Materials and Methods
We extracted all autopsies at Zurich University Hospital of stillborn fetuses with a gestational age of 32–42 weeks born between January 1975 and December 1995. All those with a clear cause of death—major malformation, chromosomal abnormality, infection, placental abruption, etc—were excluded, together with those without placenta data. As a first step, all placentas selected for study were reassessed without knowledge of the clinical events other than gestational age. They were divided into two groups. The first comprised placentas with macroscopic parenchyma loss (more than one small infarct) and evidence of normal- or even advanced-for-dates chorionic villous development on microscopy. Most also showed additional obliteration of the intervillous space by microinfarcts or perivillous fibrin deposition. The second group included macroscopically normal placentas (one or no small subchorionic infarct) showing immature vascularization of the chorionic villi on microscopy. The hallmark of this maturation defect is the reduced number of syncytiocapillary membranes, which is only one per terminal villous (mean 0.9, range 0–3, when counting the ten smallest villi in ten different fields within three different histologic slides) compared with three in a normal placenta (mean 2.9 with range 1–6) (insets in Figure 1b, c). To quantify the functional consequences of parenchyma loss and maturation defect, stillborn fetal morphometric data were scored as normal or pathologic for gestational age (above the fifth and below the 95th percentile, and below the fifth or over the 95th percentile, respectively). After conducting exhaustive tracing enquiries where necessary, we contacted the mothers concerned (n = 94) and sent them a questionnaire by mail to determine their present state of health and the number of live and stillbirths before and after the index intrauterine death. The Mann–Whitney nonparametric test was used for statistical comparisons (P ≤ .05).
Unselected placentas from a single center (Department of Obstetrics, Charite, Virchow-Klinikum, Berlin) from January 1994 to December 1998 were studied systematically without knowledge of clinical events. The 17,415 placentas accounted for 91% of all births at the center (only multiple pregnancies were excluded). The pathology reports noting severe maturation defect, defined as the absence of vessel differentiation within terminal villi in at least 50% of all areas examined, were then matched to the register of live or stillbirths.
Of the 94 women contacted, 71 responded (75.5%), including 37 whose placenta met the maturation defect criteria. The past obstetric history plus questionnaire data revealed diabetes mellitus type I during and before the index stillbirth in three of these women, including one who had developed preeclampsia during the pregnancy. A fourth woman developed diabetes mellitus type 2. None of the remaining 33 women developed diabetes after a median of 14 years (5–20 years), and none of the diabetic individuals had had another stillbirth. However, the group as a whole had two further stillbirths, giving a recurrence risk of 5.4% (95% confidence interval [CI] 0.7%, 18%), ie, a tenfold increase over the 0.5–1.1% baseline.1 The mean number of live births per maturation defect group member was 1.9; three of 37 (8.1%) remained without a live child.
Five of the 34 women in the parenchyma loss group had preeclampsia during the index pregnancy, but none of the five had a chronic disease 7–15 years later. A sixth woman had chronic hypertension diagnosed during the index pregnancy and she remains on therapy. A seventh woman had diabetes type I and hypertension diagnosed 13 years after the index stillbirth. Two further women had a second stillbirth, giving a recurrence risk of 5.8% (95% CI 0.7%, 19%) for the whole group, again approximately tenfold over baseline.1 The mean number of live births was 1.6; five of the 34 women remained childless (14.7%), significantly more than in the maturation defect group.
Of the 34 parenchyma loss stillbirths, 25 (73.5%) occurred in a first pregnancy compared with 16 of 37 (43.2%) in the maturation defect group (P < .05). Median gestational age of stillbirth also differed: 34.5 weeks (range 32–40 weeks) compared with 38 weeks (range 32–42 weeks), respectively (P < .001). Maternal age at stillbirth was similar: 29.5 years (range 20–42 years) and 31 years (range 23–43 years), respectively (P = .14).
Stillbirth growth in the maturation defect group was largely proportionate and appropriate for dates. Autopsy findings were often consistent with hypoxia, eg, superficial petechial hemorrhages on the lungs, heart, and thymus, and meningeal congestion. In contrast, in the parenchyma loss group, fetal weight distribution was strongly skewed to below normal, with most autopsies showing severely reduced internal organ weights, especially liver, thymus, and spleen; the lungs, heart, and kidneys were less affected, and brain weight was usually at the 50th percentile.
Of the 17,415 placentas, 993 (5.7%) showed maturation defect, including 2.3% associated with stillbirth. Normal placental maturation was associated with a stillbirth rate of 0.033%. Thus, although only a minority of fetuses with placental maturation defect die, the relative risk of death is 70 times that of fetuses with a normal placenta.
Intrauterine death may have a systemic fetal cause such as sepsis or hemolysis. Single organ failure is rarely responsible except for certain malformations of the heart and great vessels (eg, Ebstein's anomaly and intrauterine closure of the ductus arteriosus), although it may contribute to death (eg, bilateral renal agenesis). Some fetal organs, eg, the lungs, gut, and brain, are completely unnecessary to sustain intrauterine life. Intrauterine life in fact depends on three organs, only two of which are technically fetal: the heart, which maintains the circulation; the liver, which provides all the blood components, proteins, and, by virtue of physiologic extramedullary hematopoiesis, blood cells; and the placenta, which is responsible for all exchange functions, oxygen supply, and nutrition.
Growth arrest documented by serial ultrasound in an otherwise healthy, normally formed fetus with a normal karyotype is presumed to result from placental dysfunction; death or dystrophy may ensue if 30% or more of functioning parenchyma is lost by infarction. The remaining chorionic villi often compensate by increased vascularization, rendering the dysfunction chronic. In acute placental dysfunction, however, apart from the pallor caused by diminished vascular development, macroscopy is deceptively normal. As in adults, death from suffocation is much harder to diagnose than death from starvation. Importantly, few fetuses die from placental maturation defect, in fact only about one in 40 of the 5.7% detected in an unselected population do so. Most are rescued by birth, and to date by chance, rather than by any diagnostic or therapeutic procedure. This situation is not dissimilar from that of adults with an organ pathology. Thus hypertensive cardiomegaly and concomitant coronary artery disease may cause death at any moment, yet a fair number of patients will still be alive in a year. Why do so many fetuses survive maturation defect? Simply because a placenta with reduced oxygen transport capacity only has to maintain function for a few weeks, by which time most affected fetuses will have been rescued by birth.
The proportion of perinatal autopies that are inconclusive11 or of intrauterine deaths that are unexplained12 ranges from 24 to 31%. If these terms are reserved for cases with normal chorionic villi maturation, the proportion drops to 5–6%. Defects in chorionic villus angiogenesis and sinusoidal transformation13 have proved unconvincing explanations for intrauterine death. Our study shows that, compared with normal placental development, severe maturation defect increases the risk of fetal death 70-fold and the risk of recurrent stillbirth tenfold, thus it is a disease entity with serious functional credentials.
Lack of a clearly established etiology for the vascular maturation defect hampers acceptance of the concept of impending placental respiratory dysfunction threatening the fetus with suffocation. In our study, 11% of the women with acute placental dysfunction had diabetes mellitus in their stillbirth pregnancy. An association between diabetes and large maturation-defect placentas has long been known.14 Overt diabetes is preceded by a long period of immunomediated beta-cell destruction and the circulation of anti-islet cell antibodies. Gestational diabetes mellitus as a risk factor for subsequent diabetes can be predicted from the number of autoantibodies present during and after pregnancy.15 This fact led us to hypothesize that some of the maturation defect stillbirths would anticipate diabetes.16 However, none of the nondiabetic women in this group developed diabetes in the 5–20 years that followed the index event, nor was there any increase in the incidence of hypertension or rheumatic disease. The etiogenesis of placental maturation defect remains unknown.
Early diagnosis of placental dysfunction is a major clinical challenge. In parenchyma loss, slowing of the fetal growth curve and head/abdomen disproportion are diagnostic of chronic dysfunction. No such markers are available for maturation defect: placental dysfunction is manifested only as a terminal acute event. In theory, hypoxia as a cause of stillbirth has great potential for fetal salvage.9 Unfortunately, however, there seems no safe method for prenatal detection of children at risk. Fetal movement monitoring by trained mothers17–19 failed to show benefit in a large trial.18 Doppler flow velocity waveforms of the umbilical arteries may correlate with the size of the placental vascular tree20 but have not been able to predict maturation defect.21 Death from placental maturation defect usually occurs after 35 weeks. Early induction could be an option to prevent sibling catastrophe for mothers with a history of this underestimated pattern of placental dysfunction.
1. Fretts RC, Boyd ME, Usher RH, Usher HA. The changing pattern of fetal death, 1961–1988. Obstet Gynecol 1992;79:35–9.
2. Pitkin RM. Fetal death: Diagnosis and management. Am J Obstet Gynecol 1987;157:583–9.
3. Rayburn W, Sander C, Barr M Jr, Rygiel R. The stillborn fetus: Placental histologic examination in determining a cause. Obstet Gynecol 1985;65:637–41.
4. Gruenwald P. Fetal deprivation and placental pathology: Concepts and relationships. Perspect Pediatr Pathol 1975;2:101–49.
5. Schweikhart G, Kaufmann P, Beck T. Morphology of placental villi after premature delivery and its clinical relevance. Arch Gynecol 1986;239:101–14.
6. Benirschke K, Kaufmann P. Three-dimensional aspects of villous maldevelopment. In: Benirschke K, Kaufmann P, eds. Pathology of the human placenta. 2nd ed. New York: Springer-Verlag, 1990:114–29.
7. Kingdom J, Huppertz B, Seaward G, Kaufmann P. Development of the placental villous tree and its consequences for fetal growth. Eur J Obstet Gynecol Reprod Biol 2000;92:35–43.
8. Vogel M. Zottenreifungsstörungen. In: Vogel M. Atlas der morphologischen Plazentadiagnostik. 2nd ed. New York: Springer-Verlag, 1996:82–93.
9. Morrison I, Olson J. Weight-specific stillbirths and associated causes of death: An analysis of 765 stillbirths. Am J Obstet Gynecol 1985;152:975–80.
10. Stoltenberg C, Magnus P, Skrondal A, Lie RT. Consanguinity and recurrence risk of stillbirth and infant death. Am J Public Health 1999;89:517–23.
11. Saller DN Jr, Lesser KB, Harrel U, Rogers BB, Oyer CE. The clinical utility of the perinatal autopsy. JAMA 1995;273:663–5.
12. Incerpi MH, Miller DA, Samadi R, Settlage RH, Goodwin TM. Stillbirth evaluation: What tests are needed? Am J Obstet Gynecol 1998;178:1121–5.
13. Asmussen I. Vascular morphology in diabetic placentas. Contrib Gynecol Obstet 1982;9:76–85.
14. Hörmann G. Zur Systematik einer Pathologie der menschlichen Plazenta. Arch Gynäkol 1958;19:1297–344.
15. Füchtenbusch M. Prediction of type 1 diabetes postpartum in patients with gestational diabetes mellitus by combined islet cell autoantibody screening. Diabetes 1997;46:1459–67.
16. Emmrich P, Godel E, Dempe A. Morphology of the placenta in premanifest maternal diabetes mellitus. Z Geburtshilfe Perinatol 1978;182:79–85.
17. Neldam S. Fetal movements as an indicator of fetal wellbeing. Lancet 1980;1:1222–4.
18. Grant A, Elbourne D, Valentin L, Alexander S. Routine formal fetal movement counting and risk of antepartum late death in normally formed singletons. Lancet 1989;2:345–9.
19. Christensen FC, Rayburn WF. Fetal movement counts. Obstet Gynecol Clin North Am 1999;26:607–21.
20. Hitschold TP. Doppler flow velocity waveforms of the umbilical arteries correlate with intravillous blood volume. Am J Obstet Gynecol 1998;179:540–3.
21. Arabin B, Jimenez E, Vogel M, Weitzel HK. Relationship of utero-and fetoplacental blood flow velocity wave forms with pathomorphological placental findings. Fetal Diagn Ther 1992;7:173–9.