2,756)($1,372)

�37 weeks $725

Kirkby et al.29 32 weeks $43,667 N/A N/A N/A N/A N/A N/A N/A33 weeks $31,53534 weeks $22,575

Korvenranta et al.20 23 weeks N/A €175,490 N/A N/A N/A €147,398 €28,092 14.624e25 weeks €143,570 €120,179 €23,391 12.126e27 weeks €105,631 €88,188 €17,443 8.628e29 weeks €72,366 €56,588 €15,778 8.230e31 weeks €45,714 €35,147 €10,567 5.7�32 weeks €43,325 €32,153 €11,172 5.5Term €4,580 €1,334 €3,246 2.0

Mangham et al.30 �27 weeks N/A £136,748 £75,832 N/A N/A N/A N/A N/A28e31 weeks £101,649 £93,93532e36 weeks £56,683 £55,593(32e33 weeks)(34e36 weeks)

(£77,305) (£52,761) (£74,145) (£51,969)

�37 weeks £42,049 £41,813

McLaurin et al.31 Late PT $38,301 N/A N/A N/A N/A $26,054 $12,247 4.5Term $6,156 $2,087 $4,069 3.4

Petrou et al.32 <28 weeks N/A N/A £13,639 N/A N/A £6,569 £7,070 N/A28e31 £14,059 £6,700 £7,35932e36 £4,378 £1,922 £2,456>37 weeks £1,333 £752 £581

Petrou33 <28 weeks £17,820 N/A N/A N/A N/A N/A N/A N/A28e31 £17,75132e36 £7,394>37 weeks £1,659

Profit et al. 37 Admitted to NICU:24e26 weeks $12,00027e29 weeks $11,10030e33 weeks $7,700

Not admitted to NICU:24e26 weeks $60027e29 weeks $1,60030e33 weeks $4,700

(continued on next page)

S.Petrou,K.K

han/Sem

inarsin

Fetal&Neonatal

Medicine

17(2012)

170e178

175

Table

4(con

tinu

ed)

Referen

ceGestation

alag

e(s)

Totalco

stper

patient

includinginitial

hospitalisation(m

ean)

Totalc

ostper

survivor

includinginitial

hospitalisation(m

ean)

Totalc

ostper

live

birthincludinginitial

hospitalisation(m

ean)

Inpatientco

stsper

patient(m

ean)

Outpatientco

stper

patient(m

ean)

Birth

hospitalisation

costs(m

ean)

Reh

ospitalisation

costs(m

ean)

Reh

ospitalisation

LOS(m

ean)

Ringb

orget

al.24

�27wee

ksN/A

€68

,406

N/A

€61

,886

N/A

N/A

€6,52

0N/A

28e31

wee

ks€29

,962

€27

,074

€2,88

832

e36

wee

ks€15

,151

€13

,655

€1,49

6(32e

33wee

ks)

(34e

36wee

ks)

(€18

,524

)(€14

,177

)(€16

,917

)(€12

,713

)(€1,60

7)(€1,46

4)

�37wee

ks€6,80

1€5,54

3€1,25

8

Russelle

tal.34

<28

wee

ks$6

5,60

0N/A

N/A

N/A

N/A

N/A

N/A

N/A

28e36

wee

ks$1

2,10

0

Underwoo

det

al.36

�27wee

ks$5

,207

N/A

N/A

N/A

N/A

N/A

28e31

wee

ks$2

,267

32e36

wee

ks$1

,065

(32e

33wee

ks)

(34e

36wee

ks)

($1,35

1)($96

8)

LOS,

lengthof

stay

;N/A,n

otav

ailable;

PT,p

reterm

;NICU,n

eonatal

intensive

care

unit.

S. Petrou, K. Khan / Seminars in Fetal & Neonatal Medicine 17 (2012) 170e178176

incorporated estimates of economic costs beyond those falling onthe public sector, including direct costs borne by parents and carers,and indirect costs associated with lost productivity. Costs accruingafter the first year of life were discounted at an annual rate of 3.5%.The results of this economic modelling study are summarised inTable 5. In keeping with the broader literature, an inverse rela-tionship is observed between gestational age at birth and publicsector costs. We also report, to our knowledge for the first time,a clear inverse relationship between gestational age at birth andcosts falling beyond the public sector, including costs borne byparents and carers, and indirect costs associated with lostproductivity. The incremental societal cost per moderate pretermchild surviving to 18 years compared with a term survivor wasestimated at £36,291. The corresponding incremental societal costestimate for a late preterm child was £10,828. The bulk of theseadditional costs fell on the health and social care sectors.

4. Discussion

This paper has reviewed the recent published evidence on theeconomic consequences of moderate and late preterm birth for thehealth services, for other sectors of the economy, for families andcarers and, more broadly, for society. Although data on theeconomic consequences of moderate and late preterm birth aresparse, they consistently suggest that service provision for infantsborn between 33 and 36 weeks’ gestation is associated withsubstantial incremental costs during the initial hospital stay andthroughout childhood. Moreover, our economic modelling study isthe first, to our knowledge, to highlight the substantial incrementalsocietal costs associated with moderate and late preterm birththroughout childhood. It should be noted that existing economicresearch in this area is hampered by a paucity of epidemiologicalstudies that accurately quantify neonatal morbidity and longerterm adverse outcomes for this population or that identify riskfactors contributing to these outcomes. There is therefore a clearneed for prospective, comprehensive data collection for infantsborn between 33 and 36 weeks’ gestation to allow quantification ofthe clinical and economic outcomes for this group. This is essential,first, to highlight areas in which changes in perinatal and neonatalcare may improve outcomes and, second, to inform equitable andefficient allocation of scarce public sector resources for childrenborn at these gestational ages.

In addition to the costs highlighted byour review, bothmoderateand late preterm births are likely to have broader consequences thatrequire further evaluation from an economic perspective. A recentliterature review highlighted the significant costs that are borne byfamilies and carers among thosewho care for a childwith confirmeddisability,39 and it is likely that these categories of costs are relevantto some families and carers of children born either moderate or latepreterm. In addition to the costs of travel, child care and accom-modation that we included in our economic modelling study, otherpotential costs faced by families and informal carers include incre-mental expenditures on health goods, such as alternative therapies,and non-health goods, such as nutritional requirements, laundry,clothing, heating utilities and repairs to the home. Furthermore, nomonetary valuation of the intangible consequences ofmoderate andlate preterm birth, such as the pain, fear, suffering and emotionaland social isolation that might be experienced by the individualsthemselves and their carers, has been attempted.

Given recent evidence of an increasing incidence of pretermbirth, it is imperative that clinical decision-makers and budgetaryand service planners recognise the overall economic impact ofmoderate and late preterm birth in their service planning, as well asthe potential contribution of clinical and sociodemographic factorsto future public sector and broader societal costs. Both our research

Admitted to

neonatal care

Severe

disability

Moderate

disability

Mild disability No disability

Death

From discharge

to 2 years

Key:

Cost of health

state

Probability of moving between health states

Live birth

Figure 1. Structure of the Markov model estimating the childhood economic costs of preterm birth.

Table 5Societal costs of care through the childhood years; mean costs per survivor (£, 2006prices).

Gestational age Health andsocial care

Education Parentalexpenses

Lostproductivity

All costs

�27 weeks £98,528 £38,505 £9,626 £5,096 £151,18928e31 weeks £66,188 £35,326 £8,540 £3,300 £113,16032e36 weeks £22,123 £34,549 £7,695 £1,924 £66,291(32e33 weeks) £42,460 £34,814 £7,971 £2,440 £87,685(34e36 weeks) £18,255 £34,499 £7,642 £1,825 £62,222�37 weeks £7,757 £34,292 £7,583 £1,773 £51,394

S. Petrou, K. Khan / Seminars in Fetal & Neonatal Medicine 17 (2012) 170e178 177

and research by other investigators30,36 suggests that children bornmoderate or late preterm contribute the largest share of total publicsector costs since they account for the vast majority of pretermbirths. In addition to informing the planning of services, our datashouldbeof interest anduse to researchersplanning toevaluatenewor existing interventions from an economic perspective, particularlythose wishing to incorporate within a decision-analytic frameworkthe long term economic impact of moderate and late preterm birthand the cost-effectiveness of prevention and treatment strategies.

Conflicts of interest

None declared. The views contained in the paper are those of theauthors and not necessarily of the funding bodies.

Funding sources

TheWarwick Clinical Trials Unit benefited from facilities fundedthrough the Birmingham Science City Translational MedicineClinical Research and Infrastructure Trials Platform, with supportfrom Advantage West Midlands.

References

1. Goldenberg RL, Culhane JF, Iams JD, Romero R. Epidemiology and causes ofpreterm birth. Lancet 2008;371:75e84.

2. Anonymous. Critical care decisions in fetal and neonatal medicine: ethical issues.London: Nuffield Council on Bioethics; 2006.

3. Tucker J, McGuire W. Epidemiology of preterm birth. BMJ 2004;329:675e8.4. Lefebvre F, Glorieux J, St-Laurent-Gagnon T. Neonatal survival and disability

rate at age 18 months for infants born between 23 and 28 weeks of gestation.Am J Obstet Gynecol 1996;174:833e8.

5. O’Shea TM, Klinepeter KL, Goldstein DJ, Jackson BW, Dillard RG. Survival anddevelopmental disability in infants with birth weights of 501 to 800 grams,born between 1979 and 1994. Pediatrics 1997;100:982e6.

6. Allen MC, Donohue PK, Dusman AE. The limit of viability e neonatal outcomeof infants born at 22 to 25 weeks’ gestation. N Engl J Med 1993;329:1597e601.

7. Saigal S, Doyle LW. An overview of mortality and sequelae of preterm birthfrom infancy to adulthood. Lancet 2008;371:261e9.

8. Hack M, Friedman H, Fanaroff AA. Outcomes of extremely low birth weightinfants. Pediatrics 1996;98:931e7.

9. Petrou S, Sach T, Davidson L. The long-term costs of preterm birth and low birthweight: results of a systematic review. Child Care Health Dev 2001;27:97e115.

10. Bracewell M, Marlow N. Patterns of motor disability in very preterm children.Mental Retard Dev Disabil Res Rev 2002;8:241e8.

11. Saigal S, Ouden LD, Wolke D, et al. School-age outcomes in children who wereextremely low birth weight from four international population-based cohorts.Pediatrics 2003;112:943e50.

12. Hille ETM, den Ouden AL, Saigal S, et al. Behavioural problems in children whoweigh 1000 g or less at birth in four countries. Lancet 2001;357:1641e3.

13. Pietz J, Peter J, Graf R, et al. Physical growth and neurodevelopmental outcome ofnonhandicapped low-risk childrenbornpreterm.EarlyHumDev2004;79:131e43.

14. Blackwell MT, Eichenwald EC, McAlmon K, et al. Interneonatal intensive careunit variation in growth rates and feeding practices in healthy moderatelypremature infants. J Perinatol 2005;25:478e85.

15. Holmqvist P, Regefalk C, Svenningsen NW. Low risk vaginally born preterminfants: a four year psychological and neurodevelopmental follow-up study.J Perinat Med 1987;15:61e72.

16. Huddy CL, Johnson A, Hope PL. Educational and behavioural problems in babiesof 32e35 weeks gestation. Arch Dis Child Fetal Neonatal Ed 2001;85:F23e8.

17. Petrou S, Eddama O, Mangham L. A structured review of the recent literatureon the economic consequences of preterm birth. Archs Dis Childh Fet Neonat Ed2011;96:F225e32.

18. Chollet DJ, Newman Jr JF, Sumner AT. The cost of poor birth outcomes inemployer-sponsored health plans. Med Care 1996;34:1219e34.

19. Gilbert WM, Nesbitt TS, Danielsen B. The cost of prematurity: quantification bygestational age and birth weight. Obstet Gynecol 2003;102:488e92.

20. Korvenranta E, LinnaM, Rautava L, et al. Hospital costs and quality of life during 4years after very preterm birth. Arch Pediatr Adolesc Med 2010;164:657e63.

21. Lo JMK, Henry E, Weng H, Hopkins P, Esplin MS. The cost of prematurity: theassociation between gestational age at delivery and overall neonatal cost.Reproductive Sciences 2011;18:69Ae384A.

22. Luke B, BiggerHR, Leurgans S, SietsemaD. The cost of prematurity: a caseecontrolstudy of twins vs singletons. Am J Public Health 1996;86:809e14.

23. Phibbs CS, Schmitt SK. Estimates of the cost and length of stay changes that canbe attributed to one-week increases in gestational age for premature infants.Early Hum Dev 2006;82:85e95.

24. Ringborg A, Berg J, Norman M, Westgren M, Jönsson B. Preterm birth in Swe-den: what are the average lengths of hospital stay and the associated inpatientcosts? Acta Pædiatrica 2006;95:1550e5.

S. Petrou, K. Khan / Seminars in Fetal & Neonatal Medicine 17 (2012) 170e178178

25. Schmitt SK, Sneed L, Phibbs CS. Costs of newborn care in California: a pop-ulation-based study. Pediatrics 2006;117:154e60.

26. St John EB, Nelson KG, Cliver SP, Bishnoi RR, Goldenberg RL. Cost of neonatalcare according to gestational age at birth and survival status. Am J ObstetGynecol 2000;182:170e5.

27. Xu X, Grigorescu V, Siefert KA, Lori JR, Ransom SB. Cost of racial disparity inpreterm birth: evidence from Michigan. J Health Care Poor Underserved2009;20:729e47.

28. Clements KM, Barfield WD, Ayadi MF, Wilber N. Preterm birth-associated costof early intervention services: an analysis by gestational age. Pediatrics2007;119:e866e74.

29. Kirkby S, Greenspan J, Kornhauser M, Schneiderman R. Clinical outcomes andcost of the moderately preterm infant. Adv Neonat Care 2007;7:80e7.

30. Mangham LJ, Petrou S, Doyle LW, Draper ES, Marlow N. The cost of pretermbirth throughout childhood in England and Wales. Pediatrics2009;123:e312e27.

31. McLaurin KK, Hall CB, Jackson EA, Owens OV, Mahadevia PJ. Persistence ofmorbidity and cost differences between late-preterm and term infants duringthe first year of life. Pediatrics 2009;123:653e9.

32. Petrou S, Mehta Z, Hockley C, Cook-Mozaffari P, Henderson J, Goldacre M. Theimpact of preterm birth on hospital inpatient admissions and costs during thefirst 5 years of life. Pediatrics 2003;112:1290e7.

33. Petrou S. The economic consequences of preterm birth during the first 10 yearsof life. BJOG 2005;112:10e5.

34. Russell RB, Green NS, Steiner CA, et al. Cost of hospitalization for preterm andlow birth weight infants in the United States. Pediatrics 2007;120:e1e9.

35. Bird TM, Bronstein JM, Hall RW, Lowery CL, Nugent R, Mays GP. Late preterminfants: birth outcomes and health care utilization in the first year. Pediatrics2010;126:e311e9.

36. Underwood MA, Danielsen B, Gilbert WM. Cost, causes and rates of rehospi-talization of preterm infants. J Perinatol 2007;27:614e9.

37. Profit J, Lee D, Zupancic JA, et al. Clinical benefits, costs, and cost-effectivenessof neonatal intensive care in Mexico. PLoS Med 2010;7:e1000379.

38. Petrou SJS, Wolke D, Marlow N. The association between neurodevelopmentaldisability and economic outcomes during mid-childhood. Child: Care HealthDev (in press).

39. AndersonD, Dumont S, Jacobs P, Azzaria L. The personal costs of caring for a childwith a disability: a review of the literature. Publ Hlth Rep 2007;122:3e16.

at SciVerse ScienceDirect

Seminars in Fetal & Neonatal Medicine 17 (2012) 179

Contents lists available

Seminars in Fetal & Neonatal Medicine

journal homepage: www.elsevier .com/locate/s iny

Correspondence

Re: ‘Myth: Group B streptococcal infection in pregnancy: Comprehendedand conquered’ published in volume 16 (2011) pp 254–258

Dear Sir,

We arewritingwith reference to the article ‘Myth: Group B strep-tococcal infection in pregnancy: Comprehended and conquered’ pub-lished in volume 16 (2011) pp 254–258. We are concerned aboutthe accuracy of some of the article’s key assumptions relating toscreening.

The authors estimate that 700 babies will be affected by earlyonset group B streptococcal disease (eogbs) each year in the“absence of any antibiotic prophylaxis”. However this assumesthat there is no current management strategy. This is not thecase; an RCOG and management of at risk women.1 The figure ishypothetical and a misleading starting point. The best data wehave were published in 2004 and suggests that the rate of cultureproven eogbs is w0.5/1000 births or 380 babies per year.2 This iscomparable to the rate of culture proven eogbs in countries whichoffer screening.

The authors state that testing using an enriched culture mediumwill detect “more than 90%” of pregnant carriers. The supportingcitation is a 1996 study comparing the detection rate from directinoculation of swabs into selective media with that from delayedinoculation of swabs stored in Stuart’s transport medium. However,the results from other studies of culture-based testing have notbeen consistently as high as this. Those cited elsewhere by theauthors suggest a sensitivity of 86.6%3 and 76%4 respectively. Themost recent CDC screening guidelines report a study which foundtest sensitivity to be as low as 54%.5

A recent systematic review of test effectiveness studiesconcluded that 69% of carriers detected in late pregnancy willremain positive at term.6 This compares unfavourably with thefigure of 87% cited from a much older, single, study by the authors.3

A recent review of the experience of screening in the US suggeststhat this latter study overestimated the specificity of culturetesting. The number of babies born to women with negative testshas been higher than expected and has emerged as a cause forconcern in recent publications in the US and Europe.7–10 Clearlythere are uncertainties regarding the characteristics of this test.

In the UK w30% of eogbs is in preterm babies born too early forscreening to impact significantly on disease in this group.2 Theeffectiveness of IAP in labour is uncertain11,12 but estimated to be85 – 89% effective in preventing eogbs.5 This, combined with thealready mentioned issues relating to the characteristics of the

DOI of original article: 10.1016/j.siny.2011.03.005.

1744-165X/$ – see front matter � 2012 Elsevier Ltd. All rights reserved.doi:10.1016/j.siny.2012.02.006

test, makes it extremely doubtful that screening and IAP in labourcan prevent ‘up to’, ‘about’ or ‘more than 90% of eogbs disease inneonates’.

We are concerned that this article creates its own myths, partic-ularly in relation to the effectiveness of screening.

Yours sincerely

References

1. RCOG, Prevention of Early Onset Neonatal Group B Streptococcal Disease,Greentop Guideline 36, 2003.

2. Heath PT, Balfour G, Weisner AM, et al. Group B streptococcal disease in UK andIrish infants younger than 90 days. Lancet 2004;363:292–4.

3. Yancey MK, Schuchat A, Brown LK, Ventura VL, Markenson GR. The Accuracy ofLate Antenatal Screening Cultures in Predicting Genital Group B StreptococcalColonization at Delivery. Obstet Gynecol 1996;vol. 88(5):811–5.

4. Kaambwa B, Bryan S, Gray J, et al. Cost effectiveness of rapid tests and otherexisting strategies for screening and management of early onset group B strep-tococcus during labour. BJOG 2010;117:1616–27.

5. Centers for Disease Control and Prevention. Prevention of Perinatal Group BStreptococcal Disease: Revised Guidelines from CDC. MMWR 2010;59(RR-10):1–36.

6. Valkenburg van den Burg AW. Timing of Group B Streptococcus Screening inPregnancy: A Systematic Review. Gynecol Obstet Invest 2010;69:174–83.

7. Van Dyke MK, Phares CR, Lynfield R, et al. Evaluation of Universal AntenatalScreening for Group B Streptococcus. N Eng J Med 2009;360:2626–36.

8. Berardi A, Di Fazzio G, Gavioli S, et al. Universal antenatal screening for group Bstreptococcus in Emilia-Romagna. J Med Screen June 2011;18:60–4.

9. Albouy-Llaty M, Nadeau C, Descombes E, Pierre F, Migeot V, et al. Improvingperinatal Group B streptococcus screening with process indicators, Journal of Eval-uation in Clinical Practice 2011 Mar 18:1–7.

10. Randis TM, Polin RA. Early-onset group B Streptococcal sepsis: new recommen-dations from the Centres for Disease Control and Prevention, Arch Dis ChildFetal Neonatal Ed 2011;fetalneonatal-2011-300627 Published Online First: 4November 2011.

11. Ohlsson A, Shah VS. Intrapartum antibiotics for knownmaternal Group B strep-tococcal colonization, Cochrane Database of Systematic Reviews. (3), 2009.Article Number: CD007467.

12. Carbonell-Estrany X, Figueras-Aloy J, Salcedo-Abizanda S, de al Rosa-Fraile M.Probably early-onset group B streptococcal neonatal sepsis: a serious clinicalcondition related to intrauterine infection. Arch Dis Child Neonatal Ed 2008;93:F85–9.

John Marshall*

UK National Screening Committee, UK

Catherine PeckhamCentre for Paediatric Epidemiology and Biostatistics,

Institute of Child Health, London, UK

* Corresponding author.E-mail address: John.Marshall@imperial.nhs.uk (J. Marshall)

at SciVerse ScienceDirect

Seminars in Fetal & Neonatal Medicine 17 (2012) 180–181

Contents lists available

Seminars in Fetal & Neonatal Medicine

journal homepage: www.elsevier .com/locate/s iny

Correspondence

Reply to Letter to the Editor Re: ‘Myth: Group B streptococcal infection in pregnancy:Comprehended and conquered’ published in volume 16 (2011) pp. 254–258

Dear Sir,

Thank you for allowing us to respond to the letter from MrMarshall and Professor Peckham.

Marshall and Peckham state that in our article we estimatedthat in the ‘absence of any antibiotic prophylaxis 700 babieswill be affected by early onset group B streptococcal disease(eogbs) each year’. In fact, we referred to ‘serious GBS infections’,which includes late onset. While it is true that the Royal Collegeof Obstetricians and Gynaecologists’ (RCOG) 2003 guidelines rec-ommended risk factor screening, according to a survey commis-sioned by the National Screening Committee, carried out by theRCOG and published in 2007 (http://www.rcog.org.uk/our-profession/good-practice/audit/prevention-neonatal-group-b-streptoccocal-disease-audit), these guidelines have resulted inonly “a slight improvement in the proportion of units offeringIAP to appropriate women since the previous surveys in 1999and 2001”. Moreover, recent UK cost-effectiveness surveys haverepeatedly concluded that “Testing (only) high-risk women formaternal GBS colonisation would not be cost-effective”,1 “Thecurrent strategy of risk-factor-based screening is not cost-effective compared with screening based on culture”2 and that“screening, based on a culture test at 35–37 weeks’ gestation,with the provision of antibiotics to all women who screened posi-tive (is) most cost-effective”.3 While historically, the rate ofculture proven eogbs in the UK was low, the Health ProtectionAgency (HPA) reports that voluntarily reported cases in EnglandWales and Northern Ireland have risen from 229 in 2003 to 302in 2010. This rising trend is contrary to the major falls seen inthe many countries that have introduced culture based screening,and we see no reason why such falls could not be replicated inthe UK.

The sensitivity of tests employed for screening has to be judgedagainst a ‘gold standard’ such as that currently employed by theHPA – an enriched culture medium with two selective chromo-genic media (personal communication from Androulla Efstratiou).It is obviously true that if an insensitive test is used, the pickuprate will be low, and the CDC report quoted by Marshall and Peck-ham4 says that “when direct agar plating is used instead of selec-tive enrichment broth, as many as 50% of women who are GBScarriers have false-negative culture results”. Unfortunately, it is

DOI of original article: 10.1016/j.siny.2012.02.006.

1744-165X/$ – see front matter � 2012 Elsevier Ltd. All rights reserved.doi:10.1016/j.siny.2012.02.007

the direct agar plating method that is routinely available in NHShospitals, and part of the GBSS campaign is for the sensitive testto be used instead. Marshall and Peckham suggest that only 69%of carriers detected in late pregnancy will remain positive atterm, but the paper they quote5 reviews nine articles and saysthat “Positive predictive values for antenatal GBS cultures rangedfrom 43 to 100% (mean 69%) and negative predictive values from80 to 100% (mean 94%). GBS cultures collected in late pregnancyhad high positive predictive values for colonization duringdelivery. The negative predictive value was high and relativelyconstant regardless of GA”. Again, the sensitivity of the test usedis crucial – to get good results, one must use the most sensitivetest.

Marshall and Peckham report concerns that “The number ofbabies born to women with negative tests has been higher thanexpected”. This may in part reflect the use of suboptimal culturetechniques, but can also be attributed to the treatment paradox.Assuming a sensitivity of carriage detection as low as 85%, of100 babies who would develop eogbs without screening, 85would have mothers detected as carriers by screening. If asMarshall and Peckham concede, intrapartum antibiotic prophy-laxis is at least 85% effective, eogbs would occur in 15 cases wherethe mother was not identified as a carrier, compared with 13cases where she was. So more than 50% of cases with eogbswill result from false negative screening. But the key point hereis that the overall burden of disease will be reduced from 100to 28 – a 72% reduction. Increasing sensitivity to 90% and treat-ment efficacy to 90% results in an 81% reduction. To decry thisimpressive reduction because there remain false negatives seemsperverse.

Finally, optimal approaches to eogbs prevention involve givingintravenous penicillin to all women in preterm labour, whichrenders prior screening unnecessary.

Yours sincerely,

References

1. Colbourn T, Asseburg C, Bojke L, Philips Z, Claxton K, Ades AE, et al.Prenatal screening and treatment strategies to prevent group B strepto-coccal and other bacterial infections in early infancy: cost-effectivenessand expected value of information analyses. Health Technol Assess 2007;11(29):1–226. iii.

2. Kaambwa B, Bryan S, Gray J, Milner P, Daniels J, Khan KS, et al. Cost-effective-ness of rapid tests and other existing strategies for screening and manage-ment of early-onset group B streptococcus during labour. BJOG 2010;117(13):1616–27.

Correspondence / Seminars in Fetal & Neonatal Medicine 17 (2012) 180–181 181

3. Daniels J, Gray J, Pattison H, Roberts T, Edwards E, Milner P, et al. Rapid testingfor group B streptococcus during labour: a test accuracy study with evalua-tion of acceptability and cost-effectiveness. Health Technol Assess 2009;13(42):1. iv.

4. MMWR. Centers for Disease Control and Prevention, Prevention of PerinatalGroup B Streptococcal Disease: Revised Guidelines from CDC. 59, No. RR-10, 1–36. 2010.

5. Valkenburg-van den Berg AW, Houtman-Roelofsen RL, Oostvogel PM,Dekker FW, Dorr PJ, Sprij AJ. Timing of group B streptococcusscreening in pregnancy: a systematic review. Gynecol Obstet Invest2010;69(3):174–83.

Philip Steer*Emeritus Professor, Imperial College London, United Kingdom

Jane PlumbGroup B Strep Support, PO Box 203, Haywards Heath, West Sussex,

RH16 1GF, United Kingdom

* Corresponding author.E-mail address: p.steer@imperial.ac.uk (P. Steer)

at SciVerse ScienceDirect

Seminars in Fetal & Neonatal Medicine 17 (2012) 182

Contents lists available

Seminars in Fetal & Neonatal Medicine

journal homepage: www.elsevier .com/locate/s iny

Lessons from the current literature

Prolonged initial empirical antibiotic treatment is associated with adverseoutcomes in premature infants

Luc CornetteDepartment of Neonatology, AZ St-Jan Bruges-Ostend AV, Ruddershove 10, 8000 Bruges, Belgium

VS Kuppala, J Meinzen-Derr, AL Morrow, KR Schibler J Pediatr2011;159:720e5

Abstract

In this retrospective cohort study, the authors investigate therelationship between

� prolonged (�5 days) empirical antibiotic administration ton ¼ 365 premature infants (�32 weeks GA and �1500 g birthweight), who survived free of sepsis and necrotizing entero-colitis (NEC) for 7 days during their first week of lifeand� late onset sepsis (LOS), NEC and death.

Multivariable logistic regression is conducted, controlling e.g.for race, prolonged premature rupture of membranes, as well asdegree of respiratory support. In n ¼ 131/365 infants (36%), theprolonged course of antibiotics is independently associated withLOS (OR, 2.45 [95% CI, 1.28-4.67]) and with the combination of LOS,NEC, or death (OR, 2.66 [95% CI, 1.12-6.3]). It is concluded thatprolonged administration of empirical antibiotics to prematureinfants with sterile cultures in the first week of life is associatedwith subsequent severe outcome.

Comment

Antibiotics are the most commonly used therapeutics inneonatal intensive care units. Clinicians have a low threshold forinitiating empiric antibiotic therapy in high-risk newborns, mostlyin view of the devastating consequences of untreated sepsis.Virtually all extremely low birth weight infants receive empirical

E-mail address: Luc.Cornette@azsintjan.be.

1744-165X/$ e see front matterdoi:10.1016/j.siny.2012.02.009

antibiotics in the first postnatal days, although cultures remainsterile. However, prolonged or continued initial empirical antibiotictherapy with little if any microbiological justification may not bebenign, as antibiotic therapy itself can result in elimination ofcommensal flora as well as colonization by multi-drug resistantorganisms, including Candida. The current study by Kuppala et al.confirms such association in a rather impressive way: each extraday of antibiotic therapy is associated with a significant increase inthe odds of NEC, LOS or death.

How to rationally curtail any possible excessive use of antibi-otics in preterm infants ? In this respect, and hopefully in a not toodistant future, our decision to treat preterm infants with sterilecultures and without signs of sepsis beyond the first 3 postnataldays may be more informed with the development of adjunctivediagnostic tests such as levels of procalcitonin, cell surfacemarkers, cytokines (IL-6 and IL-8) and leukocyte adhesion factors.Sensitivity, specificity, and predictive values are promising formany of these tests, but most are not yet available in clinicalsettings. In the meantime, we must continue to do everythingpossible to ascertain whether an infant truly has an infection thatrequires antibiotic therapy. Such is firstly obtained by educatingour junior doctors towards correct blood sampling, i.e. 2 mL ofblood volume is needed to obtain reliable culture results. Suchmay be technically difficult, but it is well-known that low bloodvolume decreases the sensitivity of blood cultures. Secondly, wecould reduce overtreatment by reducing the likelihood of treatingcontaminants, e.g. by obtaining blood cultures at 2 separate sites,using meticulous sterile conditions. Finally, close adherence toGBS screening and hospital intra-partum antibiotic prophylaxisguidelines remains crucial, in order to reduce the number ofunnecessary sepsis evaluations in newborn infants, resulting inless empiric antibiotic treatment.