A structured review of placental morphology and histopathological lesions associated with SARS-CoV-2 infection


 Abstract
 
 The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), a novel coronavirus, was first identified after a cluster of cases in Wuhan, China in December 2019. Whether vertical transmission or placental pathology might occur following maternal infection during pregnancy remains unknown. This review aimed to summarise all studies that examined the placenta or neonates following infection with SARS-CoV-2, or closely related highly pathogenic coronavirus (SARS-CoV-1, or the Middle East respiratory syndrome coronavirus (MERS-CoV)). Structured literature searches found 50 studies that met the inclusion criteria.
 Twenty studies reported placental histopathology findings in third trimester placentas following maternal SARS-CoV-2 infection. Using the Amsterdam Consensus criteria to categorise the histopathology results, evidence of both fetal vascular malperfusion (35.3% of cases; 95% Confidence Interval (CI) 27.7–43.0%) and maternal vascular malperfusion (46% of cases; 95% CI 38.0–54.0%) were reported, along with evidence of inflammation in the placentas (villitis 8.7% cases, intervillositis 5.3% of cases, chorioamnionitis 6% of cases). The placental pathologies observed in SARS-CoV-2 were consistent with findings following maternal SARS-CoV-1 infection. Of those tested, a minority of neonates (2%) and placental samples tested positive for SARS-CoV-2 infection (21%).
 Limited conclusions can be drawn about the effect of maternal SARS-CoV-2 infection on placental pathology as most lack control groups and the majority of reports followed third trimester infection. Collaboration to maximise the number of samples examined will increase the reliability and generalisability of findings. A better understanding of the association between maternal SARS-CoV-2 infection and placental pathology will inform maternity care during the coronavirus pandemic.
 



Introduction 1
Although there are an increasing number of studies examining the clinical outcomes of women 24 diagnosed with COVID-19 during pregnancy, and their infants, the reasons underpinning any 25 association with an adverse outcome remain unclear. As the placenta is the maternal-fetal interface, 26 it not only has an essential role in protecting the fetus from infection but can also be affected by an 27 adverse maternal environment. Notably, severe COVID-19 infection results in sustained hypoxia [10], 28 an insult implicated in FGR, pre-eclampsia and stillbirth. Furthermore, vertical transmission of 29 infectious agents can occur either during pregnancy or childbirth. In the former case, infectious 30 agents may cross the placenta leading to fetal abnormalities (e.g. microcephaly in Zika virus infection 31 [11]); alternatively, infectious agents may directly cause placental damage (e.g. villitis from 32 cytomegalovirus [12]) leading to FGR or fetal demise. In either case, evidence of placental 33 transmission or damage is needed to establish a mechanistic link between maternal infection and 34 fetal outcome. 35 To identify evidence of either of these placental events, we conducted a structured review to 36 identify publications reporting pregnancy outcome in women with COVID-19 and which describe 37 placental findings, in order to evaluate evidence for vertical transmission and/or changes in 38 placental morphology in women with SARS-CoV-2 infection. 39 included four multiple-pregnancies [25]. There were six terminations of pregnancy including one at 111 22 weeks' gestation [18]. In another case the outcome of the pregnancy was unclear [17] Two 112 infants were described as small-for-gestational-age (SGA), but no definition was provided [31], three 113 infants had FGR or SGA, but no definitions were provided [24] and 11 birthweights were recorded as 114 <2500g, with no comment on whether the weights were appropriate for gestational age 115 [19,21,32,33]. The birthweight of 5 stillborn infants were provided: 1070g, 329g, 2895g, 680g, and 116 1430g at 28, 21, 38, 23 and 30 weeks' gestation, respectively [30]. Six studies reported perivillous fibrin deposition in 10 placentas [27,28,30,31,44,52]. Chen S et al. 143 reported various degrees of fibrin deposition inside and around the villi [52]. Increased "syncytial 144 nodules", more commonly described as syncytial nuclear aggregates (SNAs), were noted in two of 145 the placentas [52]. 146 Inflammatory changes in the placenta can be a result of maternal and fetal inflammatory response 147 often to infective microorganisms, such as chorioamnionitis [58]. Inflammatory changes were 148 reported in 10 studies. In one study, both cases born at term had CD68 + macrophages in the villi and 149 intervillous space and were diagnosed as having chronic intervillositis [23]. High grade chronic villitis 150 was detected in one of the placentas of a twin pregnancy, with an additional case of high grade 151 chronic villitis and associated avascular villi and two cases of chronic villitis were also reported [31]. 152 Villitis was reported in five placentas [29,43,44,57], whilst histiocytic villitis was found in two 153 placentas [55,57]. In one study, chorioamnionitis was reported in placentas from three stillborn 154 infants, with one also having the finding of villitis [30]. Additionally, the placentas of a 24 week twin 155 pregnancy where both infants were stillborn both had the findings of chronic intervillositis [34]. 156 of chorangiosis, one case with "acute inflammatory pathology" and one case of "chronic 159 inflammatory pathology" were reported [28]. The number of cases with "acute inflammatory 160 pathology" was significantly lower compared to both control groups (melanoma cohort p=0.001 and 161 historical cohort <0.001), whereas the number of cases with "chronic inflammatory pathology" did 162 not differ. A 38-week stillbirth placenta had evidence of acute chorioamnionitis, severe chronic 163 villitis, mild funisitis but no viral cytopathic changes [22]. However, the exact time of fetal demise 164 was unknown and there would have been a delay before tissue sampling could occur, which may 165 account for the findings. 166 Other histopathological findings have been reported in pregnancies with adverse outcomes, such as 167 umbilical cord coiling in the placentas of SGA infants and following stillbirth [

Placental histopathological findings following SARS-CoV-1 or MERS-CoV infections. 258
In total eight articles met the inclusion criteria. Characteristics of included studies are described in 259  This study is strengthened by the systematic nature of the search strategy run on multiple databases 294 and standardised data collection. To maximise information obtained, pre-prints were included; as 295 this information has not been peer-reviewed, the data may be less robust. This methodological 296 approach is hampered by publication bias, which may prevent publication of negative findings, thus 297 could result in overestimation of the frequency of placental pathology in women with COVID-19. As 298 the SARS-CoV-2 pandemic originated less than 6 months ago, it is likely that further studies of 299 placental tissue will be published and the evidence-base in this field will grow. One study identified lesions and reported evidence of viral mRNA by in situ hybridisation but, 314 critically, viral particles were only seen in the syncytiotrophoblast, rather than passing through the 315 basement membrane into the villous stroma where it would be high-risk for the virus to enter the 316 fetal circulation. We conclude from these studies that, while placental infection may occur rarely, 317 this is not apparent in the majority of cases. Notably, one of the cases where viral particles were 318 seen in the placenta, the mother had exceptionally high antibody titres [18]. In such cases, virions 319 might be phagocytosed by the placenta as immune complexes (IgG mediated) rather than by direct 320 invasion of the placental tissue. 321 Five studies reported testing neonates for the presence of IgM and IgG antibodies. IgG is the only 322 antibody known to cross the placenta [74]. IgM is the first antibody produced during a microbial    SARS-IgG detection kit, indirect immunofluorescence, ELISA on serum/plasma, Western blot on nucleocapsid "N" protein on maternal blood, cord blood, amniotic fluid, PCR on maternal blood, cord blood and amniotic fluid.
Western blot showed n protein in IgG in maternal blood, cord blood and amniotic fluid. IFA also positive.
No SARS-CoV genes detected in maternal blood, umbilical blood, or fluid using a SARS virus fluorescence qPCR.

Ng WF et al. 2006
SARS-CoV-1 7 No evidence of pregnancy complications, but two cases with oligohydramnios. Two maternal deaths. Methylprednisolone given to 4 patients. Four AGA. One pregnancy electively terminated. Two infants had suspected fetal growth restriction; both were small for gestational age. All term infants had Apgar score of 9 or 10 at 5 minutes. All the neonates were negative for SARS-CoV-1 after "investigations" Yes, tested positive 1 x same week as delivery, 1 x 1 week before, 1 x 5 weeks before, 1 x 7 weeks before, 1 x 35 weeks before delivery. Termination of pregnancy = 3 weeks before.