Postpartum Hemorrhage - Vouch · related deaths due to hemorrhage in one series8. PLACENTA PREVIA Placenta previa is defined as partial or complete insertion of the placenta onto

10PLACENTAL ABNORMALITIES

M. K. Mehasseb and J. C. Konje

INTRODUCTION

Obstetric hemorrhage is still considered to beone of the leading causes of maternal mortalityand morbidity1–7. Placental abnormalities area major contributor to obstetric hemorrhage.The common abnormalities include placentalabruption, placenta previa, morbidly adherentplacentae (accreta, increta, percreta) andretained placenta. These abnormalities, forexample, accounted for 36% of pregnancy-related deaths due to hemorrhage in one series8.

PLACENTA PREVIA

Placenta previa is defined as partial or completeinsertion of the placenta onto the lower uterinesegment after fetal viability (20 weeks indeveloped countries and 24–28 weeks in devel-oping countries). Four grades of placenta previaare recognized (Figure 1):

(1) Grade I: placenta is in the lower segmentbut its edge does not reach the internal os;

(2) Grade II: lower placental edges reach the osbut do not cover it;

(3) Grade III: edge covers the os and theplacenta is asymmetrical;

(4) Grade IV: placenta symmetrically coversthe os.

Although this classification/grading system isthe most common, others reflect the ultrasounddefinition of the placental site.

An alternative anatomical grading is providedbelow:

(1) Total placenta previa: where the internal cer-vical os is completely covered by placenta;

(2) Partial placenta previa: where the internal osis partially covered by placenta;

(3) Marginal previa: where the edge of theplacenta is at the margin of the internal osbut does not cover it;

(4) Low-lying placenta: where the placental edgedoes not reach but is in close proximity tothe internal os.

This other classification depends on the state ofthe cervix at the time of examination; for exam-ple, a low-lying placenta at 2 cm dilatation may

80

Figure 1 Grades of placenta previa: I, Encroaching on the lower segment; II, reaching the internal os;III, asymmetrically covering the internal os; IV, symmetrically covering the internal os

I II III IV

102Z:\Sapiens Publishing\A5211 - Postpartum Hemorrhage\Make-up\Postpartum Hemorrhage - Voucher Proofs #T.vp30 August 2006 14:20:02

Color profile: Generic CMYK printer profileComposite Default screen

become a partial placenta at 8 cm dilatation.Since a digital examination is not recommendedin cases of placenta previa, this alternativeclassification has limited clinical application.

Incidence and risk factors

The overall incidence is variable in differentseries, but is on average 1 in 300 deliveries9,10.Risk factors for placenta previa include:

(1) It is thought to be more common withadvanced maternal age. This may, however,be a reflection of increased parity ratherthan age. However, a rise in incidencefrom 0.3% to 0.7% over a 10-year periodhas been attributed to a shift to an olderobstetric population11.

(2) Women of higher parity have a higherincidence12.

(3) Multifetal gestation: secondary to an increasein the surface area occupied by the placentalmass13.

(4) The incidence increases with the numberof previous Cesarean section deliveries14,15. Asingle Cesarean section increases the risk by0.65%, two by 1.5%, three by 2.2% andfour or more by 10%. A previous Cesareansection in association with placenta previaincreases the risk of Cesarean hysterectomyalmost four-fold11.

(5) Smoking doubles the risk of placentaprevia13–16. This may be attributed toplacental hypertrophy secondary to carbonmonoxide hypoxemia16.

(6) Patients with placenta previa have 12 timesthe usual risk of having a recurrent previa insubsequent pregnancies.

(7) For unclear reasons, fetal anomalies areincreased with placenta previa even aftercontrol for maternal age9. It is alsouncertain if there is an association withintrauterine fetal growth restriction17,18.

Diagnosis

This can either be clinical or by imaging.

Clinical

The most characteristic feature is painlessvaginal bleeding. This is usually recurrent andunprovoked and does not commonly appear untilthe end of the second trimester. The firstepisode is usually self-limiting and is rarely soprofuse as to prove fatal. However, the earlier inpregnancy the first presentation of bleeding, themore likely is the later need for early inter-vention. ‘Fetal distress’ is unusual unless thehemorrhage is severe enough to cause maternalshock.

Abdominal palpation is not diagnostic but,where the presenting part is free in late preg-nancy or abnormal, placenta previa should besuspected.

Sometimes, especially with minor degrees ofplacenta previa, bleeding might not appear untilthe onset of labor. This may clinically mimicabruption (see below).

The possibility of placenta previa shouldalways be considered in women who presentwith bleeding in the latter half of pregnancy.The diagnosis can seldom be made solely on aclinical basis.

There is no role for digital examination in thediagnosis unless in the operating theater as partof the double set-up with adequate preparationfor proceeding to Cesarean section. Althoughuncommon, where imaging (see below) is easilyavailable and reliable, it remains useful incases where the diagnosis is in doubt and wheredouble set-up facilities are unavailable.

Imaging

The most commonly used method of placentallocalization in modern obstetrics is ultrasoundscan. It is safe, accurate and non-invasive and isthe method of choice for making the diagnosis.The gestational age at which diagnosis is madesignificantly influences accuracy. The earlier thescan is performed, the more likely the placentais to be found in the lower pole of the uterus.Consequently, routinely localization of the pla-centa at the 20–22 weeks’ gestation anomalyscan poses several questions. For example, is alow-lying placenta at this stage predictive of pla-centa previa at the time of delivery, or does suchscreening reduce the adverse outcome for the

81

Placental abnormalities



pregnancy? More importantly, should the scanbe repeated at 32–34 weeks’ gestation and doesthe asymptomatic patient have to be admittedand if so when?

About 28% of placentas in women scannedtransabdominally before 24 weeks are found tobe ‘low’ but by 24 weeks this drops to 18% andonly 3% are low-lying by term19. Conversely, afalse-negative scan for a low placenta is found inas many as 7% of cases at 20 weeks20. Suchresults are more common when the placenta isposterior, the bladder is over-filled, the fetalhead obscures the margin of the placenta, or theoperator fails to scan the lateral uterine wall21.A low-lying placenta is more common in earlypregnancy because the lower segment does notexist. This apparent ‘placental migration’ is dueto enlargement of the upper segment and for-mation of the lower segment, with many appar-ently low placentas being found to be above thelower segment. Comeau and colleagues22 andRuparelia and Chapman23 have shown that themore advanced the pregnancy is, the more accu-rate a scan diagnosis of placenta previa will be.

Transvaginal ultrasound is not only moreaccurate in diagnosing placenta previa but itis more precise in defining the relationship ofthe lower edge of the placenta to the internalos (Figure 2). Placenta previa is diagnosed ontransvaginal ultrasound scan when the placentaledge is less than 3 cm from the internal os.Where the distance between the lower edgeof the placenta and the internal cervical osis measured, the persistence of a low-lying

placenta at a later gestation is higher. Taipaleand colleagues24, for example, observed that, ifa placenta overlapped the internal os by at least25 mm at 18–23 weeks, the positive predictivevalue for previa at delivery was 40% with a sen-sitivity of 80%. Indeed, Becker and colleagues25

found that, when the lower edge overlapped theos by at least 25 mm at 20–23 weeks, a vaginaldelivery was not possible at all at term (i.e. ithad a 100% positive predictive value).

Although the routine practice of localizingthe placenta at the anomaly scan will no doubtcontinue, its limitations should be recognizedand, wherever possible, transvaginal ultrasoundscans should be offered to improve the accuracyof localization and also measure the distancefrom the os to the placental edge to help definethe degree of ‘low lying’.

Since there are no randomized, controlledtrials of the effect of routine localization versusno localization on the mother and fetus, currentpractice will have to be governed by large-cohort case studies. It is, however, easy toassume that, where there is a low-lying placenta,education of patients and carers enhances thechances of a better outcome for mother andbaby. Whether such patients should be rou-tinely admitted at a later gestation is debatable.Most units do not, however, routinely admit butrepeat the scans at 32–24 weeks’ gestation.Dashe and colleagues26 observed that persis-tence of placental previa diagnosed at 20–23weeks occurred in 34% of cases at delivery,whereas 73% of those present at 32–35 weekspersisted at delivery. A policy of routine scan-ning will therefore reduce the false-positive ratesbut will be at the expense of increasing workloadand patient anxiety. Unfortunately, none of thestudies reported on the proportion of patientswith low-lying placentas diagnosed at 32–34weeks that later presented with bleeding. Forunits not routinely scanning for placental site at20 weeks, scanning for placental site is onlyindicated with abnormal presentation, vaginalbleeding or a chance finding when ultrasoundwas undertaken in late pregnancy for other rea-sons. For such cases, a transvaginal approach isrecommended as it is associated with a betterdiagnostic accuracy, especially with posteriorplacenta previa27,28. This approach has beenshown to be safe and is well tolerated.

82

POSTPARTUM HEMORRHAGE

Figure 2 Transabdominal ultrasound scan withsuperimposed color Doppler signal showing ananterior placenta previa



Transperineal sonography has been used bysome investigators29. It allowed easy visualiza-tion of the internal os in all cases and carried apositive predictive value of 90% and a negativepredictive value of 100% for placenta previa.

Magnetic resonance imaging (MRI) hasbeen used to visualize placental abnormalitiesincluding placenta previa (Figure 3). It has theadvantages of being an objective, reproducibletest, minimizing the operator error. However,due to cost and logistic limitations, it is unlikelythat it will replace ultrasonography for routineevaluation30,31.

Management

Management depends on whether the patient issymptomatic or not. Asymptomatic patients(where the diagnosis is made on ultrasoundscan) are managed expectantly, often as forthose with mild symptoms that are non-threatening to either the mother or fetus.

Those with symptoms can be divided intofour categories depending on the maternal con-dition, severity of hemorrhage, the gestationalage and the neonatal facilities available in theunit. These categories are:

(1) Pregnancy < 37 weeks’ gestation withoutthreat to the mother;

(2) Pregnancy > 37 weeks without threat to themother;

(3) Severe life-threatening, non-stopping(continuing) hemorrhage < or > 37 weeks;

(4) Hemorrhage associated with uterinecontractions.

The management of the third and fourthcategories is immediate delivery by Cesareansection. In the presence of non-life-threateninghemorrhage after 37 weeks’ gestation, a planneddelivery is also advisable. This must, however,be with the recognition that such a hemorrhagecould very rapidly become life-threatening.For category 1, the best approach is expectantmanagement, although this must not be to thedetriment of maternal life.

Expectant management

The perinatal mortality in placenta previais directly related to gestational age atdelivery32–35. Macafee33 and Johnson and col-leagues35 introduced expectant management ofplacenta previa with the aim of achieving maxi-mum fetal maturity possible while minimizingthe risks to both mother and fetus, the overallobjective being to reduce perinatal mortality,and, at the same time, reducing maternal mor-tality. This management plan was based on theassumption that most episodes of bleeding areusually small, self-limited and are not fatal tothe fetus or mother in the absence of provokingtrauma (e.g. intercourse, vaginal examination)or labor, and that a proportion of cases, particu-larly those presenting early with lesser degreesof previa, may resolve to permit vaginal delivery.More recently, an improvement in perinatalmortality attributed mainly to prolongation ofpregnancy with expectant management has alsobeen reported36,37.

Although Macafee33, in his regimen, advo-cated that the patient remained as an inpatientin a fully equipped and fully staffed maternityhospital from the time of initial diagnosis todelivery, a policy of permitting a selection ofwomen to return home has also been advo-cated38 as part of expectant management, butremains controversial. Cotton and colleagues32

reported no difference in their perinatal andmaternal mortality rates in those sent home andthose managed in hospitals, whereas D’Angeloand Irwin39 suggested keeping the mother in

83


Figure 3 MRI of a grade IV placenta previa(completely covering the internal os)



hospital until delivery was justified, on thegrounds that neonatal mortality and morbidityand cost of treatment were reduced. Kaunitzand colleagues40, in a review of 355 maternitiesmanaged at home, however, reported oneintrapartum death from placenta previa.

This controversy is more evident in casesof asymptomatic transvaginally diagnosedplacenta previa. For this group, it is becomingincreasingly acceptable to manage them athome41,42. In a review of 15 930 deliveries inEdinburgh, Love and Wallace43 concluded that,while clinical outcomes were highly variableand cannot be predicted from antenatal events,the majority of cases with or without bleeding,irrespective of the degree of previa, could bemanaged on outpatient bases. There are norandomized, controlled trials on the differentapproaches to this aspect of placental previaand such evidence is urgently needed to enablerationale decision-making in clinical practice.

Although most experts will advocate immedi-ate delivery where there is severe hemorrhage(heavy vaginal bleeding producing maternalhypovolemia), it is, however, not considered acontraindication to expectant management43.An aggressive approach involving admissionand repeated blood transfusions improves peri-natal morbidity and mortality, especially wherethe bleeding occurs very early in pregnancy. Inone study, where approximately 20% of thewomen lost over 500 ml of blood, half of themwere managed expectantly with a mean gainin gestation of 16.8 days44. Crenshaw andcolleagues34, on the other hand, managed only43–46% of patients successfully with an aggres-sive expectant approach, whereas Cotton andcolleagues32, with an aggressive approach,successfully managed 66% of womenexpectantly.

During expectant management, pretermlabor remains a problem. Brenner and col-leagues45 found that 40% of women withplacenta previa had prelabor rupture ofmembranes, and went into spontaneous laboror other developed problems that resulted indelivery before 37 weeks’ gestation. Inhibitingcontractions in those with preterm labor wouldseem logical, but some regard antepartumhemorrhage as a contraindication to the use oftocolytics46. With vaginal bleeding and uterine

contractions, placental abruption, which iswidely regarded as a contraindication to toco-lysis, cannot be excluded. In addition, placentalabruption is said to coexist with placenta previain 10% of cases, and tocolytics cause maternaltachycardia and palpitations, features that couldbe confused with hypovolemia. Sampson andcolleagues47 advocate the use of tocolytics incases of placenta previa and uterine contrac-tions after 21 weeks and cite a reduction inperinatal mortality from 126 to 41 per 1000.

Mild blood loss in placenta previa is notassociated with a significantly high perinatalmortality. In contrast, significant blood loss isassociated with a high perinatal loss. Liberal useof blood transfusion has been reported to nullifythis effect32. Although there is no theoreticallimit to the number of blood transfusions apatient can have, most blood banks do not haveendless supplies. To optimize oxygen supply tothe fetus and protect the mother against antici-pated future blood loss, the ideal aim of trans-fusion should be to maintain a hemoglobin levelof at least 10 g/dl or a hematocrit of 30%.

Despite expectant management, 20% ofwomen with placenta previa are delivered earlierthan 32 weeks. These cases account for 73% ofperinatal deaths32. They remain a major prob-lem and, although the use of cervical cerclagehas been advocated, this is generally notused. The neonatal mortality and morbidity arereduced in this group by maternal corticosteroidadministration.

Continuous hospitalization is costly and hasan associated psychological effect of separationon families. In developing countries, this may beunaffordable to many families. However, theadvantages include easy access to resuscitationand prompt delivery and ensuring bed rest(which anecdotally has been thought todecrease the occurrence of hemorrhage) as wellas limitation of activities. With improvementin transportation facilities and ambulance ser-vices in developed countries, highly motivatedwomen who clearly understand the necessity ofrestriction of activity and are within, for exam-ple, 15–30 min of the hospital perhaps may bemonitored at home. This will only apply to casesof grades I–III placenta previa or asymptomaticgrade IV. In all cases of expectant manage-ment, cross-matched blood (two units) must be

84




available at all times. However, in many hospi-tals this requirement is the sine qua non of alimitation on therapeutic options.

Method of delivery

A diagnosis of placenta previa means deliveryby Cesarean section, but this is not inevitable,especially where the previa is to a minor degree.For the minor degree of placenta previa (grade Ior II anterior) and an engaged fetal head,pregnancy may be allowed to continue beyond37–38 weeks and vaginal delivery anticipated.In such patients, amniotomy followed bysyntocinon can be considered.

In patients with a major grade of placentaprevia (grade II posterior, grades III–IV), deliv-ery should be by elective or emergency Cesareansection. The former is ideal since emergencydelivery has a negative effect on perinatal mor-tality and morbidity, independent of gestationalage. Cotton and colleagues32 found that 27.7%of babies born as emergencies had anemiacompared to 2.9% delivered electively.

Cesarean section for placenta previa posesseveral problems. It should, therefore, never beleft to an inexperienced obstetrician. The RoyalCollege of Obstetricians and Gynaecologists inthe UK recommends that such Cesarean sec-tions are performed by consultants. Althoughgeneral anesthesia was preferred to regional inthe past, there is an increasing tendency to usingthe latter especially as Frederiksen and col-leagues48 demonstrated not only its safety but areduction in intrapartum blood loss comparedto that with general anesthesia.

Procedure

Epidural analgesia is increasingly beingadvocated for Cesarean section (in developedcountries) although Moir49 considers placentaprevia to be an absolute contraindication to anepidural. This is because epidurals, by loweringthe blood pressure, may critically reduce uterineand placental perfusion. Crawford50, however,believes that, in experienced hands, an epiduralis safe. Indeed, an increasing number ofanesthetists offer regional anesthesia to thesepatients30,31. Where the patient’s condition isstable and there is no active bleeding, epidural

or spinal anesthesia should not be regardedas contraindicated provided an experiencedanesthetist is available.

The uterine incision should be a transverselower segment incision (if possible), providedthere is a lower segment. Where the lower seg-ment is non-existent or is very vascular, someobstetricians advocate a classical or a De Lee’sincision. Scott51, however, believes that suchincisions are rarely justified because of theirconsequences and long-term disadvantages.When difficulties are encountered with trans-verse lower segment incisions, these may beconverted to inverted T-, J- or U-shapedincisions.

Where the placenta is anterior, twoapproaches are available for incising the uterus,going through the placenta or defining its edgeand going through the membranes above orbelow the placenta. The former approachrequires speed and may result in significantfetal blood loss52. The latter, however, may beassociated with undue delay in the delivery ofthe fetus, more troublesome bleeding from apartially separated placenta and therefore fetalblood loss and anoxia. Myerscough52 advisesagainst cutting or tearing through the placentabecause of the inevitable fetal blood loss thatoccurs as fetal vessels are torn. Because thelower segment is less muscular, contraction andretraction, which result in the occlusion of thesinuses of the placental bed, are inadequate,and intraoperative hemorrhage is therefore notuncommon53. Where hemostasis is difficultto achieve, bleeding sinuses could be oversewnwith atraumatic sutures51. If this is unsuccess-ful, packing the uterus is possible, but the majordisadvantage is that, by leaving the pack in situduring closure of the uterus, the bleeding maycontinue but remain concealed for some time asthe pack is soaking through. The use of balloonswith a tamponading effect on the bleedingplacenta bed or intramyometrial injection ofprostaglandin F2α has been shown to be usefulin such cases52. More recently, where the facili-ties are available, uterine artery embolizationhas been used with excellent results. The diffi-culty with this is planning to ensure that thefacilities and the interventional radiologist areavailable on the labor ward during the delivery.When the bleeding remains uncontrollable,

85




ligation of the internal iliac artery or evenhysterectomy may be necessary as the last resort(see Chapters 32 and 34).

PLACENTAL ABRUPTION

The Latin term abruptio placentae means ‘rendingasunder of the placenta’, implying and denotinga sudden accident, which is a valid clinicalcharacteristic of most cases. It represents bleed-ing due to premature separation of a normallysited placenta after fetal viability. The initialevent in abruption is bleeding into the deciduabasalis.

Incidence and risk factors

It occurs in about 1 in 200 pregnancies10,although higher incidences have beenreported54. When placentas are examined rou-tinely, the incidence is much higher at 4.5%55

suggesting that small episodes are more com-mon than those diagnosed clinically. Placentalabruption can be revealed or concealed (Figure4), the former occurring in 65–80% of cases.The concealed type is clinically more dangerousas it is often associated with more severecomplications.

Risk factors for placental abruption include:

(1) Parity: more common in women of higherparity;

(2) Age: more common in older women butthis may again be a reflection of parityrather than age;

(3) Previous placental abruption: this variesfrom 6 to 16.7% after one episode and25% after two episodes. Up to 7% of thosewith abruption severe enough to result infetal death have the same outcome in asubsequent pregnancy and 30% of allfuture pregnancies in women who havea placental abruption do not result in aliving child56–59.

(4) Premature rupture of fetal membranes: ameta-analysis of 54 studies demonstrateda three-fold increase in the risk ofabruption60 and this risk was much higherwith rupture between 20 and 36 weeks’gestation and if rupture was for longerthan 24 hours61,62;

(5) Cigarette smoking: the incidence in smokersis almost double that in non-smokers;smokers who quit have an associatedreduction in risk;

(6) Cocaine use: significantly increased riskcompared to non-users63;

86


Figure 4 Concealed and revealed placental abruption



(7) Abdominal trauma: placental abruptioncomplicates 1–6% of minor injuries andup to 50% of major injuries64;

(8) Sudden decompression of the uterus aftermembrane rupture, e.g. in twin pregnancies,external cephalic version or pregnancieswith polyhydramnios;

(9) Unexplained raised α-fetoprotein;

(10) Hyperhomocysteinemia and thrombo-philias, especially factor V Leiden65,66;

(11) Hypertensive disorders of pregnancy.

Diagnosis

Unlike placenta previa where ultrasound is themainstay of diagnosis, the diagnosis of placentalabruption is usually made on clinical grounds(Table 1). Ultrasonography may, however, behelpful in certain instances, for example, wherethere is a large retroplacental hematoma. This,however, is an uncommon finding even insevere cases. The symptoms and signs are diag-nostic in moderate to severe cases. In the mildforms, the diagnosis may not be obvious untilafter delivery when a retroplacental clot isidentified.

Placental abruption classically presents withvaginal bleeding, abdominal pain, uterinecontractions and tenderness. Vaginal bleeding,however, is a symptom in no more than 70–80%of cases28. The bleeding which occurs after the36th week of gestation in about 50% of cases28 ischaracteristically dark and non-clotting. Becauselabor is the commonest factor precipitating pla-cental separation70, nearly 50% of patients withplacental abruption are in established labor. Thepresence of uterine contractions may, however,

be difficult to distinguish from the abdominalpain of abruption which is often unremitting.Where this distinction is possible, the contrac-tions are characteristically very frequent with arate often of over five in 10 min71.

The absence of abdominal pain does notexclude placental abruption, especially wherethe placenta is posteriorly sited. This is evi-denced by the so-called ‘unsuspected or silentabruption’ referred to by Notelovitz and col-leagues72 and the higher pathological incidenceof placental abruption found by Fox55. Thepresence of pain is probably indicative ofextravasation of blood into the myometrium. Insevere cases (grade 3), the pain is sharp, severeand sudden in onset. Some patients may, inaddition, present with nausea, anxiety, thirst,restlessness and a feeling of faintness, whereasothers may complain of absent or reduced fetalmovements.

Some patients present with signs of shockwhere the blood loss is significant (tachycardiapredominates; blood pressure has a poor rela-tion with blood volume in this condition). Thepresence of hypertension may, however, masktrue hypovolemia but an increasing abdominalgirth or a rising fundal height must raise thesuspicion of significant concealed hemorrhage.Typically, the uterus is ‘woody hard’ in severecases where the fetus is difficult to palpate and acontinuous fetal heart rate monitor or real-timeultrasonography is essential to identify the fetalheart beat. The fetus may be ‘distressed’ withfetal heart rate abnormalities or it may be dead.The former occurs in grades 1–2, but, in grade3, the latter is an invariable occurrence bydefinition73. In severe cases complicated bydisseminated intravascular coagulation, theremay be absence of clotting in the vaginal bloodloss, which is dark-colored. The incidence ofcoagulopathy varies from 35 to 38%57,74 andthis occurs mainly in the severe forms.

A vaginal examination reveals blood clotsin the vagina, which is typically non-clotting.Serous fluid from a retroplacental clot maybe confused with liquor. The cervix may bedilating since 50% of cases are in labor. If themembranes are ruptured, blood-stained liquoris usually present.

Ultrasound scan is not a sensitive method ofdiagnosing placental abruption but is useful in

87


Symptom/sign Frequency (%)

Vaginal bleedingUterine tenderness or back painFetal distressPreterm laborHigh-frequency contractionsHypertonusDead fetus

78666022171715

Table 1 Clinical picture of placental abruption71

109Z:\Sapiens Publishing\A5211 - Postpartum Hemorrhage\Make-up\Postpartum Hemorrhage - Voucher Proofs #T.vp06 September 2006 16:49:22


excluding coincident placenta previa, whichis present in 10% of cases. Where the retro-placental clot is large, ultrasonography identifiesit as hyperechogenic or isoechogenic whencompared to the placenta. Such echogenicitymay therefore be misinterpreted as a thick pla-centa75. A resolving retroplacental clot appearshyperechogenic within 1 week and sonolucentwithin 2 weeks.

Though ultrasound scan is not an accuratediagnostic tool, it is useful in monitoring casesmanaged. The size of the hematoma, locationand change in size over time and fetal growthare all parameters monitored by ultrasoundscan. A Kliehauer–Betke test may be useful inmaking the diagnosis when a patient presentswith abdominal pain but without vaginal bleed-ing or even in cases of ‘unsuspected or silentabruption’.

Management

The severity of the abruption, the state of thefetus and the gestational age of the pregnancy allimpinge on management which can be dividedinto general and specific measures. Sher andStatland76 divided placental abruption intothree degrees of severity upon which manage-ment can be based. These are shown in Table 2.

General management is similar to that forany patient presenting with bleeding (see aboveunder placenta previa). The specific measuresinclude immediate delivery, expectant manage-ment and management of complications.

Immediate delivery

This depends on the severity of abruption andwhether the fetus is alive or dead. If the fetus isdead, vaginal delivery should be the goal after

maternal resuscitation, as fetal death occurscommonly in the severe variety of placentalabruption, often with coagulopathy. Onceresuscitation has been initiated, the fetal mem-branes should be ruptured to hasten the onset oflabor. This is effective in most cases but, in afew, augmentation with syntocinon may beneeded. This must be administered cautiouslyas uterine rupture could occur from anoverstimulated uterus.

Where the fetus is alive, the decision on howbest to achieve delivery is not always easy. Thisis compounded by the fact that the outlook forthe fetus is poor, not only in terms of immediatesurvival but also because studies have shownthat as many as 15.4% of liveborn infants donot survive77. However, delivering by Cesareansection when the fetus is alive has been shownin non-randomized, controlled trials to have abetter outcome than vaginal delivery (52% vs.16%78; 20% vs. 15%73). Indecision and unnec-essary delays in performing Cesarean sections70

are responsible for most poor results fromCesarean section in the last quarter of preg-nancy. Cesarean section must therefore be con-sidered in all cases where the fetus is alive,particularly if there is evidence of fetal distress.However, the presence of coagulopathy addsconsiderable risk to the mother, and morbidityand mortality could be increased by surgery.

Once the decision is to deliver and the fetusis alive, the degree of abruption and the stateof the fetus must be taken into considerationbefore delivering. When the abruption is severe,Cesarean section must be performed onceresuscitation has commenced. Such deliveryshould be performed promptly, especially asmost post-admission fetal deaths occur infetuses delivered more than 2 hours afteradmission.

88


Grade Description

01

23

Asymptomatic abruption with a small retroplacental clot (< 150 ml)Vaginal bleeding (150–500 ml); uterine tetany and tenderness may be present; no signs of maternalshock or fetal distressVaginal bleeding; no signs of maternal shock; signs of fetal distressVaginal bleeding; marked uterine tetany yielding a board-like consistency on palpation; persistentabdominal pain, with maternal shock and fetal demise; coagulopathy may be evident in 30% of cases

Table 2 Grading of placental abruption (Sher and Statland)76



If the abruption is mild to moderate, themode of delivery should be determined by thecondition of the baby, its presentation andthe state of the cervix. In the presence of abnor-mal fetal heart rate patterns, immediate deliveryby Cesarean section is the option of choice.However, if the decision is to deliver vaginally,continuous fetal monitoring should be availableto enable early identification of abnormal fetalheart rate patterns. Golditch and Boyce79,Lunan80 and Okonufua and Olatubosun78 haveall shown that the perinatal mortality is higherwith vaginal delivery in the absence of electronicfetal monitoring. There is a place for the use ofprostaglandins in the ripening of the cervixof women with mild abruption, but then thedanger of inducing tetanic contractions mustalways be borne in mind. Where amniotomy isfeasible, this often hastens delivery but, whereit is not possible, syntocinon can be used,though once again maintaining vigilance forhyperstimulation.

Expectant management

This is recommended where neither the fetusnor the mother are at risk. Unfortunately, thelack of signs of fetal compromise on monitoringdoes not guarantee absence of deterioration inthe fetal condition. With expectant manage-ment, pregnancy is prolonged in the hope ofimproving fetal maturity and therefore survival.

It is ideal for pregnancies less than 37 com-pleted weeks of gestation; however, since neo-natal survival is virtually guaranteed > 34–35weeks’ gestation, there is no place in persistingwith such an approach for pregnancies > 34weeks where fetal monitoring cannot bemaintained. Expectant management is recom-mended for patients in whom vaginal bleedingis slight, abdominal pain is mild and usuallylocalized and they are cardiovascularly stable.Once a decision has been made on conservativemanagement, the fetal condition must bemonitored closely as it may change very quickly.

Expectant management can be in the com-munity or in the hospital; admission is not asso-ciated with a better outcome. However, wherepatient education and access to hospital arepoor, admission may provide a safer option. It isperhaps in such communities that admission

may be rejected because it is expensive or causessignificant family disruptions.

During expectant management, fetal growthshould be monitored by regular ultrasound scanas fetal growth restriction is a common findingin association with placental abruption. Thetiming of delivery depends on further vaginalbleeding, the fetal condition, gestational ageand available neonatal care facilities. If thebleeding episodes are recurrent, induction at37–38 weeks is advisable, provided there is nofetal compromise. Where the initial episode issmall and self-limiting and there are no acutefeatures of fetal compromise (e.g. abnormalcardiotocography or a biophysical profile score< 6) or chronic fetal compromise (growthrestriction, oligohydramnios or abnormalumbilical artery Doppler recording), no evi-dence supports induction of labor. Despite this,it is nevertheless common for induction of laborat term to be advocated in such patients, usingthe speculative argument that some undetecteddamage might have occurred to the integrityand function of the placenta and, in the face ofsuch uncertainty, delivery at term confers moreadvantages.

In a small proportion of cases, mildabruption may co-exist with labor. Whetherabruption provoked labor, or vice-versa, isdifficult to establish in these cases. The use oftocolytics in such patients is controversial, astheir use in the presence of placental abruptionis regarded by many as contraindicated sincethey may worsen the process of abruption46.Sholl81, however, stated that a trial of tocolyticsin the presence of mild placental abruptionand labor may successfully prolong pregnancywithout jeopardizing the mother and fetus.There have as yet been no large trials to confirmSholl’s statement.

Management of the complications of placentalabruption

Complications of placental abruption include:

(1) Maternal shock: this may be disproportion-ate to the revealed blood loss. The typeof resuscitation should therefore be deter-mined by the clinical state of the patient.In most cases of shock, features of

89




disseminated intravascular coagulationmust be excluded, as their presence willrequire additional measures to replacecoagulation factors.

(2) Disseminated intravascular coagulation(DIC): treatment will require correction ofthe coagulation factor deficits, in consulta-tion with a hematologist. Monitoring ofrenal function is essential as acute tubularnecrosis is a recognized sequela.

(3) Ischemic necrosis of the distal organs (e.g.kidneys and brain): this requires adequatefluid replacement.

(4) Postpartum hemorrhage (secondary to DICor Couvelaire uterus): treatment is withuterotonic drugs and other methods ofmanaging postpartum hemorrhage.

(5) Isoimmunization: the administration ofanti-D needs to be within 72 h but thequantity administered should be deter-mined by Kleihauers–Betke test.

Management of cases with intrauterinefetal death

Where there is fetal death (in 20% of cases),placental detachment is usually greater than50%, and approximately 30% of patients showevidence of coagulopathy. Such cases shouldtherefore be classified as severe. The manage-ment should consist of the following:

Evaluation and replacement of blood loss

Blood loss > 2500 ml is common. At least 4units of blood should be cross-matched andtransfusion commenced with packed red bloodcells, regardless of the initial vital signs as theinitial hematocrit or hemoglobin levels maybe normal due to hemoconcentration. Onceresuscitation has been established, subsequenthypotension and tachycardia may then appear.

Management of coagulopathy (30% of cases)

Without evidence of excessive vaginal bleeding,no therapy is warranted even in the presenceof abnormal laboratory results. Appropriatereplacement of blood components and

preservation of the intravascular volume are thecornerstones of treatment. Heparin has no rolein the modern management of consumptivecoagulopathy. The presence of coagulopathy perse is not an indication for Cesarean delivery butrather a strong contraindication. Also, the pres-ence of an unfavorable cervix is not an indica-tion for Cesarean delivery, unless the conditionof the mother necessitates prompt delivery.The abdominal and uterine incisions can bleedexcessively when coagulation defects persist (seeChapter 25).

Delivery

Unless there is an obstetric contraindication tovaginal delivery or hemorrhage is so brisk that itcannot be safely managed with vigorous bloodtransfusion, every attempt should be madeto deliver these patients vaginally (withoutjeopardizing maternal health).

Amniotomy (artificial rupture of membranes)and syntocinon infusion should be started. Therigidity of the uterus or the presence of a highintrauterine pressure should not deter the use ofsyntocinon. If no rhythmic uterine contractionsare superimposed on the background uterinehypertonus, then syntocinon should be startedin standard doses. The benefits of achievinga vaginal delivery override the risks of usingsyntocinon. There is no evidence that its use isassociated with enhanced passage of thrombo-plastin into the maternal circulation and therebyinitiating or enhancing maternal consumptivecoagulopathy82. With intrauterine fetal demise,no time limit for delivery is necessary. Thematernal outcome is mainly dependent on thediligence of fluid and blood replacement ratherthan on the interval to delivery83. Where thecervix is unfavorable and maternal health is notin danger, prostaglandins may be used to inducedelivery.

PLACENTA ACCRETA, INCRETA ANDPERCRETA

This is a group of morbidly adherent placenta ofvarying severity. Such morbid adherence occurswhen the implantation site is lacking a sufficientamount of decidua. Consequently, the physio-logical cleavage plane through the decidual

90




spongy layer is missing. This leads to oneor more cotyledons being firmly anchoredto the decidua basalis and even to themyometrium.

The term ‘placenta accreta’ is used to des-cribe any placental implantation that is firmlyadherent to the uterine wall. Placental villi areanchored to the myometrium due to defectivedecidualization. If villi invade the myometrium,the condition is called placenta increta. If theinvasion goes as deep as reaching the serosalsurface, this is called placenta percreta (seeChapter 8).

Although uncommon, they are associatedwith a significantly high maternal morbidityand sometimes mortality primarily due tohemorrhage, uterine perforation, infectionand the associated surgical difficulties andcomplications84.

Incidence

They occur in about 1 in 2500 deliveries. Therehas been a marked increase in the last 50 years,probably secondary to the increase in Cesareansection delivery rates85.

Risk factors include implantation over thelower uterine segment overlying a previoussurgical scar or excessive uterine curettageresulting in Asherman’s syndrome. Placentaprevia is identified in one-third of cases, and25% of women have had a previous Cesareandelivery. Nearly one-quarter have previouslyundergone curettage and another quarter aregrand multigravida (five or more)86.

Diagnosis

Diagnosis is often not made until after delivery.Some patients may present with vague featureswhich include a raised maternal serumα-fetoprotein87 and bleeding before delivery,although this is usually a consequence ofplacenta previa. Uterine rupture may occurantenatally due to myometrial invasion bychorionic villi at the site of a Cesarean sectionscar88.

The use of ultrasound Doppler color flowmapping improves the diagnostic sensitivity.The two most sensitive criteria are, first, adistance less than 1 mm between the uterine

serosal bladder interface and the retroplacentalvessels, and, second, the presence of largeintraplacental lakes89.

Preliminary work suggests that the applica-tion of three-dimensional color power Dopplerultrasound can be complementary to other tech-niques for antenatal imaging. It has been shownto be superior to magnetic resonance imaging inthis context90.

Management

In most cases, problems arise after delivery ofthe baby. Most of the complications of morbidlyadherent placentas are related to the problemsof delivery or failure to deliver. Managementmust therefore aim to minimize these complica-tions (see Chapter 24).

Hemorrhage is the most common and this isassociated with attempts to detach the placentafrom the uterus. In most of these cases, unfortu-nately, the ultimate treatment is usually hyster-ectomy. Alternative approaches to managementinclude uterine/hypogastric artery ligation orangiographic embolization. Sometimes, thepercreta type might even invade the bladderbase, further complicating the surgical proce-dure required and making the control ofhemorrhage very difficult.

In cases of extensive placenta accreta(involving most of the placental surface), bleed-ing might be very limited until attempts atmanual removal are made. At times, traction onthe cord may lead to uterine inversion. Manualremoval is usually not successful as the plane ofcleavage between the uterus and the placentacannot be developed. The safest treatmentis usually hysterectomy. Attempts at uterineconservation include piece-meal removal ofas much placental tissue as possible followedby packing of the uterine cavity, but thisapproach is reported to carry an unacceptablyhigh mortality rate of 25%86. Another optionto conserve the uterus is to leave the entireplacenta in situ if there is no bleeding. Kayemdescribes a case where spontaneous resorptionof the placenta occurred over 6 months follow-ing uterine artery embolization91. Other groupsdescribe a similar approach, but using metho-trexate. The placenta spontaneously deliveredafter 4 weeks92,93.

91




RARE TYPES OF PLACENTALABNORMALITIES: SHAPE

Some anatomical variations in the shape ofthe placenta can give rise to serious postpartumhemorrhage. These include bipartite placentas,succenturiate lobes and placenta membranacea.

Bipartite placenta

Bipartite placenta occurs when the placenta isoccasionally separated into two lobes, and thedivision is incomplete with vessels of fetal originextending from one lobe to the other beforeending in the umbilical cord. Its incidence isabout 1 in 350 deliveries94.

Succenturiate lobes

In this abnormal form, one or more smallaccessory lobes develops in the membranesat a distance from the main placenta. Thesuccenturiate lobes usually connect to the latterwith vascular connections of fetal origin. Itcan be considered to be like a small version ofthe lobate placenta. The accessory lobe may beretained in the uterus after delivery, causingserious hemorrhage. Its incidence has beenreported to be as high as 5%95.

Placenta membranacea

This type of placenta develops as a thin mem-brane-like structure with the whole of the fetalmembranes covering the functioning villi. Thediagnosis can be made with ultrasound scan. Itcan give rise to serious hemorrhage as an associ-ation with placenta previa or accreta. One varia-tion is the ‘ring-shaped’ or ‘horse-shoe’ placentawhere the process does not involve the wholeplacenta, but only a central part. This mightoccur in about 1 in 6000 deliveries95.

References

1. Bonnar J. Massive obstetric haemorrhage.Baillieres Best Pract Res Clin Obstet Gynaecol2000;14:1–18

2. Gilbert TT, Smulian JC, Martin AA. Obstetricadmission to the intensive care unit: outcomesand severity of illness. Obstet Gynecol 2003;102:897

3. Hazelgrove JF, Price C, Papapachan VJ, et al.Multicenter study of obstetric admission to 14intensive care units in southern England. CritCare Med 2001;29:770

4. Zeeman GG, Wendel GDJ, Cunningham FG,et al. A blueprint for obstetric critical care. Am JObstet Gynecol 2003;188:532

5. Jegosathy R. Sudden maternal deaths in Malay-sia: a case report. J Obstet Gynaecol Res 2002;28:186

6. Rahman MH, Akhter HH, Khan ChowduryME, et al. Obstetric deaths in Bengladesh. Int JGynaecol Obstet 2002;77:161

7. Nagaya K, Fetters MD, Ishikawa M, et al.Causes of maternal mortality in Japan. JAMA2000;283:2661

8. Chichakli LO, Atrash HK, Mackay AP, et al.Pregnancy-related mortality in the United Statesdue to hemorrhage: 1979–1992. Obstet Gynecol1999;94:721

9. Crane JMG, Van Den Hof MC, Dodds L, et al.Neonatal outcomes in placenta previa. ObstetGynecol 1999;93:541

10. Martin JA, Hamilton BE, Ventura SJ, et al.Births: Final data for 2001. National VitalStatistics report. Hyattsville: National Center forHealth Statistics, 2002

11. Frederiksen MC, Glassenberg R, Stika CS, et al.Placenta previa: A 22-year analysis. Am J ObstetGynecol 1999;180:1432

12. Babinszki A, Kerenyi T, Torok O, Grazi V,Lapinski RH, Berkowitz RL. Perinatal outcomein grand and great-grand multiparity: effects ofparity on obstetric risk factors. Am J ObstetGynecol 1999;181:669–74

13. Ananth CV, Smulian JC, Vintzileos AM. Theeffect of placenta previa on neonatal mortality:a population-based study in the United States,1989 through 1997. Am J Obstet Gynecol 2003;188:1299–304

14. Gesteland K, Oshiro B, Henry E, et al. Ratesof placenta previa and placental abruption inwomen delivered only vaginally or only by cesar-ean section. J Soc Gynecol Invest 2004;11:208A

15. Gilliam M, Rosenberg D, Davis F. The likeli-hood of placenta previa with greater number ofcesarean deliveries and higher parity. ObstetGynecol 2002;93:973

16. Williams MA, Mittendorf R, Lieberman E,Monson RR, Schoenbaum SC, Genest DR.Cigarette smoking during pregnancy in relationto placenta previa. Am J Obstet Gynecol 1991;165:28–32

17. Brar HS, Platt LD, DeVore GR, Horenstein J.Fetal umbilical velocimetry for the surveillance

92


114Z:\Sapiens Publishing\A5211 - Postpartum Hemorrhage\Make-up\Postpartum Hemorrhage - Voucher Proofs #T.vp04 September 2006 13:44:57


of pregnancies complicated by placenta previa.J Reprod Med 1988;33:741–4

18. Ananth CV, Demissie K, Smulian JC, VintzileosAM. Relationship among placenta previa, fetalgrowth restriction, and preterm delivery: apopulation-based study. Obstet Gynecol 2001;98:299–306

19. Chapman MG, Furness ET, Jones WR, SheatJH. Significance of the location of placenta site inearly pregnancy. Br J Obstet Gynaecol 1989;86:846–8

20. McLure N, Dornan JC. Early identification ofplacenta previa. Br J Obstet Gynaecol 1990;97:959–61

21. Laing FC. Placenta previa: avoiding false-negative diagnoses. J Clin Ultrasound 1981;9:109–13

22. Comeau J, Shaw L, Marcell CC, Lavery JP.Early placenta previa and delivery outcome.Obstet Gynecol 1983;61:577–80

23. Ruparelia BA, Chapman MG. Early low-lyingplacentae – ultrasonic assessment, progress andoutcome. Eur J Obstet Gynecol Reprod Biol 1985;20:209–13

24. Taipale P, Hiilesmaa V, Ylostalo P. Transvaginalultrasonography at 18–23 weeks in predictingplacenta previa at delivery. Ultrasound ObstetGynecol 1998;12:422–5

25. Becker RH, Vonk R, Mende BC, Ragosch V,Entezami M. The relevance of placental locationat 20–23 gestational weeks for prediction of pla-centa previa at delivery: evaluation of 8650 cases.Ultrasound Obstet Gynecol 2001;17:496–501

26. Dashe JS, McIntire DD, Ramus RM, Santos-Ramos R, Twickler DM. Persistence of placentaprevia according to gestational age at ultrasounddetection. Obstet Gynecol 2002;99:692–7

27. Tan NH, Abu M, Woo JL, Tahir HM. The roleof transvaginal sonography in the diagnosis ofplacenta praevia. Aust N Z J Obstet Gynaecol1995;35:42–5

28. Knuppel AR, Drukker JE. Bleeding in LatePregnancy: Antepartum Bleeding. Philadelphia:Saunders, 1986

29. Hertzberg BS, Bowie JD, Carroll BA, KliewerMA, Weber TM. Diagnosis of placenta previaduring the third trimester: role of transperinealsonography. Am J Roentgenol 1992;159:83–7

30. Powell MC, Buckley J, Price H, WorthingtonBS, Symonds EM. Magnetic resonance imagingand placenta previa. Am J Obstet Gynecol 1986;154:565–9

31. Fraser R, Watson R. Bleeding During the LatterHalf of Pregnancy. London: Oxford UniversityPress, 1989

32. Cotton DB, Read JA, Paul RH, Quilligan EJ.The conservative aggressive management ofplacenta previa. Am J Obstet Gynecol 1980;137:687–95

33. Macafee CH, Millar WG, Harley G. Maternaland foetal mortality in placenta praevia. J ObstetGynaecol Br Emp 1962;69:203–12

34. Crenshaw C, Jr, Jones DE, Parker RT. Placentaprevia: a survey of twenty years experience withimproved perinatal survival by expectant therapyand cesarean delivery. Obstet Gynecol Surv 1973;28:461–70

35. Johnson HW, Williamson JC, Greeley AV. Theconservative management of some varieties ofplacenta praevia. Am J Obstet Gynecol 1945;49:398–406

36. Besinger RE, Moniak CW, Paskiewicz LS,Fisher SG, Tomich PG. The effect of tocolyticuse in the management of symptomatic placentaprevia. Am J Obstet Gynecol 1995;172:1770–5;discussion 1775–8

37. Towers CV, Pircon RA, Heppard M. Is tocolysissafe in the management of third-trimesterbleeding? Am J Obstet Gynecol 1999;180:1572–8

38. Silver R, Depp R, Sabbagha RE, Dooley SL,Socol ML, Tamura RK. Placenta previa: aggres-sive expectant management. Am J Obstet Gynecol1984;150:15–22

39. D’Angelo LJ, Irwin LF. Conservative manage-ment of placenta previa: a cost-benefit analysis.Am J Obstet Gynecol 1984;149:320–6

40. Kaunitz AM, Spence C, Danielson TS, et al.Perinatal and maternal mortality in a religiousgroup avoiding obstetric care. Am J ObstetGynecol 1984;150:826–31

41. Rosen DM, Peek MJ. Do women with placentapraevia without antepartum haemorrhagerequire hospitalization? Aust N Z J ObstetGynaecol 1994;34:130–4

42. Anon MG. Editorial comment. Aust N Z J ObstetGynaecol 1994;34:130–1

43. Love CD, Wallace EM. Pregnancies complicatedby placenta praevia: what is appropriate manage-ment? Br J Obstet Gynaecol 1996;103:864–7

44. Ananth CV, Smulian JC, Vintzileos AM. Theassociation of placenta previa with history ofcesarean delivery and abortion: a metaanalysis.Am J Obstet Gynecol 1997;177:1071–8

45. Brenner WE, Edelman DA, Hendricks CH.Characteristics of patients with placenta previaand results of ‘expectant management’. Am JObstet Gynecol 1978;132:180–91

46. Besinger RE, Niebyl JR. The safety and efficacyof tocolytic agents for the treatment of pretermlabour. Obstet Gynecol Surv 1990;45:415–40

93




47. Sampson MB, Lastres O, Tomasi AM,Thomason JL, Work BA, Jr. Tocolysis withterbutaline sulfate in patients with placentaprevia complicated by premature labor. J ReprodMed 1984;29:248–50

48. Frederiksen MC, Glassenberg R, Stika CS, et al.Placenta previa: a 22-year analysis. Am J ObstetGynecol 1999;180:1432

49. Moir DD. Obstetric Anaesthesia and Analgesia,2nd edn. London: Bailliere Tindall, 1980

50. Crawford JS. Priciples and Practice of ObstetricsAnaesthesia, 15th edn. Oxford: Blackwell, 1985

51. Scott JS. Antepartum haemorrhage. InWhitefield CR, ed. Dewhurst’s Textbook ofObstetrics and Gynaecology for Postgraduates, 4thedn. Oxford: Blackwell, 1986

52. Myerscough PR. Munro Kerr’s Operative Obstet-rics, 10th edn. London: Bailliere Tindall, 1982

53. Williamson HC, Greeley AV. Management ofplacenta praevia: 12 year study. Am J ObstetGynecol 1945;50:987–91

54. Rasmussen S, Irgens LM, Bergsjo P, Dalaker K.The occurrence of placental abruption inNorway 1967–1991. Acta Obstet Gynecol Scand1996;75:222–8

55. Fox H. Pathology of the Placenta. London:Saunders, 1978

56. McShane PM, Heyl PS, Epstein MF. Maternaland perinatal morbidity resulting from placentaprevia. Obstet Gynecol 1985;65:176–82

57. Pritchard JA, Brekken AL. Clinical and labora-tory studies on severe abruptio placentae. Am JObstet Gynecol 1967;97:681–700

58. Paterson MEL. The aetiology and outcome ofabruption placentae. Acta Obstet Gynecol Scand1979;58:31–5

59. Rasmussen S, Irgens LM, Dalaker K. The effecton the likelihood of further pregnancy of placen-tal abruption and the rate of its recurrence. Br JObstet Gynaecol 1997;104:1292–5

60. Ananth CV, Savitz DA, Williams MA. Placentalabruption and its association with hypertensionand prolonged rupture of membranes: amethodologic review and meta-analysis. ObstetGynecol 1996;88:309–18

61. Kramer MS, Usher RH, Pollack R, Boyd M,Usher S. Etiologic determinants of abruptioplacentae. Obstet Gynecol 1997;89:221–6

62. Major CA, de Veciana M, Lewis DF, et al.Preterm premature rupture of membranesand abruptio placentae: is there an associationbetween these pregnancy complications? Am JObstet Gynecol 1995;172:672

63. Addis A, Moretti ME, Ahmed Syed F,Einarson TR, Koren G. Fetal effects of cocaine:

an updated meta-analysis. Reprod Toxicol 2001;15:341–69

64. Schiff MA, Holt VL. The injury severity score inpregnant trauma patients: predicting placentalabruption and fetal death. J Trauma 2002;53:946–9

65. Kupferminc MJ. Thrombophilia and pregnancy.Curr Pharm Des 2005;11:735–48

66. Gherman RB, Goodwin TM. Obstetricimplications of activated protein C resistanceand factor V Leiden mutation. Obstet GynecolSurv 2000;55:117–22

67. Ananth CV, Oyelese Y, Yeo L, Pradhan A,Vintzileos AM. Placental abruption in theUnited States, 1979 through 2001: temporaltrends and potential determinants. Am J ObstetGynecol 2005;192:191–8

68. Ananth CV, Berkowitz GS, Savitz DA, LapinskiRH. Placental abruption and adverse perinataloutcomes. JAMA 1999;282:1646–51

69. Eskes TK. Clotting disorders and placentalabruption: homocysteine – a new risk factor. EurJ Obstet Gynecol Reprod Biol 2001;95:206–12

70. Hibbard BM. Bleeding in late pregnancy. InHibbard BM, ed. Principles of Obstetrics. London:Butterworths, 1988

71. Hurd WW, Miodovnik M, Hertzberg V, et al.Selective management of abruptio placentae:a prospective study. Obstet Gynecol 1983;61:467

72. Notelovitz M, Bottoms SF, Dase DF, LeichterPJ. Painless abruptio placentae. Obstet Gynecol1979;53:270–2

73. Page EW, King EB, Merrill JA. Abruptio placen-tae; dangers of delay in delivery. Obstet Gynecol1954;3:385–93

74. Green-Thompson RW. Antepartum haemor-rhage. Clin Obstet Gynaecol 1982;9:479–515

75. Nyberg DA, Cyr DR, Mack LA, WilsonDA, Shuman WP. Sonographic spectrum ofplacental abruption. Am J Roentgenol 1987;148:161–4

76. Sher G, Statland BE. Abruptio placentae withcoagulopathy: a rational basis for management.Clin Obstet Gynecol 1985;28:15–23

77. Abdella TN, Sibai BM, Hays JM Jr, AndersonGD. Relationship of hypertensive disease toabruptio placentae. Obstet Gynecol 1984;63:365–70

78. Okonofua FE, Olatunbosun OA. Caesareanversus vaginal delivery in abruptio placentaeassociated with live fetuses. Int J Gynaecol Obstet1985;23:471–4

79. Golditch IA, Boyce NE. Management ofabruptio placentae. JAMA 1970;212:288–93

94




80. Lunan CB. The management of abruptioplacentae. J Obstet Gynaecol Br Commonw 1973;80:120–4

81. Sholl JS. Abruptio placentae: clinical manage-ment in nonacute cases. Am J Obstet Gynecol1987;156:40

82. Clark S, Cotton DB, Gonik B, et al. Centralhemodynamic alterations in amniotic fluidembolism. Am J Obstet Gynecol 1995;158:1124

83. Brame RG, Harbert GM Jr, McGaughey HS Jr,Thornton WN Jr. Maternal risk in abruption.Obstet Gynecol 1968;31:224–7

84. Zelop CM, Harlow BL, Frigoletto FD Jr, SafonLE, Saltzman DH. Emergency peripartumhysterectomy. Am J Obstet Gynecol 1993;168:1443–8

85. Comstock CH. Antenatal diagnosis of placentaaccreta: a review. Ultrasound Obstet Gynecol2005;26:89–96

86. Fox H. Placenta accreta, 1945–1969. ObstetGynecol Surv 1972;27:475

87. Hung TH, Shau WY, Hsieh CC, Chiu TH, HsuJJ, Hsieh TT. Risk factors for placenta accreta.Obstet Gynecol 1999;93:545–50

88. Liang HS, Jeng CJ, Sheen TC, Lee FK, YangYC, Tzeng CR. First-trimester uterine rupturefrom a placenta percreta. A case report. J ReprodMed 2003;48:474–8

89. Twickler DM, Lucas MJ, Balis AB, et al. Colorflow mapping for myometrial invasion in womenwith a prior cesarean delivery. J Matern FetalMed 2000;9:330–5

90. Lam G, Kuller J, McMahon M. Use of magneticresonance imaging and ultrasound in the antena-tal diagnosis of placenta accreta. J Soc GynecolInvestig 2002;9:37–40

91. Kayem G, Davy C, Goffinet F, Thomas C,Clement D, Cabrol D. Conservative versusextirpative management in cases of placentaaccreta. Obstet Gynecol 2004;104:531–6

92. Henrich W, Fuchs I, Ehrenstein T, Kjos S,Schmider A, Dudenhausen JW. Antenatal diag-nosis of placenta percreta with planned in situretention and methotrexate therapy in a womaninfected with HIV. Ultrasound Obstet Gynecol2002;20:90–3

93. Nijman RG, Mantingh A, Aarnoudse JG. Persis-tent retained placenta percreta: methotrexatetreatment and Doppler flow characteristics. Br JObstet Gynaecol 2002;109:587–8

94. Fox H. Pathology of the placenta. Clin ObstetGynaecol 1986;13:501–19

95. Benirschke K, Kaufman P. Pathology of theHuman Placenta, 4th edn. New York:Springer-Verlag, 2000

95




Documents

Postpartum Hemorrhage - Vouch · related deaths due to hemorrhage in one series8. PLACENTA PREVIA Placenta previa is defined as partial or complete insertion of the placenta onto