Coronavirus disease 2019 (COVID-19) has been shown to have effects outside of the respiratory system. Placental pathology in the setting of maternal severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection remains a topic of great interest because earlier studies have shown mixed results.
To ascertain whether maternal SARS-CoV-2 infection is associated with any specific placental histopathology, and to evaluate the virus's propensity for direct placental involvement.
Placentas from 65 women with polymerase chain reaction–proven SARS-CoV-2 infection underwent histologic evaluation using Amsterdam consensus group criteria and terminology. Another 85 placentas from women without SARS-CoV-2 constituted the negative control group. A total of 64 of the placentas from the SARS-CoV-2–positive group underwent immunohistochemical staining for SARS-CoV-2 nucleocapsid protein.
Pathologic findings were divided into maternal vascular malperfusion, fetal vascular malperfusion, chronic inflammatory lesions, amniotic fluid infection sequence, increased perivillous fibrin, intervillous thrombi, increased subchorionic fibrin, meconium-laden macrophages (M-LMs) within fetal membranes, and chorangiosis. There was no statistically significant difference in prevalence of any specific placental histopathology between the SARS-CoV-2–positive and SARS-CoV-2–negative groups. There was no immunohistochemical evidence of SARS-CoV-2 virus in any of the 64 placentas that underwent staining for viral nucleocapsid protein.
Our study results and a literature review suggest that there is no characteristic histopathology in most placentas from women with SARS-CoV-2 infection. Likewise, direct placental involvement by SARS-CoV-2 is a rare event.
Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is the causative agent of coronavirus disease 2019 (COVID-19).1 Significant infection often manifests as acute respiratory distress syndrome, reflecting the virus's predominant tropism for the respiratory tract.2–4 However, the effects of SARS-CoV-2 infection are not limited to the lungs.4–8 SARS-CoV-2, not unlike its viral predecessors, has occasionally been implicated in poor pregnancy outcomes.9–18 As such, there have been ongoing efforts to study placentas from patients with COVID-19 in order to better understand as well as to predict the effects of SARS-CoV-2 on pregnant women and neonates.9,10,18
Placental infection by SARS-CoV-2 is seemingly rare but has been demonstrated by multiple methods, including polymerase chain reaction testing, ribonucleic acid in situ hybridization, and immunohistochemical (IHC) staining for viral nucleocapsid and spike proteins.17,19–21 Most reported cases of placental SARS-CoV-2 infection have been shown to be associated with intervillous histiocytic infiltrates with or without marked perivillous fibrin deposition.17,20–22 However, at least 1 study showed staining for SARS-CoV-2 protein and RNA in a placenta without these aforementioned histopathologic findings.23 Thus, the histologic manifestation of COVID-19 within the placenta is still a topic of great interest and ongoing research.
Here, we present our findings from the histologic and IHC evaluation of 65 placentas from women with polymerase chain reaction (PCR)–proven SARS-CoV-2 infection. Additionally, we account for and examine the relationship between the clinical characteristics as well as symptomatology of the patients and the histopathologic findings in the placenta.
MATERIALS AND METHODS
Case Selection and Stratification
This study was approved by the Maimonides Medical Center (Brooklyn, NY) Institutional Review Board. Placentas sent for pathologic evaluation and accessioned between March 25 and May 4, 2020, were identified in the Sunquest CoPathPlus anatomic pathology laboratory information system. A total of 65 placentas were identified from women who tested positive for SARS-CoV-2 using rapid, real-time PCR, constituting the positive cohort. Eighty-five placentas accessioned in the same time period from women who tested negative for SARS-CoV-2 made up the negative cohort. Placentas from the SARS-CoV-2–negative cohort were sent for pathologic evaluation based on either maternal, fetal, or placental indications or per obstetrician and/or patient request, as specified in the hospital guidelines. As per the guidelines, indications for placental examination included, but were not limited to, maternal and antenatal conditions, such as diabetes, hypertensive disorders, and maternal fever/suspected infection; fetal and neonatal conditions, such as thick meconium, congenital anomalies, and low Apgar scores; and any gross abnormality of the placenta. At our institution, maternal SARS-CoV-2 positivity via molecular methods was adopted as an indication for placental examination because the effects of the virus on the fetus and placenta were largely unknown at the time. Furthermore, there was growing evidence of increased risk of preterm delivery and need for respiratory support in symptomatic pregnant women with COVID-19.24 During the time period between March 25 and April 10, only those pregnant women with symptoms and/or history of exposure to SARS-CoV-2 were tested for the virus via nasal swab. Starting from April 10, all pregnant women admitted to labor and delivery were universally tested regardless of symptoms or exposure.24 Review of the electronic health record system was performed to determine presenting symptomatology and clinical characteristics of each patient's SARS-CoV-2 infection. In an effort to evaluate for confounding factors, we also reviewed patient history for comorbidities, such as hypertensive and diabetic disease, fetal growth restriction, and intrahepatic cholestasis of pregnancy. Patients in the SARS-CoV-2–positive group were classified as either having asymptomatic/mild/moderate disease or severe/critical disease based on the published National Institutes of Health (NIH) guidelines.25 This dichotomization was performed in order to study the effect of clinical severity of COVID-19 on placental pathology. Finally, results of SARS-CoV-2 PCR testing performed on newborns within the first the day of life would provide evidence of possible vertical transmission of the virus.
Specimen Processing and Examination
All placentas were fixed in 10% neutral buffered formalin for a minimum of 48 hours. Placental discs were weighed after removing fetal membranes and umbilical cord. Placental disc weights were considered small and large for gestational age if they weighed less than the 10th percentile or greater than the 90th percentile, respectively, using a reference table published by the American Registry of Pathology.26 Specimens were sectioned to include 2 membrane rolls, cross sections of umbilical cord, and full-thickness placental disc sections. In addition, any grossly visible lesions were submitted for histologic evaluation. Tissue blocks consisted of standard formalin-fixed, paraffin-embedded tissue. Hematoxylin-eosin–stained sections were generated and placed on glass slides for diagnostic review. All 150 cases received separate microscopic evaluation by 2 fellowship-trained perinatal pathologists (D.L. and K.L.) blinded to the patient's SARS-CoV-2 status. Any discordance in diagnosis led to case re-review by both pathologists with adoption of a consensus diagnosis. All gross and histologic evaluation followed Amsterdam consensus group criteria and terminology.27
Immunohistochemical staining was performed on 64 of 65 cases from the SARS-CoV-2–positive cohort because 1 tissue block was not in file. One section per case was chosen for singletons, and 2 blocks were selected for each of the 3 twin placentas in order to achieve representative sampling of each placental disc. Sections with histopathologic findings were chosen for staining. Staining preference was given to the following placental pathology: fetal vascular malperfusion, maternal vascular malperfusion, and chronic inflammatory lesions. Formalin-fixed, paraffin-embedded tissue blocks were generated at Maimonides Medical Center and sent to HistoWiz Inc (Brooklyn, New York) (histowiz.com) for IHC staining for SARS-CoV-2 protein. Tissue was put through a heat-induced epitope retrieval process, followed by a proprietary automatic staining assay developed for the BOND RX automated stainer (Leica Biosystems Division of Leica Microsystems Inc, Buffalo Grove, Illinois). Staining used a rabbit polyclonal primary antibody specific to the SARS-CoV-2 nucleocapsid, 1:250 dilution (NB100-56576, Novus Biologicals LLC, Centennial, Colorado). The staining process included the use of the BOND Polymer Refine Detection Kit (Leica Biosystems) for primary antibody conjugate and chromogen in accordance with the manufacturer's protocols. After staining, sections were dehydrated and film coverslipped using a Tissue-Tek Film automated Coverslipper (Sakura Finitek USA Inc, Torrance, California). Whole slide scanning (×40 objective) was performed on an Aperio AT2 digital whole slide scanner (Leica Biosystems). The positive control consisted of HistoWiz Inc's stock PCR-confirmed SARS-CoV-2–positive placental tissue, which was fixed in 10% neutral buffered formalin for a minimum of 48 hours and processed in a manner virtually identical to that of the placentas that were evaluated at our institution. In addition to IHC positivity for SARS-CoV-2 nucleoprotein, the control tissue also showed in situ hybridization evidence of SARS-CoV-2 RNA. Two negative controls were used and consisted of placental tissue from mothers who tested negative for SARS-CoV-2 by way of PCR testing on nasopharyngeal swab samples.
Descriptive analyses were performed on both groups. Maternal age and gestational age were categorized into advanced maternal age and term gestation. All categoric variables were summarized by N and percentage. Fisher exact test was performed to compare clinical characteristics and placental findings between the positive and negative groups. A planned subgroup analysis of placental findings was performed among the positive group between preeclampsia and nonpreeclampsia. All statistical analyses were performed on SAS 9.4 (SAS Institute Inc, Cary, North Carolina).
The clinicopathologic characteristics for the SARS-CoV-2–positive and SARS-CoV-2–negative cohorts are listed in Table 1. Maternal age ranged from 19 to 45 years (mean age, 30) for the SARS-CoV-2–positive group (n = 65) and 19 to 44 years (mean age, 30) for the SARS-CoV-2–negative group (n = 85). Most women in both groups were younger than 35 years (50 [77%] of the SARS-CoV-2–positive women and 67 [79%] of the SARS-CoV-2–negative women; P = .84). There were fewer (although not statistically significant) cesarean deliveries in the SARS-CoV-2–positive group than the SARS-CoV-2–negative group (20 [30%] versus 36 [42%], respectively; P = .13). There were fewer placentas from twin gestations in the SARS-CoV-2–positive group than the SARS-CoV-2negative group (3 [5%] versus 16 [19%]; P = .04). Most gestations reached the third trimester, with only 2 (3%) and 5 (6%) of the deliveries occurring in the second trimester in the SARS-CoV-2–positive and SARS-CoV-2–negative groups, respectively (P = .47). Intrauterine fetal demise occurred in both second-trimester cases in the SARS-CoV-2–positive group and in 2 of 5 second-trimester cases in the SARS-CoV-2–negative group. However, the SARS-CoV-2–positive group had a greater percentage of pregnancies carried to term than the SARS-CoV-2–negative group (53 [82%] versus 58 [68%], respectively; P = .10). Of the 85 placentas from SARS-CoV-2–negative women, 4 (5%) were small for gestational age. None of the placentas from the SARS-CoV-2–positive group were small for gestational age. Similarly, there were slightly fewer large-for-gestational age placentas in the SARS-CoV-2–positive group than there were in the SARS-CoV-2–negative group (18 [28%] versus 29 [34%], respectively; P = .85). The prevalence of preeclampsia/gestational hypertension (PEC/gHTN) was similar in both groups (9 [14%] in the SARS-CoV-2–positive group and 13 [15%] in the SARS-CoV-2–negative group; P > .99). Diabetes mellitus was less prevalent in the SARS-CoV-2–positive group (2 [3%] versus 11 [13%]; P = .04). Intrahepatic cholestasis of pregnancy was exclusively seen in the SARS-CoV-2–positive group at 5 (8%); P = .01. The prevalence of obesity did not differ between the 2 groups (SARS-CoV-2–positive group: 34 [52%] versus SARS-CoV-2–negative group: 46 [54%]; P = .73). Of the 65 SARS-CoV-2–positive patients, 59 (91%) had asymptomatic/mild/moderate disease, and 6 (9%) had severe/critical disease. All patients with severe/critical disease delivered preterm. The mode of delivery for the severe/critical cases was most frequently by cesarean delivery (5 of 6). Contrastingly, cesarean delivery was performed in a minority of women with asymptomatic/mild/moderate SARS-CoV-2 infection (15 of 59).
Pathologic findings were divided into maternal vascular malperfusion (MVM), fetal vascular malperfusion (FVM), chronic inflammatory lesions, signs of amniotic fluid infection sequence, and miscellaneous pathology. Miscellaneous pathology was further divided into increased perivillous fibrin (IPVF), intervillous thrombi (IVT), increased subchorionic fibrin, M-LMs within fetal membranes, and chorangiosis.
Prevalence of MVM can be found in Table 2. Representative photomicrographs can be found in Figure 1. Signs of MVM were seen in just a little more than a third of cases in both the SARS-CoV-2–positive and the SARS-CoV-2–negative groups of women (23 [35%] versus 31 [36%]; P > .99). The 2 most common MVM-related findings in the SARS-CoV-2–positive group were increased syncytial knots (8 [12%]) and villous infarction (VI) (8 [12%]). Villous infarction predominated in the SARS-CoV-2–negative group (14 [16%]). Distal villous hypoplasia, increased syncytial knots, villous agglutination (VA), and VI, were present at a higher percentage in the SARS-CoV-2–negative group. Decidual arteriopathy, which included cases of mural hypertrophy as well as fibrinoid necrosis of decidual arteries, was equally present in both groups (6 [9%] in the SARS-CoV-2–positive group and 8 [9%] in the SARS-CoV-2–negative group; P > .99). Prevalence of retroplacental hemorrhage was nearly equal in both groups as well (2 [3%] in the SARS-CoV-2–positive group versus 3 [3.5%] in the SARS-CoV-2–negative group; P > .99). As such, there was not one histopathologic finding of MVM that had a higher prevalence in the SARS-CoV-2–positive group. Although villous agglutination showed a statistically significant difference in prevalence between the 2 groups, that statistical significance disappeared when adjusting for the confounding variable of PEC/gHTN. Moreover, the overall prevalence of MVM was nearly identical for both the SARS-CoV-2–positive and SARS-CoV-2–negative groups (23 [35%] in the SARS-CoV-2–positive group versus 31 [36%] in the SARS-CoV-2–negative group; P > .99).
Prevalence of FVM can be found in Table 3. Representative photomicrographs can be found in Figure 2, A through C. Of the 65 placentas from SARS-CoV-2–positive women, 6 (9%) showed FVM, whereas 16 placentas (19%) from the SARS-CoV-2–negative group exhibited FVM (P = .11). Villous stromal-vascular karyorrhexis (VS-VK) was the most prevalent FVM-related pathology in the SARS-CoV-2–positive group (3 [5%]), whereas intramural fibrin deposition (IFD) was the most common lesion in the SARS-CoV-2–negative group (11 [13%]). All 4 evaluated FVM-related histopathologic findings, including VS-VK, avascular villi (AV), IFD, and stem vessel obliteration, were present at a higher percentage in the SARS-CoV-2–negative group. In the SARS-CoV-2–positive group, the only placenta with findings of large vessel obliteration was a second-trimester intrauterine fetal demise case in which the pathology was interpreted as postmortem vascular change instead of FVM. Similarly, 1 case of VS-VK in the SARS-CoV-2–negative group was also interpreted as a postmortem phenomenon. Risk factors for FVM were identified in only 3 cases—all belonging to the SARS-CoV-2–negative group. The risk factors included marginal and velamentous insertion of the umbilical cord. Like with MVM, there were not any histopathologic findings of FVM that showed a greater prevalence in the SARS-CoV-2–positive group. Only IFD showed a statistically significant difference between the 2 groups (SARS-CoV-2–positive group: 2 [3%] versus SARS-CoV-2–negative group: 11 [13%]; P = .04. However, overall, there was no statistically significant difference in FVM between the SARS-CoV-2–positive and the SARS-CoV-2–negative groups (P = .11).
Acute chorioamnionitis involving extraplacental membranes and the chorionic plate, consistent with maternal response to a likely amniotic fluid infection, was present in 32 cases (38%) from the SARS-CoV-2–negative group of women and 16 cases (25%) from the SARS-CoV-2–positive group (P = .11). Acute inflammatory infiltrates originating in the chorionic plate and umbilical cord vessels, consistent with fetal response to the offending pathogen, were found in 9 cases (11%) from the SARS-CoV-2–negative group and 5 cases (8%) from the SARS-CoV-2–positive group (P = .59). However, neither finding showed a statistically significant difference in prevalence between the SARS-CoV-2–positive and SARS-CoV-2–negative groups.
Increased perivillous fibrin was seen in 17 cases (20%) from the SARS-CoV-2–negative and 10 cases (15%) from the SARS-CoV-2–positive group (P = .53). In both groups, perivillous fibrin failed to reach the minimum degree necessary for a diagnosis of maternal floor infarction or massive perivillous fibrin deposition. Intervillous thrombi were almost equally present in both groups (15 [23%] in the SARS-CoV-2–positive group and 19 [22%] in the SARS-CoV-2–negative group; P > .99). Increased subchorionic fibrin was present in 11% of cases in both groups (SARS-CoV-2–positive group: n = 7 versus SARS-CoV-2–negative group: n = 9; P > .99). There were M-LMs present in 25 of the 85 placentas (29%) belonging to women who tested negative for SARS-CoV-2 and 11 of the 65 placentas (17%) from those who tested positive for the virus (P = .09). Chorangiosis, which implies chronic underperfusion, was found in 9 placentas (11%) from the SARS-CoV-2–negative group versus 5 placentas (8%) from the SARS-CoV-2–positive group (P = .59). There was no statistically significant difference in prevalence of these pathologic findings between the SARS-CoV-2–positive and SARS-CoV-2–negative groups.
Evaluation of all placentas for chronic inflammatory lesions revealed no difference between the 2 groups. Representative photomicrographs can be found in Figure 2, D through F. Both groups showed a similar prevalence of chronic inflammatory lesions (23 [35%] in the SARS-CoV-2–positive group versus 29 [34%] in the SARS-CoV-2–negative group; P > .99). However, low- and high-grade chronic villitis (CV-LG and CV-HG, respectively)—better known as villitis of unknown etiology—showed a trend toward a higher prevalence in the SARS-CoV-2–negative group. There was CV-LG present in 5 cases (8%) in the SARS-CoV-2–positive group versus 13 cases (15%) in the SARS-CoV-2–negative group (P = .21). There was CV-HG present in 1 case (2%) in the SARS-CoV-2–positive group versus 5 cases (6%) in the SARS-CoV-2-negative group (P = .24). Conversely, basal chronic villitis (BCV; 13 [20%] in the SARS-CoV-2–positive group versus 13 [15%] in the SARS-CoV-2–negative group; P = .52) along with chronic deciduitis (CD) (18 [28%] in the SARS-CoV-2–positive group versus 17 [20%] in the SARS-CoV-2–negative group; P = .33) were more common in placentas from the SARS-CoV-2–positive group. However, these differences in prevalence did not reach statistical significance.
Collectively, the 6 cases with severe/critical COVID-19 showed MVM, BCV, CD, acute chorioamnionitis with fetal response, IPVF, and an IVT (Table 4). One case did not show any pathologic findings. IVT, BCV/CD, and findings of amniotic fluid infection were each seen in 1 of 6 cases. IPVF was seen in 2 of 6 cases. MVM was present in 3 of 6 cases. None of the other evaluated histopathologic findings were present in the severe/critical subgroup of SARS-CoV-2–infected patients.
Results of IHC Staining for SARS-CoV-2
A total of 64 placentas from the SARS-CoV-2–positive cohort underwent IHC staining for SARS-CoV-2 nucleocapsid protein. Histopathologic findings among tissue sections chosen for staining included distal villous hypoplasia, IVT, CV-LG, CV-HG, BCV, CD, IPVF, M-LM, increased syncytial knots, retroplacental hemorrhage, VI, VS-VK, AV, IFD, increased subchorionic fibrin, and those with signs of amniotic fluid infection sequence, including acute chorioamnionitis with and without a fetal response. Despite incorporating sections with varying and unique pathology, all placental sections from SARS-CoV-2–positive women were negative for evidence of the virus via IHC staining as represented in Figure 3, A. Photomicrograph representation of the positive control can be found in Figure 3, B. Digitally scanned images (hematoxylin-eosin and IHC) can be accessed at the following hyperlink (and clicking the “guest access” button): https://app.histowiz.com/shared_orders/be0d8a50-2263-4997-b491-f6dc4ef7e718/slides/ (accessed December 27, 2020).
In our study—one of the largest of its kind to date—we compared the pathology of placentas from women with and without PCR-proven SARS-CoV-2 infection. Placentas from both groups were evaluated for 21 distinct histopathologic findings separated into 5 major groups: MVM, chronic inflammatory lesions, FVM, amniotic fluid infection sequence, and miscellaneous pathology. All pathologic findings were either equally present among the SARS-CoV-2–positive and SARS-CoV-2–negative groups or showed a greater prevalence in the SARS-CoV-2–negative group. Notwithstanding, there was no statistically significant difference in the prevalence of pathologic findings between the SARS-CoV-2–positive and SARS-CoV-2–negative groups.
Like other investigators, we accounted for confounding variables by excluding certain risk factors and etiologies for the various histopathologic findings in our study. However, for example, although Patberg et al28 included only cases without certain obstetric or medical complications (ie, cases complicated by fetal growth restriction, oligohydramnios, chronic hypertension, pregnancy-induced hypertension, preexisting or gestational diabetes, coagulopathy, or thrombophilia) in their control group, our approach was different. Attempting to exclude placentas delivered from women who experienced obstetric and medical complications prior to case selection would have significantly decreased the sample size of our SARS-CoV-2–positive group. Additionally, had we only excluded cases with confounding variables from our control group, there would have been significant overrepresentation of obstetric and medical complications in the SARS-CoV-2–positive group. Thus, we opted to adjust for confounding variables upon statistical analysis. Incidentally, aside from SARS-CoV-2 status, both groups were remarkably similar from a clinical standpoint, buffering against the effects of confounding variables. Nonetheless, we excluded cases from women with PEC/gHTN in our final statistical analysis of prevalence of MVM in both groups. In our evaluation of FVM, we made note of cases with risk factors (eg, velamentous cord insertion) for and mimics (eg, postmortem pathology) of the histologic findings often seen in aberrant fetal circulation. However, there were too few of these cases for there to be any meaningful effect on the statistical significance of the prevalence of FVM between both groups. Aside from the aforementioned obstetric and medical risk factors, it is our understanding that there were no other clinical conditions, in our study, that would have been significant confounding variables in the examination of the role of maternal SARS-CoV-2 infection on placental pathology.
Of all of the examined placental pathology, the only major group of pathologic findings that showed a potential etiologic link with maternal SARS-CoV-2 infection was that of chronic inflammatory lesions. Although Patberg et al28 showed villitis of unknown etiology to be more common in placentas from women with COVID-19, our study failed to replicate this association. Contrastingly, despite the lack of statistical significance, BCV and CD were observed more frequently in our SARS-CoV-2–positive cohort. According to the literature, BCV and CD have an association with intrauterine growth restriction and preterm labor.29 Hence, the possible link between BCV/CD and maternal SARS-CoV-2 infection may need to be examined using larger studies to establish whether a true relationship exists.
Our results differ from earlier research, which showed a link between maternal SARS-CoV-2 infection and specific placental histopathology. Results from several studies performed at the peak of the pandemic suggested a possible SARS-CoV-2–driven malperfusion within the maternal circulation of the placenta. Placental examination performed by Shanes et al30 revealed MVM in 12 of 15 placentas from mothers who tested positive for SARS-CoV-2. Histologic signs of MVM included central and peripheral VI, villous agglutination, AVM, and decidual arteriopathy (including mural hypertrophy of membrane arterioles as well as atherosis and fibrinoid necrosis of maternal vessels). There were IVTs also found in 6 of 15 placentas from women with SARS-CoV-2 infection, representing a statistically significant increase in prevalence compared with the historical control groups. Similarly, Smithgall et al31 looked at 51 placentas from women with SARS-CoV-2 infection and found an increased prevalence of MVM in this group. Histopathologic findings that reached statistical significance in the SARS-CoV-2–positive group included villous agglutination and subchorionic thrombi. In our study, we examined placentas from 65 SARS-CoV-2–positive women and 85 SARS-CoV-2–negative women and did not find a statistically significant difference in the prevalence of MVM between the 2 groups even after accounting for PEC/gHTN. Likewise, IVTs and increased subchorionic fibrin were equally present in both of our cohorts. Although 3 of 6 of the placentas (50%) from the severe/critical COVID-19 subgroup of our SARS-CoV-2-positive cohort revealed MVM, the sample size was too small to allow for any meaningful statistical analysis. Thus, overall, our findings do not support maternal SARS-CoV-2 infection as an etiology of or a risk factor for MVM.
Similarly, earlier studies suggested an association between SARS-CoV-2 and thrombosis within the placenta. Mulvey et al32 looked at 5 placentas from women with SARS-CoV-2 infection and found signs of FVM in every single one. Subsequently, Baergen and Heller33 performed histologic evaluation of 20 third-trimester placentas from women with COVID-19 and found signs of FVM in nearly half (9 of 20) of the cases. Specific findings included IFD, focal VS-VK, AV, and thrombosis. Risk factors for FVM were only identified in 1 case.33 Similarly, Prabhu et al34 found FVM in 14 of 29 placentas (48%) from women with COVID-19 versus only 11% of placentas in SARS-CoV-2–negative women. Most recently, Patberg et al28 showed FVM to be significantly more common in cases of COVID-19, with 32.5% of examined placentas exhibiting the histopathologic finding. Overall, these studies suggested a link between COVID-19 and thrombotic events within large and small fetal vessels of the placenta. Contrary to these studies, we did not find an association between FVM and maternal SARS-CoV-2 infection.
Our findings are consistent with Hecht et al,23 Gulersen et al,35 and He et al.36 The first of the aforementioned studies found no characteristic histopathologic findings in 19 placentas from women with proven SARS-CoV-2 infection, despite RNA evidence of the virus in syncytiotrophoblast in 2 of the cases.23 Gulersen et al35 examined 50 placentas from women diagnosed with SARS-CoV-2 in the third trimester and found no difference in the prevalence of various placental pathology between women with and without SARS-CoV-2 infection. Lastly, He et al36 looked at 21 placentas from SARS-CoV-2–positive women and found no specific SARS-CoV-2–related pathology.36
Although ancillary testing for SARS-CoV-2 has been undertaken by other investigators, our study represents the largest series of placentas to undergo immunohistochemical evaluation for the presence of SARS-CoV-2 within placentas from SARS-CoV-2–positive women. Using antibodies directed against the SARS-CoV-2 nucleocapsid protein, we stained 64 of 65 placentas and found no IHC evidence of the virus in any of the sections. Our findings are similar to Smithgall et al,31 who performed IHC staining for SARS-CoV-2 spike protein on 46 placentas from SARS-CoV-2–positive women and reported a negative result in every case. However, smaller series and rare case reports have described viral protein and nucleic acid localization in syncytiotrophoblast cells of chorionic villi.17,21,23 Nonetheless, the absence of IHC staining in our series and that of Smithgall et al suggests that placental infection by SARS-CoV-2 is a rare event.31
Lack of direct placental infection by SARS-CoV-2 in our series may explain why none of the neonates from our SARS-CoV-2–positive cohort showed clinical evidence of vertical transmission. All newborns tested negative for SARS-CoV-2 by PCR method within 1 day of life. Likewise, most of the published literature failed to show evidence of vertical transmission of SARS-CoV-2. Rare examples of plausible intrauterine transmission suggest that, like direct placental infection, vertical transmission of SARS-CoV-2 is a rare phenomenon.17
The main limitation of our study is the relatively small size of the SARS-CoV-2–positive cohort when it comes to establishing an association between SARS-CoV-2 and chronic inflammatory lesions at the basal plate. Similarly, the significance of pathologic findings among severe/critical cases is limited by the small number of such cases, allowing for only a descriptive analysis.
Our study suggests that there is no characteristic histopathology in most placentas from women with SARS-CoV-2 infection. Lesions characteristic of MVM and FVM were no more likely to be present in placentas from women with COVID-19 than they were in those without SARS-CoV-2 infection. However, further studies with a larger sample size may be needed to establish whether chronic inflammatory lesions—particularly BCV and CD—are more likely to be present in placentas from women with SARS-CoV-2 infection. Additionally, the absence of immunohistochemical evidence of SARS-CoV-2 nucleocapsid in 64 placentas from SARS-CoV-2 swab-positive women suggests that direct placental involvement by the virus is a rare event. Nonetheless, rare case reports of placental SARS-CoV-2 infection indicate a need for collaborative multicenter studies aimed at elucidating the histopathology and clinical implications of direct placental involvement by SARS-CoV-2.
Cheng (founder, CEO) and Mann are affiliated with and represent HistoWiz Inc. The other authors have no relevant financial interest in the products or companies described in this article.