To assess the impact of separation of SARS-CoV-2 PCR-positive mother–newborn dyads on breastfeeding outcomes.
This is an observational longitudinal cohort study of SARS-CoV-2 PCR-positive mothers and their infants at three NYU Langone Health hospitals from March 25, 2020 through May 30, 2020. Mothers were surveyed by telephone regarding pre-delivery feeding plans, in-hospital feeding, and home feeding of their neonates. Any change prompted an additional question to determine whether this change was due to COVID-19.
Of the 160 mother–newborn dyads, 103 mothers were reached by telephone, and 85 consented to participate. No significant difference was observed in pre-delivery feeding plan between the separated and unseparated dyads (P = .268). Higher rates of breastfeeding were observed in the unseparated dyads compared with the separated dyads in the hospital (p<0.001), and at home (p=0.012). Only two mothers in each group reported expressed breast milk as the hospital feeding source (5.6% of unseparated vs 4.1% of separated). COVID-19 was more commonly cited as the reason for change among the separated compared with the unseparated group (49.0% vs 16.7%, p<0.001). When dyads were further stratified by symptom status into four groups (asymptomatic separated, asymptomatic unseparated, symptomatic separated, and symptomatic unseparated), results remained unchanged.
In the setting of COVID-19, separation of mother–newborn dyads impacts breastfeeding outcomes, with lower rates of breastfeeding both during hospitalization and at home following discharge compared with unseparated mothers and infants. No evidence of vertical transmission was observed; one case of postnatal transmission occurred from an unmasked symptomatic mother who held her infant at birth.
] and American Academy of Pediatrics (AAP)[
] each published interim guidelines for management of neonates born to mothers with confirmed or suspected COVID-19, including recommendations for temporary separation of these dyads. Given a lack of evidence demonstrating SARS-CoV-2 transmission in breastmilk, both the AAP and CDC recommended expression of breastmilk after meticulous hand hygiene, and feeding of the expressed milk to separated neonates by designated caregivers.[
] In contrast, the World Health Organization (WHO) issued guidance supporting direct breastfeeding for all mothers with COVID-19, both asymptomatic and symptomatic, except in cases of severe illness or other complication that would inhibit care of the infant or interfere with breastfeeding.[
Recognizing the paucity of evidence, and with a goal of limiting exposure of neonates, the NYU Langone Health system (NYULH) issued early internal guidance recommending separating these mother–newborn dyads at birth. In line with the recommendations of the AAP and CDC, the NYULH guidelines advocated expression of breastmilk for mothers intending to breastfeed, with bottle-feeding by designated caregivers.
] Six weeks after our initial local guidelines were published, they were modified to allow asymptomatic, SARS-CoV-2 PCR-positive mothers to room-in with their infants, while wearing masks, and employing proper hand hygiene techniques. Additionally, our local policy changed to allow and encourage mothers to feed their infants directly on the breast, if desired.
Given the potential impact of policies on transmission, health, and breastfeeding behavior, we recognize the importance of validating policies, with the goal of informing future guidance. To assess the impact of our policy change surrounding mother–newborn dyad separation on breastfeeding rates, we evaluated mothers’ pre-delivery plans for feeding, and compared these with actual outcomes of breastfeeding during perinatal admission and following discharge.
Dyads were identified by NYULH Datacore services and were included in the study if all of the following inclusion criteria were met: maternal age of 18 years or more, positive maternal SARS-CoV-2 PCR test by nasopharyngeal swab, and SARS-CoV-2 PCR test by nasopharyngeal swab performed on the infant (regardless of test result). Background demographic and clinical data for these dyads was obtained through the EPIC electronic medical record system and stored, de-identified, in a secure database. Maternal baseline characteristics included age, ethnicity, race, gravidity, parity, type of delivery, reason for delivery, health status after delivery, symptoms of COVID-19, medications for COVID-19, and contraindication to breastfeeding. Neonatal characteristics included gestational age, sex, anthropometric measurements at birth, APGAR scores, admission to newborn nursery or neonatal intensive care unit (NICU), temperature during hospitalization, presence of comorbidities including respiratory distress or temperature derangements, and timing of SARS-CoV-2 nasopharyngeal swab testing. Additional baseline data collected included whether a lactation consultation was obtained, the type of isolation precautions used, and the type of separation of the dyad.
Mothers were contacted by telephone during the period from May 27, 2020 through June 17, 2020 by one of the investigators to obtain consent and authorization for voluntary participation in the telephone study. Three attempts were made to contact each mother. If contact was made, and the mother consented to participate, the investigator proceeded to ask how she had planned to feed her infant prior to delivery, how the infant had been fed during hospitalization following delivery, and how her infant had been fed since discharge from the hospital. For each question, the following four answer choices were offered: breastfeeding, expressed breastmilk, formula, or mixed feeding. If a change in feeding type between pre-delivery plan, hospital feeding, or home feeding was identified, the mother was asked about the reason for change, and whether this change was due to COVID-19.
] product limit curves, where group (NICU and newborn nursery) was the stratification variable. No data was considered censored. The two groups were compared with the log-rank test. The median total LOS was obtained from the Kaplan-Meier/Product-Limit Estimates and their corresponding 95% confidence intervals were computed using the Greenwood formula[
] to calculate the standard error. A result was considered statistically significant at the p
Table 1Maternal and infant characteristics (n = 160 unless specified)
Continuous data are presented as the mean ± standard deviation (median).
Categorical data are presented as frequency (percent).
Discrete data are presented as median [interquartile range].
Table 2Telephone survey responses by separation
Telephone respondents are presented as frequency (percent).
Table 3Telephone survey responses by separation and symptom status
Telephone respondents are presented as frequency
In this study, we found that SARS-CoV-2 infection has a significant impact on mother–newborn dyads with respect to breastfeeding outcomes, both in the hospital setting and at home. We found a statistically significant lower rate of breastfeeding among separated dyads compared with unseparated dyads. Importantly, we found no clinical evidence of vertical or horizontal transmission from asymptomatic mothers to their infants. One case of likely postnatal transmission occurred from a symptomatic mother to her neonate; the infant was found to be SARS-CoV-2 PCR positive on a nasopharyngeal swab performed on day of life five, after testing negative at birth.
] Similarly, a case series of ten neonates (including one set of twins) born to nine mothers in China reported negative SARS-CoV-2 nucleic acid testing performed on pharyngeal swabs for all ten neonates.[
] A cohort study in China described 33 infants born to mothers with COVID-19; of the 33, three were found to have early onset infection with SARS-CoV-2, with positive nasopharyngeal and anal swabs on days of life two and four.[
] The authors suggested that in light of the strict infection control measures in place during these neonates’ deliveries, vertical transmission could not be ruled out as the source of the neonates’ SARS-CoV-2 infection.[
] Importantly, these infants were not tested before day of life two, and the infection control measures in place were not described, raising the possibility of horizontal, rather than vertical transmission. Another case report, from Iran, described a 15-day-old neonate who came to attention with fever and lethargy after his mother exhibited symptoms consistent with COVID-19; the infant tested positive for SARS-CoV-2 by reverse-transcriptase PCR testing, suggesting possible horizontal transmission.[
] and AAP,[
] and no distinction was made between symptomatic and asymptomatic mothers; separation at birth was recommended for all infants born to mothers with positive SARS-CoV-2 PCR, regardless of symptoms.
] With a continued lack of evidence suggesting substantial transmission of SARS-CoV-2 via breastmilk, our policy was modified on April 20, 2020, to allow asymptomatic, SARS-CoV-2 PCR-positive mothers to room-in with their infants. Furthermore, our new policy allowed asymptomatic mothers to breastfeed while wearing masks and using strict hand hygiene. This change echoed the WHO guidance supporting direct breastfeeding, but unlike the WHO guidelines, which recommend direct breastfeeding also for symptomatic mothers, our new guidelines limit contact between symptomatic mothers and their newborns, and continue to support expression of breastmilk and bottle feeding by designated caregivers.[
] Data on impact of separation of infected mother–newborn dyads on breastfeeding outcomes has been wanting. In a commentary outlining the key literature opposing separation of mother–newborn dyads, the authors highlighted absence of evidence demonstrating the negative effect of separation during the COVID-19 pandemic.[
] The AAP Section on Neonatal-Perinatal Medicine (SONPM) is currently collecting data in a national registry about SARS-CoV-2 positive mothers and their infants, with a goal of studying transmission of the virus and outcomes of these neonates, including the impact of infection control policies including dyad separation.
Our study provides evidence that in the setting of the COVID-19 pandemic, separation of asymptomatic mother–newborn dyads has significant negative impact on breastfeeding outcomes. Our findings suggest that separation of mother–newborn dyads results in lower rates of breastfeeding both during hospitalization and at home following discharge, and higher rates of formula feeding as a substitute. Higher rates of mixed feeding type (breastfeeding, expressed breastmilk, and formula) were observed in the unseparated dyads compared with the separated dyads, suggesting that even when formula supplementation is utilized, rooming-in is associated with higher rates of being fed any breastmilk, which persisted beyond hospitalization. Many mothers reported that once reunited with their infants after separation, attempts at breastfeeding were frequently unsuccessful due to difficulty latching, and infant’s preference for bottle-feeding.
Although a significant difference was observed in the percentage of mothers reporting expressed breast milk as the sole feeding type during hospitalization (4.1% of unseparated vs. 5.6% of separated), the overall number of mothers utilizing expressed breastmilk as the feeding type was small. When considering the rates of mixed feeding, the overall rates of expressed breast milk were likely higher, as 14.3% of separated mothers, and 44.4% of unseparated mothers reported a mix of feeding types (breastfeeding, expressed breast milk, and formula) during hospitalization. Nevertheless, when the rate of mixed feeding for each group is compared with the same group’s rate of formula feeding (81.6% of separated mothers and 27.8% of unseparated mothers), it becomes evident that separated dyads had lower rates of breastmilk expression, irrespective of formula supplementation. Promotion of breastmilk expression for mothers separated from their infants due to COVID-19 is emphasized as a goal in all of the published guidelines, and our failure to do so highlights an opportunity for intervention and improvement in our support of mothers with COVID-19. Although there was no significant difference in the rate of lactation consultation utilization between the separated and unseparated dyads (40.4% vs. 40.6% respectively), perhaps this illuminates a potential opportunity for increased provision of lactation services to separated dyads in the future.
Notably, only one infant in our cohort tested positive for SARS-CoV-2 during hospitalization. The infant’s initial nasopharyngeal swab at birth was negative for SARS-CoV-2, but became positive upon repeat swab test on day of life five. Fifteen neonates in our study exhibited symptoms of fever, respiratory distress, feeding intolerance, rhinorrhea, hypothermia, or a combination of these. Although the one infant who tested positive on day of life five experienced fever, thought attributable to neonatal abstinence syndrome, the infant was otherwise asymptomatic with regards to COVID-19. The remaining 14 neonates’ symptoms were largely attributed to prematurity or environmental causes. We did not assess the impact of neonates’ symptoms, nor the impact of NICU admission on breastfeeding outcomes, and suggest that perhaps these could be the focus of future studies.
Several limitations should be noted. First, variation exists with regard to demographic characteristics of each of the three hospitals included, notably in the different baseline rates of exclusive breastfeeding among the three hospital sites. Additionally, early on, maternal testing protocols varied among the three sites. At WH, universal screening of all delivering mothers was instituted early on, with a focus on cohorting mothers for rooming purposes based on test results due to a limited number of single-occupancy rooms in the mother–baby unit. Both TH and BH initially tested only symptomatic mothers, but both have subsequently initiated universal screening protocols. These variations we addressed by pooling data from the three sites together.
The last infant respiratory sample taken for SARS-CoV-2 PCR testing was at a mean of 50 hours and likely would not reflect horizontal transmission. Further data were restricted to maternal questioning at the time of phone survey. Our relatively small sample size does not exclude low rates of infant acquisition of SARS-CoV-2 illness.
World Health Organization. WHO Director-General’s opening remarks at the media briefing on COVID-19 – 11 March 2020. World Health Organization Speeches 2020. https://www.who.int/dg/speeches/detail/who-director-general-s-opening-remarks-at-the-media-briefing-on-covid-19—11-march-2020.
New York State Department of Health. New York State Statewide COVID-19 Testing Dataset. Health Data NY 2020. https://health.data.ny.gov/Health/New-York-State-Statewide-COVID-19-Testing/xdss-u53e (accessed June 19, 2020).
New York State Department of Health. New York State DOH COVID-19 Tracker n.d. https://covid19tracker.health.ny.gov/views/NYS-COVID19-Tracker/NYSDOHCOVID-19Tracker-Map?%3Aembed=yes&%3Atoolbar=no&%3Atabs=n (accessed June 19, 2020).
National Center for Immunization and Respiratory Diseases (U.S.). Division of Viral Diseases., editor. Interim considerations for infection prevention and control of coronavirus disease 2019 (COVID-19) in inpatient obstetric healthcare settings 2020.
Puopolo K, Hudak M, Kimberlin D, Cummings J. Initial guidance: management of infants born to mothers with COVID-19 2020.
Centers for Disease Control and Prevention. Interim considerations for infection prevention and control of coronavirus disease 2019 (COVID-19) in inpatient obstetric healthcare settings 2020.
World Health Organization. Clinical management of severe acute respiratory infection (SARI) when COVID-19 disease is suspected: interim guidance, 13 March 2020. Geneva: World Health Organization; 2020.
Breastfeeding and the Use of Human Milk.
Pediatrics. 2012; 129: e827-e841https://doi.org/10.1542/peds.2011-3552
Factors Associated with Breastfeeding Initiation and Continuation: A Meta-Analysis.
The Journal of Pediatrics. 2018; 203 (): 190-196https://doi.org/10.1016/j.jpeds.2018.08.008
New York State Department of Health. Hospital Maternity-Related Procedures and Practices Statistics 2014. https://www.health.ny.gov/statistics/facilities/hospital/maternity/ (accessed June 24, 2020).
Lee ET. Statistical Methods for Survival Data Analysis. Second Edition. New York: Wiley; 1992.
Greenwood M. The Errors of Sampling of the Survivorship Table. Reports on Public Health and Medical Subjects, vol. Vol 33, London: Her Majesty’s Stationery Office; 1926.
Clinical characteristics and intrauterine vertical transmission potential of COVID-19 infection in nine pregnant women: a retrospective review of medical records.
The Lancet. 2020; 395: 809-815https://doi.org/10.1016/S0140-6736(20)30360-3
Clinical analysis of 10 neonates born to mothers with 2019-nCoV pneumonia.
Transl Pediatr. 2020; 9: 51-60https://doi.org/10.21037/tp.2020.02.06
Neonatal Early-Onset Infection With SARS-CoV-2 in 33 Neonates Born to Mothers With COVID-19 in Wuhan, China.
JAMA Pediatr. 2020; https://doi.org/10.1001/jamapediatrics.2020.0878
Novel coronavirus in a 15-day-old neonate with clinical signs of sepsis, a case report.
Infectious Diseases. 2020; 52: 427-429https://doi.org/10.1080/23744235.2020.1747634
Should Infants Be Separated from Mothers with COVID-19? First, Do No Harm.
Breastfeeding Medicine. 2020; 15: 351-352https://doi.org/10.1089/bfm.2020.29153.ams
Management of newborns exposed to mothers with confirmed or suspected COVID-19.
J Perinatol. 2020; 40: 987-996https://doi.org/10.1038/s41372-020-0695-0
Managing COVID-19-Positive Maternal–Infant Dyads: An Italian Experience.
Breastfeeding Medicine. 2020; 15: 347-348https://doi.org/10.1089/bfm.2020.0095
Breastfeeding during the COVID-19 pandemic: Suggestions on behalf of woman study group of AMD.
Diabetes Research and Clinical Practice. 2020; 165: 108239https://doi.org/10.1016/j.diabres.2020.108239
When Separation is not the Answer: Breastfeeding Mothers and Infants affected by COVID‐19.
Matern Child Nutr. 2020; https://doi.org/10.1111/mcn.13033
Centers for Disease Control and Prevention. Evaluation and Management Considerations for Neonates At Risk for COVID-19 2020. https://www.cdc.gov/coronavirus/2019-ncov/hcp/caring-for-newborns.html.
Received in revised form:
In Press Accepted Manuscript
The authors declare no conflicts of interest.
© 2020 Elsevier Inc. All rights reserved.