Fetal death from SARS-CoV-2 mediated acute placental failure

Introduction We demonstrate the nonlinear severity of symptoms of SARS-CoV-2 infection in the mother leading to fetal death after acute placental failure. Methods Careful clinical evaluation, real-time RT-PCR molecular microbiologic testing, isolation of a viable virus, and autopsy with histologic results were used to investigate the possible vertical transmission of SARS-CoV-2 infection from mother to fetus. Results Histologic changes in the placenta correlate with SARS-CoV-2 infection. Total nucleic acid isolated from vaginal swabs, fresh placental tissue, and deparaffinized tissue showed a high viral load of SARS-CoV-2. Complete genome sequencing confirmed the presence of the SARS-CoV-2 Delta variant. Discussion Several methods have been used to confirm SARS-CoV-2-mediated acute placental failure, all of which were conclusive. It should be noted that careful periodic fetal well-being checks are required in women infected with SARS-CoV-2, regardless of the severity of symptoms. Most of the cases described with fetal death occurred in the third trimester.


Introduction
The Covid-19 pandemic has contributed to more than 6 million deaths while infecting more than 504 million people worldwide (World health Organization, 2022). Europe has been the most infected region with more than 211 million confirmed cases (World health Organization, 2022). Among the infected were pregnant women. The link between SARS-CoV-2 infection and the risk of harmful outcomes for the mother and baby has been established (Di Mascio et al., 2020). SARS-CoV-2 infected mothers with severe symptoms were more likely to have a preterm birth, pre-labor cesarean section, as well as gave birth to a baby that was stillborn or admitted to a neonatal intensive care unit (Di Mascio et al., 2020). There is evidence of vertical transmission from mother to fetus when the infection occurs in the third trimester. Estimated vertical transmission is approximately 3.2 % detected via infant nasopharyngeal swab testing or 7.7 % via placental sample analysis (Kotlyar et al., 2021). The study by Husen et al. revealed that the SARS-CoV-2 placental signature seemed to correlate with fetal distress but not with the symptom severity of infected mothers (Husen et al., 2021). We present a 27-year-old woman who had a stillbirth in 29th gestational week, 10 days after she tested positive for SARS-CoV-2, in an otherwise uneventful pregnancy.

Case
We present the case of the 27-year-old nulliparous woman with fetal death during mild symptomatic Covid-19 disease. She had not been vaccinated against Covid-19. She had controlled hypothyroidism and gestational diabetes, which was treated with diet. Her pregnancy was uneventful until 28 2/7 gestational weeks when she tested positive for SARS-CoV-2. A viral load concentration of 1.6 × 106 copies RNA/ul was detected in the nasopharyngeal swab. She suffered from a headache and cough. Five days after the positive test, she was examined by an obstetrician at the perinatal outpatient clinic of UMC Ljubljana because of suspected preterm labor. On admission, maternal vital signs were normal. Ultrasound (US) examination revealed normal fetal movements of a normally grown fetus, normal appearance of the placenta, and normal amniotic fluid index. The length of the cervix was 30 mm, without funneling.
The patient was reexamined four days later at 29 4/7 gestational weeks as fetal movements were absent for two days. Maternal vital signs were normal. Ultrasound examination confirmed fetal death. A male fetus with a birth weight appropriate for gestational age (1210 g) and no visible abnormalities was born on the same day. A vaginal swab taken during labor and a placental sample tested positive for SARS-CoV-2 with high viral load concentration. The patient was discharged one day after delivery. The postpartum course was unremarkable.
Tests performed after delivery did not reveal any other usual cause of fetal death. Tests for thrombophilia and antiphospholipid syndrome were negative. TSH and HbA1c were within the normal ranges. There was no evidence of fetomaternal hemorrhage. Serologic testing revealed no recent toxoplasma, human cytomegalovirus, or parvovirus B19 infection.

Histology
Four tissue samples were collected from the central part of the native placenta and areas with different macroscopic appearance than normal, including hemorrhage. One tissue sample was collected from the chorioamniotic membranes and one from the umbilical cord. Samples were fixed in 10 % formalin and underwent standard histological processing. Slides were observed using light microscopy.

Immunohistochemistry
Two immunohistochemical (IHC) stains were performed on a 4month-old formalin-fixed paraffin tissue sample from the central area of the placenta. SARS-CoV nucleocapsid protein antibody from Novus-Bio and SARSCoV2 spike protein antibody from Abcam were used for detection of SARS-CoV-2.
For viral isolation, the homogenized placental tissue was inoculated on VERO E6 cells (ATCC-LGC; CRL-1586) in Dulbecco's modified eagle medium (DMEM) containing 4 % fetal bovine serum (FBS) and antibiotics. The inoculated cells were observed daily, and a passage was performed on day 5, when an intense cytopathic effect occurred. After the second passage, the virus was collected and characterized by complete genome sequencing.
Completed genome sequencing was performed directly from placental tissue, and the virus was isolated on cell culture. In brief, PCR amplicons were prepared according to the ARTIC V3 RT-PCR protocol (nCoV-2019 sequencing protocol v2 (GunIt) (protocols.io)). Amplicon NGS libraries were prepared using the Nextera XT Library Preparation Kit (Illumina, San Diego, CA, USA) according to the manufacturer's instructions. Prepared samples were sequenced using the NextSeq 500/ 550 High Output Kit v2.5 (300 cycles) on the Illumina NextSeq 550 instrument. A consensus sequence was generated with iVAR and lineage assignment was performed using Phylogenetic Assignment of Named Global Outbreak Lineages (Pangolin), which implements the dynamic nomenclature of SARS-CoV-2 lineages. Both sequences were deposited to GISAID under the accession numbers: EPI_ISL_14327818 and EPI_ISL_14327819.

Autopsy
An autopsy of a male fetus and the corresponding placenta was performed. The fetus was normally developed, weighing 1190 g and the length of 42 cm. Fetal measurements of body proportions and organ weights were appropriate for gestational age, all between the 10th and 90th percentiles (Archie et al., 2006). The autopsy of the fetus revealed no abnormalities. Fetal membranes were bright to dark red and opaque. The placenta (measured with fetal membranes and without the umbilical cord) weighed 340 g, responding to the 25th and 50th percentiles (Almog et al., 2011). The umbilical cord was 44 cm long with 10 turns, normally coiled, a smooth surface without knots and 3 blood vessels without thrombi. Insertion of the umbilical cord was central. The fetal side of the placenta was smooth and bright pink. The maternal side was bright to focally dark red. On cross sections, the placental parenchyma was red to brown with focally small yellow to gray firm areas, diffuse small hemorrhages, and congested blood vessels. The percentage of the placental disc involved is hard to evaluate because of the diffuse spread of small and quite subtle changes, nevertheless the rough estimation was around 10 %.

Histology and immunohistochemistry
The most outstanding pattern was diffuse necrosis of cyto-and syncytiotrophoblast with perivillous fibrin deposition. The pattern was present in approximately 70 % of the examined slides, affecting full thickness of the placental disc. Chorionic villi were almost all avascular, focally necrotic. There was an intense mixed cellular, predominantly neutrophilic infiltrate (Ma et al., 2021) with karyorrhexis in the basal decidua that was spreading into the intervillous space. Throughout the parenchyma were a couple of small hemorrhages and thrombohematomas (Huynh et al., 2022), some of the larger measuring in diameter up to 10 mm ( Fig. 1; A-B). Between the chorion and the membranous decidua was mild to intermediate mixed cellular infiltrate with sparse neutrophils in the amniotic connective tissue. There was no amniotic epithelial necrosis (Redline et al., 2021) (Fig. 2; A-D). There were no signs of fetal inflammatory response. There were no other changes that would correspond to fetal or maternal vascular malperfusion (Redline et al., 2021).
To determine if SARS-CoV-2 has spread from the placenta to the fetus, the same IHC stains for SARS-CoV-2 were performed on the left lower lung lobe sample (least affected from autolysis) which showed normal histology for gestational age on H&E staining. Both stains were completely negative (Best Rocha et al., 2020) ( Fig. 4; A-B).

Microbiological SARS-CoV-2 diagnostics
Since the patient tested positive for SARS-CoV-2 10-days prior to the complications, a complete SARS-CoV-2 diagnostic was performed. Viral load concentrations of 4.3 × 10 2 copies and 7.0 × 10 7 copies RNA/ul were detected in the vaginal swab and placental tissue, respectively. In addition, SARS-CoV-2 RNA was also amplified from formalin-fixed paraffin (FFP) sections of placental tissue with a similar high viral load concentration of 1.7 × 10 7 copies RNA/ul. The FFP fetal lung sample was SARS-CoV-2 negative. In addition, the infectious virus was successfully isolated from a fresh placental sample collected after labor (Fig. 5). A complete genome sequencing confirmed infection with the Delta SARS-CoV-2 genetic variant and showed no difference between the virus sequenced directly from placental tissue and the virus isolate.

Discussion
Fetal death occurred within 10 days after a positive SARS-CoV-2 test in the third trimester in unvaccinated mother, although the mother had a relatively mild form of Covid-19. Our results are consistent with the findings of Husen et al (Husen et al., 2021). SARS-CoV-2 binds primarily to angiotensin-converting enzyme 2 (ACE2) (Komine-Aizawa et al., 2020). The receptor is highly expressed in syncytiotrophoblast, cytotrophoblast, endothelial cells, and the vascular smooth muscle cells of the primary and secondary villi (Komine-Aizawa et al., 2020). It is known that the novel coronavirus canonically utilizes the ACE2 receptor and the serine protease TMPRSS2 for cell entry, however that data gathered with examining the expression of ACE2 and TMPRSS2 throughout the pregnancy and third trimester chorioamnionitic membranes by Pique-Regi et al. suggests that the placenta minimally expresses the canonical cell-entry mediators for SARS-CoV-2 (Pique- Regi et al., 2020). In their small study, Roberts et al. have found that TMPRSS2 expression was rarely detectable in villous endothelium and very rarely detectable in the villous syncytiotrophoblast throughout the pregnancy. They have showed that ACE2 expression varied throughout the pregnancy with circumferential syncytiotrophoblast expression more common in early gestations and polarized expression more common in later gestation (Roberts et al., 2021a). However, Shook et al. has observed sex differences in placental TMPRSS2 but not ACE2 in the setting of maternal SARS-CoV-2 infection. TMPRSS2 expression was found significantly correlated with ACE2 expression in males, which may have implications for fetal vulnerability to placental infection such as in our case (Shook et al., 2021).
IHC staining for SARS-CoV-2 was significantly positive in the cytoand syncytiotrophoblast, confirming infection of these cells (Komine-Aizawa et al., 2020). The most commonly described histologic pattern in placentas affected by SARS-CoV-2 is extensive perivillous fibrin deposition with fibrin thrombi formation in the intervillous space and histiocytic infiltration (Sharps et al., 2020), suggesting intrauterine transplacental transmission of SARS-CoV-2 virus (Schwartz and Morotti, 2020). In our case, the placenta had diffuse cyto-and syncytiotrophoblast necrosis with perivillous fibrin deposition, avascular and focally necrotic villi and intermediate mixed cellular, predominantly neutrophilic infiltrate (Ma et al., 2021) that spread from the basal decidua to the surrounding villi and nonspecific subchorionitis. The signs of maternal or fetal vascular malperfusion were absent.
According to the guidelines (Roberts et al., 2021b), we amplified SARS-CoV-2 RNA from fresh placental samples and successfully isolated the infectious virus. However, we were unable to detect the presence of the virus in fetal tissue using IHC staining or real-time RT-PCR on a formalin-fixed paraffin sample of the fetal lung.
Even early on during the COVID-19 epidemics, stillbirth was found to be 2-times more likely in SARS-CoV2 infected women (WAPM World Association of Perinatal Medicine Working Group on COVID-19, 2021). The accumulation of safety data and increasing evidence of substantial harms of COVID-19 to a pregnant woman and fetus led many professionals and professional organizations to recommend vaccination during pregnancy (Martins et al., 2021;$author1$, 1 et al., 2021;Kalafat et al., 2022). In a large study from Israel, including 7 530 women COVID-19 vaccinated during pregnancy and 7 530 unvaccinated women, the odds for fetal death were lower in vaccinated women (OR: 0.49; 95 % CI 0.01-9.60), albite not significant due to the overall low stillbirth rate (Goldshtein et al., 2021). But, in a meta-analysis including 66 067 COVID-19 vaccinated and 424 624 unvaccinated pregnant women, the risk of stillbirth was significantly lower in the vaccinated cohort by 15 % (pooled OR 0.85; 95 % CI 0.73-0.99) . Despite the substantial evidence for the safety and benefits, many women, like in our case, did not decide on vaccination during pregnancy.  Based on clinical, pathological, and microbiological correlation supported by specific histological findings and IHC staining, we conclude that the cause of fetal death in a woman with mild Covid-19 disease was acute placental failure triggered by SARS-CoV-2 infection.

Funding
Molecular diagnostics and virus isolation were funded by the Institute of Microbiology and Immunology, Faculty of Medicine, University of Ljubljana; by the Slovenian Research Agency (grants P3-0083); by the European Virus Archive-GLOBAL project, which has received funding from the European Union's Horizon 2020 research and innovation program under grant agreement no. 871029; and by Network of infrastructure the Centers of University of Ljubljana (MRIC-UL-IC-BSL3+, IP-022), financed by the Slovenian Research Agency.

Declaration of Competing Interest
The authors whose names are listed immediately below certify that they have NO affiliations with or involement in any organisation or certify with any financial interest (such as honoraria; educational grants; participation in speakers' bureaus; membership, employment, consultancies, stock ownership, or other equity interest; and expert testimony or patent-licensing arrangements), or non-financial interest (such as personal or professional ralationship, affiliations, knowledge or beliefs) in the subject matter or materials discussed in this manuscript.