Stillbirths with placental hemorrhagic endovasculitis: a morphologic assessment with clinical implications

Hemorrhagic endovasculitis (HEV) is a vasodisruptive alteration affecting fetal-placental blood vessels of all calibers. Hemorrhagic endovasculitis is found in association with stillbirth and abnormalities of growth and development in livebirths. The role of HEV in the pathogenesis of these conditions is not known.

To further understand these events, we compare clinicopathologic features of HEV-affected placentas from stillbirths with those from livebirth pregnancies. Additionally, we assess the relationship of morphologic forms of HEV to clinical events and time of fetal death in utero and evaluate the significance of extensive versus localized HEV lesions in placentas of stillbirths.

We reviewed the clinical records and slides from 119 stillbirths with placentas affected by HEV classified above a specified severity level (cases) and 119 matched stillbirths with placentas not affected by HEV (controls). A subset of 21 stillbirth placentas exhibiting focal HEV lesions was similarly evaluated. Slides were graded for HEV, villitis of unknown etiology, chorionic thrombi, villous fibrosis, erythroblastosis, and lesions indicative of maternal hypertension. Hemorrhagic endovasculitis was subcategorized into active, bland, and healed forms and clustered capillary lesions (hemorrhagic villitis). Focal, segmental, and diffuse patterns of villous fibrosis were delineated. Interlesional relationships were established by matching HEV severity indices with severity indices of co-existing lesions. Timing of fetal death was determined by published criteria. Data were analyzed for significance using chi2 and t tests. Results were compared with published analyses of livebirths with placental HEV.

Lesions occurring with significant frequency in HEV-affected (case) placentas include villitis of unknown etiology, chorionic thrombi, villous fibrosis, erythroblastosis, and meconium staining. Interlesional relationships were evident between HEV and villous fibrosis, villitis of unknown etiology, and chorionic thrombi. Growth restriction was more common in case versus control infants (P = .02). A segmental pattern of villous fibrosis predominated in cases versus controls and within the case group (P < .001). Time to delivery after fetal death was longer in cases than controls. Active-vasodestructive forms of HEV correlate with shorter intervals of intrauterine retention, whereas bland forms correlate with longer intervals (P = .04). Placentas with focal HEV were associated with coexisting chorionic thrombi and villous fibrosis but not with fetal growth restriction.

Patterns of interlesional interplay are similar in HEV-affected placentas of livebirths and stillbirths. This suggests that the pathogenesis of infant morbidity and mortality is similar in both groups. Active-vasodestructive forms of HEV may precede whereas bland forms may follow intrauterine demise. The segmental pattern of villous fibrosis and high incidences of growth restriction, erythroblastosis, and meconium in cases suggests a chronicity of adverse intrauterine events that may precede fetal loss. Stillbirths with focal HEV lesions are probably not at risk.