The Infant of the Diabetic Mother Richard M. Cowett ... · DOI:€10.1542/neo.3-9-e173 NeoReviews 2002;3;173 Richard M. Cowett The Infant of the Diabetic Mother

DOI: 10.1542/neo.3-9-e173 2002;3;173 NeoReviews

Richard M. Cowett The Infant of the Diabetic Mother

http://neoreviews.aappublications.org/cgi/content/full/neoreviews;3/9/e173located on the World Wide Web at:

The online version of this article, along with updated information and services, is

Online ISSN: 1526-9906. Illinois, 60007. Copyright © 2002 by the American Academy of Pediatrics. All rights reserved. by the American Academy of Pediatrics, 141 Northwest Point Boulevard, Elk Grove Village,it has been published continuously since 2000. NeoReviews is owned, published, and trademarked NeoReviews is the official journal of the American Academy of Pediatrics. A monthly publication,

. Provided by McMaster University on April 3, 2010 http://neoreviews.aappublications.orgDownloaded from

http://neoreviews.aappublications.org/cgi/content/full/neoreviews;3/9/e173

http://http://neoreviews.aappublications.org

The Infant of theDiabetic MotherRichard M. Cowett, MD* Objectives After completing this article, the reader should be able to:

1. List the most important factors producing a normal neonate without untowardcomplications for the diabetic mother.

2. Explain the “mixed nutrient” hypothesis.3. Describe the most probable cause of congenital malformations.4. Describe the morbidities that have been reported in the infants of mothers who have

uncontrolled diabetes.

IntroductionThe infant of the diabetic mother (IDM) is the premier metabolic example of themorbidity that may exist in the neonate due to maternal disease (diabetes). From adevelopmental standpoint, the normal neonate is in a transitional state of glucose ho-meostasis. The fetus is completely dependent on the mother for glucose delivery, and theadult is considered to have control of glucose homeostasis because plasma glucoseconcentration is regulated precisely. (1) In contrast, maintenance of glucose homeostasis maybe a major problem even for the normal neonate. (2) The precarious nature of this equilibriumis emphasized by the numerous morbidities producing or associated with neonatal hypo- andhyperglycemia. Although many IDMs have an uneventful perinatal course, there is still anincreased risk of complications. Many of these can be minimized, but not eliminated, withappropriate obstetric and pediatric intervention. Because a recent analysis indicated that thereis still much room for improvement due to the multiplicity of factors that affect any specificpregnancy, (3) the obstetrician and pediatrician must be knowledgeable about these potentialproblems. This discussion enumerates many of the difficulties that the IDM may encounterand evaluates the pathophysiologic basis of their occurrence.

Perinatal Mortality and MorbidityTheoretically, closer metabolic control of the pregnant woman who has diabetes results ina greater potential for producing a normal neonate. The physician responsible for the careand delivery of the mother must inform the physician responsible for the care of theneonate of the mother’s condition well in advance of delivery. Certainly, the pregnancy ofthe diabetic mother should be considered high risk. Knowledge of the character of thematernal diabetes, prior pregnancy history, and complications occurring during pregnancyallows the physician caring for the neonate to anticipate many potential fetal and neonatalcomplications and to be present at delivery (Table 1).

Studies of perinatal morbidity and mortality from diverse centers attest to the improvingsuccess of managing the pregnant diabetic woman with appreciation of the high-risk

nature of the pregnancy. A number of years ago, Pedersenand associates (4) published a review of their experiencesover a 26-year period with an analysis of 1,332 diabeticpregnancies. Perinatal mortality varied directly with maternalseverity of diabetes, as judged by two commonly used ma-ternal classification schema: White’s original classification ofdiabetes in pregnancy and Pedersen’s Prognostically BadSigns in Pregnancy (PBSP) classification. White’s revisedclassification (Table 2) is based on duration of diabetes and

* Professor of Pediatrics, Northeastern Ohio Universities College of Medicine, Rootstown, Ohio.

Abbreviations

IDM: infant of the diabetic motherPBSP: Prognostically Bad Signs in PregnancyRDS: respiratory distress syndromePTH: parathyroid hormone

Article endocrinology

NeoReviews Vol.3 No.9 September 2002 e173. Provided by McMaster University on April 3, 2010 http://neoreviews.aappublications.orgDownloaded from


the presence of late vascular complications; (5) the PBSPclassification (Table 3) includes abnormalities of the cur-rent pregnancy.

A report from the Joslin Clinic supports the impor-tance of these factors, especially preeclampsia (pretox-emia), as a significant morbidity in the pregnant diabeticwoman. (6) Of 420 patients in the series who had type 1diabetes, 110 (26.2%) delivered before 37 weeks’ gesta-tion compared with an incidence of 9.7% in the nondia-betic population. One third of the preterm deliverieswere related to preeclampsia. The authors concludedthat a major problem of the diabetic pregnancy relates tothe problem of preeclampsia that produces prematurity.The risk to the fetus was increased when the PBSPclassification was “added” to the White classification.The importance of preeclampsia recently was reiteratedas a significant clinical issue in the care of the diabeticpregnancy. (7)

The relationship between PBSP and preeclampsia wasemphasized by Diamond et al, (8) who studied 199pregnancies. They noted that the presence of PBSPincreased the perinatal mortality rate from 7.3% to 17.1%and was predictive of general pulmonary morbidity(31.6% versus 16.3%). The investigators concluded thatthe combination of the two classification systems re-mained as predictive as had been noted by Pedersen.

Although these investigators noted an improvement innondiabetic pregnancy outcome during this same pe-riod, they emphasized that the improved classificationschema combined with increased experience were themajor reasons for the improved results in the diabeticpregnancy. This improved perinatal mortality has beenconfirmed at many centers in the United States and inEurope. The frequency of macrosomia has decreased,but the rate remains higher than that in the neonate bornto the woman who does not have diabetes. In a survey ofmacrosomic neonates (ie, large-for-gestational age,�95th percentile weight for gestational age), most wereborn to obese mothers, not all of whom had glucoseintolerance. (9)(10) Nevertheless, the woman who hasgestational diabetes with glucose intolerance during latepregnancy may remain undiagnosed and may deliver aneonate who has a greater risk for perinatal complica-tions.

Table 2. White’s Classification ofDiabetes in Pregnancy (Modified)

Gestationaldiabetes

Abnormal glucose tolerance test, buteuglycemia maintained by diet alone;if diet alone insufficient, insulinrequired

Class A Diet alone, any duration or age ofonset

Class B Age of onset: >20 y; duration: <10 yClass C Age of onset: 10 to 19 y; duration:

10 to 19 yClass D Age of onset: <10 y; duration: >20 y;

background retinopathy orhypertension (not preeclampsia)

Class R Proliferative retinopathy or vitreoushemorrhage

Class F Nephropathy, with >500 mg/dproteinuria

Class RF Criteria for both classes R and Fcoexist

Class H Arteriosclerotic heart disease clinicallyevident

Class T Prior renal transplantation

Table 3. Prognostically Bad Signsof Pregnancy (PBSP)● Chemical pyelonephritis● Precoma or severe acidosis● Toxemia● “Neglecters”

Table 1. Morbidities in the IDM● Asphyxia● Birth injury● Caudal regression● Congenital anomalies● Double-outlet right ventricle● Heart failure● Hyperbilirubinemia● Hypertrophic obstructive cardiomyopathy● Hypocalcemia● Hypoglycemia● Hypomagnesemia● Increased blood volume● Macrosomia● Neurologic instability (short- and long-term)● Organomegaly● Polycythemia and hyperviscosity● Renal vein thrombosis● Respiratory distress● Respiratory distress syndrome● Septal hypertrophy● Shoulder dystocia● Small left colon syndrome● Transient hematuria● Truncus arteriosus

endocrinology diabetes

e174 NeoReviews Vol.3 No.9 September 2002. Provided by McMaster University on April 3, 2010 http://neoreviews.aappublications.orgDownloaded from


Maternal glucose variability has been studied in 154pregnant diabetic patients who were hospitalized for 1month prior to delivery. (11) There was a significantassociation between maternal glucose variability and en-hanced neonatal outcome (ie, decreased incidence ofcomplications) and no correlation between maternal glu-cose variability and the birthweight of the neonate. Theauthors acknowledged that absence of glucose variabilitywould not ensure prevention of neonatal complications.A variation on this theme was reported by Mello et al,(12) who suggested that overall daily glucose concentra-tions of 95 mg/dL (5.3 mmol/L) or less were requiredin the second and third trimesters to avoid alterations (ie,excesses) in fetal growth.

Roberts and Pattison (13) reported on a 20-yearexperience involving 1,528 pregnancies of diabeticwomen of whom 571 had type 1 diabetes and 957 hadgestational diabetes. The perinatal mortality rate de-creased from 15.2% to 2% in those who had type 1 andfrom 6.7% to 0.5% for those who had gestational diabe-tes. The investigators related the improvement in mor-tality to improved glucose control. They reported, asnoted by others, that the major outstanding problem wasthe persistent high incidence of congenital malforma-tions.

Coustan and Imarah (14) first attempted to use pro-phylactic insulin in women who had gestational diabetesto reduce the incidence of macrosomia, operative deliv-ery, and birth trauma. There was a partial decline ofcomplications with tightened maternal metabolic con-trol. Subsequently, the same investigators evaluated in-sulin pump or intensive conventional therapy by ran-domizing 22 pregnant diabetic women to the twotreatments. (15) No significant differences were foundwith either regimen, and excellent glucose control wasachieved with both. Howorka et al (16) reported thenormalization of pregnancy outcome with the use offunctional insulin treatment through individualization ofinsulin dosages during the day.

The use of insulin therapy was also reported in 108women who had gestational diabetes and were random-ized to receive diet plus insulin or diet alone to maintainglycemic control. (17) Among the patients treated for atleast 6 weeks with diet plus insulin, the mean birth-weight, the incidence of macrosomia, and the ponderalindex were reduced. No patient who weighed less than200 lb (90 kg) and maintained euglycemic control deliv-ered a neonate who weighed more than 4,000 g. Theinvestigators concluded that maternal obesity or failureto achieve glycemic control should alert the clinician toan increased risk of macrosomia.

Norlander and associates (18) found that perinatalmorbidity was significantly more frequent in the womenwho had gestational diabetes (23%) than in the controlgroup (13%). The occurrence of large-for-gestational ageneonates did not differ between groups. Of those born tothe women who had gestational diabetes, the neonateswho presented with morbidities were of earlier gesta-tional age at delivery, were delivered more frequently bycesarean section, and had mothers who had higherprepregnancy weights and greater area under the glucosetolerance curve. Gestational age at delivery and maternalprepregnancy weight were the most significant factors.The investigators concluded that factors beyond bloodglucose control during pregnancy were critical to neona-tal outcome in this population.

A population-based study compared 68 women whohad diabetes requiring insulin treatment and 403 treatedwith diet alone with a random sample of 1 in 12 of 893nondiabetic women who delivered in one regional hos-pital. (19) No relationship was noted between maternalglycosylated hemoglobin at delivery and the neonatalbirthweight. At each week of gestation, the infants bornto the diabetic women were heavier than the infants bornto the nondiabetic women (P�0.05). No differenceswere noted in maternal glycosylated hemoglobin be-tween the two groups throughout pregnancy. Factorsaffecting birthweight included diabetes, ethnic origin,and parity. The investigators concluded that substratesother than glucose, which induce hyperinsulinemia, maybe related to the higher birthweight of the neonate.

Finally, Hod and colleagues (20) reported data eval-uating the effect of patient compliance, fasting plasmaglucose on the oral glucose tolerance test, maternal bodyconstitution, and method of treatment on perinatal out-come of patients who had gestational diabetes mellitus.A total of 470 patients were compared with 250 controlnondiabetics. Patient compliance reduced the rate ofmacrosomia (14.4%) and neonatal hypoglycemia (3.4%),but not to the level of the control population (5.2% and1.2%, respectively). Intensified insulin treatment reducedthe rate of perinatal complications in the obese parturi-ents, but again, not to the level of the control group.

Although most investigators have agreed on the im-portance of maintaining euglycemia, the optimal clinicalmethod has not been established. deVeciana et al (21)compared the efficacy of postprandial and preprandialmonitoring to achieve glycemic control in the gestationaldiabetic woman. Sixty-six women who were at at least30 weeks’ gestation were treated with insulin therapyfollowing either preprandial monitoring or postprandialmonitoring 1 hour after a meal. The change in glycosy-




lated hemoglobin was greater in the postprandial group,and the birthweights of the neonates were lower. Simi-larly, there was a lower rate of neonatal hypoglycemia inthe large-for-gestational age neonates and a lower deliv-ery rate by cesarean section.

Maintenance of a normal metabolic state, includingeuglycemia, should diminish, but will not completelyeradicate the increased perinatal and neonatal mortalitiesand morbidities noted in the diabetic pregnancy.

Pathogenesis of the Effects of MaternalDiabetes on the FetusAs yet, no single pathogenic mechanism has been clearlydefined to explain the diverse problems observed in theIDM. Nevertheless, many of the effects can be attributedto maternal metabolic (ie, glucose) control. Pedersenoriginally emphasized the relationship between maternalglucose concentration and neonatal hypoglycemia (Ta-ble 4). (22) His simplified hypothesis recognized thatmaternal hyperglycemia resulted in fetal hyperglycemia,which stimulated the fetal pancreas, resulting in islet cellhypertrophy and beta cell hyperplasia with increasedinsulin availability. Following delivery, the neonate, whono longer was supported by placental glucose transfer,developed neonatal hypoglycemia.

Hyperinsulinemia in utero affects diverse organ sys-tems, including the placenta. Insulin acts as the primaryanabolic hormone of fetal growth and development,resulting in visceromegaly, especially of heart and liver,and macrosomia. In the presence of excess substrate,such as glucose, fat synthesis and deposition increasesduring the third trimester. Fetal macrosomia is reflectedby increased body fat, muscle mass, and organomegaly,but not by an increased size of the brain or kidney.(23)(24) After delivery, there is a rapid fall in plasmaglucose concentration, with persistently low concentra-tions of plasma free fatty acids, glycerol, and beta-

hydroxybutyrate. In response to an intravenous glucosestimulus, plasma insulin-like activity is increased, as isplasma immunoreactive insulin, determined in the ab-sence of maternal insulin antibodies and plasmaC-peptide concentration. (25) The insulin response tointravenous arginine is also exaggerated in the infant of awoman who has gestational diabetes. (26)

In a follow-up study using the chronic hyperinsuline-mic fetal rhesus monkey, Susa and associates (27) studiedneonatal insulin secretion following delivery. Comparedwith controls, the experimental group evidenced ablunted insulin and C-peptide response to infusion of300 mcg/kg glucagon to stimulate insulin secretion.The investigators suggested that fetal hyperinsulinemiainhibited insulin synthesis and secretion in utero and thatthese alterations persisted after birth.

MacFarlane and Tsakalakos (28) suggested that theinitial increase in fetal size due to fetal hyperinsulinemiaproduced developing hypoxemia. The limitation in fetaloxygen availability altered differential utilization of glu-cose and increased alpha-glycerophosphate synthesis inthe fetal adipocyte, which resulted in fetal adiposity.

Schwartz et al (29) evaluated whether macrosomia inthe fetus of the diabetic mother is related to fetal hyper-insulinemia and whether hyperinsulinemia and macroso-mia are related to maternal metabolic control. A total of95 nondiabetic pregnant women were compared with155 insulin-treated pregnant women who were subdi-vided according to the White classification, the presenceof hypertension, the birthweight, and the mode of deliv-ery. Optimal care was provided, and the neonates wereevaluated. Macrosomia (�97th percentile) was noted in10% to 27% of the diabetic groups and was correlatedwith umbilical total insulin, free insulin, and C-peptideconcentrations. Glycosylated hemoglobin was only aweak predictor of birthweight and fetal hyperinsulin-emia. The investigators concluded that the cause ofmacrosomia essentially remains unexplained, but hyper-insulinemia is the major stimulus for excessive fetalgrowth.

In an evaluation of specific material factors, adiposityin 119 term neonates born to women who had gesta-tional diabetes (57 large for gestational age, 62 appropri-ate for gestational age) was compared with a controlgroup of term neonates (74 large for gestational age, 69appropriate for gestational age). (30) Multiple regressionanalyses suggested that pregnancy weight and weightgain were significant predictors of neonatal body massindex for both groups. Second and third trimester glu-cose concentrations were significant predictors for thebody mass index of infants born to women who had

Table 4. Components for theHypothesis of“Hyperinsulinism” in the IDM● Islet hyperplasia and beta-cell hypertrophy● Obesity and macrosomia● Hypoglycemia with low free fatty acid concentration● Rapid glucose disappearance rate

—Higher plasma insulin-like activity after glucose—Umbilical vein reactive immunoinsulin increase

● Elevated C-peptide and proinsulin concentrations




gestational diabetes; a significant glucose screen pre-dicted body mass index in the control group. Increasedadiposity in the IDMs was related to increased neonatalblood pressure.

Lipase activities were measured in the placentae of ratsmade diabetic by streptozotocin treatment and in theplacentae of women identified as having type 2 diabetes(ie, impaired glucose tolerance) or type 1 diabetes withor without vascular complications. (31) Normals wereevaluated for comparison. At pH 4, lipase activity in-creased in both the rat and the human placentae com-pared with controls. In the women, placental lipase ac-tivity at pH 4 correlated with birthweight in patients whohad impaired glucose tolerance, suggesting that in-creased activity may contribute to increased fetal weightby adding to fat transfer across the placenta.

An interesting variation on the theme of the effects ofmaternal diabetes on the fetus was addressed by Homkoet al, (32) who evaluated the relationship between eth-nicity and gestational diabetes relative to macrosomia inan urban academic diabetes program. Between 1991 and1994, gestational diabetes mellitus was diagnosed in 103African-American women and 36 Latino women. Allfactors being equal, macrosomia developed in 50% of theneonates of the Latino women versus 19% of the neo-nates of the African-American women. The investigatorssuggested that, at least in this series, there may be anethnic variation in fetal growth.

Finally, the cause of macrosomia in the IDM wasfurther evaluated by the National Institute of ChildHealth and Human Development-Diabetes in EarlyPregnancy Study, which recruited insulin-dependent di-abetic and control women before conception and pro-vided an opportunity to evaluate the relationship be-tween maternal glycemia and percentile birthweight.(33) Data were analyzed from 323 diabetic and 361control women. Fasting and nonfasting venous plasmaglucose concentrations were measured on alternateweeks in the first trimester and monthly thereafter. Gly-cosylated hemoglobin was measured weekly in the firsttrimester and monthly thereafter. More infants of thediabetic women were at or above the 90th percentile forbirthweight than infants of control women (28.5% versus13.1%, P�0.001). The third-trimester nonfasting glu-cose concentration, adjusted for data in prior trimesters,was the stronger predictor of percentile birthweight(P�0.001). After adjusting for maternal hypertension,smoking, and ponderal index, the investigators con-cluded that monitoring of the nonfasting glucose con-centration rather than the fasting concentration, which is

more commonly monitored in clinical practice, is neces-sary to prevent macrosomia.

Metabolic AnalysesIn vivo kinetic analysis has been used by numerousinvestigators to evaluate the IDM metabolically. An earlystudy employing stable nonradioactive isotopes used[1-13C]glucose and the prime constant infusion tech-nique. (34) The investigators measured systemic glucoseproduction rates in five infants of normal (nondiabetic)mothers and five infants of insulin-dependent diabetics at2 hours of age. As expected, the IDMs had a lowerglucose concentration during the study. For the firsttime, the investigators reported that the IDM had alower systemic glucose production rate. They suggestedthat decreased glucose output was related to inhibitedglycogenolysis and speculated that increased insulin anddecreased glucagon concentrations and catecholamineresponses resulted in decreased systemic output. For thetime studied (late 1970s), the women who had diabeteswere considered to be in excellent control. They hadbeen hospitalized during the last 4 weeks of the preg-nancy to achieve strict metabolic control, with maternalblood glucose levels maintained at 50 to 150 mg/dL(2.8 to 8.3 mmol/L). Yet, the systemic glucose produc-tion rates of these neonates were lower than that of thecontrol neonates.

A further evaluation of the IDM was reported by thesame group 5 years later. (35) Again focusing on theneonate of the mother in “strict control,” the investiga-tors evaluated systemic glucose production in five infantsof insulin-dependent mothers, one neonate of a womanwho had gestational diabetes, and five neonates born towomen who did not have diabetes. The blood glucosedata were in a more restrictive range of 36 to 104 mg/dL(2.0 to 5.8 mmol/L), and the glucose levels were con-trolled in the hospital setting for 3 to 4 weeks prior todelivery. In this series, the systemic glucose productionrate was similar in all infants. However, these investiga-tors, like other groups (36), carried their analysis a sig-nificant step further. They infused exogenous glucose,which can diminish endogenous glucose production,because of the precise control known to be the hallmarkof the adult. The IDM did not evidence as great asuppression of endogenous glucose production as theadult. The investigators concluded that altered regula-tion of glucose production may be due to intermittentmaternal hyperglycemia, even in the woman whose glu-cose levels are strictly controlled.

These studies paralleled the work of Cowett et al (37),who have studied glucose kinetics in the neonate. Using




78% enriched D[U-13C] glucose, 16 infants of diabeticwomen (10 of whom were insulin-dependent and 6 ofwhom were chemical-dependent) were compared withfive infants of nondiabetic women. Four infants ofinsulin-dependent mothers and five infants of chemicaldiabetic mothers received 0.45% saline as the stable iso-topic tracer diluent to determine basal endogenous glu-cose production. All of the mothers were evaluated rela-tive to control mothers by use of hemoglobin A1C andmaternal plasma glucose or cord vein glucose at delivery.None of the women were maintained in the hospitalprior to study. There was a similarity between the basalglucose production rate in the neonates who received noexogenously infused glucose. The investigators con-cluded that good metabolic control of the maternaldiabetic state would help maintain euglycemia. How-ever, in a subsequent analysis in which neonates of non-diabetic mothers received glucose exogenously to main-tain euglycemia, heterogeneity continued to exist in theability of the neonate to depress endogenous glucoseproduction. (38) These latter data parallel other workfrom the same group that reflect the transitional natureof glucose metabolism in the term and preterm infant,born to both diabetic and nondiabetic mothers. (36)(39)

Baarsma et al (40) followed 15 mother-infant pairsfrom the beginning of pregnancy until birth, measuringglucose kinetics on the first day after birth with stableisotopic dilution as well as plasma free fatty acids andketones. The neonates received 3.4�0.7 mg/kg�1 perminute�1 glucose during the study. No relationship ex-isted between maternal control and glucose kinetics inthe neonate. Total production was 5.2�1.1 mg/kg�1

per minute�1, and endogenous glucose production was1.8�1.1 mg/kg�1 per minute�1 following subtractionof the glucose infusion. Endogenous glucose productionwas significantly lower in the neonates studied at the endof the first day after birth. The lower production rate wasassociated with an increased concentration of ketonebodies, which suggested increased production. The in-vestigators concluded that glucose kinetics in the infantsof women whose diabetes is tightly controlled are prob-ably normal.

The realization that neonatal glucose homeostasis isin a transitional state and is not the only factor affectingneonatal birthweight is further supported by studies inwhich maternal control was evaluated in a group ofwomen who had gestational diabetes relative to thebirthweight of the neonate. (41) If the Pedersen hypoth-esis were correct, neonatal birthweight should correlatewith the degree of diabetic control of the mother duringthe pregnancy. There was a lack of correlation between

birthweight and mean maternal plasma glucose concen-tration during the third trimester of pregnancy in thisgroup of women who had gestational diabetes (Fig. 1).This lack of correlation further supports the heterogene-ity of the diabetic state and suggests that both control ofglucose homeostasis and control of fetal growth aremultifactorial. Similar conclusions led Freinkel (42) andothers to decide that mixed nutrients (ie, amino acids,free fatty acids) other than glucose are important infetal-neonatal metabolic control (Fig. 2). This concept isimportant for ongoing research.

Support for this concept has been provided by Kalk-hoff and associates (43), who studied the relationshipbetween neonatal birthweight and maternal plasmaamino acid profiles in lean and obese nondiabetic womanand in women who had type 1 diabetes. HbA1c, plasmaglucose concentration, and total amino acid profiles wereelevated in the diabetic patients compared with controls,but there were no differences between obese and leancontrol groups. Plasma glucose concentrations and pro-files of HbA1c did not correlate with relative weight ofthe neonate; average total plasma amino acid concentra-tions did. The investigators concluded that maternalplasma amino acid profiles may influence fetal weightgenerally and affect the development of neonatal macro-somia.

Some further investigation of protein metabolism inpregnancy was performed by Whittaker and associates.(44) They evaluated protein kinetics in six nondiabeticand five insulin-dependent diabetic women during andafter pregnancy using stable isotopes and thehyperinsulinemic-euglycemic clamp and amino acid in-fusions. Evaluation of protein breakdown, as measuredby leucine kinetics, was not higher than normal, andneither pregnancy nor type 1 diabetes altered insulinsensitivity to amino acid turnover. The data were inter-preted to suggest that alterations may enable conserva-tion of amino acids for protein synthesis and accretion inlate pregnancy.

Other investigators evaluated the relationship be-tween neonatal birthweight and plasma triglyceride con-centrations. (45) They drew plasma samples for a num-ber of metabolic parameters 1 hour after administrationof a 50-g glucose load in 521 randomly selected nega-tively screened individuals, 264 positively screened indi-viduals who had negative findings on glucose tolerancetests, and 96 positively screened individuals who hadpositive glucose tolerance test results at 24 to 28 weeks ofgestation. Plasma triglyceride concentration was the onlytest elevated in the women who had gestational diabetes,but not in the negative glucose tolerance test group.




Only plasma triglyceride concentration was significantlyrelated to birthweight ratio, with birthweight correctedfor gestational age and to glucose intolerance besidesglucose itself. The investigators concluded that plasmatriglyceride may be a physiologic contributor to neonatalbirthweight (Fig. 3).

This concept was further analyzed by Kitajima et al(46), who evaluated plasma triglyceride concentrationsbetween 24 and 32 weeks’ gestation in women who hadpositive diabetic screening results but negative glucosetolerance test results. Plasma triglyceride concentrationsof more than 259 mg/dL (2.9 mmol/L) represented thesignificant factor related to the development of macroso-mia rather than maternal obesity or maternal plasmaglucose concentrations at these weeks of gestation.Clearly, metabolites rather than glucose contribute tothe propensity of fetal/neonatal macrosomia and need tobe evaluated prospectively.

Congenital AnomaliesAlthough most of the morbidity and mortality data forthe IDM improve with time, congenital anomalies re-main a significant unresolved problem. In a population-based study of 7,958 infants over 12 years, Becerra et al(47) documented differences in congenital malformationbetween the IDM and the nonIDM. The relative rate ofmajor malformations in the neonate born to the motherwho had insulin-dependent diabetes mellitus was 7.9compared with the neonate of the nondiabetic mother.Similarly, the relative risks for central nervous system andcardiovascular system defects were 15.5 and 18.0, re-spectively. Interestingly, the neonate of the mother whohad gestational diabetes and required insulin therapy was20.6 times as likely to have a cardiovascular malformationas the neonate of the nondiabetic mother. This finding isfrom a time when centers were reporting no difference inperinatal mortalities among offspring of women who hadinsulin-dependent diabetes and women who did nothave diabetes after correction for death due to congenitalanomalies. (48)

These data were refined in a recent series reviewingthe relationship between specific birthweight and thepresence of major congenital malformations. (49) A totalof 8,226 neonates in the Texas Birth Defects Monitoringfiles were compared with a control group of almost1,000,000 neonates who did not have defects. Infantswho had 45 specific defects were more likely to havebirthweights greater than 4,500 g. In contrast, investiga-tors in another series reported that the odds ratio ofdelivering a neonate who had congenital heart diseasewas higher in nondiabetic women if their prepregnancy

weights indicated obesity. (50) Both of these data setsdeserve further evaluation and confirmation.

The pathogenesis of the increase in congenital anom-alies among IDMs has remained obscure, although sev-eral causes have been proposed, including hyperglycemia(either preconceptional or postconceptional), hypogly-cemia, fetal hyperinsulinemia, uteroplacental vasculardisease, and genetic predisposition. A review has co-gently summarized the relevant data obtained from in-vestigations during the 1980s. (51) There are data tosupport each of these proposed mechanisms, but currentevidence seems strongest for postconceptional hypergly-cemia.

If a preconceptional influence of hyperglycemia orhypoglycemia or a genetic predisposition for congenitalanomalies were operative, the offspring of the diabeticfather as well as of the diabetic mother would be expectedto have an increased incidence of anomalies. This as-sumes that the sperm and egg are equally affected by thephysiologic and biochemical permutations of maternaldiabetes. In a careful hospital chart review by Neave (52)of 1,262 offspring of diabetic fathers, only a slight in-crease in anomalies of questionable significance wasfound compared with matched controls. In this samestudy, a marked increase in anomalies was found amongthe offspring of the diabetic mother compared with boththe offspring of the diabetic father and an independentcontrol group.

One of the first studies of normalization of bloodglucose concentration before conception in the diabeticwoman was reported from a large European diabeticpopulation in Karlsburg, Germany, that included non-pregnant women. (53) These women were cared for inan ongoing diabetic outpatient program to normalizeblood glucose concentration. The infants of the mothersin the study group had a markedly lower incidence ofcongenital anomalies compared with infants of a simul-taneously studied group of women who received noprenatal therapeutic diabetic regimen.

These conclusions were confirmed in a group of 75insulin-dependent diabetic women of whom 44 werefollowed preconceptually. (54) Glycemic control wasobtained by intensified insulin therapy. There were nocongenital malformations in the intensively followedgroup compared with a frequency of 9.6% in the controlgroup, whose diabetes was not managed preconceptu-ally.

In a series of women who had gestational diabetes,136 underwent preconceptual counseling at least 2months before the onset of pregnancy. (55) Evaluationincluded oral glucose tolerance testing, mean blood glu-




cose measurement, assessment of glycosylated hemoglo-bin, and management by self-monitoring as well as nu-tritional counseling. These women were compared witha group of 154 women who did not undergo this pro-gram. Those who participated in the preconceptualcounseling delivered their neonates without congenitalmalformations.

In recent years, a number of studies have reported thestrong association of preconceptional testing and glyce-mic control with a marked diminution in the incidence ofcongenital anomalies. These data parallel other reportssuggesting that control after the first trimester did notresult in a decreased incidence of congenital malforma-tions, although other morbidity did decrease. In fact, theneonatal malformation rate rose and was not influencedby maternal age or diabetic class. (56)

Hypoglycemia may play a teratogenic role in thediabetic pregnancy. Symptomatic hypoglycemia duringthe first trimester is a frequently observed morbidity inthe insulin-dependent diabetic, although quantificationhas been difficult. The injection of insulin into the chickembryo has induced “rumplessness,” (57) but data indi-cate that the primate placenta probably acts as a completebarrier to maternal insulin from midgestation onward.(58) The failure of insulin to cross the placenta in the ratduring the critical period of organogenesis was noted byWidness and associates (59) using iodinated insulin.

One rare congenital defect that is increased in fre-quency in the IDM is the small left colon syndrome. (60)The cause of this deformity is obscure. With conservativemedical management, the condition usually resolvesspontaneously within the neonatal period.

The association of maternal diabetes and cardiovascu-lar malformations was reported from the BaltimoreWashington Infant study, a population-based case-control study of cardiovascular malformations. (61) Thestrongest associations with overt type 1 diabetes were fordouble-outlet right ventricle and truncus arteriosus. Noassociations were noted with gestational diabetes.

Weber and colleagues (62) suggested that hypertro-phic cardiomyopathy and abnormal ventricular diastolicfilling in the IDM is related to poor maternal glycemiccontrol. These investigators evaluated the fetuses of well-controlled mothers and of nondiabetic women. Cardiacgrowth and ventricular diastolic filling were evaluated20 to 26, 27 to 33, 34 to 40, and 48 to 72 hours afterbirth. No differences were noted between groups duringany period, although progression of diastolic filling isabnormally delayed in the fetus of the diabetic womanand is presumably more likely in the poorly controlleddiabetic.

Finally, a number of publications have raised thepossibility that vitamin E may provide a protective effectagainst diabetic embryopathy through its well-knownantioxidant effect. (63)(64)(65) Further research is re-quired to determine the clinical applicability of thesedata.

Macrosomia, Birth Injury, and AsphyxiaAt birth, the infant of the poorly controlled diabeticoften appears macrosomic in contrast to the infant bornto either the well-controlled diabetic or the nondiabetic,nonobese mother. One consequence of undetected fetalmacrosomia may be a difficult vaginal delivery due toshoulder dystocia, with resultant birth injury or asphyxia.These potential birth injuries include cephalhematoma,subdural hemorrhage, facial palsy, ocular hemorrhage,brachial plexus injuries, and clavicular fracture. Injury tothe brachial plexus may have a variety of presentationsbecause the nerves of the brachial plexus may be variablydamaged. In addition to the obvious injury to the nervesof the arm, diaphragmatic paralysis occurs when thephrenic nerve is injured. Because of associated organo-megaly in the IDM, hemorrhage in the abdominal or-gans is possible, specifically the liver and the adrenalgland. Hemorrhage into the external genitalia of thelarge neonate has been reported.

In a study of the incidence and predisposing factorsfor clavicular fracture in the neonate, 46 of 3,030 neo-nates were diagnosed as having clavicular fracture. (66)When compared with a control group of 52 neonateswho did not have a fracture, the affected neonates hadhigher birthweights, older mothers, longer second stageof labor, and higher rates of instrumented delivery andshoulder dystocia. Eighty percent of the neonatesweighed less than 4,000 g. Multivariate analysis revealedtwo significant predisposing variables: birthweight morethan 3,500 g and maternal age older than 29 years.

Because IDMs are at high risk, intrapartum monitor-ing is essential to minimize potential complications.Clearly, early identification of macrosomia is critical.Mintz et al (67) suggested that shoulder soft tissuemeasurements and abdominal circumstance may be thebest individual predictors of the potential for macroso-mia. The combination of the abdominal circumferencegreater than the 90th percentile and the shoulder softtissue width greater than 12 mm was the best predictor,with a sensitivity of 96%, specificity of 89%, and accuracyof 93%.

Although the specific cause of asphyxia is unclear, itmay be due to difficulty in the intrapartum period be-cause of relative macrosomia. Asphyxia may have diverse




consequences for the neonate. Acutely, it may affectrespiratory, renal, and central nervous system function.Decreased fluid intake usually is recommended until thedegree of injury to the renal and central nervous systemscan be ascertained. An important complication of as-phyxia in the neonate may be later respiratory difficulty.For infants of women who have insulin-dependent dia-betes, investigators have suggested that poor glycemiccontrol in the third trimester, diabetic vascular disease,preeclampia, and smoking are significant risk factors forperinatal asphyxia. (68) In this prospective study of 162infants born to 149 diabetic mothers of White classB-R-T, 44 neonates (27.2%) had evidence of asphyxia.Its presence did not correlate with third trimester controlor the other factors listed, but it did correlate withnephropathy occurring in pregnancy, maternal hypergly-cemia before delivery, and prematurity. The investigatorsconcluded that in the pregnant diabetic woman, mater-nal and subsequently fetal hyperglycemia before deliveryleads to fetal hypoxemia.

A rare occurrence of gangrene in an upper extremityof an IDM with massive muscle necrosis of the forearmrequiring debridement was noted in one report. (69) Itwas postulated that gangrene developed from a propen-sity for thrombosis in the IDM. Another case was re-ported of an IDM who was diagnosed in utero as havinga brachial artery thrombosis. (70) Of 32 neonates re-ported in the literature, seven who had peripartum limbgangrene were IDMs. The investigators speculated thatthe IDM may be at increased risk for thrombosis if anumbilical artery catheter was placed.

Identification of maternal diabetes and maintenanceof good metabolic control in the pregnant diabeticshould diminish the frequency and magnitude of macro-somia and its attendant complications. Careful obstetricmanagement should prevent birth injury and asphyxia.Ogata et al (71) reported data that seem to confirm thisconcept. Serial studies to estimate fetal biparietal diame-ter and abdominal circumference were used as differen-tial inducers of intrauterine growth in fetuses of motherswho were White class A through C compared with fe-tuses of nondiabetic mothers. Biparietal diameter wassimilar in fetuses of both groups, but abdominal circum-ference was normal or enhanced in the fetuses of diabeticmothers. The group that had enhanced abdominal cir-cumference had an increased insulin concentration,weighed more at birth, and had more subcutaneous fat.The investigators concluded that ultrasonography wasuseful in the preliminary detection of macrosomia.

A recent study evaluated whether macrosomia is asso-ciated with increased perinatal morbidity and mortality

by assessing whether the birth of a previous macrosomicneonate heralded the risk for a subsequent macrosomicneonate. (72) A population-based cohort study usingbirth data from Washington state between 1984 and1990 identified a 7.0 times risk in this clinical situation.The overall prevalence of macrosomia was 22% and wasinterpreted to suggest that a mother who has one mac-rosomic neonate is at markedly increased risk to deliver asecond in a subsequent pregnancy.

Respiratory Distress SyndromeRespiratory distress, including respiratory distress syn-drome (RDS), may be a relatively frequent and poten-tially severe complication in the IDM. Although theincreased susceptibility to RDS has been suspected in theIDM, a definitive retrospective analysis evaluated therelative risk in a large series of diabetic pregnancies fromthe Joslin Clinic and the Boston Hospital for Women anumber of years ago. (73) The relative risk of RDS in theIDM was higher compared with infants of nondiabeticmothers. If specific confounding variables were ex-cluded, including gestational age, delivery by cesareansection, presence of labor, birthweight, gender, Apgarscore at 5 minutes, hemorrhage, presence of hydramnios,maternal anemia, and maternal age, the relative risk was5.6 times higher in the IDM. This effect was primarilyconfined to the neonate whose gestational age was38 weeks or less. Present obstetric management hasmarkedly reduced the frequency of RDS.

Warburton et al (74) provided a biochemical correla-tion of the association between RDS and diabetes in afetal lamb model. Infusion of 14 mg/kg�1 per minute�1

glucose before 113 days’ gestation resulted in 1.5-foldhigher pulmonary desaturated phosphatidyl cholinecompared with controls. After this point in gestation, itincreased in the controls but not in the glucose-infusedfetuses. Lung stability to air inflation was 2.0-fold greaterin the control. The investigators concluded that desatu-rated phosphatidyl choline and lung stability were com-parable following chronic glucose infusion in fetal lamblung, with a predisposition of the fetus to develop RDS.

Surfactant production increases near term from acti-vation of the pathway for dipalmitoyl lecithin, which maybe mediated through increases in fetal plasma cortisolconcentration. Although plasma cortisol productionrates are normal in the IDM, insulin can interfere withincorporation of choline into lecithin, even when cortisolis present. (75) Neufeld and associates (76) has reportedthat incorporation of labeled glucose and fatty acid resi-dues into saturated phosphatidyl choline was reduced infetal rabbit lung slices in the presence of insulin. Endog-




enous insulin, known to be increased in the fetus of thepoorly controlled pregnant diabetic woman, may play asignificant role in delaying pulmonary maturation. Thespecific biochemical mechanisms are not completely un-derstood, but these data correlate with the clinical situa-tion in which pulmonary maturation is not only delayedin the IDM, but RDS may be noted with lecithin/sphingomyelin ratios of at least 2.0.

A extension of these concepts was reported by Curetet al, (77) who analyzed samples of amniotic fluid forphospholipid content and correlated them with the inci-dence of RDS. The incidence of RDS was 4.5% in IDMsand 5.3% in infants of nondiabetic mothers. If phosphati-dyl glycerol was present, no neonate developed RDS, butRDS occurred in 16.7% and 14.4% of infants of diabeticand nondiabetic women, respectively, if phosphatidylglycerol was absent. After 37 weeks’ gestation, RDS didnot occur in IDMs and only in 0.6% in infants of nondi-abetic mothers. The investigators suggested that datingof the pregnancy decreases the need for phospholipidanalysis.

HypoglycemiaA decline in plasma glucose concentration followingdelivery is characteristic of the IDM, especially amongneonates who are either large- or small-for-gestationalage or whose mothers had poor glycemic control duringtheir pregnancies. Other factors, besides hyperinsulin-emia, that may contribute to the development of hypo-glycemia include defective counterregulation by cat-echolamines or glucagon.

The neonate exhibits transitional control of glucosemetabolism, which suggests that a multiplicity of factorsaffect homeostasis. Many of the factors are similar tothose that influence homeostasis in the adult. What dif-fers in the neonate are the various stages of maturation.Prior work in conjunction with glucose infusion studiessuggest that there is blunted splanchnic (hepatic) respon-siveness to insulin both in the IDM and in the pretermand term neonate of the nondiabetic mother comparedwith the adult. (38) What have not been studied, but areof particular interest, are the many contrainsulin hor-mones that influence metabolism. If insulin is the pri-mary glucoregulatory hormone, contrainsulin hormonesassist in balancing the effect of insulin and other factors.

The contrainsulin hormones that have been of partic-ular interest in the IDM are those of the sympathoadre-nal neural axis. Results of evaluations of epinephrine andnorepinephrine concentrations in the IDM are variable.A very early study involved 11 infants of diabetic moth-ers, only two of whom had gestational diabetes. (78)

Urinary excretion of catecholamines was measured andcompared with excretion from 10 infants of nondiabeticmothers. Urinary norepinephrine and epinephrine con-centrations did not increase in the infants of diabeticwomen who had severe hypoglycemia, but they didincrease in the neonates whose mothers had mild hypo-glycemia. Other investigators have corroborated this re-port. (79)(80)(81)

Other factors related to sympathoadrenal activity inthe neonate may be of importance. In a continuingevaluation of the transitional nature of neonatal glucosemetabolism, both of insulin and contrainsulin factors,epinephrine was infused in two doses (50 mg or 500 mg/kg�1 per minute�1 in a newborn lamb model), andglucose kinetics (turnover) were measured with[6-3H]glucose. (82)(83) The newborn lamb showed ablunted response to the lower dose of epinephrine. Theinvestigators speculated that the newborn lamb evi-denced blunted responsiveness to this important contra-insulin stimulus. This tendency was reaffirmed by recentdata from the same laboratory. (84) If this occurs in thediabetic state, this would partially account for the clinicalpresence of hypoglycemia.

Thus, the IDM is a prime example of the potential ofglucose disequilibrium in the neonate. Because of thetransitional nature of glucose homeostasis in the neona-tal period, accentuation of the disequilibrium may beenhanced in the IDM due to metabolic alterations in thediabetic mother. Substantial work is necessary to appre-ciate fully the operative physiology.

The IDM is generally asymptomatic, with a relativelylow plasma glucose concentration. This may be due tothe initial brain stores of glycogen, although the exactbiochemistry is undefined. Signs and symptoms that maybe observed in the asymptomatic neonate are nonspecificand include tachypnea, apnea, tremulousness, sweating,irritability, and seizures. The asymptomatic neonate gen-erally does not require parenteral treatment to maintaincarbohydrate homeostasis.

Glucagon has been administered within 15 minutesafter delivery to prevent hypoglycemia, but because themajority of neonates are asymptomatic, this is probablynot warranted in most patients. Furthermore, glucagonmay stimulate insulin release, which may exaggerate thetendency to hypoglycemia.

Prompt recognition and treatment of symptomaticneonates has minimized sequelae. To date, there is noconsensus about the potential for long-term sequelaedue to hypoglycemia in the neonate. (85)




Hypocalcemia and HypomagnesemiaHypocalcemia ranks as one of the important metabolicderangements observed in the IDM. (86) Serum calciumconcentration is elevated following an increase in para-thyroid hormone (PTH) concentration by three mecha-nisms: mobilization of bone calcium, reabsorption ofcalcium in the kidney, and increased absorption of cal-cium in the intestine through the action of vitamin D. Incontrast, serum calcium concentration is decreased fol-lowing an increase in calcitonin, which antagonizes theaction of PTH. Serum calcium concentration may beincreased by vitamin D (1,25 dihydroxy vitamin D),which improves both absorption of calcium in the intes-tine after feeding and reabsorption from bone.

During pregnancy, calcium is transferred frommother to fetus, concomitant with an increasing hyper-parathyroid state in the mother. Calcium concentrationis higher in the fetus than in the mother. This hyperpara-thyroid state functions as homeostatic compensation torestore the maternal calcium that is diverted to the fetus.Neither calcitonin nor parathyroid hormone crosses theplacenta.

At birth, because of the concentration of calcitoninand 1,25 dihydroxy vitamin D, serum calcium concen-tration declines following interruption of maternal-fetalcalcium transfer. Increases in PTH and 1,25 dihydroxyvitamin D as early as 24 hours after birth ensure correc-tion of the low serum calcium concentration.

Tsang and associates (87)(88)(89) have shown thatthe neonate is prone to hypocalcemia, particularly thepreterm neonate, the one who is asphyxiated, and theIDM. Noguchi et al (90) evaluated parathyroid functionin the hypocalcemic versus the normocalcemic IDM. Inthe hypocalcemic IDM, serum PTH concentration didnot increase in response to low serum calcium concen-tration, but PTH concentration did increase in the nor-mocalcemic IDM in response to a slight decline in serumcalcium concentration. These data in the IDM are inter-preted to suggest that maternal diabetes may be anindependent factor related to suppressed neonatal para-thyroid function.

Martinez et al (91) measured osteocalcin and PTHserum concentrations in 41 insulin-dependent diabeticpregnant women throughout pregnancy, in the umbilicalcord, and in infants during the first days after birth. Thedata suggested that diabetes decreases bone turnoverduring pregnancy in the mother and during the perinatalperiod in the offspring.

More recently, Namgung and Tsang (92) reportedthat several factors have a significant impact on newbornbone mineral content. The IDM whose mother had poor

glycemic control in the first trimester evidenced de-creased bone mineral content at birth. They suggestedthat the poor bone mineral content was related to de-creased transplacental mineral transfer.

Hypomagnesemia (�1.5 mg/dL [0.6 mmol/L]) hasbeen found in as many as 33% of IDMs. As with hypocal-cemia, the frequency and severity of clinical symptomsare correlated with the maternal status. Noguchi et al(90) have correlated neonatal magnesium concentrationwith that of the mother as well as with the maternalinsulin requirement and concentration of intravenousglucose administered to the neonate. They speculatedthat hypocalcemia in the IDM may be due to decreasedhypoparathyroid function resulting from the hypomag-nesemia. In a subsequent evaluation, these investigatorscorrelated decreased maternal serum magnesium con-centration with adverse fetal outcome in the insulin-dependent diabetic woman. They speculated that de-creased magnesium concentration may contribute to thehigh spontaneous abortion and malformation rate seenin such women. (93)

Hyperbilirubinemia and PolycythemiaHyperbilirubinemia is observed more frequently in theIDM than in the normal neonate. Although a number ofhypotheses have been suggested, the pathogenesis re-mains uncertain. Red blood cell lifespan, osmotic fragil-ity, and deformability have not been found to be appre-ciably different in the IDM, and an increased umbilicalcord bilirubin concentration or an increased postnatalrate of hemoglobin decline has not been demonstrated.Results of one evaluation suggested that only the macro-somic IDM was at risk for hyperbilirubinemia and thatincreased hemoglobin turnover was a significant factor inits pathogenesis. (94) However, Stevenson et al (95)(96)initially suggested that delayed clearance of the bilirubinload was a factor, as measured by pulmonary excretion ofcarbon monoxide, which served as an index of bilirubinproduction.

The polycythemia frequently observed in the IDMmay be the most important factor associated with hyper-bilirubinemia. Indirect evidence for fetal hypoxia in theIDM may explain the neonatal preponderence for poly-cythemia and hyperbilirubinemia. Umbilical cord eryth-ropoietin concentration, measured at birth using a highlysensitive and specific radioimmunoassay, which is stimu-lated by hypoxia, was found to be above the narrow rangefor the control neonate in one third of a group of 61IDMs. (97) There was an association with relative hyper-insulinemia at birth. The fetal monkey that is hyperinsu-linemic in the last third of gestation in the absence of




maternal diabetes has been shown to have markedlyelevated plasma erythropoietin concentrations as well asother evidence of increased fetal erythropoiesis, such asan elevated reticulocyte count. (98) In addition, chron-ically catheterized fetal sheep that have been made hy-perglycemic have been found to have increased oxygenconsumption and decreased distal aortic arterial oxygencontent. (99) Similar speculation was suggested by Mi-mouni et al (100) following their finding of a 29.4%incidence of polycythemia in the IDM versus 5.9% in thecontrol neonate.

Another consideration is potential ineffective erythro-poiesis in the IDM, which is suggested by findings froma study of gestational age-matched controls and IDMs inwhom increased carbon monoxide excretion, derivedfrom heme metabolism, was observed. (95)(96) Hemo-globin concentration was not significantly higher in theIDM. Hemolysis was not present, as confirmed by theabsence of Coombs-positive blood group incompatibil-ity. Increased ineffective erythropoiesis, defined as ery-throid precursors harbored in body organs such as theliver and spleen and not released into the peripheralcirculation, was postulated as a cause for the observedincreased bilirubin concentration in the IDM. In relateddata, Perrine and associates (101) reported delay in thefetal globin switch in the IDM. The mechanism of thisdelay is unknown.

In a study of 32 mother-infant pairs, Green andcolleagues (102) showed a correlation between maternaltotal glycosylated hemoglobin at delivery and neonatalhematocrit. They concluded that improved maternal gly-cemic control during late gestation may decrease theincidence of neonatal polycythemia.

In a study of the mechanisms involving neonatalpolycythemia in the IDM, a complete blood count,serum iron, transferrin, and ferritin concentrations wereevaluated in samples from the umbilical cords of neo-nates born to 9 mothers who had gestational diabetes, 21mothers who had type 2 diabetes, and 8 mothers whohad type 1 diabetes. (103) There were no differences inserum iron concentration, but transferrin concentrationwas higher and ferritin concentration was significantlylower in the IDMs compared with the control popula-tion. The investigators concluded that iron storage isreduced in the fetus of the diabetic mother.

Renal Vein ThrombosisRenal vein thrombosis is a severe, life-threatening, butrare occurrence in the perinatal period. (104) Its occursmore frequently in association with maternal diabetesmellitus. Although Pedersen failed to mention this con-

dition in his monograph, (22) in one postmortem surveyof 16 cases of neonatal renal vein thrombosis, five werefound in IDMs. (105) Seven other affected neonateswere born to mothers without known diabetes, but theinfants had fetal macrosomia and pancreatic beta-cellhypertrophy and hyperplasia. Another center reported acase of an IDM who had nearly a totally occlusive throm-bosis in the umbilical vein. (106)

The pathogenesis of this lesion remains obscure, al-though most of the speculation has centered around thepossible role of polycythemia. Sludging of red bloodcells, combined with a further reduction in cardiac out-put as a result of diabetic cardiomyopathy, may be acontributing factor. Stuart and colleagues (107) havesuggested that because platelet endoperoxides are in-creased in the IDM, the normal balance between proag-gregatory platelets and antiaggregatory vascular prosta-glandins is disrupted, favoring the development ofthrombosis. In a subsequent evaluation, this group eval-uated abnormalities in vascular arachidonic acid metab-olism in the IDM, noting that decreased prostacyclinformation has been suggested as a cause of an athero-thrombotic tendency in the adult. (108) The level of6-ketoprostaglandin F1 was normal in umbilical cordsamples from IDMs if the mothers were in good glycemiccontrol. Inhibition of 6-keto F1 alpha was noted if themother’s glycosylated hemoglobin was elevated, indicat-ing poor diabetic control. The investigators suggestedthat the correlation observed between plasma 6-keto F1

alpha prostaglandin formation and endogenous vascularprostaglandin formation in the IDM indicated an in vitrodeficiency of prostacyclin formation that reflected a con-comitant in vivo abnormality. Why this lesion showsselectivity for the kidney is not known. Birth trauma is anunlikely initiating factor because this lesion has beenobserved in both the stillborn infant and the IDM deliv-ered by cesarean section. Another case has been reportedin a stillborn IDM whose mother received oxytocininduction. (109)

Long-Term Prognosis and Follow-upOf equal concern to problems primarily encounteredduring the neonatal period and perhaps of greater ulti-mate importance to the IDM are the long-term effects ongrowth and development, psychosocial intellectual capa-bilities, and the risk of subsequently developing diabetes.One of the most important factors influencing long-termprognosis is the improvement in management of thepregnant diabetic patient and her neonate. Assumingthat many of the deleterious effects of the diabetic preg-nancy are being modified by normalization of metabolic




status in both the pregnant woman and her conceptus,the poor prognoses that have been reported in previousretrospective studies should be ameliorated in futureprospective evaluations.

Gerlini et al (110) evaluated infant body weight,length, and head circumference from birth through 48months of age. In the IDM, no specific differences wererelated to the White classification. Children of mothersexhibiting poor glycemic control during pregnancyshowed higher values for weight and the weight-to-height ratio in infancy compared with neonates of well-controlled diabetic mothers. Female offspring contrib-uted most to the observed differences.

Silverman et al (111) tested the hypothesis that long-term postnatal development may be modified by the inutero metabolic experience. They enrolled the offspringof women who had type 1 diabetes mellitus, type 2diabetes mellitus, and gestational diabetes in a prospec-tive study from 1977 through 1983. Fetal beta-cell func-tion was assessed by measuring amniotic fluid insulin at32 to 38 weeks’ gestation. Postnatally, plasma glucoseand insulin concentrations were measured yearly from1.5 years of age after fasting and 2 hours after a1.75 g/kg oral glucose tolerance test. Control subjectshad a single oral glucose challenge at 10 to 16 years ofage. In the offspring of diabetic mothers, the prevalenceof impaired glucose tolerance was 1.2% at less than 5years of age, 5.4% at 5 to 9 years, and 19.3% at 10 to 16years. The 88 offspring of diabetic mothers (12.3�1.7y), when compared with 80 control subjects of the sameage and pubertal stage, had higher 2-hour glucose andinsulin concentrations. Impaired glucose tolerance wasnot associated with the cause of the mother’s diabetes ormacrosomia at birth. Impaired glucose tolerance wasrecorded in only 3.7% of adolescents whose amnioticfluid insulin was normal and 33.3% of those who hadelevated concentrations. The investigators concludedthat impaired glucose tolerance in the offspring is along-term complication of maternal diabetes. Excessiveinsulin secretion in utero, as assessed by amniotic fluidinsulin concentration, is a strong predictor of impairedglucose tolerance in childhood. These investigators sub-sequently restated the significance of these data. (112)

The outcome of children at 1, 3, and 5 years of agewas evaluated by Stehbens and associates. (113) Psycho-logical evaluations suggested that at 3 and 5 years of age,the IDM was more vulnerable to intellectual impairment,especially if the neonate was born small-for-gestationalage or if the pregnancy was complicated by acetonuria.This concept was reinforced by the work of Petersen et al,(114) who studied early growth delay in diabetic preg-

nancies related to psychomotor development at age 4years. Their studies of 99 consecutive insulin-dependentand 101 nondiabetic pregnant women led them to con-clude that children who have a history of growth delay inearly diabetic pregnancy should be screened at 4 to 5years of age by the Denver Developmental ScreeningTest for possible developmental impairment.

The presence of hypoglycemia per se has not beenrelated to later neuropsychological defects. Persson andcolleagues (115) found no evidence that asymptomatichypoglycemia leads to intellectual impairment by 5 yearsof age. No obvious relationship was found betweenmaternal acetonuria during pregnancy, infant birth-weight, blood glucose levels during the first hours afterbirth, or neonatal complications and the intellectual quo-tient (IQ) of the children. There was a correlation be-tween maternal and child IQ. Hadden et al (116) studied123 children of mothers who had type 1 diabetes and124 children of nondiabetic mothers. No differenceswere found following pediatric assessment or by a psy-chologically based maternal and teacher questionnaire ofthe emotional states or academic achievements of thechildren.

Questioning the extent to which maternal metabo-lism during pregnancy affects cognitive and behavioralfunction of the offspring by altering brain development,Rizzo et al (117) correlated measures of metabolism inpregnant diabetic and nondiabetic woman with intellec-tual development of their offspring. Of 223 pregnantwomen, 89 had pregestational diabetes mellitus, 99 hadgestational diabetes, and 35 had normal carbohydratemetabolism. Carbohydrate and lipid metabolism wereevaluated with respect to two measures of infant devel-opment: the Bayley Scale administered at age 2 years andthe Stanford-Binet administered at ages 3, 4, and 5 years.Bayley scale results at 2 years of age correlated inverselywith the mother’s third trimester plasma beta hydroxylbutyrate concentration, and the Stanford-Binet resultscorrelated inversely with third trimester plasma beta hy-droxybutyrale and free fatty acid concentrations. Theinvestigators concluded that ketoacidosis and acceleratedstarvation should be avoided in pregnancy because ofpotential long-term adverse consequences.

More recently, Ornoy et al (118) studied the neu-robehavorial effects that maternal pregestational and ges-tational diabetes might have on offspring studied atschool age compared with children of control mothers.Pregestational and gestational diabetes were reported toaffect attention span and motor functions adversely, butnot cognitive ability. The effects were negatively corre-lated with the degree of maternal glycemic control and




were more pronounced when observed in the youngerchild.

After documenting the decline in perinatal mortalityfrom 23% prior to 1961 to 14% from 1961 to 1975 andsubsequently to about 4%, Warram and associates (119)evaluated the number of neonates who subsequentlydeveloped diabetes in a cohort of 1,391. Type 1 diabeteshad developed in 21 of the children, a risk of 2.1�0.5%,by 20 years of age. The risk of diabetes in offspring of thediabetic mother was increased in the young mother andwas independent of the risk factors for perinatal mortal-ity. This is one third of the risk previously reported foroffspring of fathers who have type 1 diabetes. The inves-tigators speculated that exposure in utero to an affectedmother may protect the fetus from developing type 1diabetes later in life.

It is apparent that both immediate and potentiallylong-term effects are seen in the offspring of women whohave type 1 or gestational diabetes. Further researchclearly is required to refine the operative pathophysiol-ogy, thereby allowing the development and study ofenhanced treatment modalities. Clearly, the goal is toafford the pregnant women who has diabetes eitherbefore or during her pregnancy the opportunity to de-liver as normal and unaffected a neonate as possible.

References1. Wolfe RR, Allsop J, Burke JF. Glucose metabolism in man:responses to intravenous glucose. Metab Clin Exp. 1979;28:210–2202. . Cowett RM, Farrag HM. Neonatal glucose metabolism. In:Cowett RM, ed. Principles of Perinatal Neonatal Metabolism. NewYork, NY: Springer Verlag; 1998:686–7223. Carrapato MR, Marcelino F. The infant of the diabetic mother:the critical developmental windows. Early Pregnancy. 2001;5:57–584. Pedersen J, Molsted-Pedersen L, Andersen B. Assessors of fetalperinatal mortality in diabetic pregnancy. Analyses of 1332 preg-nancies in the Copenhagen series 1946–1972. Diabetes. 1974;23:302–3055. Hare JW, White P. Gestational diabetes and the White classifi-cation. Diabetes Care. 1980;3:3946. Greene MF, Hare JW, Krache M, et al. Prematurity amonginsulin requiring diabetic gravid women. Am J Obstet Gynecol.1989;161:106–1117. Persson B, Hanson U. Neonatal morbidities in gestational dia-betes mellitus. Diabetes Care. 1998;21(suppl 2 B):9–848. Diamond MD, Salyer SL, Vaughn WK, et al. Reassessment ofWhite’s clarification and Pedersen’s prognostically bad signs ofdiabetic pregnancies in insulin dependent diabetic pregnancies.Am J Obstet Gynecol. 1987;156:599–6049. Pollak A, Brehm R, Havelec L, et al. Total glycosylated hemo-globin in mothers of large for gestational age infants: a postpartum

test for undetected maternal diabetes? Biol Neonate. 1983;40:129–13510. Widness JA, Schwartz HC, Zeller WP, et al. Glycohemoglobinin postpartum women. Obstet Gynecol. 1983;57:414–42111. Artal R, Golde SH, Dorey F, et al. The effect of plasma glucosevariability on neonatal outcome in the pregnant diabetic patient.Am J Obstet Gynecol. 1983;147:537–54112. Mello G, Parretti E, Mecacci F, et al. What degree of maternalmetabolic control in women with type 1 diabetes is associated withnormal body size and proportions in full term infants? DiabetesCare. 2000;23:1494–149813. Roberts AB, Pattison NS. Pregnancy in women with diabetesmellitus, twenty years experience: 1968–1987. N Z Med J. 1990;103:211–21314. Coustan DR, Imarah J. Prophylactic insulin treatment of ges-tational diabetes reduces the incidence of macrosomia, operativedelivery, and birth trauma. Am J Obstet Gynecol. 1984;150:836–84215. Coustan DR, Reece EA, Sherwin RS, et al. A randomizedclinical trial of the insulin pump vs intensive conventional therapy indiabetic pregnancies. JAMA. 1986;108:329–33016. Howorka K, Pumpria J, Gabriel M, et al. Normalization ofpregnancy outcome in pregestational diabetes through functionalinsulin treatment and modular outpatient education adapted forpregnancy. Diab Med. 2001;18:965–97217. Thompson DJ, Porter KB, Gunnells DJ, et al. Prophylacticinsulin in the management of gestational diabetes. Obstet Gynecol.1990;75:960–96418. Norlander E, Hanson U, Persson B. Factors influencing neo-natal morbidity in gestational diabetic pregnancy. Br J Obstet Gy-necol. 1989;96:671–67819. Fraser D, Weitzman S, Leiberman JR, et al. Factors influencingbirth weight in newborns of diabetic and non-diabetic women.A population based study. Eur J Epidemiol. 1990;6:427–43120. Hod M, Rabinerson D, Kaplan B, et al. Perinatal complicationsfollowing gestational diabetes: how sweet is ill ? Acta Obstet GynecolScand. 1996;75:809–81521. deVeciana M, Major CA, Morgan MA, et al. Postprandialversus preprandial blood glucose monitoring in women with gesta-tional diabetes mellitus requiring insulin therapy. N Engl J Med.1995;333:1237–124122. Pedersen J. The Pregnant Diabetic and Her Newborn. 2nd ed.Baltimore, Md: Williams & Wilkins; 197723. Naeye RL. Infants of diabetic mothers: a quantitative morpho-logic study. Pediatrics. 1964;35:980–98824. Susa JB, McCormick KL, Widness JA, et al. Chronic hyperin-sulinemia in the fetal rhesus monkey. Effects on fetal growth andcomposition. Diabetes. 1979;28:1058–106325. Block MD, Pildes RS, Mossabhou NA, et al. C-peptide immu-noreactivity (CRP): a new method for studying infants of insulin-treated diabetic mothers. Pediatrics. 1974;53:923–92826. King KC, Adam PAJ, Yamaguchi K, et al. Insulin response toarginine in normal newborn infants and infants of diabetic mothers.Diabetes. 1974;23:816–82027. Susa JB, Boylan JM, Sehgal P, et al. Impaired insulin secretionin the neonatal rhesus monkey after chronic hyperinsulinemia inutero. Proc Soc Exp Biol Med. 1990;194:209–21528. MacFarlane CM, Tsakalakos N. The extended Pedersen hy-pothesis. Clin Physiol Biochem. 1988;6:68–7329. Schwartz R, Gruppuso PA, Pelzold K, et al. Hyperinsulinemia




and macrosomia in the fetus of the diabetic mother. Diabetes Care.1994;17:640–64830. Vohr BR, McGarvey ST, Coll CG. Effects of maternal gesta-tional diabetes and adiposity on neonatal adiposity and bloodpressure. Diabetes Care. 1995;18:467–47531. Kaminsky S, Silbley CP, Maresh M. The effects of diabetes onplacental lipase activity in the rat and human. Pediatr Res. 1991;30:541–54332. Homko CJ, Sivan E, Nyirjesy P, et al. The interrelationshipbetween ethnicity and gestational diabetes in fetal macrosomia.Diabetes Care. 1995;18:1442–144533. Jovanovic-Peterson CM, Peterson GF, Reed R, et al. Maternalpostprandial glucose levels and infant birth weight: The Diabetes inEarly Pregnancy Study. Am J Obstet Gynecol. 1991;164:103–11134. Kalhan SC, Savin SM, Adam PAJ. Attenuated glucose produc-tion rate in newborn infants of insulin-dependent diabetic mothers.N Engl J Med. 1977;296:375–37635. King KC, Tserng KY, Kalhan SC. Regulation of glucose pro-duction in newborn infants of diabetic mothers. Pediatr Res. 1982;16:608–61236. Cowett RM, Susa JB, Giletti B, et al. Variability of endogenousglucose production in infants of insulin dependent diabetic moth-ers. Pediatr Res. 1980;14:570A37. Cowett RM, Susa JB, Giletti B, et al. Glucose kinetics in infantsof diabetic mothers. Am J Obstet Gynecol. 1983;146:781–78638. Cowett RM, Oh W, Schwartz R, et al. Persistent glucoseproduction during glucose infusion in the neonate. J Clin Invest.1983;71:467–47339. Cowett RM, Andersen GE, Maguire CA, et al. Ontogeny ofglucose kinetics in low birth weight infants. J Pediatr. 1988;112:462–46540. Baarsma R, Reijngoud DJ, van Asselt WA, et al. Postnatalglucose kinetics in newborns of lightly controlled insulin dependentdiabetic mothers. Pediatr Res. 1993;34:443–44741. Widness JA, Cowett RM, Coustan DR, et al. Neonatal mor-bidities in infants of mothers with glucose intolerance in pregnancy.Diabetes. 1985;34(suppl 2):61–6542. Freinkel N. Of pregnancy and progency. Banting Lecture.Diabetes. 1980;29:1023–103543. Kalkhoff RK, Kandaraki E, Morrow PG, et al. Relationshipbetween neonatal birth weight and maternal plasma amino acidprofiles in lean and obese nondiabetic women and in type I diabeticpregnant women. Metabolism. 1988;37:234–23944. Whittaker PG, Lee CH, Taylor R. Whole body protein kineticsin women: effect of pregnancy and IDDM during anabolic stimu-lation. Am J Physiol Endocrinol Metab. 2000;279:E978–E98845. Knopp RH, Magee MS, Walden CE, et al. Prediction of infantbirth weight by GDM screening tests: importance of plasma triglyc-eride. Diabetes Care. 1992;15:1605–161346. Kitajima M, Oka S, Yasuchi I. et al. Maternal serum triglycerideat 24–32 weeks’ gestation and newborn weight in nondiabeticwomen with positive diabetic screens. Obstet Gynecol. 2001;97:776–78047. Becerra JE, Koury MJ, Cordero JF, et al. Diabetes mellitusduring pregnancy and the risks for specific birth defects: a popula-tion based case control study. Pediatrics. 1990;85:1–948. Jovanovic L, Druzin M, Peterson CM. Effects of euglycemiaon the outcome of pregnancy in insulin-dependent diabetic womenas compared with normal control subjects. Am J Med. 1980;68:105–11249. Waller DK, Keddie AM, Canfield MA. Do infants with major

congenital anomalies have an excess of macrosomia? Teratology.2001;64:311–31750. Watkins ML, Botto LD. Maternal prepregnancy weight andcongenital heart defects in offspring. Epidemiology. 2001;12:439–44651. Metzger BE, Buchanan TA, eds. Diabetes and birth defects:insights from the 1980’s, prevention in the 1990’s. Diabetes Spec-trum. 1990;3:149–18952. Neave C. Congenital malformation in offspring of diabetics.Perspect Pediatr Pathol. 1984;8:213–22253. Fuhrmann K, Reiher H, Semmler K, et al. Prevention ofcongenital malformations in infants of insulin dependent diabeticmothers. Diabetes Care. 1983;6:219–22354. Goldman JA, Dicker D, Feldberg D, et al. Pregnancy outcomein patients with insulin-dependent diabetes mellitus with precon-ceptional diabetic control: a comparative study. Am J Obstet Gy-necol. 1986;155:293–29755. Dicker D, Feldberg D, Yeshaya A, et al. Pregnancy outcome ingestational diabetes with preconceptional diabetes counseling. AustN Z J Obstet Gynaecol. 1987;27:184–18756. Ballard JL, Holroyde J, Tsang RC, et al. High malformationrates and decreased mortality in infants of diabetic mothers man-aged after the first trimester (1956–1978). Am J Obstet Gynecol.1984;148:111–11857. Landauer W. Rumplessness in chicken embryos produced bythe injection of insulin and other chemicals. J Expz Ool. 1945;98:65–7758. Adam PAJ, Teramo K, Raiha N, et al. Human fetal insulinmetabolism early in gestation. Diabetes. 1969;18:409–41659. Widness JA, Goldman AS, Susa JB, et al. Impermeability of therat placenta to insulin during organogenesis. Teratology. 1983;28:327–33260. Davis WS, Allen RP, Favara BE, et al. Neonatal small left colonsyndrome. Am J Roent Rad Therap Nucl Med. 1974;120:327–32961. Ferencz C, Rubin JD, McCarter RJ, et al. Maternal diabetesand cardiovascular malformations: predominance of double outletright ventricle and truncus arteriosus. Teratology. 1990;41:319–32662. Weber HS, Botti JJ, Baylen BG. Sequential longitudinal eval-uation of cardiac growth and ventricular diastolic filling in fetuses ofwell controlled diabetic mothers. Pediatr Cardiol. 1994;15:184–18963. Viana M, Herrera E, Bonet B. Teratogenic effects of diabetesmellitus in the rat: prevention by vitamin E. Diabetologia. 1996;39:1041–104664. Sivan E, Reece EA, Wu YK. Dietary vitamin E prophylaxis anddiabetic embryopathy. Morphologic and biochemical analyses.Am J Obstet Gynecol. 1996;175:793–79965. Fosberg H, Borg LA, Cagliero E, et al. Altered levels ofscavenging enzymes in embryos subjected to a diabetic environ-ment. Free Radical Res. 1996;24:451–45966. Marry A, Brenner SH, Yaron Y. Prospective study of incidenceand predisposing factors for clavicular fracture in the newborn. ActaObstet Gynecol Scand. 1996;75:378–38167. Mintz MC, Landon MB, Gabbe SG, et al. Shoulder soft tissuewidth as a predictor of macrosomia in diabetic pregnancies. Am JPerinatol. 1989;6:240–24368. Mimouni F, Miodounik M, Siddigi TA, et al. Perinatal as-phyxia in infants of insulin dependent diabetic mothers. J Pediatr.1988;113:345–35369. Hsi AC, Davis DJ, Sherman FC. Neonatal gangrene in the




newborn infants of diabetic mothers. J Pediatr Orthoped. 1985;5:358–36070. VanAllen MI, Jackson JC, Knopp RH, et al. In utero throm-bosis and neonatal gangrene in an infant of a diabetic mother. Am JMed Genet. 1989;33:323–32771. Ogata ES, Sabbagha R, Metzger BE, et al. Serial ultrasonog-raphy to assess evolving fetal macrosomia. Studies in 23 pregnantwomen. JAMA. 1980;243:2405–240872. Davis R, Woelk G, Mueller BA, et al. The role of previousbirthweight on risk for macrosomia on a subsequent birth. Epide-miology. 1995;6:607–61173. Robert MD, Nel RK, Hubbell JP, et al. Association betweenmaternal diabetes and the respiratory distress syndrome in thenewborn. N Engl J Med. 1976;294:357–36074. Warburton D, Parton L, Buckley S, et al. Effects of glucoseinfusion on surfactant and glycogen regulation in fetal lamb lung.J Appl Physiol. 1987;63:1750–175675. Smith BT, Giroud CJP, Robert M, et al. Insulin antagonism ofcortisol action on lecithin synthesis by cultured fetal lung cells.J Pediatr. 1975;87:953–95576. Neufeld ND, Sevanian A, Barrett CT, et al. Inhibition ofsurfactant production by insulin in fetal rabbit lung slices. PediatrRes. 1979;13:752–75477. Curet LB, Tsao FH, Zachman RD, et al. Phosphatidyl glyc-erol, lecithin/sphingomyelin ratio and respiratory distress syn-drome in diabetic and nondiabetic pregnancies. Int J Obstet Gynecol.1989;30:105–10878. Light IJ, Sutherland JM, Loggie JM, et al. Impaired epineph-rine release in hypoglycemic infants of diabetic mothers. N EnglJ Med. 1967;277:394–39879. Artel R, Platt LD, Kummula RK, et al. Sympatho-adrenalactivity in infants of diabetic mothers. Am J Obstet Gynecol. 1982;42:436–43980. Artal R, Doug N, Wu P, et al. Circulating catecholamesus andglucagon in infants of strictly controlled diabetic mothers. BiolNeonate. 1988;53:121–12581. Broberger U, Hansson U, Lagercrantz H, et al. Sympatho-adrenal activity and metabolic adjustment during the first 12 hoursafter birth in infants of diabetic mothers. Acta Pediatr Scand.1984;73:620–62582. Cowett RM. Decreased response to catecholamines in thenewborn: effect on glucose kinetics in the lamb. Metabolism. 1988;37:736–74083. Cowett RM. Alpha adrenergic agonists stimulate neonatalglucose production less than beta adrenergic agonists in the lamb.Metabolism. 1988;37:831–83684. Cowett RM, Rapoza RE, Gelardi NL. Insulin counter regula-tory hormones are ineffective in neonatal hyperinsulinemic hypo-glycemia. Metabolism. 1999;48:568–57485. Cornblath M, Schwartz R, Aynsley-Green A, et al. Hypogly-cemia in infancy: the need for a rational definition. Pediatrics.1990;85:834–83786. Tsang RC, Brown DR, Steichen JJ. Diabetes and calciumdisturbances in infants of diabetic mothers. In: Merkatz IR, AdamPAJ, eds. The Diabetic Pregnancy. A Perinatal Perspective. NewYork, NY: Grune and Stratton; 1979:207–22587. Tsang RC, Kleinman L, Sutherland JM. Hypocalcemia ininfants of diabetic mothers: studies in Ca, P and Mg metabolism andin parathyroid hormone responsiveness. J Pediatr. 1972;80:384–39588. Tsang RC, Light IJ, Sutherland JM, et al. Possible pathoge-

netic factors in neonatal hypocalcemia of prematurity. J Pediatr.1973;82:423–42989. Tsang RC, Chen I, Atkinson W, et al. Neonatal hypocalcemiain birth asphyxia. J Pediatr. 1974;84:428–43390. Noguchi A, Erin M, Tsang RC. Parathyroid hormone inhypocalcemic and normocalcemic infants of diabetic mothers. J Pe-diatr. 1980;97:112–11491. Martinez ME, Catalan P, Lisbona A, et al. Serum osteocalcinconcentrations in diabetic pregnant women and their newborns.Horm Metab Res. 1994;26:338–34292. Namgung R, Tsang RC. Factors affecting newborn bone min-eral content: in utero effects on newborn bone mineralization. ProcNutr Soc. 2000;59:55–6393. Mimouni F, Miodovnik M, Tsang RC, et al. Decreased mater-nal serum magnesium concentration and adverse fetal outcome ininsulin-dependent diabetic women. Obstet Gynecol. 1987;70:85–8894. Peevy KJ, Landaw SA, Gross SA. Hyperbilirubinemia in infantsof diabetic mothers. Pediatrics. 1980;66:417–41995. Stevenson DF, Ostrander CR, Cohen RS, et al. Pulmonaryexcretion of carbon monoxide in the human infant as an index ofbilirubin production. Eur J Pediatr. 1981;137:255–25996. Stevenson DR, Ostrander CR, Hopper AO, et al. Pulmonaryexcretion of carbon monoxide as an index of bilirubin production.IIa. Evidence for possible delayed clearance of bilirubin in infants ofdiabetic mothers. J Pediatr. 1981;98:822–82497. Widness JA, Susa J, Garcia JF, et al. Increased erythropoiesisand elevated erythropoietin in infants born to diabetic mothers andin hyprinsulinemic rhesus fetuses. J Clin Invest. 1981;67:637–64298. Carson BS, Philipps AF, Simmons MA, et al. Effects of asustained insulin infusion upon glucose uptake and oxygenation ofthe ovine fetus. Pediatr Res. 1980;14:147–15299. Philipps AF, Widness JA, Garcia JF, et al. Erythropoietinelevation in the chronically hyperglycemia fetal lamb. Proc Soc ExpBiol Med. 1982;170:42–47100. Mimouni F, Miodovnik M, Siddiqi TA, et al. Neonatal poly-cythemia in infants of insulin-dependent diabetic mothers. ObstetGynecol. 1986;68:370–372101. Perrine SP, Greene MF, Faller DV. Delay in the fetal globinswitch in infants of diabetic mothers. N Engl J Med. 1985;312:334–338102. Green DW, Khoury J, Mimouni F. Neonatal hematocrit andmaternal glycemic control in insulin dependent diabetes. J Pediatr.1992;120:302–305103. Murata K, Toyoda N, Ichio T, et al. Cord transferrin andferritin values for erythropoiesis in newborn infants of diabeticmothers. Endocrinol Jpn. 1989;36:827–832104. Avery ME, Oppenheimer EH, Gordon HH. Renal veinthrombosis in newborn infants of diabetic mothers. N EnglJ Med.1957;265:1134–1138105. Takeuchi A, Benirschke K. Renal vein thrombosis of thenewborn and its relation to maternal diabetes. Biol Neonate. 1961;3:237106. Fritz MA, Christopher CR. Umbilical vein thrombosis andmaternal diabetes mellitus. J Reprod Med. 1981;26:320–323107. Stuart MJ, Sunderji-Shirazali G, Allen JB. Decreased prosta-cyclin production in the infant of the diabetic mother. J Lab ClinMed. 1981;98:412–416108. Stuart M, Sunderje SC, Walenga RW, et al. Abnormalities invascular arachidonic acid metabolism in the infant of the diabeticmother. Br Med J. 1983;290:1700–1702109. Al-Samarrai SF, Kato A, Urano Y. Renal vein thrombosis in




stillborn infants of diabetic mothers. Acta Pathol Jpn. 1983;34:1411–1417110. Gerlini G, Arachi S, Gori MG, et al. Developmental aspects ofthe offspring of diabetic mothers. Acta Endocrinol (Supp). 1986;277:150–155111. Silverman BL, Metzger BE, Chon H, et al. Impaired glucosetolerance in adolescent offspring of diabetic mothers. Relationshipto fetal hyperinsulinemia. Diabetes Care. 1995;18:611–617112. Silverman BL, Rizzo TA, Cho NH. Long term effects of theintrauterine environment. The Northwestern University Diabetesin Pregnancy Denter. Diabetes Care. 1998;(suppl 2B):142–149113. Stehbens JA, Baker GL, Kitchell M. Outcome at ages 1, 3,and 5 years of children born to diabetic women. Am J ObstetGynecol. 1977;127:408–413114. Petersen MB, Pedersen SA, Greisen G, et al. Early growthdelay in diabetic pregnancy: relation to psychomotor developmentat age 4. Br Med J. 1988;296:598–600

115. Persson B, Gentz J. Follow up of children of insulin depen-dent and gestational diabetic mothers. Neuropsychological out-come. Acta Paediatr Scand. 1983;73:349–358116. Hadden DR, Bryne E, Trotter S, et al. Physical and psycho-logical health of children of type 1 (insulin dependent) diabeticmothers. Diabetologia.1983;26:250–254117. Rizzo T, Metzger RE, Burns WJ, et al. Correlations betweenantepartum maternal metabolism and child intelligence. N EnglJ Med. 1991;325:911–916118. Ornoy A, Ratson N, Greeenbaum C, et al. School-age chil-dren born to diabetic mothers and to mothers with gestationaldiabetes exhibit a high rate of inattention and fine and gross motorimpairment. J Pediatr Endocrinol Metab. 2001;14(suppl 1):681–689119. Warram JH, Krolewski AS, Kahn CR. Determinants of IDMand perinatal mortality in children of diabetic mothers. Diabetes.1988;37:1328–1334

NeoReviews Quiz

3. A 26-year-old primigravida is seen at a prenatal clinic. Her history is significant for type 1 diabetesmellitus initially detected at 21 years of age. She has no complications related to diabetes. Of thefollowing, the class of diabetes in pregnancy according to the White classification most consistent with thispatient is:

A. Class A.B. Class B.C. Class C.D. Class D.E. Class R.

4. Pregnancy in a patient who has diabetes mellitus often is complicated because of perturbations of glucosehomeostasis. Of the following, the most significant morbidity associated with diabetes during pregnancy is:

A. Metabolic acidosis.B. Obesity.C. Preeclampsia.D. Pyelonephritis.E. Sepsis.

5. Infants born to diabetic mothers are at a substantially greater risk for major malformations than those bornto nondiabetic mothers. Of the following, the mechanism most likely implicated in the pathogenesis ofmalformations in a diabetic pregnancy is:

A. Antioxidant deficiency.B. Fetal hyperinsulinism.C. Genetic predisposition.D. Postconceptional hyperglycemia.E. Uteroplacental vascular disease.




DOI: 10.1542/neo.3-9-e173 2002;3;173 NeoReviews

Richard M. Cowett The Infant of the Diabetic Mother

& ServicesUpdated Information

;3/9/e173http://neoreviews.aappublications.org/cgi/content/full/neoreviewsincluding high-resolution figures, can be found at:

Supplementary Material

;3/9/e173/DC1http://neoreviews.aappublications.org/cgi/content/full/neoreviewsSupplementary material can be found at:

Permissions & Licensing

http://neoreviews.aappublications.org/misc/Permissions.shtmltables) or in its entirety can be found online at: Information about reproducing this article in parts (figures,

Reprints http://neoreviews.aappublications.org/misc/reprints.shtml

Information about ordering reprints can be found online:

. Provided by McMaster University on April 3, 2010 http://neoreviews.aappublications.orgDownloaded from

http://neoreviews.aappublications.org/cgi/content/full/neoreviews;3/9/e173

http://neoreviews.aappublications.org/cgi/content/full/neoreviews;3/9/e173/DC1

http://neoreviews.aappublications.org/misc/Permissions.shtml

http://neoreviews.aappublications.org/misc/reprints.shtml


Documents

The Infant of the Diabetic Mother Richard M. Cowett ... · DOI:€10.1542/neo.3-9-e173 NeoReviews 2002;3;173 Richard M. Cowett The Infant of the Diabetic Mother