Study of placental histopathology in COVID-19 mothers at a dedicated COVID-19 hospital in India

ABSTRACT

Introduction:

Maternal infections can affect the placenta, which acts as a fetomaternal barrier. This study aimed to determine the spectrum of morphologic alterations in the placentas of pregnancies complicated by SARS-CoV-2 infection and the impact on fetal or neonatal outcomes.

Materials and Methods:

This is a prospective case-control study. One hundred SARS-CoV-2 positive pregnant women and an equal number of SARS CoV-2 negative pregnant women in their third trimester who delivered at our tertiary care center between December 2020 and November 2021 were enrolled in the study. This study was conducted at the end of the first wave and during the second COVID-19 wave. Histopathological examination of the placentas was done using Amsterdam consensus criteria. We observed for evidence of maternal vascular malperfusion [MVM], fetal vascular malperfusion, and inflammation in the placenta.

Results:

The clinical findings were compared between the cases and controls. Evidence of MVM was seen in comparable numbers between the cases and controls, but as it involved less than 30% of the placental disc, it was considered an insignificant finding. Deciduitis was seen in equal proportions in both groups. The comparison between the asymptomatic and symptomatic groups failed to show any difference in placental pathology between both groups. There was no adverse fetal outcome seen in the pregnancies complicated by SARS-CoV-2 infection.

Conclusion:

Placental injury at the microscopic level was observed but was neither significant nor specific to the SARS-CoV-2 infection. SARS CoV-2 infection did not influence the placental pathology. Also, no adverse neonatal outcomes were observed.

Introduction

The quiescent yet still unfolding COVID-19 pandemic, a zoonotic disease caused by the novel coronavirus SARS-CoV-2, led to a great deal of morbidity and mortality worldwide, thereby challenging the existing healthcare infrastructure around the globe. India faced a second wave in February 2021 due to the emergence of the B.1617.2 delta variant and a third wave in December 2021 due to the B.1.1.529 Omicron variant. Mathews M et al.[1] on demographic comparison of the three waves observed in-hospital mortality rate to be 9.4%, 15.83%, and 10.47% in the first, second, and third waves respectively. They concluded that the second wave was more severe in terms of symptoms, complications, the requirement of ventilators, and higher mortality due to mutations in the virus, a lack of immunity, and vaccination. A recent surge in cases due to a subvariant of omicron BF.7 has been reported in Japan, Hong Kong, China, and Taiwan. India reports an average of 210 new cases per day.

Several studies conducted to determine the maternal and fetal outcome of SARS-CoV-2 infection observed an increased risk of maternal mortality, spontaneous abortions, preterm birth, pre-eclampsia, and stillbirth.[2,3,4,5,6] Pregnant women in their 3^rd trimester have the highest risk of critical illness, high chances of ICU admission, and may need mechanical ventilation. Maternal age >40 yr., obesity, and diabetes were associated with high risk in pregnant women. Vertical transmission has been demonstrated in the first-trimester placenta with fetal demise in an asymptomatic mother.[7] Unequivocal transplacental transmission is demonstrated by the presence of viral RNAs in the placenta, amniotic fluid in utero prior to the onset of labor, cord or neonatal blood, body fluids, or respiratory samples, or the demonstration of viral particles by electron microscopy, immunohistochemistry, or the in situ hybridization method in fetal or placental tissues.[8,9] Histopathological studies of the placenta have been done by several researchers with very varied results.[10,11,12,13] Therefore, we undertook this prospective case-control study to study the histopathologic spectrum of changes seen in the placentas of 19 COVID-positive mothers, to ascertain the significance of various changes seen by comparing the histologic parameters between the cases and the controls, and to correlate the placental changes with maternal and fetal outcomes.

Materials And Methods

A prospective case-control study was carried out at our institute, a dedicated COVID-19 hospital in Mumbai, India. Our study group consisted of 100 SARS-CoV-2 positive women who delivered at ≥34 gestational weeks (i.e third trimester) while the control group comprised an equal number of SARS-CoV-2 negative women who delivered at ≥34 gestational weeks under emergency situations at our hospital from December 2020 to November 2021. This study was conducted at the end of the first wave and in the second wave. Institutional ethics committee approval was obtained from ECARP ref no ECARP/2020/171 dated 25^th March 2021. All women who had obstetrical complications like diagnosed and/or suspected to have uteroplacental insufficiency, fetal growth restriction, oligohydramnios (AFI <5), hypertensive disorders of pregnancy, gestational diabetes mellitus, medical complications like diabetes mellitus, coagulopathy, thrombophilia or any other co-infections and stillbirths were excluded from the study. The SARS-CoV-2 test was done on all the pregnant women as per national testing. Maternal disease severity was classified based on defined criteria – mild cases were those who had symptoms of SARS CoV-2 infection but without dyspnoea or abnormal CT findings, moderate cases had evidence of lower respiratory tract infection on clinical examination and on imaging along with blood oxygen saturation of >93% on room air while severe cases had respiratory rate >30 breaths/min, blood oxygen saturation <93% on room air and lung infiltrates >50% of lung parenchyma or with respiratory failure, septic shock or multi-organ failure.

Data collection

Sociodemographic and clinical data like parity, gestational age, medical history, obstetrical history, COVID-19-related symptoms, laboratory investigations, radiologic investigations, details of pharmacologic management and oxygen requirement, time of infection, details of labor, postpartum complications were noted from the indoor files. Neonatal data consisting of neonatal weight, APGAR at 1 min and 5 min, need for NICU admission, and neonatal swab reports were collected. The data was entered into an excel sheet and stored in a secure computer with access only to the investigators. Data were periodically reviewed by the project investigators.

Processing of placenta

Placentas were received in a properly labeled leak-proof wide-mouthed container with an adequate amount of buffered 10% formalin. A biohazard safety label was stuck on the containers. The placentas were fixed in formalin for 36 to 48 hrs and processed later with utmost precautions. After separating the membranes and the umbilical cord, the placenta was weighed. Gross and microscopic examination was done according to the Amsterdam Placental Workshop Group 2014 classification. We studied 2 sections from the membranes, 1 from the umbilical cord, and 3 sections from the placental disc with fetal and the maternal plate. Additional sections were taken if any abnormal gross findings like grey-white areas, calcific areas, or focal hemorrhagic areas were seen. A routine Hematoxylin and Eosin [H and E] stain was performed on all the sections. Histopathology reporting of the placental sections was done by the subject histopathology expert [author no. 2] who was not blinded to the patient’s clinical details or the SARS-CoV-2 status.

The histologic parameters that were assessed included changes in maternal vascular malperfusion [MVM] like infarction, increased/massive perivillous fibrin deposition [PVFD], accelerated villous maturation, Tenney-Parker change, decidual vasculopathy, retroplacental hemorrhage [RPH] and intervillous thrombosis [IVT] were noted. Fetal vascular malformation [FVM] like thrombi in the fetal circulation, avascular villi, karyorrhexis, delayed villous maturation, and chorangiosis were looked for. We also observed for inflammatory changes in the placenta like chorioamnionitis, chronic villitis, chronic deciduitis, subchorionitis, choriovasculitis, fetal vasculitis, and other findings like marginal insertion of the umbilical cord, hypercoiling of the umbilical cord, phagocytosis of meconium and diffuse villous oedema. In the literature search we did not find any uniform criteria to quantify the inflammatory lesions hence we subjectively classified chorioamnionitis as mild, moderate, and severe. Also, for chronic deciduitis, we established the following criteria.

Mild deciduitis – Presence of isolated lymphocytes, macrophages, and plasma cells in the maternal plate. Moderate deciduitis – Presence of 3 or more clusters of lymphocytes, macrophages, and plasma cells in the maternal plate (Each cluster of 10 or more cells). Severe deciduitis – Diffuse infiltration of lymphocytes, macrophages, and plasma cells in the maternal plate. If placental infarction, retroplacental hemorrhage, calcification, and fibrinoid necrosis were more than 30% of placental disc it was considered a significant finding. Perivillous fibrin deposition was considered massive if it involved more than 50% of the placental disc.[14]

Statistical analysis

The study was an observational prospective study between the cases and an equal number of controls. The data was saved in an excel sheet and continuous variables were expressed in terms of mean, median, and interquartile range. The data were analyzed using a statistical package for the social sciences SPSS version 18.0. Chi-square test was used for statistical comparison. A P value of less than 0.05 was considered statistically significant.

Results

A total of 100 placentas delivered to SARS CoV-2 positive mothers and an equal number of gestational age-matched controls were studied. The median maternal age in cases was 28 years and 25 years in controls. The gestational ages at delivery in cases and controls were 38.75+/-1.47 weeks and 38.56+/-1.48 weeks, respectively. Dry cough was the most common symptom in the SARS CoV-2 positive cases, followed by fever and dyspnea. 80% of the cases were asymptomatic, 17% had mild disease, and 3% had severe disease. [Table 1] Oxygen therapy was given to three mothers with SARS-CoV-2 infection. The most common comorbidities observed in cases were hypothyroidism and tuberculosis. There were 12 percent of neonates in both groups who required NICU admission.

Table 1.

Demographic information and clinical characteristics of pregnant women with SARS-CoV-2 infection

Characteristics	Cases, n=100	Controls, n=100	P
Age in years, Median (IQR)	28 (24–31)	25 (23–29)	0.018 (S)@
Comorbidities
Hypothyroidism	8	2	0.049 (S)#
Tuberculosis	1	0	0.316 (NS)#
Other Comorbidity	4	1	0.408 (NS)#
Gestational age at delivery (weeks)	39 (38–40)	39 (38–39)	0.323 (NS)#
Primigravida	30	29	0.782 (NS)$
Multigravida	70	71	0.782 (NS)$
Maternal and Neonatal Outcome
Vaginal Delivery	61	68	0.338 (NS)$
Cesarean Section	40	32	0.368 (NS)$
Livebirth	99	100	0.316 (NS)#
5-minute APGAR Score	9 (8–9)	8 (8–9)	0.032 (S)@
NICU admission	12	12	0.893 (NS)$
Pregnancy complications
Preterm delivery	3	10	0.049 (S)#
PROM	15	21	0.045 (S)$
Preterm Labour	2	9	0.024 (S)#
Investigations
Haemoglobin (<12.5 g/dL)	91	98	0.042 (S)$
Hemoglobin levels (g/dl))	10.6 (9.7–11.9)	10.2 (9.8–11)
WBC (3500–9500/µL)	43	54	0.128 (NS)$
WBC count (/µL))	10100 (8000–13800)	9000 (7500–11100)
Platelet Count (125–350×103/µL)	89	79	0.054 (NS)$
Platelet count (/µL)	223 (171–287)	254 (220–321)	0.044 (S)$

@Kruskal Wallis test, #Fisher’s Exact Test, $Chi Square Test, (S)=Significant, (NS)=Not Significant. APGAR, Appearance, Pulse, Grimace, Activity, and Respiration; NICU – neonatal intensive care unit; PROM, Premature Rupture of membrane; WBC -White blood cell

Placental studies in both groups showed that the placental weight was comparable between both groups. Perivillous fibrin deposition, fibrinoid necrosis of the villi, placental infarction, retroplacental hemorrhage, intervillous thrombus, and calcification was seen in almost all the cases and controls. [Figures 1 and 2] But since they constituted less than 10% of the placental disc, they were considered insignificant findings, as these changes may not have an adverse impact on the fetal or neonatal outcome. Only 1 placenta from the control group had a significant intervillous thrombus occupying 75% of the placental disc. Acute intervillositis was seen in 5 cases and 7 controls but was not statistically significant. [Table 2 and Figure 3].

Figure 1.

(a) slices of placenta showing retroplacental haemorrhage, ill-defined white foci of PVFD; (b) slice of placenta showing a well-defined intervillous thrombus with lamellations; (c) slice of placenta with lamellated intervillous thrombus at periphery of placental disc and focal areas of retroplacental haemorrhage

Figure 2.

10X HE section showing (a) intervillous thrombus. (b) foci of calcifications; (c) crowding and necrosis of villi suggestive of placental infarction; (d) increased perivillous fibrin deposition. (Inset e) shows a 40X HE image with pink intervillous fibrin deposition

Table 2.

Histopathological changes in placenta in RT-PCR confirmed Covid-19 pregnant women

Histopathology Findings	Changes in placenta	SARS CoV 2 positive cases n=100	Controls n=100	P
Placental weight (gms)		420±106	380±92.7	0.027 (S)*
Perivilllous fibrin deposition (PVFD)	Significant	0	0	0.003 (S)$
	Not significant	73	92
Fibrinoid necrosis of villi	Significant	1	0	<0.001 (S)$
	Not significant	52	86
Infarction	Significant	0	0	0.019 (S)$
	Not significant	10	02
Intervillous thrombus	Significant	0	1	0.558 (NS)$
	Not significant	8	7
Retroplacental hemorrhage	Significant	0	0	0.613 (NS)$
	Not significant	7	9
Acute intervillositis	Significant	5	7	0.362 (NS)$
	Not significant (very focal)	3	3
Calcification	Significant	1	0	0.221 (NS)$
	Not significant	35	45
Chronic villitis		3	3	0.549 (NS)#
Chronic intervillositis		0	3	0.218 (NS)#
Fetal vessel vasculitis		1	3	0.403 (NS)#
Deciduitis-absent		40	38	0.920 (NS)$
Mild chronic deciduitis		32	44	0.214 (NS)$
Moderate chronic deciduitis		26	18	0.258 (NS)$
Severe chronic deciduitis		2	0	0.520 (NS)#
Chorioamnionitis (CA)- Absent		56	42	0.192 (NS)$
Mild chronic CA		16	22	0.421 (NS)$
Moderate chronic CA		8	11	0.628 (NS)$
Severe chronic CA		1	0	1.0 (NS)#
Acute CA		19	25	0.459 (NS)$

*t test, #Fisher’s Exact Test, $Chi Square Test, (S)=Significant, (NS)=Not Significant

Figure 3.

(a) 10X HE showing immature intermediate villi with loose reticular stroma; (b) 40X HE showing thrombus in fetal vessel; (c) 10X HE showing features suggestive of acute intervillositis; (d) 40X HE showing infiltration by neutrophils and few histiocytes in intervillous space

Moderate-to-severe deciduitis was seen in 28% of cases and in 18% of controls but did not meet statistical significance. [Figure 4] Chronic as well as acute chorioamnionitis was seen in 58% of controls versus 44% of cases. One case of acute chorioamnionitis showed moderate chronic deciduitis with fetal vasculitis and thrombi in fetal vessels. Chronic villitis was seen in 3 cases each in cases as well as in controls.

Figure 4.

(a) showing apoptotic bodies in membranes; (b) 40X HE showing features suggestive of moderate deciduitis; (c) 10X HE showing a dense inflammatory cell infiltrate suggestive of chorioamnionitis; (d) 40X HE showing a dense infiltrate of neutrophils in subchorial plate suggestive of acute chorioamnionitis

The histopathological changes observed failed to meet any statistical significance between asymptomatic and symptomatic cases [Table 3].

Table 3.

Comparison of histopathological changes in placenta among pregnant women with SARS-CoV-2 infection (asymptomatic vs symptomatic)

Histopathology Findings	Changes in placenta	Asymptomatic n=80	Symptomatic n-20	P
Perivilllous fibrin deposition (PVFD)	Significant	0	0
	Not significant	60 (75%)	13 (65%)	0.016 (S)$
Fibrinoid necrosis of villi	Significant	0	1 (5%)
	Not significant	41 (51.25%)	11 (55%)	0.013 (S)$
Infarction	Significant	0	0
	Not significant	7 (8.75%)	3 (15%)	0.404 (NS)#
Intervillous thrombus	Significant	0	0
	Not significant	6 (7.5%)	2 (10%)	0.712 (NS)#
Retroplacental hemorrhage	Significant	0	0
	Not significant	7 (8.75%)	0	0.031 (S)#
Acute intervillositis	Significant	3 (3.75%)	2 (10%)	0.566 (NS)#
	Not significant (very focal)	3 (3.75%)	0	0.78 (NS)#
Calcification	Significant	0	1 (5%)	0.89 (NS)#
	Not significant	25 (31.25%)	10 (50%)	0.115
Chronic villitis		3 (3.75%)	3 (15%)	0.068 (NS)#
Chronic intervillositis		0	0
Fetal vessel vasculitis		1 (1.25%)	0	0.396 (NS)#
Deciduitis-absent		37 (46.25%)	3 (15%)	0.010 (S)#
Mild chronic deciduitis		25 (31.25%)	7 (35%)	0.757 (NS)$
Moderate chronic deciduitis		22 (27.5%)	4 (20%)	0.012 (S)#
Severe chronic deciduitis		1 (1.25%)	1 (5%)	0.371 (NS)#
Chorioamnionitis (CA)- Absent		45 (56.25%)	11 (55%)	0.982 (NS)#
Mild chronic CA		14 (17.55)	2 (10%)	0.493 (NS)#
Moderate chronic CA		8 (10%)	0	0.048 (S)#
Severe chronic CA		1 (1.25%)	0	0.396 (NS)#
Acute CA		16 (20%)	3 (15%)	0.610 (NS)#

#Fisher’s Exact Test, $ChiSquare Test, (S)=Significant, (NS)=Not Significant

Discussion

Our study did not reveal any significant placental changes in the form of MVM or FVM that can be attributed to SARS-CoV-2 infection. In our study population, 80% of mothers were asymptomatic, 17% had mild disease, and 3% had severe disease, which could have been the reason for the lack of specific and significant findings seen in placental pathology in our study group. Although we did see features of maternal vascular malperfusion, since they did not involve >30% of the placental disc, they were considered to be insignificant findings.[14]

Several case reports, case series, and studies published have described conflicting results of the involvement of the placenta in mothers with SARS-CoV-2 infection.

The earliest studies conducted by Shanes et al.[15] and Facchetti et al.[16] showed an increased occurrence of features of decidual arteriopathy, maternal vascular malperfusion in the form of intervillous thrombus, infarction, increased perivillous fibrin, and subchorial hematoma. MVM is thought to occur due to a systemic inflammatory response triggered by COVID-19-derived hypoxia. Taglauer et al.[17] observed the presence of inflammation and fibrin deposition in the intervillous and subchorionic regions of the placentas of SARS-CoV-2-positive mothers. Baergen et al.[11] studied 20 placentas of SARS-CoV-2-infected positive mothers, in which 50% of the cases showed some features of FVM, or fetal vascular thrombosis, which they felt were related to the hypercoagulability associated with COVID-19. However, in the above studies, there was no mention of the percentage of the placental volume affected by these lesions. Most of the above changes are seen even in term placentas during normal gestation and do not affect the fetal outcome if less than 30% of the placental volume is involved.[14] None of our cases showed features of FVM except for one case, which showed fetal vessel thrombus but was associated with acute chorioamnionitis and fetal vasculitis. In our study, the predominant histopathologic finding was chronic deciduitis, which was comparable in both groups. An exhaustive study by Leonardo Resta et al.[18] on 83 placentas from SARS-CoV-2-infected mothers and 142 historical controls did not reveal a significant difference in FVM in both groups. Luca Bertero et al.[19] studied showed a higher frequency of chronic villitis, chronic deciduitis, microvascular thrombosis, placental infarction, and accelerated villous maturation in placentas delivered to SARS-CoV-2-infected mothers. Twenty-seven placentas from asymptomatic patients or those with mild symptoms of SARS CoV-2 infection when compared with an equal number of controls by Jaiswal et al.[20] found evidence of placental injury in the form of MVM, FVM, and placental inflammation in the study group. Although maternal viral infections are the most common cause of chronic villitis, there is very little data available on the association between placental pathology and coronavirus infection. In our study, chronic villitis was seen in an equal number of cases as well as in controls. Chronic villitis is usually related to an underlying maternal infection, but an association with a specific aetiology is not possible; it is termed “villitis of unknown significance” and is seen in only 2 to 33.8% of placentas.[21] A structured review by Sharps et al.[22] of 20 studies on placental histopathology reported FVM in 35.3% of cases, MVM in 46% of cases, and inflammation in 8.7% of cases. A meta-analysis published by Raffaella Di Girolamo et al.[23] revealed MVM in 30.7% of placentas, FVM in 27.08% of cases, and acute and chronic inflammatory pathologies in 22.68% of cases.

Our study is in coherence with several studies done on placentas of SARS CoV-2 positive mothers, which did not reveal any specific gross or histologic findings attributable to SARS CoV-2 maternal infection. Studies conducted by Hecht et al.,[13] Zhang et al.[24] and Moti Gulersen et al.[25] on placentas from SARS-CoV-2, positive mothers failed to reveal any specific histologic changes associated with SARS-CoV-2 infection. Tasca et al.,[26] in a multicentre case-control study, found no significant histologic differences between the groups when comparing treated vs. not treated and symptomatic vs. non-symptomatic patients. A review article by Caio Ribeiro Vieira Leal et al.[27] suggests that there is no specific placental histopathologic feature characteristic of SARS-CoV-2 infection; however, the most common findings in the placentas of infected women were fibrin deposition and severe inflammatory cell infiltrates.

The severity of the infection did not seem to influence the histopathological changes seen in the placenta of asymptomatic vs. symptomatic cases. Our study is in coherence with the study conducted by Husen et al.[28] on 36 SARS-CoV-2 positive mothers, who did not observe any correlation between the histopathological changes observed in the placenta and the severity of the disease.

The novelty and strength of the study lie in the fact that our study was conducted in the first dedicated COVID-19 hospital in India in pregnant and postpartum women admitted during the first and second waves of COVID-19. In mid-April 2020, our hospital was declared a COVID-19 hospital for the Mumbai Metropolitan Region and surrounding districts. The study population included a wide spectrum of COVID-19 in pregnant women and covered a comparatively larger sample size (compared to any published Indian data) over the first two waves of the COVID-19 pandemic. The study population was therefore representative of the majority of pregnant women with COVID-19 in the MMR region with a 26 million population. Our admission policy was uniform during both waves of the pandemic. As per the admission policy of our hospital pregnant women with COVID-19 who are near-term, need obstetric interventions, or are symptomatic were admitted.[29]

All women in labor or near term were tested for SARS-CoV-2. Therefore, we can confirm that the study population had a positive SARS-CoV-2 report within 48 hours of childbirth and the control population had a negative COVID-19 report within 48 hours of childbirth. Also, the control population included women who had never suffered from COVID-19 during or before getting pregnant. Therefore, we can confirm that the infection was still present at the time of delivery in the study population. However, we could not compare data between the two waves because of the small percentage of patients in the first wave and the non-availability of genome sequencing data.

The limitations of our study were that the reporting subject expert pathologist was not blinded to the SARS-CoV-2 status of the patient or the clinical details while reporting, which could have led to bias. Identification of SARS-CoV-2 in placental or fetal tissue with RT-PCR or ISH assays [with appropriate positive and negative controls] is considered more definitive evidence of infection. We did not perform any immunohistochemistry, in situ hybridization [ISH], or real-time PCR on placental tissue for SARS CoV-2 virus on placental sections, as the sensitivity for IHC is a mere 53.4% as per our literature search.[30] The second reason for not doing these tests is financial constraints. The non-availability of genome sequencing data to show the direct association of B.1.617 leading to adverse outcomes and the single-center study are the limitations of our study.

Conclusion

The histologic changes observed in the third-trimester placenta in SARS-CoV-2 infection cannot be attributed to the infection as they are even seen in normal-term placentas. The percentage of the placental disc involved in these changes also needs to be evaluated, as ≤30% involvement of the placental disc may be a non-significant finding that does not lead to an adverse fetal outcome.

Whether infection in the first or second trimester causes significant morphologic alterations in the placenta is not known. Identification of placental molecular markers specific to SARS-CoV-2 infection along with histologic examination of placentas from women infected in the first and second trimesters would help us understand the effect of the virus on placental pathophysiology.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.