Association of Maternal Perinatal SARS-CoV-2 Infection With Neonatal Outcomes During the COVID-19 Pandemic in Massachusetts

This cohort study examines the risks, incidence, and outcomes of SARS-CoV-2 among neonates who were exposed to the virus within 2 weeks before and after birth.


Introduction
Biologically plausible routes of perinatal SARS-CoV-2 transmission include transplacental, contact with infected secretions during delivery and with respiratory droplets after delivery, and breast milk.
Low rates of virus positivity in relevant biological specimens [1][2][3] suggest that perinatal transmission is uncommon, but accumulating evidence indicates that some neonates who are born to mothers with SARS-CoV-2 do obtain positive test results for the virus. 1 Systematic reviews of case series of mothers with SARS-CoV-2 reported a neonatal test result positivity rate of 3.1% to 9.1%, [4][5][6] but such reviews included mainly small studies from single centers, limiting the generalizability of the findings and raising the possibility of selection and/or publication bias. 7Larger multihospital cohorts from New York City included up to 149 mothers with SARS-CoV-2 and reported a test result positivity rate of less than 1% in neonates. 8,9Overall, considerable uncertainty remains about the true incidence of neonatal test result positivity, which can serve as a proxy for perinatal transmission.
Given the low incidence of neonatal test result positivity for SARS-CoV-2, identifying the risk factors associated with its occurrence is challenging.Mothers who present with COVID-19 symptoms may have higher SARS-CoV-2 viral loads and be more likely to transmit the virus than mothers who are identified by routine screening. 10Clinical practices that were initially recommended to reduce perinatal transmission included elective cesarean delivery, mother-newborn separation, and temporary suspension of breastfeeding. 11Subsequently, based on accumulating evidence, the American Academy of Pediatrics promoted rooming-in and breastfeeding with precautions. 12In addition to clinical factors, social adversity may play a role given that the COVID-19 pandemic has disproportionately affected racial/ethnic minority populations.The specific pathways by which social disadvantage might affect mother-to-child transmission of SARS-CoV-2 include differential access to care and clinician bias. 13Discrimination also may be a factor in chronic stress, which diminishes antiviral immune responses. 14,15To our knowledge, no epidemiologic study to date has examined the risk factors for neonatal test result positivity.
Several studies have reported short-term health outcomes for neonates who were born to mothers with SARS-CoV-2, including the need for neonatal intensive care unit admission and respiratory support, 16,17 but little is known about the specific factors associated with neonatal clinical or viral testing outcomes beyond the birth hospital discharge.One New York City hospital routinely followed perinatally exposed newborns and found that none had clinically significant signs of SARS-CoV-2 infection and 6 had negative test results. 18Also in New York City, a 3-hospital cohort study of 120 neonates who were exposed to SARS-CoV-2 reported 0 positive result from nasopharyngeal swabs taken at 5 to 7 days and 14 days of life during follow-up, and no clinical concerns in telehealth evaluations at 1 month of age were reported. 9Although these previous studies did not suggest that substantial clinical illness in newborns commonly followed perinatal exposure to SARS-CoV-2, more data are needed.(2) indicators of adverse health during the birth hospitalization, including preterm birth (<37 weeks) or low birth weight (<2500 g), very preterm birth (<32 weeks) or very low birth weight (<1500 g), delivery room resuscitation (positive pressure ventilation, intubation, and/or chest compressions), continuous positive airway pressure (CPAP) or mechanical ventilation, therapeutic hypothermia, and length of stay; and (3) clinical signs and viral testing, which were identified through EMR documentation of nonroutine health care visits through 30 days after hospital discharge.Test result positivity for SARS-CoV-2 was defined as at least 1 positive result on a specimen obtained by nasopharyngeal swab using a polymerase chain reaction-based method.

Statistical Analysis
We performed a comparison between neonates with any positive SARS-CoV-2 test results and those with negative test results on the basis of sociodemographic characteristics, pregnancy and delivery factors, and newborn care practices while accounting for hospital clustering using Cochran-Mantel-Haenszel χ 2 tests for categorical variables.A 2-tailed P < .05 was considered statistically significant.
For multivariable analyses, we selected covariates for parsimonious models according to a priori hypotheses and associations observed in unadjusted analyses, and we used mixed-effects logistic regression models to account for hospital clustering.Because of the small number of neonates with positive SARS-CoV-2 test results, we used a robust sandwich estimator to estimate odds ratios (ORs) with 95% CIs.
We hypothesized that the following variables would increase the risk of the neonatal test result positivity: maternal symptomatic SARS-CoV-2 infection, vaginal delivery mode, any rooming-in practice, Black race or Hispanic ethnicity, language status (non-English vs English), and SVI of 90th percentile or higher.Furthermore, we hypothesized that delivery prompted by worsening maternal SARS-CoV-2 infection-related clinical status would be associated with indicators of adverse neonatal health outcomes.
If outcome data were missing, the participant was removed from the analysis.We used SAS, version 9.4 (SAS Institute Inc) for all analyses.

Birth Hospital Data
From the 11 participating hospitals, we identified 255 neonates who were born to 250 mothers with positive SARS-CoV-2 test results.Maternal demographic characteristics are outlined in Table 1.The mean (SD) maternal age was 30.4 (6.3) years, and 121 (48.4%) were Hispanic mothers of any race, 50 (20.0%)were non-Hispanic White mothers, and 46 (18.4%) were non-Hispanic Black mothers.For the zip code-derived SVI, 68 mothers (27.2%) had an overall score of 90th percentile or higher.A total of 170 mothers (68.0%) were asymptomatic when tested for SARS-CoV-2, which was a consequence of the implementation of universal surveillance testing.Of the 79 women with symptomatic COVID-19, 23 (29.1%) required hospitalization and/or received medication for COVID-19 treatment before delivery.Worsening COVID-19 illness prompted delivery in 23 mothers (9.2%), of which 20 (87.0%) were cesarean deliveries.In 52 of 113 cesarean deliveries (46.0%), rupture of membranes occurred at birth, which limited the exposure of the fetus to maternal genital tract flora before delivery.2. The mean (SD) gestational age at birth was 37.9 (2.6) weeks; 13 neonates (5.1%) were small for gestational age (<10th percentile), 25 and 62 (24.3%) were delivered either at low birth weight or preterm.Among the newborns, 49 (19.2%)required resuscitation at birth, 88 (34.5%) were separated from their mothers, and 152 (59.6%) were directly breastfed.Four neonates (1.6%) had neonatal encephalopathy and underwent therapeutic hypothermia; all 4 had negative SARS-CoV-2 test results, and their mothers had only mild COVID-19 symptoms.We observed 1 newborn death that was secondary to severe hypoxic ischemic encephalopathy; this newborn's mother had mild SARS-CoV-2 infection-related symptoms, and the newborn had 2 negative SARS-CoV-2 test results at 24 and 48 hours.
During their hospital stay, 225 neonates (88.2%) were tested for SARS-CoV-2 and 5 (2.2%) had positive results (Table 2).Thus, the test result positivity rate was 2.2% (95% CI, 1.0-5.1)among newborns tested before hospital discharge.A total of 124 newborns (55.1% of those tested) underwent only 1 test during their birth hospitalization.Including all 6 neonates with positive SARS-CoV-2 test results within the first week of life (eTable 1 in the Supplement), the test result positivity rate was 2.7% among those who were tested.Two neonates who presented with respiratory distress were delivered preterm: 1 newborn had nasal congestion, and 3 newborns were asymptomatic.We found no substantial differences in maternal or neonatal characteristics between neonates who were tested and those who were not during their hospitalization (eTable 2 in the Supplement).

Risk Factors Associated With Neonatal Test Result Positivity
Characteristics of neonates with negative and those with positive SARS-CoV-2 test results are shown in eTable 3 in the Supplement.In unadjusted analyses, neonates with positive test results were more likely to be born to mothers with symptomatic COVID-19 (OR, 1.84; 95% CI, 0.51-6.58;P = .35),less likely to be delivered vaginally (OR, 0.39; 95% CI, 0.12-1.22;P = .11),and less likely to room-in (OR, 0.26; 95% CI, 0.02-3.07;P = .29),but none of these results was statistically significant (Table 3).
Individual-level racial/ethnic minority status or non-English-speaking status was not associated with a higher risk of the newborn having a positive test result (OR, 0.81; 95% CI, 0.09-7.66;P = .85).
Adjusting for maternal symptoms, delivery mode, and rooming-in practice, mothers with high SVI (Ն90th percentile) were more likely to have neonates with positive SARS-CoV-2 test results (adjusted OR, 4.95; 95% CI, 1.53-16.01;P = .008)(Table 3).symptoms, but we did not otherwise collect data on the indication for preterm delivery.Among the neonates who were delivered for worsening maternal illness, 17 (73.9%)were delivered preterm.

Newborn Follow-up Data 30 Days After Hospital Discharge
Of the 255 neonates who were exposed to SARS-CoV-2, 151 (59.2%) had at least 1 postdischarge medical encounter documented in the EMR (Table 5).The demographic characteristics were similar for neonates with and those without postdischarge EMR information except the month of birth differed significantly between the 2 groups (eTable 4 in the Supplement).Most encounters were for routine care, whereas 28 were nonroutine visits, including 18 visits to urgent care or an emergency department.None of the who received routine care only underwent testing for SARS-CoV-2, whereas 7 neonates with nonroutine encounters were tested; 1 had a positive SARS-CoV-2 test result on day 5 of life during an emergency department visit for nasal congestion (eTable 1 in the Supplement).Four neonates were rehospitalized in the first 30 days after discharge, and none was for conditions directly associated with SARS-CoV-2 infection (eTable 5 in the Supplement).f The denominator for this variable is a fraction of the total cohort.

Discussion
To our knowledge, this study presented data for the largest US cohort of neonates who were born to mothers with positive SARS-CoV-2 test results.Among the 255 mother-newborn dyads from 11 hospitals in Massachusetts, we found a 2.2% test result positivity rate in neonates who underwent SARS-CoV-2 testing during birth hospitalization.In addition, we identified maternal social .17  Adverse health outcomes among neonates who were born to mothers with positive SARS-CoV-2 test results were associated with delivery prompted by worsening maternal COVID-19 illness, whereas the health outcomes of neonates with positive SARS-CoV-2 test results were largely favorable.We identified minimal neonatal health burden that could be directly associated with SARS-CoV-2 infection within 30 days after hospital discharge.
Previous studies reported a wide range in the percentage of test result positivity among neonates who were born to mothers with positive SARS-CoV-2 test results.Systematic reviews reported a 3.1% to 9.1% neonatal test result positivity, but these reviews were prone to publication bias and included studies that were published before the wide implementation of maternal surveillance testing, potentially overestimating the true incidence of positive SARS-CoV-2 test results in neonates. 5,6In contrast, 3 New York City cohorts with larger sample sizes reported virtually no neonatal test result positivity despite high rates of rooming-in and direct breastfeeding. 8,9,18Given that 30 newborns in the present cohort were not tested for SARS-CoV-2, it is possible that the neonatal test result positivity rate is even lower than 2.2%.Overall, the literature suggests low rates of acquired infection among New York City and Massachusetts neonates who were born to mothers with positive SARS-CoV-2 test results during the first wave of the pandemic. 9,18e COVID-19 pandemic has disproportionately affected Hispanic and Black communities with higher infection, morbidity, and mortality rates. 26,27In children, higher SVI, Hispanic ethnicity, and Black race independently increased the risk of multisystem inflammatory syndrome. 28We found that high social vulnerability, defined by the maternal zip code, was associated with a nearly 5-fold higher risk for neonatal test result positivity, although individual-level race/ethnicity and language status were not associated with a higher risk for neonatal test result positivity.Previous studies have identified the built environment and other neighborhood variables as factors associated with SARS-CoV-2 infection in pregnant women 29 and adverse perinatal outcomes in general, 30,31 but we could find no published studies that examined the sociodemographic risk factors for test result positivity among neonates who were exposed to SARS-CoV-2.The association of maternal social vulnerability with neonatal test result positivity was only slightly attenuated by adjustments for maternal illness severity, suggesting that nonclinical factors may be at play.We speculate that living in a socially disadvantaged neighborhood may be a factor in stress-mediated alterations in the maternal and/or fetal immune response, facilitating SARS-CoV-2 transmission. 14,15,32Given that 4 of 6 neonates with positive results were born at the same hospital, it is possible that hospital-level factors, such as air flow or building design, were also at play, although we minimized the impact of hospital-level factors by accounting for clustering in the multivariate model.
Newborns with positive SARS-CoV-2 test results appeared to have minimal burden of illness that was directly associated with a viral infection.However, those who were born in the context of delivery prompted by worsening maternal COVID-19 symptoms were more likely to be preterm births, which led to a need for resuscitation in the delivery room, CPAP or mechanical ventilation, and longer length of stay.These results indicate that maternal SARS-CoV-2 infection has an association with neonatal health, which is brought about not through viral transmission from the mother to the neonate but rather through the impact of preterm delivery undertaken because of the mother's worsening illness.
Few previous studies have ascertained the neonatal outcomes beyond the birth hospitalization.
We leveraged EMR data to identify nonroutine newborn health care visits possibly related to SARS-CoV-2 infection.Reassuringly, we found very few such encounters.The findings in this study complement those in the US-based PRIORITY (Pregnancy Coronavirus Outcomes Registry) study, which involved maternal reporting of newborn outcomes through 6 to 8 weeks of age. 17In the PRIORITY study, 2 of 80 neonates presented with upper respiratory tract infection symptoms, 0 had pneumonia, and 0 was rehospitalized; 2 had positive SARS-CoV-2 test results during the follow-up period. 17

Strengths and Limitations
This study has some strengths.We included 11 academic and community hospitals, representing more than 50% of all births in Massachusetts, but the findings may not generalize to nonacademic and level I or II hospitals.Racial and ethnic diversity of the study population was commensurate with the population in a recent report by the Centers for Disease Control and Prevention of pregnant women in the US with SARS-CoV-2, 33 suggesting generalizability to other US perinatal populations.
We leveraged active hospital-level surveillance for COVID-19 but may have missed a small number of dyads.
This study has some limitations.Like other studies, 34 the present study had limited ability to differentiate transient colonization from true positive test results in newborns because of a lack of repeated neonatal testing.In addition, some neonates were not tested during the birth hospitalization, and few were tested after discharge.Because of our reliance on clinical SARS-CoV-2 testing data, we could not determine the exact timing of maternal infection, especially in mothers with asymptomatic COVID-19.Although the study sample was large compared with samples in other published studies, we had limited ability to examine multiple factors simultaneously because only 6 newborns had positive results; residual confounding was possible.Practices evolved during the study period such that, by month 3, rooming-in and breastfeeding were standard in most, if not all, hospitals in Massachusetts.Furthermore, the evolution of these practices did not vary by social factors. 23We were not able to ascertain 30-day outcome data for all neonates because of the limitations of EMR-based follow-up.Demographic characteristics were similar in mothers and neonates who had available EMR encounters vs those who did not, but the data on 30-day outcomes may not be missing at random, preventing firm conclusions.

Conclusions
The neonatal test result positivity rate for SARS-CoV-2 during the birth hospitalization was 2.2% in a statewide perinatal cohort.Maternal social vulnerability was associated with an increased risk for neonatal test result positivity, whereas individual-level maternal race/ethnicity and language status was not.Newborns who had exposure to SARS-CoV-2 were at risk for both direct and indirect adverse health outcomes, whereas preterm delivery owing to worsening maternal COVID-19 illness was associated with substantial neonatal morbidity.The findings support ongoing surveillance of the virus and long-term follow-up.
Large, geographically defined cohort studies that track newborns after hospital discharge are needed to accurately define the incidence of neonatal test result positivity for SARS-CoV-2 and to identify the factors associated with increased positive test results.Within a statewide cohort, we conducted a cohort study with the following objectives: (1) to ascertain the percentage of neonates who were born to mothers with positive SARS-CoV-2 test results during the birth hospitalization, (2) to identify clinical and sociodemographic factors associated with neonatal test result positivity, and JAMA Network Open | Pediatrics Association of Perinatal SARS-CoV-2 With Neonatal Outcomes During the Pandemic Outcomes Primary neonatal outcomes were (1) positive SARS-CoV-2 test result during the birth hospitalization;

Newborn
Abbreviations: IQR, interquartile range; LBW, low birth weight; LOS, length of stay; VLBW, very low birth weight.a P values account for hospital-level clustering.b Delivery room resuscitation includes positivepressure ventilation, intubation, and/or chest compressions.c Respiratory support is continuous positive airway pressure or mechanical ventilation.

Table 4
shows that all adverse neonatal health outcomes (preterm or low birth weight, very preterm or very low birth weight, delivery room resuscitation, CPAP or mechanical ventilation, and length of stay) were increased among neonates who were born to mothers whose worsening COVID-19 illness

Table 1 .
Maternal Demographic and Pregnancy and Delivery Characteristics (continued) The denominator for this variable is a fraction of the total cohort.
b c Other includes American Indian or Alaska Native, Native Hawaiian or Pacific Islander.

Table 3 .
Factors Associated With Positive SARS-CoV-2 Test Results Among 226 Neonates a

Table 4 .
Adverse Neonatal Health Outcomes by Delivery Indication

Table 5 .
Follow-up Data of Neonates in the First 30 Days After Hospital Discharge Association of Perinatal SARS-CoV-2 With Neonatal Outcomes During the Pandemic defined by zip code, as a risk factor for neonatal test result positivity, whereas maternal COVID-19 symptoms, delivery mode, and rooming-in practice were not significant factors.