key: cord-303407-n7j56sci
authors: Popofsky, Stephanie; Noor, Asif; Leavens-Maurer, Jill; Quintos-Alagheband, Maria Lyn; Mock, Ann; Vinci, Alexandra; Magri, Eileen; Akerman, Meredith; Noyola, Estela; Rigaud, Mona; Pak, Billy; Lighter, Jennifer; Ratner, Adam J.; Hanna, Nazeeh; Krilov, Leonard
title: Impact of Maternal SARS-CoV-2 Detection on Breastfeeding Due to Infant Separation at Birth
date: 2020-08-10
journal: J Pediatr
DOI: 10.1016/j.jpeds.2020.08.004
sha: 
doc_id: 303407
cord_uid: n7j56sci

OBJECTIVE: To assess the impact of separation of SARS-CoV-2 PCR-positive mother–newborn dyads on breastfeeding outcomes. STUDY DESIGN: 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. RESULTS: 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. CONCLUSION: 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.

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 J o u r n a l P r e -p r o o f 3 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.

The novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), responsible for coronavirus disease-2019 (COVID- 19) , spread globally, reaching pandemic status on March 11, 2020. [1] Cases of COVID-19 in New York State reached a peak in April, 2020, with over 386,000 cases and 24,000 deaths by June, 2020. [2, 3] Early in the COVID-19 pandemic, limited data existed regarding the risk of adverse outcomes for pregnant women infected with SARS-CoV-2, and the risk of vertical or horizontal transmission to their newborns was unknown. Given the uncertainty surrounding potential transmission from infected mother to neonate, early guidance relied on a cautious approach and recommended separation of mother-newborn dyads to minimize the risk of transmission. The Centers for Disease Control and Prevention (CDC) [4] and American Academy of Pediatrics (AAP) [5] 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. [5, 6] 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. [7] 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 J o u r n a l P r e -p r o o f and CDC, the NYULH guidelines advocated expression of breastmilk for mothers intending to breastfeed, with bottle-feeding by designated caregivers.

We recognize the importance of breastfeeding, and advocate for supportive environments and policies to facilitate breastfeeding, including early skin-to-skin contact and rooming-in. [8, 9] 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.

For this observational longitudinal cohort study, we studied mother-newborn dyads at three NYU Langone Health hospitals from March 25, 2020 through May 30, 2020. The Institutional Review Board of NYU Langone Health approved this study. Tisch Hospital (TH) is a university-based tertiary and quaternary hospital in Manhattan with over 6,000 births annually, NYU-Winthrop Hospital (WH) is a tertiary hospital in Nassau County (a suburb of New York City) with over 5,000 annual births, and NYU Langone Hospital-Brooklyn (BH) is an academic hospital in Brooklyn with over 4,000 annual births. [10] All three hospitals are designated as J o u r n a l P r e -p r o o f Baby-Friendly Hospitals through the Baby Friendly Hospital Initiative, led jointly by the World Health Organization and UNICEF, and both TH and WH are designated by New York State as regional perinatal centers. The published baseline breastfeeding rates of infants being fed any breastmilk and exclusive breastfeeding during hospitalization are as follows: 97.7% and 89.8%, respectively at TH; 85.5% and 44.2%, respectively at WH; and 89.8% and 38.6%, respectively at BH. [10] 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 J o u r n a l P r e -p r o o f 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 predelivery 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.

Descriptive statistics (mean ± standard deviation or median [25th, 75th percentiles] for continuous variables; frequencies and percentages for categorical variables) were calculated for the sample of mothers and neonates separately. The chi-square test or Fisher exact test, as deemed appropriate, was used to compare those who were separated from those who were not separated for categorical variables. The analysis of total length of stay (LOS) was accomplished by applying standard methods of survival analysis, i.e., computing the Kaplan-Meier [11] 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 [12] to calculate the standard error. A result was considered statistically significant at the p<0.05 level of significance. All analyses were performed using SAS version 9.4 (SAS Institute Inc., Cary, NC).

A combined total of 160 mother-newborn dyads from the three hospitals met inclusion criteria and were included in the baseline characteristic cohort. Of these, 80 dyads were identified at WH, 33 dyads were identified at TH, and 47 dyads were identified at BH. Maternal baseline characteristics are included in Table I . Mean maternal age was 30.8 years, and 59 (36.9%) of the mothers were symptomatic during perinatal hospitalization, with fever (over 37.7°C), cough, shortness of breath, or a combination of these. Twenty-five mothers (15.6%) had been symptomatic prior to the perinatal admission but were no longer symptomatic during hospitalization. In total, 149 (93.1%) of the mothers were characterized as being well for breastfeeding following delivery, whereas 11 (6.9%) were ill and unable to breastfeed. Overall, 148 (92.5%) deliveries were expected, and the remaining 12 were preterm deliveries due to various indications; 120 (75%) infants were born via spontaneous vaginal delivery, 38 (23.7%) were born via cesarean delivery due to various indications, and two (1.3%) were born extramurally. A lactation consultation was initiated for 64 (40%) mothers, of which 38 received lactation consultation services during hospitalization. Only one mother had a contraindication for breastfeeding, due to maternal opioid dependence and neonatal abstinence syndrome.

Neonatal baseline characteristics are presented in Table 1 . Fifteen (9.4%) neonates required resuscitation with positive pressure ventilation; 145 (90.6%) of the newborns were admitted to the newborn nursery. Among the 15 (9.4%) neonates who were symptomatic, four had fever, 11 showed respiratory distress, four experienced feeding intolerance, one exhibited rhinorrhea, and seven had hypothermia. At the time of writing, one neonate remains hospitalized in the NICU, with the remainder having all been discharged. For neonates admitted to the newborn nursery, the median length of stay was 2 days (95% CI 1-2), compared with 3 days J o u r n a l P r e -p r o o f (95% CI [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] for those admitted to the NICU. Only one infant had a positive SARS-CoV-2 PCR test (on day of life five, after a negative test at birth); the remainder had negative tests throughout. This infant was held by an unmasked symptomatic mother immediately after birth.

Telephone contact was made with 103 (64.4%) mothers. Of these, 85 (82.5%) consented to participation in the telephone survey. Date of telephone call ranged from a minimum of 10 days after birth to a maximum of 77 days after birth, with a median average of 45 days. Survey responses are presented in the Figure. In total, 30 mothers (35.3%) indicated that COVID-19, and specifically the separation and subsequent difficulty latching, was the cause for change in feeding plan from pre-delivery to hospital or home. No change in feeding occurred from predelivery plan to hospital or home feeding for 23 mothers (27.1%).

There was no statistically significant difference in pre-delivery feeding plan between the separated and unseparated dyads (p=0.268) ( Table 2 ). Hospital feeding type differed significantly for every single feeding type between the separated and unseparated dyads (p<0.001), with higher rates of breastfeeding among unseparated dyads compared with separated dyads (22.2% vs 0%) and higher rates of formula feeding among separated dyads compared with unseparated dyads (81.6% vs 27.8%) ( Table 2 ). Only two mothers in each group reported using expressed breast milk as the sole feeding source during hospitalization (5.6% in the unseparated group vs. 4.1% in the separated group). A higher percentage of mothers in the unseparated group reported a mix of feeding types, with a combination of breastfeeding, expressed breast milk, and formula, compared with the separated group (44.4% vs 14.3%). Home feeding type also differed significantly for every single feeding type between the separated and unseparated dyads (p=0.012), again with higher rates of breastfeeding among unseparated dyads compared with J o u r n a l P r e -p r o o f separated dyads (27.8% vs 12.2%) and higher rates of formula feeding among separated dyads compared with unseparated dyads (34.7% vs 8.3%) ( Table 2) . Again, the unseparated mothers reported higher rates of expressed breastmilk than the separated mothers (5.6% vs 2.0%), as well as higher rates of mixed feeding (58.3% vs 51.0%). Reason for change in feeding type from predelivery plan to hospital feeding type and/or home feeding type differed significantly between the two groups (p<0.001), with a higher percentage reporting a change due to COVID-19 among the separated group compared with the unseparated group (49.0% vs 16.7%). There was no difference in the rate of lactation consultation between the separated vs. unseparated dyads (40.4% vs. 40.6% respectively, p<0.98).

When the separated and unseparated dyad groups were further stratified into four groups by symptom status (asymptomatic separated, asymptomatic unseparated, symptomatic separated, and symptomatic unseparated), the results remain unchanged. No statistical differences were observed among any of the four groups for pre-delivery feeding plan (p=0.698) ( Table 3) .

Hospital feeding type differed significantly for every single feeding type among all four groups of dyads (p<0.001), with the highest rate of breastfeeding in the asymptomatic unseparated group (22.6%), and the highest rate of formula feeding in the symptomatic separated group (83.3%) (Table 3) . Home feeding type also differed significantly for every single feeding type among all four groups of dyads (p=0.018), with the highest rate of breastfeeding in the asymptomatic unseparated group (22.6%), and the highest rate of formula feeding in the asymptomatic separated group (36.8%) ( Table 3 ). The highest rate of mixed feeding was observed among the asymptomatic unseparated group, both during hospitalization (45.2%), and at home (64.5%).

Reason for change in feeding type from pre-delivery plan to hospital feeding type and/or home feeding type differed significantly among the four groups (p<0.001). A higher percentage J o u r n a l P r e -p r o o f reported a change due to COVID-19 in the asymptomatic separated group (57.9%) and the symptomatic separated group (43.3%) compared with both the asymptomatic unseparated group (16.1%) and the symptomatic unseparated group (20.0%).

In this study, we found that SARS-CoV-2 infection has a significant impact on mothernewborn 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.

Early published data during the COVID-19 pandemic highlighted variable disease severity and outcomes for neonates and underscored a lack of clear evidence or understanding of transmission surrounding spread from infected mothers to their infants, either through vertical or horizontal transmission. A case series of nine pregnant women with confirmed COVID-19 during the third trimester of pregnancy in China suggested that in utero, vertical transmission of SARS-CoV-2 to neonates did not occur, as samples of amniotic fluid, cord blood, and neonatal throat swabs tested at birth for SARS-CoV-2 were all negative. [13] 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. [14] 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 J o u r n a l P r e -p r o o f days of life two and four. [15] 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. [15] 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. [16] During a tumultuous period with rapid spread of SARS-CoV-2 and inconclusive evidence to guide evolving practices surrounding childbirth and post-partum neonatal care, our institution implemented guidelines supporting separation of mother-newborn dyads with the goal of limiting exposure and infection of neonates. Our guidelines mirrored those of the CDC [4] and AAP, [5] 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.

As time progressed through the pandemic, NYULH frequently reviewed and revised our policies to address the needs of our patients and to reflect the most up-to-date knowledge and evidence surrounding COVID-19. It became evident that separation of mother-newborn dyads was particularly stressful for many mothers and their newborns, and the impact of separation on breastfeeding could be harmful. [17] 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 J o u r n a l P r e -p r o o f 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. [7] At the time of writing, several centers around the world have published their experiences and recommendations surrounding management of SARS-CoV-2 positive mother-newborn dyads during the pandemic. [18] [19] [20] Data on impact of separation of infected mother-newborn dyads on breastfeeding outcomes has been wanting. In a commentary outlining the key literature 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 J o u r n a l P r e -p r o o f 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 bottlefeeding.

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.

Although our local revised guidelines support rooming-in only for asymptomatic dyads, some symptomatic mothers still favor rooming-in after education about the risks of transmission.

The sample size for this group of symptomatic unseparated dyads is small (n=5), but when the separated and unseparated groups were further stratified to account for symptom status, our findings still demonstrate significant differences in feeding type both in the hospital and at home, with higher rates of breastfeeding among the unseparated dyads, and higher rates of formula feeding among the separated groups. The risks of transmission must always be weighed against the impact on breastfeeding as a result of separation. In keeping with the CDC recommendations highlighting the risk of transmission through respiratory droplets from symptomatic mothers, [22] we continue to separate dyads in cases where the mother is symptomatic with cough. 

World Health Organization. WHO Director-General's opening remarks at the media briefing on COVID-19 -11

New York State Statewide COVID-19 Testing Dataset. Health Data NY 2020

Interim considerations for infection prevention and control of coronavirus disease 2019 (COVID-19) in inpatient obstetric healthcare settings

Initial guidance: management of infants born to mothers with COVID

Interim considerations for infection prevention and control of coronavirus disease 2019 (COVID-19) in inpatient obstetric healthcare settings

Clinical management of severe acute respiratory infection ( SARI) when COVID-19 disease is suspected: interim guidance

Breastfeeding and the Use of Human Milk

Factors Associated with Breastfeeding Initiation and Continuation: A Meta-Analysis

Hospital Maternity-Related Procedures and Practices Statistics

Statistical Methods for Survival Data Analysis

The Errors of Sampling of the Survivorship Table

Clinical characteristics and intrauterine vertical transmission potential of COVID-19 infection in nine pregnant women: a retrospective review of medical records

Clinical analysis of 10 neonates born to mothers with 2019-nCoV pneumonia

Neonatal Early-Onset Infection With SARS-CoV-2 in 33 Neonates Born to Mothers With COVID-19 in Wuhan, China

Novel coronavirus in a 15-day-old neonate with clinical signs of sepsis, a case report

Should Infants Be Separated from Mothers with COVID-19? First, Do No Harm

Management of newborns exposed to mothers with confirmed or suspected COVID-19

Managing COVID-19-Positive Maternal-Infant Dyads: An Italian Experience

Breastfeeding during the COVID-19 pandemic: Suggestions on behalf of woman study group of AMD

When Separation is not the Answer: Breastfeeding Mothers and Infants affected by COVID 19

Centers for Disease Control and Prevention. Evaluation and Management Considerations for Neonates At Risk for COVID-19 2020