Perinatal death is an increasingly important problem as the coronavirus disease 2019 (COVID-19) pandemic continues, but the mechanism of death has been unclear.
To evaluate the role of the placenta in causing stillbirth and neonatal death following maternal infection with COVID-19 and confirmed placental positivity for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2).
Case-based retrospective clinicopathologic analysis by a multinational group of 44 perinatal specialists from 12 countries of placental and autopsy pathology findings from 64 stillborns and 4 neonatal deaths having placentas testing positive for SARS-CoV-2 following delivery to mothers with COVID-19.
Of the 3 findings constituting SARS-CoV-2 placentitis, all 68 placentas had increased fibrin deposition and villous trophoblast necrosis and 66 had chronic histiocytic intervillositis. Sixty-three placentas had massive perivillous fibrin deposition. Severe destructive placental disease from SARS-CoV-2 placentitis averaged 77.7% tissue involvement. Other findings included multiple intervillous thrombi (37%; 25 of 68) and chronic villitis (32%; 22 of 68). The majority (19; 63%) of the 30 autopsies revealed no significant fetal abnormalities except for intrauterine hypoxia and asphyxia. Among all 68 cases, SARS-CoV-2 was detected from a body specimen in 16 of 28 cases tested, most frequently from nasopharyngeal swabs. Four autopsied stillborns had SARS-CoV-2 identified in internal organs.
The pathology abnormalities composing SARS-CoV-2 placentitis cause widespread and severe placental destruction resulting in placental malperfusion and insufficiency. In these cases, intrauterine and perinatal death likely results directly from placental insufficiency and fetal hypoxic-ischemic injury. There was no evidence that SARS-CoV-2 involvement of the fetus had a role in causing these deaths.
The emergence of new viral diseases has always created anxiety among persons at risk for infection, but perhaps this is most true for pregnant women, who fear not only for themselves but also for their unborn children. An important aspect of the current coronavirus disease 2019 (COVID-19) pandemic caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is its effect on pregnant women, the fetus, and the newborn. Previous experiences with the pathogenic coronaviruses severe acute respiratory syndrome coronavirus (SARS-CoV or SARS-CoV-1) and Middle East respiratory syndrome coronavirus (MERS-CoV), as well as other RNA respiratory viruses, had indicated that transplacental infections were either absent or rare.1,2 Studies performed at the beginning phase of the current pandemic found that although pregnant women in China could develop infection with the newly identified coronavirus, the large majority of infected mothers had either mild or nonexistent symptoms and did not become more ill than did nonpregnant women of the same age, and that, except for a reported increase in premature delivery, there was little or no excess perinatal mortality.3–6 As the virus spread throughout the world, the genome of SARS-CoV-2 developed mutations resulting in new genetic strains, with the most worrisome labeled as variants of concern. These included the alpha (B.1.1.7), beta (B.1.351), gamma (P.1), and delta (B.1.617.2) strain variants.7,8 Eventually, COVID-19 was found to be associated with adverse pregnancy outcomes including severe maternal illness as well as neonatal complications.9,10 However, until recently, studies from multiple countries11–16 failed to demonstrate any statistically significant association between COVID-19 in pregnant women and the occurrence of stillbirth. With the increasing spread of these new viral strains during successive waves of infection, anecdotal experiences by pathologists and clinicians together with some published reports suggested that increasing numbers of pregnant women infected with SARS-CoV-2 were having stillbirths.17–20 This was supported in April 2021 when a cluster of 6 stillborn fetuses and 1 miscarriage occurred in mothers with COVID-19 from Ireland,17–21 and then in May 2021 when a population-based cohort study from England demonstrated an increased risk among pregnant women infected with SARS-CoV-2 for having a fetal death.22 The association of SARS-CoV-2 infection and stillbirth was confirmed on November 26, 2021, when the US Centers for Disease Control and Prevention reported a population-based study showing that pregnant women with COVID-19 had an increased risk for stillbirth compared with uninfected women, and that the strength of this association was highest during the period of the SARS-CoV-2 B.1.617.2 (delta) variant predominance.23
Stillbirth can occur as a result of maternal infection with several viruses, collectively termed TORCH (an acronym for Toxoplasma, other, rubella, cytomegalovirus, herpes) agents, which include a variety of infectious agents including several new members and Ebola and Zika viruses.24–26 In such cases, the mechanism leading to death typically results from transplacental passage of the virus following maternal viremia and placental involvement, culminating in fetal infection, intrauterine fetal demise, or neonatal death. Although it has now been established that SARS-CoV-2 can cause fetal deaths, the mechanism(s) remains largely unknown. To understand the cause(s) of fetal and neonatal demise following maternal infection from COVID-19, we analyzed 64 stillbirth and 4 neonatal death cases originating in 12 countries in which the placentas were proven to be infected with SARS-CoV-2.
MATERIALS AND METHODS
In this multinational case-based retrospective study the inclusion criteria were (1) women having a positive test result for SARS-CoV-2 during pregnancy using reverse transcriptase polymerase chain reaction (RT-PCR) prior to delivery; (2) an obstetric outcome of either stillbirth or early neonatal death; and (3) the placenta having been submitted for pathology examination and diagnosed with SARS-CoV-2 infection by PCR of placental tissues, direct visualization of fetal-derived placental cells using immunohistochemistry for SARS-CoV-2 antigens, RNA in situ hybridization for SARS-CoV-2 nucleic acid, fluorescence in situ hybridization (FISH), or a combination of these techniques.
For all 68 cases occurring from the 12 countries that comprised this study group, the perinatal pathologists, clinical specialists including obstetricians and pediatricians, and others involved with these patients were personally contacted by one of the authors (D.A.S.) for confirmation of the clinical, laboratory, and pathology findings. A unique and important aspect of this study was that the placentas were evaluated to determine the percentage of involvement by destructive tissue elements of SARS-CoV-2 placentitis as previously identified and defined; these consisted of chronic histiocytic intervillositis, increased perivillous fibrin deposition including massive perivillous fibrin deposition (MPFD), and villous trophoblast necrosis.27–29 Clinical data, laboratory testing, and pathologic data, including the results of autopsy (when performed), were collected on forms designed specifically for the study. All contributors approved of the clinical, laboratory, and diagnostic details of their cases as described in this report.
All data are listed in tabular format for stillbirth cases in Tables 1 through 6 and for neonatal deaths in Table 6. Basic maternal demographic data include age and gestational age at delivery. Significant maternal conditions not related to SARS-CoV-2 infection are noted and listed as table footnotes. To the best of our knowledge, all mothers in this cohort were unvaccinated. In the case of neonatal deaths, Apgar scores and the day of life during which death occurred are listed. The status of SARS-CoV-2 infection and results of laboratory testing for the coronavirus are listed for the mother, stillborn, or neonate where available.
Characteristics of Stillborn Fetuses and Placentas From Pregnant Women With SARS-CoV-2 Infection (Cases 1–12)

Characteristics of Stillborn Fetuses and Placentas From Pregnant Women With SARS-CoV-2 Infection (Cases 13–24)

Characteristics of Stillborn Fetuses and Placentas From Pregnant Women With SARS-CoV-2 Infection (Cases 25–36)

Characteristics of Stillborn Fetuses and Placentas From Pregnant Women With SARS-CoV-2 Infection (Cases 37–48)

Characteristics of Stillborn Fetuses and Placentas From Pregnant Women With SARS-CoV-2 Infection (Cases 49–61)

Characteristics of Stillborn Fetuses (Cases 62–64), Neonatal Deaths (Cases 65–68), and Placentas From Pregnant Women With SARS-CoV-2 Infection

Placentas were weighed and examined grossly, and multiple representative sections were taken on site. The major diagnoses were performed and recorded using routine hematoxylin and eosin–stained slides. The presence of SARS-CoV-2 was evaluated in the majority of placentas using immunohistochemistry for SARS-CoV-2 antigens. In a few cases, RNA in situ hybridization for viral messenger RNA or FISH evaluation for SARS-CoV-2 was performed. Evaluation of placentas was conducted in some cases using RT-PCR on tissues that were either fresh, flash frozen, or formalin fixed and paraffin embedded. All testing was conducted according to locally approved methods in the pathology department at the hospital site.
The extent of placental pathology involvement was estimated using a synthesis of findings based upon the gross inspection of the placenta that was confirmed thorough microscopic analysis of a minimum of 4 representative sections of placental parenchyma. The number of tissue blocks submitted exceeded the minimum recommended in the Amsterdam Placental Workshop Group Consensus Statement.30 The pathologists in this study reported the estimated percentage of placental involvement in 2 ways: either as a single percentage metric representing the combination of all destructive lesions, or as a metric that was specific for a given microscopic finding(s). Site pathologists estimated the placental tissue involvement as either a single figure or a range of percentages.
In those placentas that had previously had some aspect of the case published, the references were provided. Pathologists at all study sites adhered to the placental pathology diagnostic criteria recommended in the Amsterdam Placental Workshop Group Consensus Statement.30 Because the diagnostic criteria for MPFD have varied among investigators, in this study a minimum of 30% of placental fibrin deposition in the characteristic pattern was necessary to make the diagnosis.
In all cases there was either approval received from the local institutional review boards or institutional waiver and parental permission obtained, and there was compliance with the Declaration of Helsinki for Human Research.
RESULTS
Analysis of SARS-CoV-2 Placentitis Abnormalities
SARS-CoV-2 placentitis, as defined by the coexistent occurrence of 3 microscopic findings—chronic histiocytic intervillositis, increased fibrin deposition, and trophoblast necrosis—was identified in 65 of 68 placentas (97%) in this study (Tables 1 through 6). Two of the 3 cases that did not have all 3 constituents of SARS-CoV-2 placentitis diagnosed (cases 42 and 46) were preterm deliveries (20 5/7 and 29 weeks, respectively) lacking chronic histiocytic intervillositis but having MPFD and trophoblast necrosis. The third case, case 60, did not have MPFD, but had massive recent infarcts and decidual vessel thrombi present together with trophoblast necrosis and chronic histiocytic intervillositis.
Increased fibrin deposition was diagnosed in all 68 placentas (100%) (Figures 1 through 3). Among 68 placentas with increased fibrin, MPFD was diagnosed in 63 cases (93%), not being diagnosed in cases 19, 20, 22, 31, and 60. In the 63 placentas having MPFD, it occurred together with trophoblast necrosis in all 63 cases (100%) and with chronic histiocytic intervillositis in 61 (98%) (Figures 4, A and B, and 5).
Serially sectioned placenta from case 62 showing appearance of SARS-CoV-2 placentitis. Microscopic examination showed massive perivillous fibrin deposition, chronic histiocytic intervillositis, and trophoblast necrosis. The extent of pathology resulting from these destructive lesions was greater than 90% and led to placental insufficiency and stillbirth.
Figure 2. Gross pathology appearance of massive perivillous fibrin deposition that occurred with SARS-CoV-2 placentitis from a stillborn fetus. Intervillous thrombohematomas can be seen.
Figure 3. Sectioned placental specimen from case 61 illustrating SARS-CoV-2 placentitis. There was 70% involvement of placental tissue with this destructive process.
Serially sectioned placenta from case 62 showing appearance of SARS-CoV-2 placentitis. Microscopic examination showed massive perivillous fibrin deposition, chronic histiocytic intervillositis, and trophoblast necrosis. The extent of pathology resulting from these destructive lesions was greater than 90% and led to placental insufficiency and stillbirth.
Figure 2. Gross pathology appearance of massive perivillous fibrin deposition that occurred with SARS-CoV-2 placentitis from a stillborn fetus. Intervillous thrombohematomas can be seen.
Figure 3. Sectioned placental specimen from case 61 illustrating SARS-CoV-2 placentitis. There was 70% involvement of placental tissue with this destructive process.
A and B, Placenta from a stillborn fetus demonstrating the features of SARS-CoV-2 placentitis including massive perivillous fibrin deposition, chronic histiocytic intervillositis, and syncytiotrophoblast necrosis (hematoxylin-eosin, original magnifications ×4 [A] and ×10 [B]).
Figure 5. An area of intervillositis in a placenta from a stillborn fetus (case 64). This placenta also had massive perivillous fibrin deposition and necrosis of the syncytiotrophoblast (hematoxylin-eosin, original magnification ×20).
Figure 6. Placenta from a stillbirth (case 9) demonstrating positive staining for SARS-CoV-2 in the syncytiotrophoblast using RNA in situ hybridization (original magnification ×20).
A and B, Placenta from a stillborn fetus demonstrating the features of SARS-CoV-2 placentitis including massive perivillous fibrin deposition, chronic histiocytic intervillositis, and syncytiotrophoblast necrosis (hematoxylin-eosin, original magnifications ×4 [A] and ×10 [B]).
Figure 5. An area of intervillositis in a placenta from a stillborn fetus (case 64). This placenta also had massive perivillous fibrin deposition and necrosis of the syncytiotrophoblast (hematoxylin-eosin, original magnification ×20).
Figure 6. Placenta from a stillbirth (case 9) demonstrating positive staining for SARS-CoV-2 in the syncytiotrophoblast using RNA in situ hybridization (original magnification ×20).
Chronic histiocytic intervillositis was present in 66 of 68 placentas (97%). It was not diagnosed in case 42, in which no other inflammatory process was present, and in case 46, which had 50% of placental involvement with villitis. Among the 66 placentas with chronic histiocytic intervillositis, 62 (94%) had concurrent MPFD.
Villous trophoblast necrosis was present in all 68 placentas (100%) from stillbirths and neonatal deaths.
Additional Placental Findings
Except for the findings that constitute SARS-CoV-2 placentitis, the most frequent pathology finding present in this cohort was intervillous thrombi or hemorrhages, present in 25 placentas (37%). Villitis was the next most frequent abnormality, occurring in 22 of 68 placentas (32%). These were followed by findings of maternal vascular malperfusion in 12 placentas (18%), antemortem fetal vascular malperfusion in 7 (10%), and acute chorioamnionitis in 9 (13%). Less common findings included placental infarcts, umbilical vessel thrombi, chorangiosis, and chronic chorioamnionitis.
There were 23 placentas that measured below the 10th percentile of weight stratified for gestational age.
Percentage Placental Involvement by SARS-CoV-2 Placentitis
In each placenta the contributing pathologist(s) carefully estimated the percentage of placental tissue involvement of representative sections for the destructive components of SARS-CoV-2 placentitis in correlation with the gross features of the placenta. These included intervillous fibrin deposition, chronic histiocytic intervillositis, and trophoblast necrosis. In some placentas a percentage range of placental involvement was provided, and in these cases the mean of the range of placental involvement was used in calculating the average placental involvement for the entire data set. Some cases estimated the percentage of placental involvement as greater than a specific number (for example >80%), and in these cases the stated percentage metric (for example 80%) was used.
Among the 68 placentas, the mean extent of tissue involvement by SARS-CoV-2 placentitis was 77.7%. Both the median and mode values for the extent of placental involvement were 80%, with a range between 35% and 100%. The interquartile range was 15%, with outliers of 35%, 37.5%, and 40%.
Identification of SARS-CoV-2 Involvement and Distribution in the Placenta
Among the 68 placentas from 64 stillborn fetuses and 4 neonatal deaths in this study, there were differing laboratory methods used to identify SARS-CoV-2 involvement of the placenta (Tables 1 through 6). All 68 placentas had at least 1 testing modality positive for SARS-CoV-2. The most frequent method used was immunohistochemical staining with antibody to SARS-CoV-2 antigen, which was performed in 53 of 68 placentas (78%), either alone or along with another type of testing. It was performed as the only test to detect SARS-CoV-2 in 38 of 68 placentas (56%). Immunohistochemistry was used in combination with other tests in 15 of 68 placentas (22%): together with RNA in situ hybridization in 6 placentas, in combination with PCR in 6 placentas, with FISH and PCR in 1 case, and with RNA in situ hybridization and PCR in 2 cases. RNA in situ hybridization (Figure 6) was used as the only test to detect SARS-CoV-2 placental involvement in 5 of 68 placentas (7%). PCR testing of fresh, frozen, or fixed placental tissues was performed as the sole test to detect SARS-CoV-2 in 10 of 68 placentas (15%).
The most common placental cell to be involved with SARS-CoV-2 was the syncytiotrophoblast, which stained positive in all 58 placentas (100%) in which testing was performed that could localize the virus to specific cell types. In a minority of cases there were additional cell types identified to be positive for the virus; these included cytotrophoblast in 7 of 58 placentas (12%), Hofbauer cells in 3 of 58 placentas (5%), villous stromal cells (not otherwise specified) in 3 of 58 placentas (5%), maternal cells (macrophages) in the intervillous space in 3 of 58 placentas (5%), villous capillary endothelial cells in 2 of 58 placentas (3%), and extravillous trophoblast in 1 placenta (2%).
Timing of Fetal and Neonatal Demise
Among the 64 stillborn fetuses in this study, death occurred at a mean gestational age of 30 weeks, with a modal value of 30 weeks 1 day. Delivery of the 64 stillbirths ranged from 15 to 39.2 weeks gestation. Eight stillbirth cases (13%) were delivered at full term (>37 weeks gestation).
The 4 cases of neonatal death were all delivered preterm at a mean gestational age of 30.8 weeks and survived for an average of 3.5 days following delivery.
Autopsy Pathology Findings
Autopsy examination was performed on 30 of the 68 cases (44%)—29 stillborns and 1 neonatal demise. The majority of the autopsies (19 of 30; 63%) revealed no fetal significant abnormalities. The most frequent pathologic findings that were identified related to intrauterine hypoxia and asphyxia, present in 5 cases (cases 13, 14, 19, 22, and 59). These findings of hypoxia included petechial hemorrhages in fetal organs, persistence of nucleated fetal red blood cells, and acute organ hemorrhages. There were 2 cases of thymic involution (cases 1 and 5) and 1 case each with aspiration of intrauterine contents (case 1), microvesicular steatosis (case 4), thrombosis of umbilical vein and atrium (case 13), hand malformation (case 17), unilateral renal agenesis (case 19), mild lymphocytic interstitial pulmonary infiltrates (case 61), and atelectasis with multiple organ hemorrhages (case 62). There were no gross or microscopic abnormalities identified in the 30 autopsies that related to significant tissue inflammation or necrosis that could be attributed to viral infection.
Identification of SARS-CoV-2 in the Stillborn Fetus and Neonate
Among all 68 fetuses and neonates in this study, SARS-CoV-2 was detected from a body specimen in 16 of 28 cases tested (57%). These included 10 cases in which the virus was identified by PCR of nasopharyngeal swabs alone, 2 cases having positive PCR and immunohistochemistry from multiple visceral organs, and 1 case each having positive nasopharyngeal, gastric, and mouth swabs; positive throat swab; positive PCR in a nasopharyngeal swab and lung tissue; and a positive PCR from a lung swab.
Intrauterine SARS-CoV-2 Transmission in Stillborn Fetuses
The World Health Organization criteria for evaluating intrauterine SARS-CoV-2 transmission in stillborn fetuses were used.31 Intrauterine SARS-CoV-2 infection in the case of fetal demise requires both evidence of maternal SARS-CoV-2 infection anytime during pregnancy and detection of SARS-CoV-2 in fetal tissue, amniotic fluid, or placental specimens. In addition to positive maternal testing for SARS-CoV-2, the following criteria have been proposed to identify either confirmed, possible, or unlikely cases of maternal-fetal transmission. Confirmed maternal-fetal transmission requires fetal tissue from a sterile site to test positive for SARS-CoV-2 using either RT-PCR or in situ hybridization. Possible transmission can be evaluated using 2 sets of criteria. In those cases where the fetal tissue was not tested for SARS-CoV-2 via RT-PCR and in situ hybridization, there is possible transmission if one or more of the following tests are positive for SARS-CoV-2: (1) fetal tissue immunohistochemistry or microscopy or fetal swab RT-PCR; (2) amniotic fluid; and (3) placental tissue (RT-PCR, in situ hybridization, immunohistochemistry or microscopy) or placental swab RT-PCR. In cases where the fetal tissue was tested for SARS-CoV-2 using RT-PCR or in situ hybridization and was negative, possible transmission may have occurred if the amniotic fluid is positive for SARS-CoV-2. Unlikely transmission criteria include fetal tissue testing negative for SARS-CoV-2 by RT-PCR or in situ hybridization together with one or more of the following tests being positive for SARS-CoV-2: fetal tissue immunohistochemistry or microscopy or a fetal swab RT-PCR, or placental tissue (RT-PCR, in situ hybridization, immunohistochemistry or microscopy) or placental swab RT-PCR. These criteria are not optimal, as they do not address the significance of negative immunohistochemical staining of fetal organs for SARS-CoV-2 in the absence of additional tissue analysis using RNA in situ hybridization staining or PCR. Thus, for the purposes of this study, we consider that negative staining of fetal organs for SARS-CoV-2 using immunohistochemistry makes maternal-fetal transmission unlikely in the absence of molecular testing of these organs.
Applying the World Health Organization criteria and our caveats to these data and considering that all mothers and placentas were positive for SARS-CoV-2, the results of fetal organ testing were the determining covariable in assessing the likelihood of maternal-fetal transmission. Among the 64 stillbirths, maternal-fetal transmission of SARS-CoV-2 was confirmed in 2 cases (cases 47 and 61), possible in 49 cases, and unlikely in 13 cases. In the 4 cases of neonatal death, 3 cases had possible in utero transmission and in 1 case it was unlikely. There were no clinical or pathologic findings that viral infection of fetal tissues had any significant role in causing a fetal or neonatal death in this cohort.
DISCUSSION
Even prior to the COVID-19 pandemic, stillbirth was a persistent global public health problem. As a result of deficiencies and inconsistencies in the global surveillance and reporting of stillbirths, the number that occur annually is unknown, but it has been estimated to be between 2 and 6 million.32
Maternal infections with infectious agents, especially those of the TORCH group, can result in placental infection and transmission of the agent to the fetus that results in pathologic changes to organs causing stillbirth or neonatal death.33–36 A major concern at the start of the COVID-19 pandemic was the effect of the virus on pregnant women and their offspring.2,37–39 Placental pathology has been useful in the understanding of maternal-fetal infection and adverse obstetric outcomes with previous emerging infections, but early studies from mothers with SARS-CoV-2 infection were inconclusive, as the majority of placentas came from newborns and placentas that tested negative for SARS-CoV-2 infection.37–41 In examining a series of placentas that were found to be positive for SARS-CoV-2 using immunohistochemistry or RNA in situ hybridization, Schwartz and Morotti27 found that placentas infected with the virus had a significantly different pattern of pathologic findings than did uninfected placentas, regardless of the infection status of the neonate. Additional studies40–48 found that placentas testing positively for SARS-CoV-2 were typically characterized by a spectrum of destructive findings that included villous trophoblast necrosis, chronic histiocytic intervillositis, and increased fibrin up to the level of MPFD. A study of 11 stillborn and live-born babies having placental involvement with SARS-CoV-2 confirmed that the microscopic findings present in these cases were risk factors for intrauterine viral transmission and perinatal morbidity and mortality.29 When occurring in a placenta delivered from a mother with COVID-19, the triad of findings of histiocytic intervillositis, perivillous fibrin deposition, and trophoblast necrosis has been termed SARS-CoV-2 placentitis.28
Placental abnormalities are the leading identifiable cause of stillbirth.49–51 As a result, pathology examination of the placenta is a critically important tool for the determination of the cause of perinatal mortality.52–54 Placental disease can cause malperfusion that results in placental insufficiency and stillbirth.50,55,56 In this present study, we have documented a consistent pattern of abnormalities from 68 placentas having confirmed SARS-CoV-2 involvement that were associated with stillbirths and/or neonatal deaths. The major pathology lesions that were present—fibrin deposition, trophoblast necrosis, and chronic histiocytic intervillositis—are all destructive lesions that are associated with SARS-CoV-2 maternal infection.29,57–60 These placental abnormalities can, when occurring by themselves, have deleterious effects of placental function, and recent research suggests that they can occur independent of the severity of maternal infection.58
All 68 of the placentas in this cohort were demonstrated to be positive for SARS-CoV-2 using either molecular or immunohistochemical methods. In those placentas where the virus was localized using either immunohistochemistry or RNA in situ hybridization, the syncytiotrophoblast was involved in all cases. Previous studies have indicated that although the syncytiotrophoblast is the most common placental cell type to be involved with SARS-CoV-2,59 other villous cells, including cytotrophoblasts,61 Hofbauer cells,60 and endothelial cells,60 can also stain positively for the virus. In our series, cytotrophoblasts, Hofbauer cells, and villous stromal and endothelial cells were occasionally found to stain positively for SARS-CoV-2.
The most frequent abnormality in this cohort of placentas was abnormally increased fibrin deposition, occurring in 100% of cases including stillborn fetuses and neonatal deaths. Fibrin deposits occur in placentas under normal circumstances to a certain degree, and are found beneath the chorionic plate, in the intervillous space and adjacent to chorionic villi, and at the basal plate. In pathologic conditions, a spectrum of placental disorders characterized by an abnormal increase in fibrin can develop; these include increased fibrin deposition, fibrinoid plaque, infarcts, and the 2 most severe abnormalities, MPFD and maternal floor infarction.
MPFD is a highly unusual abnormality characterized by an excessive deposition of fibrin/fibrinoid material in the intervillous space. The fibrin/fibrinoid obstructs normal perfusion and gas-nutrient exchange and entraps the chorionic villi, resulting in villous ischemia and necrosis that causes placental insufficiency.62–64 Long before the COVID-19 pandemic, MPFD had been recognized as a cause of perinatal morbidity and mortality due to fetal hypoxic injury that included spontaneous abortion, intrauterine growth restriction, preterm delivery, stillbirth, neonatal death, neurologic disease in surviving infants, and significant recurrence risk.62–66 Cases of MPFD have been described in which autopsy pathology indicated that the cause of death was from placental insufficiency.63 The published diagnostic criteria for MPFD have been variable, ranging from a high of 50% involvement67 to a lower percentage of involvement of greater than 20% and 25%.68–70 In this present report, we used a criterion of fibrin deposition of 30% or greater in a characteristic pattern for MPFD. Using this criterion, MPFD was present in 63 of the 68 placentas (94%) in this study. In all 63 of these cases (100%), it coexisted with at least 1 other placental finding of SARS-CoV-2 placentitis. Trophoblast necrosis was universally present in placentas having MPFD. In 61 of the 63 placentas (98%) with MPFD, chronic histiocytic intervillositis was also present.
Chronic histiocytic intervillositis occurred in 97% of the placentas in this cohort, but prior to the COVID-19 pandemic it was rarely seen and had an unknown etiology since it was first described by Labarrere and Mullen71 in 1987. It was found to be associated with a high recurrence rate and adverse pregnancy outcomes that included miscarriage, intrauterine fetal demise, preterm birth, and intrauterine growth restriction.71–73 Its exact prevalence is unknown, but it was believed to occur in approximately 6 of 10 000 second- and third-trimester placentas (0.6%) prior to the COVID-19 pandemic.72,74 Chronic histiocytic intervillositis is characterized by the accumulation of mononuclear inflammatory cells (predominantly histiocytes) in the intervillous space of the placenta, and may be accompanied by lymphocytes and occasionally neutrophils.75 Chronic histiocytic intervillositis was noted to occur together with MPFD before the COVID-19 pandemic,75–77 where it resulted in either intrauterine fetal demise or a pregnancy termination. In cases of SARS-CoV-2 placentitis it may be misleading to retain the term chronic in describing this intervillositis, as the development of placental pathology appears not to be long in duration. In our study, all 66 placentas with chronic histiocytic intervillositis had increased fibrin deposition, and 94% had concurrent MPFD.
Foremost among other pathology abnormalities identified were intervillous thrombi, occurring in 37% of placentas. Intervillous thrombi are not typically associated with adverse birth outcomes unless they are large or multiple; however, in placentas that are already compromised because of the destructive effects of SARS-CoV-2 placentitis, they likely exacerbate the malperfusion. Roberts and colleagues (written communication, December 2021) recently found parenchymal thrombohematomas (intervillous thrombi or hemorrhages) to be associated with SARS-CoV-2 placentitis and stillbirth. Among our cohort of 68 placentas, villitis occurred in 22 (32%). In all cases but 1, villitis was present together with chronic histiocytic intervillositis, and it remains to be determined exactly what the relationship is between these 2 inflammatory conditions.
In understanding the combined effects of the abnormalities that constitute SARS-CoV-2 placentitis in producing placental insufficiency, it is important to remember that studies conducted prior to the COVID-19 pandemic demonstrated a direct relationship between the number of placental abnormalities in any given placenta and the development of perinatal morbidity and mortality, arguing for a synergistic effect among multiple lesions.78,79 This phenomenon is well illustrated in SARS-CoV-2 placentitis, which, unlike placental infection from other TORCH agents, constitutes a simultaneous grouping of destructive placental lesions occurring in the same pregnancy. After examination of the microscopic effects of SARS-CoV-2 placentitis on the placental tissues, it is apparent that these lesions can result in obstruction of maternal and fetal blood flow through the placenta, as well as causing irreversible damage and necrosis of placental tissues and reduction of the functional capacity of the tertiary villous capillary bed, leading to significant malperfusion and placental insufficiency.
Perhaps the most important finding in this study relates to the degree of involvement of the placentas from the destructive lesions that constitute SARS-CoV-2 placentitis. The average placenta in this cohort had 77.7% involvement with SARS-CoV-2 placentitis. This extent of placental damage and consequent malperfusion is striking, and far exceeds the degree of placental involvement and destruction that is typically seen with other viral TORCH agents. At these high levels of placental damage, the placenta cannot function at the level necessary to provide sufficient oxygen and nutrients to the fetus to sustain life. In examining the results of this study, and in consideration of not only the destructive nature of the individual placental abnormalities of SARS-CoV-2 placentitis but also the occurrence of additional placental pathology findings including intervillous thrombi, villitis, and maternal vascular malperfusion, it can be reasonably concluded that placental insufficiency was occurring together with fetal hypoxia, which produced a hypoxic-ischemic fetal or early neonatal demise. Among these 68 cases of stillbirth and neonatal death, there were no other significant potential etiologies identified for perinatal demise from either a clinical or pathologic perspective.
The extent of placental damage and the nature of the pathology findings in these cases leads to questions regarding the timing of these processes and their terminology. Both increased fibrin and MPFD have not previously been considered to represent acute pathology processes and were believed to develop long before labor and delivery, based upon several factors including morphology, extent and severity of the disease process, and association with intrauterine growth restriction.62,80 The occurrence of chronic histiocytic intervillositis was also consistent with a pathologic process of some duration. However, when it occurs with COVID-19 there are data that indicate a more accelerated process, as nearly all reported infections (based on onset of symptoms or date of positive COVID-19 test) occur within approximately 2 weeks or less of the diagnosis or delivery of the stillbirth.28,58,80–82 We believe that our pathology data are strongly suggestive of a process that is occurring during a period ranging from several days up to 2 weeks after onset of maternal symptoms or positive COVID-19 testing. Because of this, it may be appropriate to use the term histiocytic intervillositis in place of chronic histiocytic intervillositis in these cases. In addition, we recommend that pregnant women with an acute SARS-CoV-2 infection be closely monitored for those first 2 to 3 weeks for fetal well-being to hopefully avoid intrauterine fetal demise.
The findings in the present study have additional important clinical ramifications. Placental insufficiency was the apparent cause of fetal and neonatal demise among these 68 cases. Although there are no standard criteria or agreed-upon consensus for the diagnosis of placental insufficiency,83 it is generally agreed that it represents a pathologic process where there is ongoing and continual deterioration in placental functioning, resulting in decreasing transfer of maternal-derived oxygen and nutrients to the fetus through the placenta, resulting in intrauterine fetal hypoxia, hypoxemia, and acidosis.83–86 In contrast to many other TORCH agents, our cases did not demonstrate evidence that the SARS-CoV-2 virus was causing mortality by inducing fetal somatic organ damage following placental infection and transplacental transmission. Instead, the tissue damage appeared to be confined to the placenta, where it was extensive and highly destructive. Given the nature and extent of the placental injury and the technologic improvements in noninvasive obstetric diagnostic methods, it may be possible that obstetric ultrasound could be used for screening in those mother-fetus dyads at risk. Doppler ultrasound including superb microvascular imaging has been demonstrated to be a useful method for evaluating both fetal and placental circulations, and magnetic resonance imaging of the placenta using advanced methods such as T2-weighted rapid acquisition with relaxation enhancement imaging has been used to detect placental vascular abnormalities, including hemorrhages and infarctions.87–90 An additional clinical consideration arises with the improvement in methods for vaccination and specific antiviral treatments. As our study indicates that the major cause of perinatal deaths among fetuses and neonates having placentas compromised by SARS-CoV-2 is placental insufficiency, and not direct viral infection of the fetal organs, reducing maternal SARS-CoV-2 viral burden through either immunization or antiviral therapy could conceivably decrease the risk of developing SARS-CoV-2 placentitis.
This study has several limitations, most of which were inherent in conducting a large retrospective clinical and pathologic investigation involving multiple geographically dispersed study sites and investigators. Protocols used for the clinical evaluation of mothers with COVID-19 were not uniform, although all clinicians in this study were experienced in the care and management of pregnant women having COVID-19. The nature of this study precluded providing detailed maternal clinical histories, but when significant maternal disease was present that was not related to COVID-19 it is listed as a table footnote, and no mothers had severe disease requiring intensive care or mechanical ventilation. There was also expected site-to-site variation in some laboratory methods, sampling of the placentas, and performance of immunohistochemical and molecular diagnostic methods at the different study locations in 12 countries. However, all testing was performed in accredited laboratories and in accordance with approved practices. Interobserver pathology diagnosis was minimized because all pathologists involved in this study either were experienced perinatal, pediatric, or placental pathologists or had a special interest in this field, and all adhered to diagnostic criteria from the Amsterdam Placental Workshop Group Consensus Statement.30 This system is used globally and has become the standard basis for clinical and research activities in the field. Because of the large sample size of placentas and autopsies, an exhaustive listing of the minor pathology findings could not be provided, and only the relevant diagnoses are listed.
Our data from these 68 cases support previous case reports suggesting that placental insufficiency is responsible for perinatal deaths occurring with SARS-CoV-2 placentitis.58,91–97 In summary, we found that SARS-CoV-2 placentitis can cause extensive placental damage as a result of destructive lesions, and that the damage can be further exacerbated by additional pathology abnormalities. Increased fibrin and MPFD, chronic histiocytic intervillositis, and trophoblast necrosis result in sizable destruction of the villous capillary bed accompanied by obstruction of the intervillous space, causing placental malperfusion and insufficiency that are incompatible with intrauterine survival. The fetal hypoxia that ensues can lead to a hypoxic-ischemic fetal demise or neonatal death. It is very fortunate that this sequence of events develops in only a small percentage of pregnant women having COVID-19.
The authors also gratefully acknowledge David Gisselsson Nord, MD, PhD, and Rodrigo Mitev Munoz, MD, in Lund, Sweden, for analyzing the placenta specimens and sharing analysis data. We would also like to thank Dominic Bewley, BS, Oana Popescu, MD, and Jessica Lee, BA. Katie Giesen at the College of American Pathologists provided invaluable expertise and guidance in the final review and production of this manuscript for publication.
References
Author notes
Babál received support from Slovak Research and Development Agency grant PP-COVID-20-051. Colson received support from the Belgian Fund for Scientific Research (FNRS-F.R.S., grant number 40002773) and the Fetus for Life charity. The other authors have no relevant financial interest in the products or companies described in this article.