A Clinical Multi-center Study of Pregnant Women with COVID-19 in Hubei, China

Background: Coronavirus disease 2019 (COVID-19) has rapidly spread to more than 200 countries. Thus far, reports regarding multi-center data from throughout gestation in women with COVID-19 and newborn outcomes are scarce. Methods: We retrospectively reviewed data from 92 pregnant women with COVID-19 (PW-COVID-19) and their 78 newborns from 27 hospitals in 12 regions of Hubei, China. The demographic, epidemiological, clinical, laboratory, and therapeutic data and pregnancy, perinatal, and neonatal outcomes were analyzed. Follow-up was censored until April 7, 2020. Findings: Median maternal age was 31.0 years (IQR 28·0-33·0), with nine patients in the rst trimester, ve in the second trimester, and 78 in the third trimester. None of the patients died, and most (92·4%) recovered and were discharged. Seventy-ve deliveries (including three sets of twins) comprised 66 cesarean sections and nine vaginal deliveries, with 21 preterm and 57 full-term infants. Seventeen live births had radiological ndings of pulmonary infection. One newborn tested positive for SARS-CoV-2 nucleic acid, and three newborns were viral antibody-positive: two IgG (+) and IgM (-), and one IgG (+) and IgM (+). The median suspected duration of virus exposure was 7 days (IQR 0 to 27). Interpretation: Compared to the pregnant women with other viral infections, such as SARS, MERS, and Zika virus infection, PW-COVID-19 had similar manifestations and relatively better outcomes. The termination time and delivery mode in PW-COVID-19 should be evaluated based on both the maternal and fetal situations. The possibility of maternal-to-fetal transmission of SARS-CoV-2 requires further investigation. Authors Shujie Liao and Renjie Wang contributed equally to this work. and therapeutic data and pregnant, perinatal, and neonatal outcomes of 92 pregnancy cases with COVID-19 and a follow-up of 78 neonates born to these mothers from 27 hospitals in 12 regions of Hubei province, China. Meaning: Compared to the pregnant women with other viral infections, such as SARS, MERS, and Zika virus infection, PW-COVID-19 had similar manifestations and relatively better outcomes. illustrate the incidence, risk factors, and possible prophylactic measures for intrauterine transmission of SARS-CoV-2.


Introduction
In December 2019, the novel coronavirus disease 2019 (COVID-19) of unknown origin was identi ed in Wuhan, the capital city of Hubei province, China. 1 As of April 30, 2020, a total of 3,090,445 patients with COVID-19 had been con rmed and 217,769 deaths reported worldwide, 2 and the World Health Organization (WHO) had declared COVID-19 a pandemic. 3 The virus, o cially named severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), belonged to the same β-coronavirus genus as severe acute respiratory syndrome coronavirus (SARS-CoV) and Middle East respiratory syndrome coronavirus (MERS-CoV), which were reported to have high morbidity and mortality, particularly adverse pregnancy outcomes in pregnant women. 4,5 According to a nationwide epidemiological survey in China on February 18, 2020, at least 252 pregnant women were suspected of having COVID-19, including 115 con rmed cases, 75 (65.2%) of which occurred in Hubei province. 6 Till April 2, 2020, one of epidemiological survey had showed no more than 150 cases of COVID-19 in pregnant women around China. 7 To date, several reports have been published regarding patients with COVID-19, but reports focusing on pregnant women remain scarce and have been limited by small sample sizes, the nature of single-center studies, few data on clinical course, pregnancy, and neonatal outcomes, and lack of information on characteristics throughout gestation, especially those in the rst and second trimester. 8,9 Whether pregnant women with COVID-19 (PW-COVID-19) have a different clinical course and adverse outcomes compared to non-pregnant patients and other virus infection diseases such as SARS, MERS, Zika, etc., or whether suffering from COVID-19 was necessary for the immediate termination of pregnancy or cesarean delivery, or whether the suspected duration of virus exposure could induce an adverse impact on the fetus remain unknown. To address these problems, we performed the rst comprehensive retrospective multi-center study on the whole gestation period in PW-COVID-19 in Hubei province, China.

Research in context
Evidence before this study We searched PubMed and China National Knowledge Infrastructure database on April 7, 2020, for articles that described the features of patients in pregnancy infected with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), using the search terms ("novel coronavirus" OR "COVID-19" OR "SARS-CoV-2") AND ("pregnancy" OR "maternal outcome" OR "fetal infection") with no language or time restrictions. We found several articles and case reports on pregnant women with COVID-19. However, no published works were found about the multi-center data analysis based on the whole gestation period with COVID-19 and considering the outcomes among pregnancy women, fetus and newborns.

Added value of this study
We retrospectively reviewed the demographic, epidemiological, clinical, laboratory, and therapeutic data and pregnant, perinatal, and neonatal outcomes of 92 pregnancy cases with COVID-19 throughout the gestation period and a follow-up of 78 neonates born to these mothers from 27 hospitals in 12 regions of Hubei province, China. None of the patients died, and most (92.4%) recovered and were discharged. Seventy-ve deliveries (including three sets of twins) comprised 66 cesarean sections and nine vaginal deliveries, with 21 preterm and 57 full-term infants.

Study design and participants
This retrospective study included 92 hospitalized PW-COVID-19 from 27 hospitals in Hubei province, China, between January 15 and April 7, 2020. The case sources were listed in the appendix (Table S1). Follow-up was censored until April 7, 2020. COVID-19 was diagnosed based on the New Coronavirus Pneumonia Prevention and Control Program (5th-7th trial edition) published by the National Health Commission of of fever for at least three days, remission of respiratory symptoms, substantial improvement in the lungs on imaging, and two consecutive negative nucleic acid tests (NATs) at least 24 hours apart. 10

De nitions
The severity of COVID-19 was de ned as mild (mild clinical symptoms and no pneumonia on imaging), moderate (presence of fever, respiratory symptoms, and pneumonia on imaging), severe (respiratory distress, respiratory rate ≥ 30 times/min, oxygen saturation ≤ 93% at rest, or partial arterial oxygen pressure [PaO 2 ]/oxygen absorption concentration [FiO 2 ] ≤ 300 mmHg [1 mmHg = 0·133 kPa]), or critical (respiratory failure requiring mechanical ventilation, or shock or failure of other organs requiring ICU treatment). 10 The composite endpoint was based on admission to an intensive care unit (ICU), the use of mechanical ventilation, or death. 13 The severity of COVID-19 was also categorized as severe or non-severe based on the American Thoracic Society (ATS) guidelines for community-acquired pneumonia. 14 The linkage to Wuhan was de ned as living in Wuhan, visiting Wuhan recently, or having contact with Wuhan residents. The incubation period was de ned as the interval between the potential earliest exposure to the transmission source (wildlife, or suspicious or con rmed patients) and the earliest onset of symptoms. Incubation periods <1 day were excluded because of the continuous exposure. The suspected duration of virus exposure was de ned as the interval between the potential earliest exposure to the transmission source, or the earliest onset of symptoms or the date of disease con rmation, and the date of delivery or the last date of follow-up. Family cluster of COVID-19 was de ned as two or more cases with con rmed COVID-19 reported within a given family in two weeks. Fever was de ned as an axillary temperature of 37·5°C or higher. Puerperium is de ned as the 6-week period following childbirth during which the woman's body almost returns to its prepregnancy state. 15 Fetal heart monitoring was performed weekly for pregnancies after 28 weeks. Fetal growth was assessed by serial ultrasound scans every two weeks. The gestation period was calculated from the rst day of the last menstrual period and was approximately 280 days (40 weeks), clinically divided into three periods: <14 weeks is the rst trimester, 14 weeks to 27+6 weeks is the second trimester, and ≥28 weeks is the third trimester. 16 Statistical analysis Statistical analysis was done with R 3·6·2 (2019-12-12) and Excel 2016, and p values of <0·01, <0·05 and <0·10 were considered as the presence of statistical signi cance. Continuous variables were directly expressed as medians and interquartile ranges or ranges. Categorical variables were expressed as number and percentages. Chi-square test and single-ratio z-test separately were used for comparison of categorical variables, while single population t-test was used for continuous variables. The ArcGIS, version 10·2·2 was used to plot the numbers of con rmed patients in each region on the map.

Role of the funding source
The funder of the study did not participate in study design, data collection, data analysis, data interpretation, or writing of the report. The corresponding authors had full access to all the data in the study and had nal responsibility for the decision to submit for publication.

Results
Clinical characteristics, laboratory results, treatments, and outcomes Of the 92 PW-COVID-19 included in the study, 47 (51·1%) were residents of Wuhan. Other patients (n=45) were distributed in 11 regions outside Wuhan, which accounted for 12/17 of the regions in Hubei province ( Figure 1, Table S1). The demographic, epidemiological, and clinical characteristics, radiological ndings, treatments, and outcomes are given in Table 1. The median patient age was 31·0 years (IQR 28·0 to 33·0). Nine pregnant women were in the rst trimester, ve in the second trimester, and 78 in the third trimester. The information of COVID-19 of all their family members was followed up until April 7, 2020. Fifteen patients (18·1% of 83 cases) had family clusters, and 77 patients (85·6% of 90 cases) had a link to Wuhan. The median incubation period was 8·0 days (IQR 6·75 to 14·25), which was longer than in patients with COVID-19 in the general population (GP-COVID-19). 13 22 patients (25·3% of 87 cases) had at least one comorbidity (e.g., hypertension, diabetes mellitus, hypothyroidism).
Among 91 cases (one pregnant woman refused the imaging examination), 89·0% had abnormalities on chest CT. Ground-glass opacity was found in 44·0% of cases and local patchy shadowing in 59·3% of cases, whereas interstitial abnormalities presented in only one case.
Most patients were categorized as moderate cases (82·6% of 92 cases), but ve patients (5·4%) and one patient (1.1%) were severe and critical cases, respectively. The distribution of COVID-19 severity in different age groups and gestation periods is shown in Figure S1 and Figure S2. Based on the ATS guidelines, only two patients were classi ed as severe, which was signi cantly lower than GP-COVID-19 (2·2% vs. 15·7%, P<0·0001). 13 As for the treatment and clinical outcomes, PW-COVID-19 were more likely to receive oxygen therapy (59·0% vs. 41·3%, P=0·0034) and antibiotic therapy (93·6% vs. 58·0%, P<0·0001), whereas antiviral therapy (including interferon alpha, lopinavir/ritonavir, ribavirin, or arbidol) (19·5% vs. 35·8%, P<0·0001), systemic glucocorticoids (6·4% vs. 18·6%, P=0·010), and intravenous immune globulin (3·8% vs. 13·1%, P=0·027) were used less often than in GP-COVID-19. 13 Mechanical ventilation was used in two patients. The median hospital stay was 9·5 days (IQR 5·25 to 15·00), which was a little shorter than reported for GP-COVID-19. 13 No patient died, and most of the patients (92·4% of 92 cases) recovered and were discharged from the hospital by the end of follow-up. The composite endpoint occurred in 17/78 patients (21·8%); 17 patients were admitted to the ICU, and two required mechanical ventilation. These patients had a similar clinical course compared to those without a composite endpoint. Regarding the two patients requiring mechanical ventilation, one was in the rst trimester and ended up having an induced abortion, the other was in the third trimester and underwent full-term delivery by cesarean section (CS). Both cases had mild symptoms at admission and no complications at delivery. Two patients recovered well, and one has discharged from the hospital by the date of data collection.

Pregnancy outcomes of suspected duration of virus exposure on PW-COVID-19
The median suspected duration of virus exposure for PW-COVID-19 was 7 days (IQR 0 to 27). Decreased gestational weight gain was noted in 7/58 cases (12·1%). The blood pressure, fundal height, abdominal circumference, and fetal heart rate in each group were within the reference range. 16 The overall rate of complications in all patients was less common than in the literature. 16,[18][19][20] Seven of the nine patients (77·8%) who presented in the rst trimester had induced abortions without any pathological indications for pregnancy termination; all were based on the patients' requests and after being informed of the risk and signing informed consent. One case in the second trimester was stillborn; the fetus was in fetal distress and exhibited fetal growth restriction before the potential earliest exposure to the transmission source, and there was no direct relationship between the fetal death and COVID-19. Twenty-one patients in the third trimester (26·9%) underwent preterm delivery. By April 7, 2020, data has been collected on 22 patients who received antenatal care. Data on antenatal care from 13 fetus and neonates born to PW-COVID-19 after virus exposure are shown in Table S4.
The median gestational age at delivery in preterm infants, full-term infants, vaginal delivery, and CS was 36+1 weeks, 38+4 weeks, 38 weeks, and 39+6 weeks, respectively. As shown in Table 4, we found no signi cant difference in regards to abnormal fetus and appendages, neonatal disease, the morbidity of COVID-19, or imaging ndings between term and preterm infants, or between vaginal delivery and CS infants.
SARS-CoV-2 NAT was positive in one newborn (#69). The viral antibody test was positive in three newborns: two (#36 and 57) with IgG (+) and IgM (-), and one (#68) with IgG (+) and IgM (+). For newborn #69, 36 hours after emergency CS at 40 weeks, the SARS-CoV-2 NAT was positive. The infant eventually recovered and was discharged 16 days after delivery. Patient #36 underwent a full-term vaginal delivery, and patient #57 delivered by CS at 38+1 weeks. The SARS-CoV-2 NATs were negative, but both had signs of pneumonia on imaging. For newborn #68, 19 hours after CS at 38 weeks, viral antibody tests were IgG (+) and IgM (+). The newborn met the criteria for discharge and was discharged from the hospital on March 20. Pulmonary infections were observed on imaging in 17/34 cases (50·0%). We had two assisted reproductive technology (ART) PW-COVID-19 cases. Their maternal and fetal outcomes were not signi cantly different from spontaneous pregnancy ( Table 4).

Discussion
In the current study, we collected and analyzed the clinical data and outcomes from 92 PW-COVID-19 and 78 neonates born to these mothers, which is the rst comprehensive multi-center study regarding maternal, fetal, and neonatal outcomes throughout gestation.
Pregnancy is considered to have greater susceptibility to certain infections because of great changes in the respiratory system due to hormonal and physiological variations and the special immunosuppressive state. However, tolerance to the semi-allogeneic fetus does not mean an overall systemic immune response; in contrast, the tolerance in the uterus is transient hypo-responsiveness, and the systemic immune system appears to be well-modulated during pregnancy. 23,24 Previous studies have demonstrated poor outcomes in pregnant women infected with SARS, MERS, and H1N1. 4,25,26 Binding with similar a nities to angiotensin-converting enzyme 2 (ACE2) and having 85% sequence similarity with SARS, patients infected with SARS-CoV-2 are considered to have similar adverse outcomes. 27 However, in our studies, PW-COVID-19 appear to have similar clinical characteristics, 1,13 milder severities, better clinical course, and better outcomes with a 0% mortality rate, and PW-COVID-19 also have a lower proportion of maternal death, pregnancy loss, and fetal death than SARS and MERS. Based on our studies, there is little evidence to indicate that PW-COVID-19 has more adverse outcomes. Some research has hinted that SARS-CoV-2 has similar a nities as SARS-CoV for binding the same receptor, ACE2, but mutation of the protein induced a higher incidence of infection and lower pathogenicity. 1,27 The different outcomes of various virus-related diseases in pregnant women require more research of their pathogenesis to nd potential targeted prophylactic and therapeutic strategies.
Sixty-six pregnant women (88·0% of 75 cases) underwent CS and nine underwent a vaginal delivery. In addition, we found no signi cant difference in the maternal, perinatal, and neonatal outcomes between vaginal delivery and CS ( Figure S3). Therefore, vaginal delivery could be considered for patients with a mild infection, stable condition, normal birth canal and satis ed labor stage, no perinatal complications, and negative NAT in vaginal discharge. 6,17 The median suspected duration of virus exposure was 7 days. The relationship between suspected duration of virus exposure and pregnancy outcomes is provided in Table 3, Table 4, Table S5, and Figure S4. In addition, our data seem to indicate that the occurrence of fetal distress, meconium-stained amniotic uid, and PROM correlates with the duration of exposure. However, whether the duration of virus exposure had an impact on the adverse outcomes of pregnancy requires more investigations.
Until now, no direct evidence has been available on maternal-to-fetal transmission of SARS-CoV-2 in pregnancy. 8,9 Some research has shown that the predominant receptor of sensitive cells for SARS-CoV-2 is expressed at very low levels in the human maternal-fetal interface. 27,28 In our research, a very important nding is the possible infection of newborns (Table 4). One newborn (#69) had a positive for SARS-CoV-2 NAT, and three newborns (#36, 57, and 68) were positive for viral antibody. After the rst positive SARS-CoV-2 NAT, newborn #69 tested negative in subsequent NATs of nasopharyngeal and anal swabs, recovered well, and was discharged from the hospital 16 days after delivery. The newborns' antibody test was still negative 46 days after delivery, whereas the mother was IgG (+) and IgM (-). The CT scan of the mother on April 1 showed further absorption compared to that on February 24. Considering the lack of NATs of intrauterine tissue samples and series of negative tests, intrauterine transmission remained uncon rmed. Newborns #36 and 57 both had signs of pneumonia on imaging but were negative on the SARS-CoV-2 NAT. Mother #68 had a positive NAT at 33+6 weeks, which was negative in the subsequent three repeats.
Newborn #68 was viral antibody IgG (+) and IgM (+) in the cord blood and neonatal peripheral blood within the rst day after CS at 38 weeks. Nasopharyngeal swabs, anal swabs, urine, peripheral blood, gastric juice before feeding, and intrauterine tissue samples, including amniotic uid, cord blood, and placenta were collected and SARS-CoV-2 NATs were all negative. However, the viral antibody tests remained positive in the subsequent ve repeats in neonatal peripheral blood. A possible reason for this nding may be that IgG is transmitted vertically from the mother through the fetal blood supply, and the fetus may acquire defense against the virus to some extent. The IgM in cord blood and peripheral blood within the rst day after delivery indicates the possibility of vertical transmission. However, we need more direct evidence. Our study has some limitations. First, this study was retrospective and, given the variation in different electronic databases at multiple centers, some laboratory tests were incomplete. Second, lack of appropriate control groups which were matched for age, gender, and comorbidities, so it was hard to conclude the authentic different characteristics between PW-COVID-19 and GP-COVID-19. Third, because the times of the SARS-CoV-2 NATs varied among patients, and obvious symptoms or clear exposure history were lacking in some patients, the calculation of suspected duration of virus exposure may not be entirely accurate. Fourth, the limited number of patients in some subgroups (e.g., patients in the rst and second trimester, patients delivered vaginally) require more relevant cases research.
In conclusion, our study showed that PW-COVID-19 have similar manifestations but relatively mild severity compared to other viral infections during pregnancy. Suffering from COVID-19 may not be an absolute indication for immediate termination of pregnancy and cesarean delivery. The most appropriate time and delivery mode for the termination of pregnancy should be determined by a comprehensive assessment of both maternal and fetal conditions. The possible presence of maternal-to-fetal transmission of SARS-CoV-2 requires further investigation. The relationship between COVID-19 and ART requires more in-depth research. Family cluster of PW-COVID-19 (see Table 1) should also be considered. Close contact tracing and family cluster surveillance are recommended for better protection of pregnant women. These data will help support clinical risk-strati cation and personalized treatment strategies. College, Huazhong University of Science and Technology) drafted the manuscript. All authors agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Declarations
(accessed March 20 2020).   11. Fever was defined as an axillary temperature of 37.3°C or higher.
12. The data was quoted from Jamieson DJ, et al. 6 .