Non-Neoplastic Maternal Gestational Diseases

Published on 30/06/2015 by admin

Filed under Pathology

Last modified 22/04/2025

Print this page

This article have been viewed 2166 times

Non-Neoplastic Maternal Gestational Diseases

Eoghan E. Mooney and Emma M. Doyle

Chapter Outline

Gross examination of the placenta by the obstetrician or midwife is a routine practice at the time of delivery. Inspection of the membranes, umbilical cord, and cotyledons can help determine whether there is any retained tissue that may result in postpartum hemorrhage, or any gross abnormalities that might prompt a closer evaluation of the newborn baby. In most institutions the decision to request a pathologic evaluation of the placenta is made by the delivering clinician.

Chapter 32 emphasizes the pathologic findings. This chapter focuses on the placental findings in common clinical scenarios. Points of particular relevance in the gross examination of the placenta are highlighted. While substantial overlap exists for some entities, e.g., intrauterine growth restriction (IUGR) and uteroplacental insufficiency, the scenarios are presented separately for ease of discussion.

Presentation of the findings in a fashion intelligible to clinicians is of major importance. The terminology proposed by the Society of Pediatric Pathology has been referred to.1–3 A synopsis of the findings may be given as a ‘one-liner’ in the pathology report, and may help clinicians to categorize the histologic findings (Table 33.1).

Table 33.1

Classification System for Placental Reporting

1	Normal placenta
2	Placenta with chorioamnionitis
3	Placenta with villitis
4	Placenta with materno-placental circulatory disorder
5	Placenta with fetal–placental circulatory disorder
6	Placenta with maturation disturbances
7	Placenta with findings suggestive of gene aberration
8	Placenta with placentation defects
9	Placenta with other pathology

Courtesy of Professor Borghild Roald, Oslo, Norway.⁴

Early Pregnancy Loss (Spontaneous Miscarriage)

The term ‘abortion’ is often replaced by the term ‘miscarriage’ in clinical practice. There is considerable variation in the definition of miscarriage, but 500 g birthweight or 24 weeks of gestation are frequently used as the upper limits in the definition of miscarriage (vs stillbirth). The incidence of miscarriage is difficult to ascertain as many women miscarry before they are aware of the pregnancy: as many as 78% of human conceptions do not result in term delivery. Approximately 15–25% of recognized pregnancies will end in a miscarriage.

The report on the products of conception from a miscarriage should (1) state that a pregnancy can be proven (by the presence of tissues of fetal origin, i.e., embryonic parts, chorionic villi, or trophoblast); (2) state if it is intrauterine (by the presence of decidua with implantation site); and (3) exclude a serious disease process, especially gestational trophoblastic disease.

Selected tissue from all miscarriages should be examined microscopically. Tissues felt to be present macroscopically are sometimes absent histologically, and the converse is also true. Fetal parts, when present, may be examined in accordance with parental wishes. Suction termination disrupts the fetus, but measurements of a hand or foot permit comparison against tables of growth rates. Whole fetuses and large embryos may be examined according to published protocols.

Chromosomal abnormalities in the conceptus are more frequently identified in earlier than in later fetal losses, with 65–75% of first trimester miscarriages having an abnormal karyotype.5,6 Most will be sporadic events, but a small minority result from balanced translocations in one of the parents and hence may recur. Some of the more commonly encountered karyotypic abnormalities are trisomies, especially 16 and 13, and monosomy X (45,X). Even when the karyotype is normal, there is a clear association between congenital abnormality and miscarriage. The incidence of neural tube defects and other minor abnormalities, such as cleft lip and palate or polydactyly, is higher in abortuses than in live births. Abnormalities of chromosomes 16 and 22 are prominent in early pregnancy loss (Figure 33.1A and B), and these chromosomes are preferentially involved by confined placental mosaicism.⁷

Figure 33.1 Villous changes seen in miscarriages with (A) trisomy 16 and (B) trisomy 22.

There have been several attempts to determine the etiology of early fetal loss by morphologic examination. Villous features thought to represent karyotypic abnormalities include vesicular change, irregular villous contour, presence of trophoblastic pseudoinclusions, trophoblastic hyperplasia, abnormal stromal cells, and, in the case of monosomy X, villous fibrosis and hypoplasia. Recognition of abnormalities at the maternal–embryonic interface (i.e., absence of interstitial trophoblast columns, absence of intravascular trophoblast plugs, and absence of physiologic changes in spiral arteries) increases the sensitivity of histology for detection of chromosomal abnormalities. Recurrent and sporadic miscarriages have similar findings, and in the majority of cases it is difficult to ascertain the cause.⁸ Elegant light and electron microscopic studies support the concept that many early pregnancy losses result from premature onset of the maternal–placental circulation secondary to inadequate trophoblast invasion.

Uterine abnormalities such as leiomyomas, cervical incompetence, and congenital abnormalities of müllerian duct fusion may all contribute to early pregnancy loss. Infection becomes increasingly important later in gestation as the influence of abnormal karyotype wanes. Early on, Campylobacter sp., Listeria monocytogenes, Toxoplasma gondii, rubella, cytomegalovirus (CMV), and syphilis are important. Molecular analysis suggests that coxsackie virus is one of the most common infectious agents, but placental findings are nonspecific.⁹ Similarly, human papillomavirus (HPV) is more commonly demonstrated in spontaneous pre-term delivery.¹⁰

The finding of a placental site trophoblastic reaction is important as it generally excludes that the pregnancy was ectopic in the fallopian tube and merely became displaced into the uterus following miscarriage. The presence of chorionic villi and decidua by themselves theoretically do not achieve this according to some reports in the literature,¹¹ but neither of the authors has encountered an example of chorionic villi in the uterus where the origin proved to be an ectopic pregnancy. In summary, examination of the early pregnancy loss will:

• Confirm intrauterine gestation in most cases

• Exclude hydatidiform mole in the vast majority of cases

• Identify an occasional case of chronic histiocytic intervillositis

• Not identify the etiology in most cases.

Mid to Late Pregnancy Loss

The 9–12 week period sees the highest rate of miscarriage. Miscarriages may be retained for some time before detection, when advanced maceration can make precise dating and definitive diagnosis difficult.

If a miscarriage at 18–22 weeks is fresh, infection is a likely cause and examination should focus on the membranes. A small placenta is frequently found in trisomies and should prompt cytogenetic analysis, as karyotypic abnormalities can be expected from over 85% of placental cultures even when maceration is present.¹² Second trimester miscarriages may occur following abnormal implantation, with ischemic changes up to and including infarction. Any infarction in the second trimester is likely to be clinically significant, and should be confirmed histologically. Placental infarction is associated with cerebral ischemia in the fetus, particularly in growth-restricted infants.¹³ The mean gestational age of pregnancy loss due to intervillositis in our practice is 25 weeks, and any increase in fibrin should prompt careful evaluation for this entity. Villitis of unknown etiology is uncommon before the third trimester.

Caution must be exercised in making the diagnosis of fetal thrombotic vasculopathy in macerated fetuses, unless there is unequivocal evidence of antemortem vascular pathology, such as clearly defined avascular villi (Figure 33.2). This is usually not possible, but, where present, enables antemortem pathology to be distinguished from the secondary effects of intrauterine death.

Figure 33.2 Despite changes of intrauterine death, avascular villi can be discerned (left of picture).

Abnormal cord coiling, assessed as both under- and overcoiled cords, is associated with fetal death.14,15 Cord coiling is frequently presented as an explanation for fetal death. Features of vascular compromise such as avascular villi, vascular ectasia, and mural fibrin thrombi should be sought in this context.

The role of an individual placental finding in stillbirth may be problematic. A National Institute of Child Health and Human Development workshop in 2007 evaluated maternal, fetal, and placental conditions as a cause of stillbirth.¹⁶ The proceedings also include useful overviews of the role of infectious organisms in stillbirth.

Abruption

Abruption, a condition in which the placenta detaches if not tears away from the uterine wall, can be a dramatic obstetric event. Although abruption and retroplacental hemorrhage are sometimes used interchangeably, abruption signifies a clinical event with signs and symptoms, whereas retroplacental hemorrhage is a pathologic finding, often with no clinical correlation (see Chapter 32). Occurring in 1% of pregnancies, abruption is a leading cause of vaginal bleeding in the latter half of pregnancy, and an important cause of perinatal morbidity. It accounts for 12% of perinatal deaths.¹⁷ The effect on the mother depends primarily on the severity of the abruption, whereas the effect on the fetus depends both on the severity of the abruption as well as on the fetus’s gestational age. Risk factors include prior abruption, smoking, trauma, multifetal gestation, hypertension, pre-eclampsia, thrombophilias, advanced maternal age, preterm premature rupture of the membranes, intrauterine infections, and polyhydramnios.¹⁸ Cocaine abuse may also be a cause. From the pathologist’s viewpoint, chorionic villus hemorrhage and villus edema are more frequent in cocaine users.

Questions frequently posed to the pathologist include: does an abruption exist, how extensive is it, and is it recent or old? More often than not, none of these questions can be answered solely on the pathologic examination, leading the pathologist to come away disappointed at not being able to add much of clinical use. Most are peripheral (Figure 33.3) and, if recent, will not be detected by pathologic examination, since the hemorrhage will have escaped peripherally and presented as vaginal bleeding. Retroplacental hemorrhage toward the center is far more significant (Figure 33.4), since the displaced placenta can no longer support fetal functions. In a study of over 53,000 pregnancies¹⁹ the extent of placental separation determined the chance of stillbirth. With a 75% separation, the risk of fetal death was increased 31-fold. Even a 50% separation showed profound effects. The risk for preterm delivery was also substantially increased when the separation was mild (with a separation of 25%, the relative risk was 5). The amount of bleeding is best gauged by the volume of retroplacental clot found at cesarean section, or secondarily by the amount of blood clot that accompanies the placenta to the pathology laboratory. The duration of the bleeding is best gauged by whether the parenchyma has been displaced by the blood, whether the junction of the blood and parenchyma shows discoloration from pigment breakdown, and whether the parenchyma near the abruption is infarcted.

Figure 33.3 Peripheral and probably clinically insignificant abruption/retroplacental hemorrhage.

Figure 33.4 Abruption with parenchymal compression nearly half the placental thickness.

Macroscopically, the location, diameter of the involved area, depth of the crater, and thickness of the normal placenta above should be recorded, as these measurements will help determine the significance of the abruption. Microscopic sections should be taken at several areas of the junction between the blood and placental parenchyma, as the tissue reaction can help date the abruption. Neutrophils in the basal plate above the hemorrhage are the earliest finding, occurring in <4 hours.²⁰ Nearly a day will have passed before the villi degenerate and an inflammatory response appears at the edge of the hemorrhage (Figure 33.5). Within several days there will be evidence of hemoglobin breakdown. Some placentas will show different changes at different areas. Placental abruption is generally regarded as an acute event, but it is often the end result of chronic processes that have begun earlier in pregnancy, even near the time of conception.²¹ In a cohort of extremely low birth weight infants (<1000 g), placental abruption was found more commonly with histologic chorioamnionitis and funisitis.²²

Figure 33.5 Abruption, clinically of 1 day’s duration. An acute inflammatory infiltrate has developed at the interphase of the bleeding and parenchyma.

Hypertensive Disorders

Pre-Eclampsia

Pre-eclampsia, a disorder unique to pregnancy, occurs in about 5–8% of pregnant women. Risk factors include nulliparity, women who are at the extremes of reproductive age, multiple gestation, and chronic diseases such as diabetes mellitus, kidney disease, and chronic hypertension. It usually begins insidiously after week 20 of pregnancy with excessive weight gain, marked fluid retention, increase in maternal systemic blood pressure, and the appearance of proteinuria. Pre-eclampsia is diagnosed when the blood pressure is sustained at or above 140 mmHg systolic or 90 mmHg diastolic and the urinary protein excretion exceeds 300 mg/day, but classification systems differ somewhat.²³

Progression of mild pre-eclampsia can be slow or rapid and can manifest in many different ways. With increasing blood pressure and declining renal function the resulting fluid retention may result in pulmonary edema. The systemic vasospasm that occurs is particularly noticeable in organs with microvasculature. Hence the classic signs and symptoms of pre-eclampsia, including: visual disturbances from retinopathy, headache from cerebral vasculature changes, chest pain and shortness of breath from cardiopulmonary changes, right upper quadrant discomfort from hepatic dysfunction, proteinuria and glomerulopathy from decreased renal blood flow, systemic edema from peripheral vasospasm, and oligohydramnios and non-reassuring fetal testing from decreased placental perfusion.

Pre-eclampsia can be fatal for both the mother and the fetus. Maternal complications include eclamptic seizure, the syndrome of hemolysis, elevated liver enzymes, and low platelets (HELLP syndrome), subcapsular liver hematoma, pulmonary edema, acute renal failure, intracerebral hemorrhage, and death. The disseminated intravascular coagulation that sometimes develops is not part of the pre-eclampsia itself as much as a consequence of the developing HELLP syndrome, abruption, or hematoma. Fetal complications include non-reassuring fetal testing, placental abruption, oligohydramnios, IUGR, and death. Pre-eclampsia is treated with magnesium sulfate and antihypertensive agents, but the definitive therapy is the removal of the placenta, usually by induction of labor or by cesarean delivery.

Pre-eclampsia is classified as early onset (<34 weeks) or late onset. While late onset is by far the more common, it is not associated with placental disease and usually does not result in fetal growth restriction. The points below relate to early onset pre-eclampsia, which has been described as ‘a disease of failed interaction between two genetically different organisms.’²⁴ Along with understanding of the role of implantation has been its description as a two-stage disease.²⁴ Stage 1 is inadequate implantation and stage 2 is the clinical manifestations of endothelial activation in the third trimester. The sequence of events is outlined in Figure 33.6. Decidual natural killer cells, which constitute the majority of decidual white cells in the first trimester, play a key role in trophoblast invasion and successful implantation.²⁵ Persistence of smooth muscle in uterine spiral arteries results in irregular pulsatile flow, with consequent mechanical and oxidative stress to the placenta. This results in increased shedding of trophoblast by necrosis and aponecrosis, which facilitates an inflammatory response in the mother.²⁶ The complex interaction of abnormal placental flow and pre-eclampsia suggests that impaired invasion of deep spiral arteries might result from, rather than cause, maternal flow defects.²³

Figure 33.6 Pathogenesis of pre-eclampsia and eclampsia. (Data from Robboy SJ, Duggan M, Kurman RT. The female reproductive system. In: Rubin E, Farber J, editors. Pathology. 3rd ed. Philadelphia: Lippincott; 1999. p. 962–1028.)

Examination of the placenta from early onset pre-eclampsia should include evaluation of acute and chronic hypoxic changes, with a comment as to the degree of severity. Expected findings include a small placenta with accelerated maturation, ischemic villous crowding, and infarction. Atherosis (Figure 33.7) will be seen in only 50% of cases, but nontransformed vessels (i.e., those whose muscularis has not been replaced by trophoblast and fibrin) (Figures 33.8 and 33.9), and hypertrophic decidual arteriopathy may be present. Laminar decidual necrosis is common. Thrombotic disease in the fetal circulation is three to four times more common when there is uteroplacental ischemic disease, and may be focal or extensive.

Figure 33.7 Maternal hypertension. Untransformed vessel wall, but with atherosis.

Figure 33.8 Maternal hypertension with a vessel showing atherosis (left) contrasted with untransformed vessels (right).

Figure 33.9 Maternal hypertension with untransformed vessel wall.

Intrauterine Growth Restriction

Intrauterine Growth Restriction (IUGR), also known as fetal growth restriction, is where the infant fails to achieve its biologic growth potential. This category overlaps with those who are small for gestational age. Both entities are commonly defined as fetal weight below the 10th percentile for gestational age, but approximately 70% of infants in this group are constitutionally small and have no pathology. Data on race, birth order, and parental body characteristics can help improve identification of true growth restriction.²⁷ Agreement between customized standards and a population standard on growth restriction can be expected in approximately two-thirds of births. Birth weight below the 3rd percentile shows the strongest correlation with perinatal mortality.

IUGR is a late stage manifestation of several disease states. The etiologies can be divided into maternal causes and fetal factors. Chronic maternal hypoxia is an important cause, seen in high altitude, pulmonary hypertension, anemia, hemoglobinopathies, and cardiac failure.²⁸ Fetal factors include fetal genetic disorders, chromosomal abnormalities, and congenital malformations. Mothers who have a history of a prior pregnancy with IUGR are also at risk for recurrence in subsequent pregnancies, including an increased risk of stillbirth.²⁹

Pre-eclampsia is accompanied by IUGR in up to one-third of cases, but IUGR may occur in normotensive pregnancies. Differences in the pathways of spiral artery remodeling may result in the development of IUGR alone, or PET with IUGR.²⁵

Placental pathology may cause or contribute to IUGR. The fetal surface should be examined for vascular thrombosis and the maternal surface for infarcts. The cut surface may show infarcts or increased perivillous fibrin. However, many diseases responsible for causing IUGR can only be diagnosed microscopically, such as villitis. In our practice over a recent 3 year period, placental findings in 126 cases of IUGR below the 3rd percentile were uteroplacental ischemia 44%, fetal thrombotic vasculopathy 8%, villitis 14% (with two-thirds of these being moderate to severe), and intervillositis or increased perivillous fibrin 4%. Ten percent of placentas were small, but histologically normal.

Morphologic findings in IUGR in some cases reflect the effects of stress on the placenta. Placental function is compromised by oxidative stress and complement activation, causing a decrease in the functioning trophoblast mass. This is manifest by syncytiotrophoblast apoptosis, syncytial knots, and increased perivillous fibrin.³⁰

Early onset IUGR is less common than late onset. It is characterized by absence or reversal of end-diastolic flow in umbilical artery Doppler studies. Histologic examination shows poor peripheral villous development with nonbranching angiogenesis and increased syncytial knots.³¹ This cohort with persistent absence or reversal of end-diastolic flow represents a severely affected group in whom early intervention is warranted. Other abnormalities of Doppler flow include abnormalities of the systolic/diastolic ratio and of the pulsatility index. These correlate strongly with placental abnormalities, not just those reflecting maternal underperfusion, but also villous abnormalities and fetal vascular obstruction.³² Epigenetic regulation of the placenta occurs, with imprinted and nonimprinted genes influencing development.³³ Overexpression of PHLDA2, a paternally imprinted gene that regulates placental growth, correlates with histologic features of maternal vascular underperfusion.³⁴

Confined Placental Mosaicism

Confined placental mosaicism occurs when the placenta’s karyotype differs from that of the fetus. This may cause both IUGR and fetal loss. In one study of IUGR with matched controls, confined placental mosaicism was significantly higher (15.7% vs 1.4% in controls).³⁵ While these placentas often showed decidual vasculopathy and infarction, there are no criteria for diagnosis by light microscopy. The diagnosis depends on awareness of the condition and ancillary tests such as fluorescent in situ hybridization or comparative genomic hybridization. Involvement of the cytotrophoblast as well as the mesenchymal core is required for adverse outcomes and low birth weight.⁷

Adverse Neurologic Outcome: Neonatal Encephalopathy and Cerebral Palsy

The area of neonatal neurologic disability is a contentious one, and the erroneous attribution of such damage to ‘birth asphyxia’ has led to obstetricians becoming uninsurable in many Western jurisdictions. The incidence of cerebral palsy has remained between 2 and 3 per 1000 live births over the last three decades, with similar figures from many centers.³⁶ Only 4% (2 of 46) of cases of cerebral palsy in one center were related to birth asphyxia when the objective American College of Obstetricians and Gynecologists/American Academy of Pediatrics criteria were used.³⁷

Placental examination, when combined with relevant clinical information, can add much to the understanding of a bad outcome. Any center wishing to examine such placentas must first have a mechanism in place for identifying, retrieving, and processing them, without producing an impossibly large workload. A diagnosis of cerebral palsy may often not be made until the nonprogressive nature of the impairment is clear, and, as such, placental analysis may have to proceed with a less definitive label. For term infants, neonatal encephalopathy is the best predictor of long-term neurologic disability. Over 90% of placentas from term infants who later develop cerebral palsy have a finding that may have contributed to the abnormality.³⁸ IUGR may offer protection against the development of brain injury.³⁹

Macroscopic description should include any findings (and relevant negatives) of the cord, membranes, fetal surface vessels, parenchyma, and basal plate. At a minimum, a section of parenchyma, cord, and membranes should be available for microscopy. Gross examination of the fixed tissue by the trained observer may identify foci of avascular villi that can be missed at a busy grossing station and may require additional sections. The key lesions to be sought, graded, and reported are those of inflammation and thrombosis.40–42 The presence of both, with lesions of different age and duration, appears to increase the risk of cerebral palsy. The coexistence of subacute and chronic pathology was significantly more likely to be identified in a cohort of medicolegal cases than in controls (24% vs 2%, respectively).⁴² In a series of 12 cases of neonatal stroke from a national registry, placental pathology was found in 10 (83%), with five (42%) showing mixed lesions.⁴³ Infection plays an important role in the genesis of cerebral palsy. Clinical and histologic chorioamnionitis both increase the risk of development of cerebral palsy.⁴⁴ Bacteria and other microorganisms may not be cultured, due either to antibiotic treatment or to their fastidious nature. Not uncommonly, infectious agents may exist in the placenta without histologic findings—nearly three-fourths of neonates with poor outcomes in one study had viral or bacterial disease (coxsackievirus 46%, bacteria 38%, herpes 8%, parvovirus 4%, and picornavirus 4%) when tested with in situ hybridization or reverse transcriptase polymerase chain reaction (RT-PCR), compared with none of the controls.⁴⁵ In a population-based study, perinatal exposure to neurotropic viruses, in particular herpes B viruses, showed an association with cerebral palsy.⁴⁵ Given the prevalence of viral nucleic acids in the control population (almost 40%), the need for a cofactor or trigger might be necessary to result in damage. Histologic findings may suggest chronic infection, with capsular deciduitis associated with periventricular leukomalacia in infants born between 23 and 34 weeks of gestation.⁴⁶

Diabetes Mellitus

Both placental weight and macrosomia increase with poor control. Delayed villous maturation (villous immaturity) is significantly associated with pregestational and gestational diabetes.⁴⁷ Other changes include chorangiosis. Maternal hypertensive disorders also impact on the placental morphology.⁴⁸ With good control and without macrosomia, placental weight may be normal and there may be no discernible changes by routine light microscopy. With stereologic assessment, more subtle changes may be detected, even in type 1 diabetics with good glycemic control, with enhanced angiogenesis, and increased capillary length and volume.^49,50

Hydrops Fetalis (Maternal Rhesus Isoimmunization)

Maternal rhesus (Rh) isoimmunization and to a lesser extent anti-Kell, Duffy, and other minor antibodies and ABO incompatibility produce similar placental changes. However, with modern obstetric practice, full-blown cases of immune hydrops fetalis are uncommon. The Rh-negative mother may be sensitized in her first pregnancy, either before birth from fetal–maternal hemorrhage or during birth. The risk of sensitization increases in cases of operative removal of the placenta.

Macroscopically, the placenta ranges from normal to enlarged and bulky. The villi may show appropriate or delayed maturity, with an increase in immature intermediate villi. Generally, the placental changes are inconstant in any given placenta, appearing as a mosaic of normal areas admixed with others that are edematous. Cytotrophoblasts are mildly increased and there is focal basement membrane thickening. Up to 30% of cases show increased numbers of intervillous thromboses and fibrinoid necrosis of villi. The villous capillaries (which tend to be sparser than normal) contain nucleated fetal red blood cells. The villous stroma is edematous and macrophages (Hofbauer cells) are easily identified. The cause of the edema is believed to be a consequence of fetal anemia leading to high-output cardiac failure. These changes are not pathognomonic for Rh isoimmunization.

Placenta Creta

Placenta accreta, which occurs in about 0.9% of deliveries,⁵¹ substantially increases the risk of preterm delivery and small for gestational age babies. Previous cesarean section is the most common cause of a morbidly adherent placenta (Figure 33.10). The disorder is also associated with prior manual removal of the placenta, cornual implantation, leiomyomas, and prior endometrial scarring. It is frequently diagnosed clinically when there is a difficult delivery of the placenta and relates to undue adherence of the placenta to the uterine myometrium. Ultrasound with Doppler imaging of the placental vasculature and MRI of the placenta can confirm these suspicions, thus helping to guide management of the delivery. The term ‘accreta’ is used to mean an attached placenta without myometrial invasion. Increta implies a moderate degree of myometrial penetration and percreta means penetration by chorionic villi through the entire wall to the level of the serosa (Figure 33.11), sometimes with invasion of adjacent pelvic organs. In practice, all types are sometimes referred to as ‘placenta creta.’

Figure 33.10 Placenta accreta: note the old cesarean section scar. (Reproduced with permission from Kelehan P, Mooney EE. Pathology of the uterus. In: Sir S. Arulkumaran et al., editors. A comprehensive textbook of postpartum hemorrhage. 2nd ed. Kirkmahoe, UK: Sapiens Publishing; 2012.)

Figure 33.11 Placenta percreta. The placenta penetrates the myometrium to the level of the serosal covering.

The primary defect is deficiency of the decidua basalis with consequent apposition of chorionic villi and myometrium. The significance of myometrial cells in the basal plate in the absence of clinical evidence of placenta accreta is controversial, and the frequency of this finding will be influenced by sampling. The term ‘placenta adhesiva’ has been used for an abnormally adherent placenta that has a clear plane of separation on manual removal.⁵² Myometrial fibers in the basal plate were found in 78% of these placentas and multinucleate trophoblast is reduced. The term ‘occult placenta accreta’⁵³ has been used for cases where there are myometrial fibers in the basal plate without intervening decidua. An increase in intermediate trophoblast was seen and there was an association with features of placental hypoxia.

Morbid adherence of the placenta is one of the most common indications for postpartum hysterectomy and this scenario is most likely when there is a prior cesarean section and current placenta previa. The placenta will have a markedly disrupted maternal surface and careful dissection is needed to identify chorionic villi adjacent to or admixed with myometrial fibers. The finding of only trophoblast admixed with superficial myometrial fibers is insufficient for the diagnosis, as this occurs normally in the uncomplicated state. In a hysterectomy specimen with placenta accreta the chorionic villi are found lying directly on the surface of the myometrium or in the uterine wall.

Alterations in the myometrial wall seen following cesarean section include distortion and widening, inflammation, and adenomyosis. A grossly visible defect may be present in the anterior wall. Implantation on either a normally healed or a diseased scar will not have the protective effect provided by the presence of fundal decidua, and normal postpartum separation cannot occur. Implantation in the lower segment (adjacent to the defect) can cause expansion of the defect, dehiscence of the wall, and the formation of a sac that will further enlarge and progress with growth of the placenta.

Postpartum Hemorrhage and Subinvolution

Postpartum hemorrhage may be immediate or delayed. Immediate causes include atony, where there are few or no pathologic findings. Implantation on a cesarean section scar leading to placenta accreta will show morbid adherence. Delayed hemorrhage may be due to retained products of conception. Subinvolution of the blood vessels of the placental bed is an important and probably under-recognized cause of secondary postpartum hemorrhage.

Normal arterial involution involves a decrease in the lumen size, disappearance of trophoblast, thickening of the intima, regrowth of endothelium, and regeneration of internal elastic lamina. These changes occur within 3 weeks of delivery. With subinvolution, arteries remain distended and contain red cells or fresh thrombus (Figure 33.12), and trophoblast persists in a perivascular location. In some cases, endovascular trophoblast may be present. Hemorrhage from subinvolution is maximal in the second week postpartum, although it may occur up to several months later. It is more common in older, multiparous women and may recur in subsequent deliveries. Subinvolution is not related to the method of delivery and may be regarded as a specific entity, possibly due to an abnormal immunologic relationship between trophoblast and the uterus.⁵⁴ The changes may be recognized on curettage specimens, and retained products may or may not be present. The hysterectomy specimen will show a uterus that is soft and larger than expected. As normally involuted vessels may be present adjacent to subinvoluted ones, multiple blocks of placental bed should be taken to exclude this process.

Figure 33.12 Subinvolution: affected vessels are dilated and filled with blood, in contrast with the normally involuted vessel (center left).

Maternal Sickle Cell Trait and Disease

The heterozygous trait (sickle cell trait) appears not to be associated with placental pathology. In the homozygous form, maternal anemia and veno-occlusive disease result in a preplacental type of fetal hypoxia. Alterations in the endothelium of the umbilical vein, demonstrable by scanning and transmission electron microscopy, may also be due to hypoxia.⁵⁵ The initial clue that the disease exists may be from the finding of sickled cells in the intervillous space of the placenta (Figure 33.13).

Figure 33.13 Sickled red cells. Compare the sickled red cells in the perivillous space with the perfectly round and smooth fetal red cells in chorionic villi.

Twin Pregnancy

As discussed in Chapter 32, establishing chorionicity (e.g., monochorionic or dichorionic) is critical in assessing the pathology that may be present in a twin pregnancy. Twins have a higher incidence of perinatal morbidity and mortality than singleton pregnancies, and monochorionic twins are at higher risk than dichorionic. Ten percent of twins with dichorionic placentas are monozygotic, and fetal outcome is related to chorionicity rather than to zygosity. Monoamniotic monochorionic twins (MoMo) have a 20% fetal mortality rate.⁵⁶ This is usually attributed to prematurity, congenital abnormalities, and rarely twin–twin transfusion syndrome. Cord entanglement is seen in 40% of MoMo twins. Monochorionic twins weigh less than twins from dichorionic gestations. Severe growth discordance between the twins (>25%) has an overall prevalence of approximately 9% in twin pregnancies⁵⁷ but is more common in MoMo twins where it is associated with an increased perinatal mortality. MoMo placentas have significantly greater numbers of both superficial and deep anastomoses than do uncomplicated monochorionic diamniotic pregnancies, which may provide a vascular basis for the twin–twin transfusion syndrome that occurs on rare occasions in monoamniotic pregnancies. Monochorionic twins have an increased risk of fetal death and neurologic injury.⁵⁸

Assessing growth restriction in twins may be problematic, and use of a twin nomogram with allowance for chorionicity is advised. Peripheral cord insertion reflective of a spatially limited intrauterine compartment and avascular villi indicative of occluded fetal vessels in the placenta are associated with moderate growth discordance (>15% of birth weight) or low birth weight (<10th percentile).⁵⁹ Decreased placental weight, velamentous insertions, and a single umbilical artery have also been associated with severe growth discordance.⁶⁰ Other findings such as thrombotic lesions differ according to chorionicity; nonocclusive mural thrombi in fetal vessels are more common in monochorionic than in dichorionic twins or in singletons, and are associated with avascular villi in the last two groups.⁶¹

A systematic review of observational studies over a 15 year period has shown that, following death of one twin, monochorionic and dichorionic twins have a death risk of 12% and 4%, respectively, with a risk of neurologic disability of 18% and 1%.⁶² The likelihood of survival of the co-twin decreases with increasing gestational age of the first fetal death, with mortality higher for same-sex twins.

Prolonged Pregnancy

Prolonged pregnancy is generally defined as a gestation, based on menstrual age, exceeding 42 weeks. With the estimated date of confinement, or due date, for normal pregnancies calculated as 38 weeks after conception, or 40 weeks from the first day of the last normal menstrual period, about 18% of pregnancies in the United States extend 1 week beyond the normal due date and 7% beyond 2 weeks. (Some differences between statistics in the United States and Canada seem to relate to whether menstrual or clinical dates are used.63,64) In general, the risks of post-term pregnancy include increased meconium aspiration, macrosomia, and perinatal death, with small for gestational age infants at increased risk.⁶⁵ Postmature placentas may be normal. Delayed villous maturation is important to recognize in this cohort. It typically permits normal growth, but by unknown mechanisms may culminate in a normally formed macerated term stillbirth.

Maternal Infections and the Placenta

The possible effects of exposure to infections during pregnancy are a cause of great anxiety. In addition to serologic evaluation during pregnancy, examination of the placenta may help to confirm or refute intrauterine infection.

Cytomegalovirus

Cytomegalovirus (CMV) is considered one of the most common and important congenital infections in developed nations with approximately 1% of pregnant women acquiring a primary infection.⁶⁶ It is also the most common agent of the toxoplasmosis, other infections, rubella, CMV infection, and herpes simplex (TORCH) group to infect the placenta and fetus.

With 40% of the women transmitting the infection to their fetuses, the risk of serious fetal injury is great. While the risk of transmission is greatest if the infection is acquired during the third trimester, the risk of serious fetal injury is greatest if acquired during the first or early part of the second trimester. The virus is transmitted in blood and body fluids, including saliva. Infection in a pregnant woman is usually asymptomatic. The virus can be transmitted following a primary infection in the mother during pregnancy or a recrudescence of a latent infection acquired at a time before pregnancy. Most (75–85%) infants suffer no adverse effects. Those that do may suffer from IUGR and serious neurologic impairment, including deafness and ocular damage. Co-infection with HIV may enhance CMV transmission and lead to more severe tissue damage.

The fetus and placenta may be hydropic. The combination of a lymphoplasmacytic villitis (Figure 33.14), villous sclerosis, and hemosiderin deposition (Figure 33.15) is commonly found. The characteristic deposition of hemosiderin pigment in sclerotic villi reflects the virus’s endotheliotropic nature, although syncytiotrophoblasts are also infected.⁶⁷ Intranuclear inclusions (Figure 33.16) may be seen adjacent to necrotic debris in the villous stroma, although these may be few or absent, even with immunohistochemistry (Figure 33.17).

Figure 33.14 CMV villous edema and chronic inflammatory cells.

Figure 33.15 CMV plasma cells and hemosiderin prominent in villus.

Figure 33.16 CMV inclusion.

Figure 33.17 CMV (immunoperoxidase stain).

The diagnosis can be otherwise confirmed by serologic testing, viral culture, and PCR assay to detect CMV DNA in fetal and placental tissues.⁶⁸ Demonstration of CMV and parvovirus B19 DNA was more common in placentas from stillbirths than controls, but does not prove causality.⁶⁹

Placentas from cases with fetal transmission of HIV-1 subtype E (the type most frequently seen in Southeast Asia) show chorioamnionitis, chronic deciduitis, and decidual necrosis.⁷³ No specific features are seen.⁷⁴ Histologic chorioamnionitis is the most common pathologic finding in placentas of HIV-1 infected women, and this may increase the risk of viral transmission to the fetus.⁷⁵

Human Papillomavirus

Human Papillomavirus (HPV) is a DNA virus trophic for squamous epithelium, but may be found in the placenta. In vitro studies show that HPV can infect extravillous trophoblast and decrease extracellular matrix invasion.¹⁰ Infection with HPV was demonstrated in 6% of cases between 11 and 13 weeks’ gestation,⁷⁶ and a higher rate of spontaneous abortion (60% vs 20% of controls) has been reported in HPV-positive women. HPV has been demonstrated in cord blood:⁷⁷ the placentas were macroscopically normal, and no microscopic features of HPV infection have been described.

Toxoplasmosis

Toxoplasmosis, an infection caused by the protozoan parasite T. gondii, may be acquired by eating contaminated meat or from exposure to contaminated soil or cat litter. Immunocompetent individuals are usually asymptomatic. Acute infections in pregnant women can be transmitted to the fetus, although approximately 75% of congenitally infected newborns are asymptomatic.⁷⁸ Consequences include mental retardation, blindness, and epilepsy, but, today, the classic triad of chorioretinitis, hydrocephalus, and cerebral calcifications is relatively rare. The risk of maternal–fetal transmission increases with gestational age at the time of exposure, whereas the incidence of severe disease decreases. If infection occurs just before or during the first trimester, it may cause miscarriage, intrauterine death, or severe neurologic lesions, whereas fetal infection occurring late in pregnancy may result in either congenital disease or a subclinical state. The classic finding in the placenta is villitis (Figure 33.20), but manifestations range from no sign of inflammation to necrotizing inflammation. Toxoplasma cysts may be seen in the cord and membranes (Figures 33.21 and 33.22), and fetal vessel calcification may be found.⁷¹ Toxoplasmosis is a rare cause of granulomatous villitis.⁷⁹

Figure 33.20 Toxoplasmosis manifest on low power as villitis.

Figure 33.21 Toxoplasma cysts in amniotic membrane (arrow).

Figure 33.22 Toxoplasma (immunoperoxidase stain).

Malaria

Pregnant women from endemic areas (especially primiparous women, who have more severe disease and significantly higher prevalence rates of malarial infection) are more prone to develop malaria than non-pregnant women. Infection in pregnancy is associated with extensive parasitic infection of the placenta, maternal anemia, a reduction in birth weight, miscarriage, preterm birth, and neonatal and maternal death.⁸⁰

A mononuclear inflammatory infiltrate (intervillositis), malarial pigment, and an increase in perivillous fibrin is characteristic (Figure 33.23). In active infection, the parasitized cells attach to villi in areas of syncytiotrophoblast damage. Fibrinoid necrosis, basal membrane thickening, and increased numbers of syncytial knots are features also associated with malarial infection. Chronic infections show the most severe changes whereas placentas with acute infections exhibit a mild increase in inflammatory cells only and those with past infections show only minimal differences compared with noninfected placentas. A histologic grading system assessing the presence of pigment and inflammation in Plasmodium falciparum infection shows a correlation with birth weight.⁸¹ HIV co-infection will increase both viral and parasitic loads and decrease birth weight.