key: cord-0821427-jn1w8h2p authors: Walker, Kate F; O’Donoghue, Keelin; Grace, Nicky; Dorling, Jon; Comeau, Jeannette L; Li, Wentao; Thornton, Jim G title: Maternal transmission of SARS‐COV‐2 to the neonate, and possible routes for such transmission: A systematic review and critical analysis date: 2020-06-12 journal: BJOG DOI: 10.1111/1471-0528.16362 sha: b6a02bbf603965dd8020bc0a9cd1a5c669ecd469 doc_id: 821427 cord_uid: jn1w8h2p BACKGROUND: Early reports of COVID‐19 in pregnancy described management by caesarean, strict isolation of the neonate and formula feeding, is this practise justified? OBJECTIVE: To estimate the risk of the neonate becoming infected with SARS‐COV‐2 by mode of delivery, type of infant feeding and mother‐infant interaction SEARCH STRATEGY: Two biomedical databases were searched between September 2019 ‐ June 2020. SELECTION CRITERIA: Case reports or case series of pregnant women with confirmed COVID‐19, where neonatal outcomes were reported. DATA COLLECTION AND ANALYSIS: Data was extracted on mode of delivery, infant infection status, infant feeding and mother‐infant interaction. For reported infant infection a critical analysis was performed to evaluate the likelihood of vertical transmission. MAIN RESULTS: We included 49 studies which included 666 neonates and 655 women where information was provided on the mode of delivery and the infant’s infection status. 28/666 (4%) neonates had confirmed COVID‐19 infection postnatally. Of the 291 women who delivered vaginally, 8/292 (2.7%) neonates were positive. Of the 364 women who had a Caesarean birth, 20/374 (5.3%) neonates were positive. Of the 28 neonates with confirmed COVID‐19 infection, 7 were breast fed, 3 formula fed, 1 was given expressed breast milk and in 17 neonates the method of infant feeding was not reported. CONCLUSIONS: Neonatal COVID‐19 infection is uncommon, uncommonly symptomatic, and the rate of infection is no greater when the baby is born vaginally, breastfed or allowed contact with the mother. Many early reports of COVID-19 in pregnancy described management by caesarean, isolation of the neonate from the mother at birth and formula feeding. The reasons included previous experience of the severity of other coronavirus infections in pregnancy as well as an intention to protect the neonate from infection. Of 12 pregnant women with SARS-CoV in the 2002-03 pandemic (1, 2) , three mothers died, four women miscarried in the first trimester, two neonates were growth restricted and four delivered preterm. Among 11 pregnant women infected with MERS-CoV (3), three mothers and three neonates died. Another factor may have been that the pandemic began in China where Caesarean rates are often over 40% and obstetricians are used to responding to problems by recommending birth by this route (4) . Expert guidelines have cautiously recommended vaginal birth in the absence of maternal respiratory failure or fetal compromise, as well as breast feeding with other precautions to minimise maternal to neonate transmission (5) . Although the number of mothers and neonates included in scientific reports of COVID-19 pregnancies now number 655 mothers and 666 neonates; many of these reports include the same or overlapping cases (6) . This may be a particular problem with reports from China. In the City of Wuhan alone, population 12M, there are 50 hospitals, 19 of which have had cases of COVID-19 in pregnancy (7) . The data are complicated by a number of other factors. The mothers involved may have been symptomatic, or asymptomatic, laboratory confirmed or not, and the babies may have been positive or negative on testing, or not tested. The latter group are sometimes assumed to be negative if they were otherwise healthy. Additional complicating factors are different testing modalities available and used across jurisdictions, including RT-PCR or serology, each with its own limitations with respect to sensitivity and specificity. We have attempted to disentangle duplicate reports. We have used the data extracted to make three comparative estimates for pregnant women with COVID-19 of the risk of the neonate becoming infected: This article is protected by copyright. All rights reserved 2. After breast or formula feeding. 3 . After rooming in with the mother or mother-baby isolation. Other systematic reviews have been published on this topic (8) (9) (10) (11) (12) (13) (14) . Our paper is unique in that we have made a concerted effort to report duplicate reports and have critically analysed the risk of neonatal infection by mode of delivery, infant feeding and mother-infant interaction. A protocol for this study was written once data extraction was underway (Appendix S3). Studies were eligible for inclusion if they, were case reports or case series, of pregnant women with confirmed COVID-19 infection. There was no language restriction. We only included cases where either the mother had confirmed COVID-19 based on a positive swab, or a high clinical suspicion of COVID-19 where a swab had not been taken e.g. symptoms and radiographic evidence in an area of high COVID-19 prevalence. We identified all scientific case reports and case series of confirmed or suspected maternal COVID-19 in pregnancy. The basis of the list was a curated list kept by the senior author on his personal blog since March 22 nd . This (Appendix S1), is a curated list of primary sources based on a daily PubMed search (Appendix S2) supplemented by alerts from colleagues on social media. After April 8 th this list was supplemented by formal daily searches by KO and KW. The search was undertaken between 8 th April to May 2020 through the following electronic bibliographic databases (Medline, Embase and Maternity and Infant Care Database) and citation tracking on relevant studies. The search terms associated with COVID-19 used in bibliographic databases were adapted in database-specific filters. The searches were re-run just before the final analyses and further studies retrieved for inclusion. The date of the last search was 05/06/2020. The search strategy is shown in Appendix S2. For assessing cases of possible vertical transmission we attempted to apply the criteria developed by Shah et al (15) in order to rank the likelihood of vertical transmission to confirmed, probable, possible, unlikely, or not infected. From these we created three tables indicating the rates of baby infection by mode of birth Caesarean or vaginal, rates of infection by breast or formula feeding, and rates by baby rooming in or isolation. Titles and abstracts identified by the search strategy were assessed for inclusion by two reviewers (KW, KO). If there was disagreement about whether a report should be included, full text was obtained for that report. For all potentially eligible studies full text copies were sought, and independently assessed for inclusion by two reviewers (KW, KO). Disagreements were resolved by discussion, and if agreement could not be reached the study was independently assessed by a third reviewer (JGT). Data on study quality and content were extracted onto an Excel spread sheet, and checked (KW, JGT). Where data was missing, the first author of the paper was contacted by email (n=4). Data was collected on maternal and neonatal outcomes, infant feeding, maternal-neonatal interaction and for cases with possible vertical transmission detailed data was collected on virological testing. Each included study was judged for the representativeness of the included mothers to three populations of women: all pregnant women with SARS-CoV2, all pregnant women with COVID-19 (i.e. symptomatic), all pregnant patients with COVID-19 admitted to hospital. We also judged the representativeness of the reported babies to the populations of all babies born to women with Covid-19. The results are shown in Table S2 . We described the flow of studies through the review ( Figure S1 ), with reasons for being removed or excluded, using the PRISMA guidance (16) . Characteristics of each study were described and tabulated. No statistical analyses were anticipated. This article is protected by copyright. All rights reserved Patients were not involved in the development of this research and a core outcome set has not been utilised. The details of the disambiguation of the reports from Chinese hospitals are shown in Table 1 . From the list in Appendix S1 we created a database of studies reporting non-duplicated reports as follows. For studies from Western countries we judged whether cases were likely to be duplicates by reviewing the hospital and time periods of recruitment. If they overlapped we excluded the smaller or less informative report as appropriate. For studies from Wuhan this was complicated by the issue of translating Chinese names and by some hospitals having multiple English names. We therefore disambiguated centres in the city of Wuhan by using the Global Research Identifier Database (GRID) available here https://www.grid.ac/ Accessed [01/05/2020]. From each report we extracted the English name for the hospital in which the patients had been cared for or delivered and entered this in the GRID "disambiguator" and retrieved the hospital GRID identifier. One of the referees who had lived in Wuhan felt that there were mistakes in the GRID database. We therefore invited a co-author WL, who had also worked in Wuhan for some years to join us. He manually checked the initial GRID centre disambiguation and made corrections. Once this manual check was complete, we grouped all reports which included patients delivered in the same hospital and reported the largest series available with useful information. For two hospitals "Wuhan Union Hospital" grid.412839.5, and "Renmin Hospital of Wuhan University" grid.412632.0 we identified two papers where there was internal evidence of non-overlap from which useful data could be extracted. The details are described in the footnotes to Table 1 . For hospitals in cities other than Wuhan without GRID Identifiers we recorded the hospital name as given in the paper and assumed no duplication with Wuhan cases. If the hospital in which patients were treated was not specified in the report, we attempted to deduce this from the affiliations of the first, last or corresponding author. However, it soon became clear that this method led to ambiguous results and added little to the reported identification. Since it was impossible to ascertain whether or not these hospitals were duplicates, they were excluded. Following disambiguation, we included 49 studies from China, USA, Europe, Honduras, Korea, Australia, Peru, Canada, UK and Iran. These studies included 666 neonates and 655 women where information was provided on the mode of delivery and the infant's infection status. Ten women in the included studies underwent Caesarean birth for twins and one woman had a vaginal birth of both twins. The risks of neonatal infection after vaginal and Caesarean birth are shown in Table 2 , of infection after breast or formula feeding or expressed milk in Table 3 , and after rooming in or separation in Table 4 . 28/666 neonates had confirmed COVID-19 infection: full details are provided in Table S1 . Due to a lack of virological testing at birth or in the first 12 hours of life, it was impossible to apply the classification proposed by Shah et al (15) . Only eight had symptoms and of these in four neonates the symptoms may have been related to prematurity. In Table 2 , data is shown on mode of delivery and neonate's infection status for 666 neonates as eleven women delivered twins. Of the 291 women who delivered vaginally, 8/292 (2.7%) neonates were found to be positive for COVID-19. Of the 364 women who had a Caesarean birth, 20/374 (5.3%) neonates were found to be positive for COVID-19. Of the 28 neonates with confirmed COVID-19 infection, 7 were breast fed, 3 formula fed, 1 was given expressed breast milk and in 17 neonates the method of infant feeding was not reported. Overall, of the 666 neonates reviewed, 148 were breastfed, 56 formula fed, 5 given expressed breast milk and for 460 neonates the method of infant feeding was not reported. Of the 28 neonates with confirmed COVID-19 infection, 7 were kept isolated from their mother, 5 were cared for the in the same room as their mother and for 16 neonates it was not reported what approach was taken. Overall 52 neonates were kept isolated from their mother, 107 were cared for the in the same room as their mother and for 502 neonates it was not reported what approach was taken. We have shown firstly that there has been a significant amount of duplicate reporting of cases of COVID-19 from China. Secondly neonatal COVID-19 infection is uncommon, almost never symptomatic, and the rate of infection is no greater when the baby is born vaginally, breastfed or allowed contact with the mother. Very This article is protected by copyright. All rights reserved few infections have been reported in the newborns of COVID-19 positive mothers. Two were reported to have occurred despite isolation from the mother and in two it was not possible to tell what approach was taken to isolation. Some babies were born prematurely and eight infants were stillborn, two twins and two singletons died in the neonatal period but were COVID-19 negative. To date there have been 28 cases (Table S1) published where the possibility for vertical transmission to have occurred have been reported. To confirm definite vertical transmission, it has been proposed that detection of the virus by PCR in either umbilical cord blood, neonatal blood collected within the first 12 hours of birth, or amniotic fluid collected prior to rupture of membranes is needed (15) . In no cases reported to date have these criteria been met although some report negative testing. A few cases deserve special mention: case 9 (Study 79) reports a positive nasopharyngeal swab in the neonate on the day of birth. The authors do not describe any procedure or care taken to clean the infant's oropharynx / mouth/nares / face prior to procuring the swab and we speculate that the presence of the virus may be due to contamination by maternal stool. Of note, the virus was not detected on repeat swab and the infant remained well. The presence of IgG would be maternal, so again not diagnostic. The UKOSS study reports 12/24 cases of possible vertical transmission. Limited information is given for the 12 neonates but 6/12 infants tested positive for COVID-19 within 12 hours of birth. It is unclear what method of testing was used and if this was a nasopharyngeal swab without precautions to clean the infant prior to testing, may again be a result of contamination. In case 23 (Study 103) a positive nasopharyngeal swab in the neonate on the day of birth occurred after careful separation of the baby and cleansing of the baby prior to taking the swab. Newborn infants can be infected in the first few hours of life, but as very few are severely affected it is likely that the benefits of contact with the mother and the ability to breast feed outweigh the potential benefits of separation. For cases where the mother has suspected or confirmed COVID-19 and the baby does not require care on the neonatal unit, guidelines including those in the UK and Canada advise skin-to-skin contact and breastfeeding if the mother uses hand hygiene precautions and (ideally) wears a surgical face mask (5, 17) . The UNICEF guideline strongly recommends breastfeeding for all babies including preterm and sick babies. Our data support such recommendations. Maintaining physical separation of more than two metres at other times is also recommended (17, 18) . Despite having taken steps to remove duplicate reports, the present review is much larger than previous ones. The precision of our estimates is therefore greater. Reassuringly, our data after disambiguation for China agrees broadly with two recent multiple hospital reports from that country, one from Wuhan only (7) and the other from a range of hospitals both inside and outside Wuhan (19) . The studies analysed include a considerable number of case reports and hospital-based series. Such reports have a high risk of being biased towards cases or findings of interest and it is important to reiterate that not all neonates born to COVID-19 positive women were tested for COVID-19 infection. For example, studies may differentially report infected babies, or uninfected babies. However, we are reassured to find that our data are broadly in line with the two regional series reported so far, (Lombardy (20) and Netherlands (https://www.nvog.nl/actueel/registratie-van-covid-19-positieve-zwangeren-in-nethoss/)). It is disappointing that the details of outcome and care of so many neonatal cases born to COVID-19 positive mothers have not been fully reported. This is a missed opportunity to confirm for neonatal and paediatric teams that babies are not likely to be vertically infected. It may be judged likely that babies would have been reported if there was a poor outcome, but the general lack of rigour around taking samples at delivery or in the first few hours of life undermines this conclusion. Authors frequently failed to describe how the baby was looked after, often did not give details of testing, in particular not of the timing and only occasionally were samples reported that were obtained at or shortly after birth. Timings described as 'day 0', 'day 1' and '24 hours' also make it hard to accurately determine when samples were actually taken. We report relatively few data from women with COVID-19 infection acquired postnatally. It is plausible that neonates of such mothers may be at increased risk of infection as they will not have received passive IgG transfer across the placenta. While we have presented the data from a robust search of the literature for 655 women and 666 neonates, this still only includes 28 infected neonates and COVID-19 is a new virus, so we caution the reader to interpret the data in light of this. This article is protected by copyright. All rights reserved The finding of low rates of neonatal infection after caesarean birth are in accord with the very first report of COVID-19 in pregnancy (21). Other systematic reviews have been published on this topic [9-15] and support our contention that vaginal delivery, breast feeding and maternal-infant interaction are safe in the context of COVID-19 disease. Our data suggest that COVID-19 disease should not be an indication for Caesarean birth, formula feeding or isolation of the infant from the mother. Caesareans should continue to be performed for the normal obstetric indications. Mothers who breastfeed and room-in with their infants should continue to observe COVID-19 hygiene precautions and wear a fluid-resistant surgical face mask, if available, while feeding or caring for the baby. There is no evidence that isolating the baby away from the mother is beneficial if such precautions are taken, and encouraging the baby to spend time with its mother is likely to help with breastfeeding and bonding. We recommend that separation only occurs where this is necessary for clinical indications. Although further hospital-based series and case reports will surely be published, better estimates of the risks of neonatal infection after different types of care are likely to come from registry studies which, as far as possible, include all cases in a geographical region or area. Such studies should indicate whether their cases are likely to overlap with other reports by listing the geographic, and hospital sources of their cases. In effort to provide confirmatory evidence on whether vertical transmission occurs in COVID-19, sites seeing infants being born to mothers with COVID-19 should take samples from the mother and baby shortly after birth, as described by Shah et al and report these in the medical literature. Neonatal COVID-19 infection is uncommon, uncommonly symptomatic, and the rate of infection is no greater when the baby is born vaginally, breastfed or allowed contact with the mother. None. Completed disclosure of interest forms are available to view online as supporting information. This article is protected by copyright. All rights reserved No ethical approvals were required for this retrospective study No funding was received for this work Note * Data from S2a and S6 are included despite being from the same hospital as follows. S6 reported 10 Caesarean births and 1 vaginal, with all babies healthy but no further details. S2a reported 3 Caesarean births of which two were at term, and one preterm. All babies healthy and all three pharyngeal swabs negative. We have made the conservative assumption that in total at Union hospital there were 11 mothers of whom 10 were This article is protected by copyright. All rights reserved delivered by caesarean and one vaginally of which one of the Caesareans was preterm. Of the 11 babies, three, including the preterm one, were negative and eight were not tested. Note ** Cases from S36 and S37 are included despite being delivered from the same hospital because S36 includes 17 women all delivered by Caesarean, and S37 three women all delivered vaginally. TOTAL 148 7 139 2 0 56 3 39 14 2 460 17 396 28 2 5 1 4 0 0 Note: Study 43 excluded from Table 3 This article is protected by copyright. All rights reserved Pregnancy and perinatal outcomes of women with severe acute respiratory syndrome Pregnancy: What obstetricians need to know Middle East Respiratory Syndrome Coronavirus Report of two cases & review of the literature Rate of caesarean section is alarming in China Coronavirus (COVID-19) Infection in Pregnancy. Information for Health Care Professionals Editorial Concern-Possible Reporting of the Same Patients With COVID-19 in Different Reports Clinical Characteristics of Pregnant Women with Covid-19 in Wuhan, China Coronavirus in pregnancy and delivery: rapid review. Ultrasound in obstetrics & gynecology : the official journal of the International Society of Ultrasound in Obstetrics and Gynecology Maternal and perinatal outcomes with COVID-19: A systematic review of 108 pregnancies Outcome of Coronavirus spectrum infections (SARS, MERS, COVID 1 -19) during pregnancy: a systematic review and meta-analysis A systematic scoping review of COVID-19 during pregnancy and childbirth COVID19 during pregnancy: a systematic review of reported cases Coronavirus disease 2019 (COVID-19) and pregnancy: a systematic review Vertical Transmission of Coronavirus Disease 19 (COVID-19) from Infected Pregnant Mothers to Neonates: A Review Classification system and case definition for SARS-CoV-2 infection in pregnant women, fetuses, and neonates Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement Breastfeeding when mothers have suspected or proven COVID-19 Statement on infant feeding on neonatal units during the coronavirus (COVID-19) outbreak 2020 Coronavirus disease 2019 (COVID-19) in pregnant women: A report based on 116 cases Vaginal delivery in SARS-CoV-2 infected pregnant women in Northern Italy: a retrospective analysis Clinical characteristics and intrauterine vertical transmission potential of COVID-19 infection in nine pregnant women: a retrospective review of medical records This article is protected by copyright. All rights reserved Accepted Article