* Dr A. L. Bane, Department of Pathology & Laboratory Medicine, Mt Sinai Hospital, Toronto M5G 1X5 Canada.
Objective To establish the incidence, recurrence rate and consequences of massive perivillous fibrinoid.
Design Retrospective analysis of the histology of all placentas with a diagnosis of massive perivillous fibrinoid between 1991 and 1998, together with the maternal case records.
Setting The histopathology department of the Rotunda Hospital, Dublin, Ireland.
Population A relatively homogeneous group of pregnant women in the northern part of Dublin City, which is the catchment area for the Rotunda Hospital, delivered between 1991 and 1998.
Methods Retrospective review of archival placental pathology and maternal charts.
Main outcome measures The incidence of massive perivillous fibrinoid, perinatal outcome and recurrence rate.
Results The incidence of massive perivillous fibrinoid was 0.028%, with a recurrence rate of approximately 18%. All the infants suffered intrauterine growth restriction; there was a 31% fetal loss rate and a 33% preterm delivery rate.
Conclusions Massive perivillous fibrinoid is associated with intrauterine death, intrauterine growth restriction and preterm delivery. It has a significant recurrence rate and both the clinical findings of intrauterine growth restriction and the postmortem findings imply a syndrome of chronic placental insufficiency.
Massive perivillous fibrinoid, otherwise known as massive perivillous fibrin deposition or maternal floor infarction, is a condition characterised by enmeshment of chorionic villi in a fibrinoid material and in placental insufficiency of varying degrees. The consequences include a significant risk of intrauterine growth restriction, intrauterine death and preterm delivery. There is a high risk of recurrence.
This pathological condition is known as maternal floor infarction in North America, where it has been studied extensively, and as massive perivillous fibrinoid or massive perivillous fibrin deposition in the United Kingdom. These names are thought to represent the same condition, but there is no universal agreement on this point. The aim of the present study was to establish the incidence, recurrence rate and consequences of massive perivillous fibrinoid in a relatively homogeneous Irish population of pregnant women in the northern part of Dublin, which is the catchment area for the Rotunda Hospital, Dublin. The women were delivered between 1991 and 1998. A critical review of the literature was undertaken.
A retrospective review of all placentas with a diagnosis of massive perivillous fibrinoid was undertaken from the archival material of the Histopathology Department of the Rotunda Hospital for 1991–1998. Thirteen cases from 11 mothers were identified. All the haematoxylin and eosin slides were reviewed; a selection of slides stained with martius scarlet blue and periodic acid Schiff was studied in each case. Six cases were examined with a Hitachi 7100 transmission electron microscope (Hitachi, Japan). An estimate was made of the extent of involvement of the placenta by the condition as a percentage of the total placental surface area. The maternal case records, where available, were reviewed for additional information, with specific reference to maternal age at diagnosis, parity, gestational age and associated medical illnesses and smoking history. The incidence of fetal death, intrauterine growth restriction, preterm delivery and recurrence rate was then determined.
Thirteen cases of massive perivillous fibrinoid of the placenta were diagnosed in 11 women between 1991 and 1998 (Table 1). During the same period the total number of live births greater than 500 g was 47,022. The estimated incidence of massive perivillous fibrinoid is 0.28 per 1000 live births. Twelve of the 13 placentas were suspected macroscopically of being involved to a significant degree by perivillous fibrinoid and were described as having a diffuse variegated grey-white cut surface on the maternal side of the placenta (Fig. 1). The average placental area involved was 50% (range 20–90%).
Histological examination of the placentas confirmed the presence of eosinophilic fibrinoid material on the maternal side of the placenta and extending to engulf chorionic villi, resulting in atrophy and sclerosis of the affected villi (Fig. 2) without evidence of villitis or perivillitis. The fibrinoid material was focally positive for the routine laboratory fibrin stain martius scarlet blue, and the periodic acid Schiff stain served to highlight the sclerosis of the villi.
Electron microscopy showed that the ‘fibrinoid’ material had a longitudinal filamentous substructure (Fig. 3) identical to that described for true fibrin1. Focal trophoblast degeneration was additionally identified in one case (Fig. 4), dense lysosomal-type lamellar concentric bodies were identified in the cytoplasm of the trophoblast cell resembling, in part, the electron microscopic findings in a lysosomal storage disease.
Of the 13 births, four of the fetuses died in utero as a direct consequence of the placental pathology (31% fetal loss rate). Three of these four fetuses underwent a complete postmortem examination. All three fetuses showed definite histological evidence of hypoxia. All the remaining nine infants (69%) were live born (one trisomy 21) but their weights were all less than the 10th centile, consistent with a diagnosis of intrauterine growth restriction. Three of the nine live born infants (33%) were born before 37 weeks of gestation (two spontaneous labour and one caesarean section for poor fetal biophysical profile at 30 weeks of gestation). Two of the women experienced a recurrence of the condition in subsequent pregnancies. The mean maternal age at delivery was 30.9 years (range 23–42 years) and their mean parity was 1.9 (range 0–7).
Three mothers were smokers; one had a prior history of discoid lupus, and one a history of recurrent urinary tract infections and the third suffered from gestational diabetes and gestational hypertension. The other women had unremarkable past medical and antenatal histories.
The name massive perivillous fibrinoid or massive perivillous fibrin deposition attempts to describe the pathological condition which is characterised by heavy deposition of fibrin in the placenta, engulfing the chorionic villi. The alternative name of maternal floor infarction is actually misleading, for ischaemic necrosis is absent and the lesion is not a true infarct. While the names are often used interchangeably in the literature, some authors believe these to be entirely different lesions.
The incidence of this condition in our study was 0.028%, which contrasts with the high incidence of 0.5% recorded by Naeye2 from the Collaborative Perinatal Study, and 0.09% quoted by Andres et al.3. The difference in incidence may reflect a broader acceptance of what entails massive perivillous fibrinoid by Naeye2 as there are no accepted standardised criteria for the diagnosis of this condition. Alternatively, massive perivillous fibrinoid may have a variable expression in different ethnic groups, and this study simply highlights a true lower incidence of this condition among women of Celtic descent. Our incidence may be an under-estimate as placentas from normal pregnancies and deliveries are not routinely examined. The criteria for examination of placentas in the Rotunda Hospital include intrauterine death, intrauterine growth restriction, fetal distress, fetal malformations and maternal medical disease.
The consequences of this condition are both severe and recurrent. The perivillous fibrinoid deposition essentially strangles the chorionic villi. This in turn leads to a marked reduction of fetal blood flow in the affected villi and secondary stromal fibrosis. Thus, functional villi are reduced in number, and if the process is sufficiently large, placental function is compromised. The degree of placental compromise is variable, but can lead to intrauterine growth restriction, preterm delivery and, in extreme cases, fetal death. In our study, 4 out of the 13 pregnancies resulted in fetal death (31%; 95% CI 6–56%) which is consistent with the frequency recorded by Naeye2 (17%) and by Andres et al.3 (40%). All of our cases experienced intrauterine growth restriction; the frequency quoted by Naeye2 was 51% and by Andres et al.3 was 54%. The mean placental area involved by massive perivillous fibrinoid in the cases that resulted in intrauterine death was 52.5%, similar to the percentage involved in cases of intrauterine growth restriction without intrauterine death (44.22%). It is known that 20–30% of villi can be entrapped by fibrin without any detriment to overall placental function, and it is not clear why three of our cases suffered the consequences of reduced placental function with 30% or less of the placenta involved by massive perivillous fibrinoid. Perhaps the deposition of fibrin is an epiphenomenon and the true abnormality lies with the trophoblast. The degree of intrauterine growth restriction is variable, the most extreme example from our study being a 535 g female infant born at 30 weeks of gestation by emergency caesarean section following signs of fetal distress. There are several reports of recurrent fetal growth restriction in women with recurrent massive perivillous fibrinoid4, the most extreme example being a description of nine fetal losses5. Two of the 11 women experienced a recurrence of the condition in a subsequent pregnancy, a similar proportion to that reported by Andres et al.3.
The high risk of recurrence and the adverse consequences of massive perivillous fibrinoid have prompted obstetricians to try to identify this condition antenatally. Mandsager et al.6 reviewed the clinical course of 13 women retrospectively and three women prospectively and suggested that a number of ultrasonic and serum abnormalities taken together with a previous history could alert the obstetrician to the possibility of massive perivillous fibrinoid. These include fetal growth restriction, oligohydramnious, a hyperechoic appearance of the placenta on ultrasound and elevated maternal serum α-fetoprotein. The usefulness of these criteria has not been formally studied.
The aetiology of massive perivillous fibrinoid is unknown. The high risk of recurrence may suggest a genetic origin. However, in our study there were two twin pregnancies, one monochorionic and one dichorionic; in only one twin in each case was the placental area affected by massive perivillous fibrinoid. Because all monochorionic placentas are monozygotic, this precludes the simpler Mendelian genetics of autosomal dominant or recessive inheritance being of importance. The genetic features of this disorder have yet to be elucidated.
Bendon and Hommel7 reported two cases of massive perivillous fibrinoid in women with autoimmune disorders. One of the 11 women in our study suffered from discoid lupus; however, her serological status was unknown, and antibody tests were not performed in any of the other women.
Vernof et al.8 postulated abnormalities of the X-cell and pregnancy-associated major basic protein in the pathology of massive perivillous fibrinoid. The X-cell is a term used to describe the non-villous trophoblast, the increased proliferation of which has been noted in massive perivillous fibrinoid. Major basic protein is a cationic protein that is virtually identical to eosinophil granule major basic protein, which is known to be cytotoxic to parasites and numerous mammalian cells. The authors suggest that the X-cell either manufactures both pregnancy-associated major basic protein and fibrin or that major basic protein induces the formation of fibrin secondary to its known cytotoxic effects.
Naeye2 proposes that inadequate blood flow through decidual branches of the spiral arteries could cause focal necrosis of the decidua capsularis. This is found in 20% of cases, twice the usual frequency. The focal necrosis could become a nidus for fibrinoid deposition. Robb et al.9,10 located herpes antigen in most placentas affected by massive perivillous fibrinoid. They suggested that the disorder may be a manifestation of latent herpes simplex virus infection.
The number of putative aetiologies for this condition serve to illustrate its enigmatic nature. We have demonstrated by electron microscopy that the ‘fibrinoid’ material is in fact true fibrin; this, together with the foci of trophoblast degeneration also illustrated on electron microscopy, leads us to believe that the deposition of fibrin is a secondary response to trophoblastic cellular injury. The degeneration of the trophoblast would expose the basement membrane to the circulating intervillous blood and trigger fibrin deposition, by activation of the extrinsic limb of the coagulation cascade. The cause of this trophoblastic injury is not apparent but may be of ischaemic, infective or autoimmune in origin, or perhaps secondary to the cytotoxic effects of major basic protein. The lack of inflammation evident in placentas affected by massive perivillous fibrinoid also raises the possibility that the trophoblast may spontaneously degenerate, perhaps due to an inherent defect in the trophoblast cell itself. Indeed, what we recognise morphologically as the single entity of massive perivillous fibrinoid may have a number of aetiologies, the common denominator for which being trophoblast injury, or degeneration and fibrin deposition.
This case series shows that massive perivillous fibrinoid is a rare but dangerous condition with a significant risk of recurrence and the number of speculative aetiologies emphasises our poor understanding of its pathogenesis. Thus, we need to develop objective diagnostic criteria in order to understand better its pathophysiology.