Six-Month Outcomes of Infants Born to People With SARS-CoV-2 in Pregnancy

Data were obtained through the Surveillance for Emerging Threats to Mothers and Babies Network (SET-NET).¹⁴  Participating state, territorial, and local health jurisdictions provided data on pregnant people with laboratory-confirmed (polymerase chain reaction [PCR] positive) SARS- CoV-2 infection and their infants. Maternal data during pregnancy and at delivery and infant information through 6 months of age were collected from existing sources (eg, electronic health records, vital statistics, laboratory reports, and health department investigations and case reporting). Jurisdictions searched laboratory databases for infant SARS-CoV-2 testing information from birth through 6 months. Data abstracted from the electronic health records of newborn (first visit after discharge, occurring before 1 month of age), 2 month, and 6 month well-child visits, included breastmilk feeding, type of well-child visit (in-person or telehealth), jaundice requiring phototherapy, hospitalization, and death. We categorized reasons for hospitalization from clinical notes and International Classification of Diseases, 10th Revision codes.

This analysis included live-born infants who met the following criteria: (1) born to people residing in SET-NET jurisdictions with a positive SARS-CoV-2 PCR test result from January 20 to December 31, 2020 in at least 1 clinical specimen at any point during pregnancy, up to and including the day of delivery; and (2) age 6 months before March 1, 2022 with at least 1 reported well-child visit between birth and age 9 months or had laboratory testing information ascertained from birth through 6 months. This activity was reviewed by the Centers for Disease Control and Prevention and was conducted consistent with applicable federal law and Centers for Disease Control and Prevention policy.

We analyzed data reported from 10 jurisdictions, including the city of Houston (TX), the territory of Puerto Rico, and the states of Kansas, Minnesota, Nebraska, New Jersey, New York (excluding New York City), South Carolina, Tennessee, and Washington. Because of the large number of cases among pregnant people and limited capacity to conduct medical record abstraction (MRA), 2 jurisdictions sampled cases for MRA at the end of pregnancy and provided data on all their liveborn infants, 1 jurisdiction provided data on all pregnancies and sampled cases for infant follow-up, 1 jurisdiction sampled for both end of pregnancy and infant follow-up, whereas the other 6 jurisdictions provided data on all pregnancies and infants meeting inclusion criteria. Sampling weights accounting for selection probability and loss to follow-up were calculated for both end of pregnancy and infant follow-up time points, as appropriate in sampling jurisdictions.¹⁵  Infant SARS-CoV-2 testing data through 6 months were available from 8 of 10 jurisdictions.

We assessed the timing of maternal infection (first positive test result) in 2 ways. First, because of the potential effects of gestational age at infection and the interval between onset of infection and delivery on maternal antibody concentrations and transfer across the placenta, we categorized the timing of maternal infection into 3 timepoints: (1) <20 weeks’ gestation and >14 days before delivery, (2) ≥20 weeks’ gestation and >14 days before delivery, and (3) ≤14 days before delivery regardless of gestational age. With category 1, we hypothesized that maternal antibody concentrations may have waned leading to a lower placental transfer; with category 2, that maternal antibody concentrations were still high with optimal transfer before delivery; and with category 3, that maternal antibody concentrations were not yet high with suboptimal transfer. Second, as a proxy for whether the pregnant person was infectious at delivery, we dichotomized maternal infection at >14 days before delivery and ≤14 days before delivery.

We estimated the incidence of PCR-positive SARS-CoV-2 infections (classified as any positive testing, only negative testing, or not known to be tested) among infants by timing of maternal infection relative to gestational age and delivery. We examined the early postnatal period, defined as birth to 14 days of age and the late postnatal period, defined as 15 days to 6 months of age, separately because SARS-CoV-2 testing guidelines may only be applied to infants of people believed to be infectious during the peripartum timepoint^16,17  We also assessed the following outcomes: breastmilk feeding, type of newborn well-child visit, jaundice requiring phototherapy, all-cause hospitalization, and all-cause mortality, overall and by timing of maternal infection relative to delivery. In a sensitivity analysis, we estimated percent positivity by limiting the analysis to infants known to be tested with positive or negative results.

Unweighted counts and weighted percentages are reported with 95% confidence intervals with Taylor series variance and finite population correction to account for the complex sampling. Rao-Scott χ² tests were used for comparisons. We used weighted Kaplan-Meier curves to examine the proportion of infants testing positive between birth and 6 months of age, stratified by timing of maternal infection relative to gestational age and delivery. Preterm birth is associated with increased SARS-CoV-2 testing⁶  and adverse health outcomes. As such, to minimize confounding, infants born <34 weeks’ gestation were excluded from analyses stratified by timing of maternal infection. All analyses were conducted in SAS 9.4 (SAS Institute, Cary, NC).

Of 13 180 infants selected from 10 jurisdictions for follow-up, 3967 (30.1%) had data from medical record abstraction. From 8 jurisdictions with complete electronic laboratory data, 6601 (50.1%) had information through 6 months of age as of March 1, 2022, although only 1695 (12.9%) were known to be laboratory tested for SARS-CoV-2 (Fig 1).

Maternal age was similar between those with infections occurring >14 days and ≤14 days before delivery. However, pregnant people identified as non-Hispanic white and those who had private insurance represented greater proportions among those with infections occurring >14 days before delivery. Most reported maternal infections occurred at ≥20 weeks’ gestation (73.3% [95% confidence interval: 69.6–76.9]). Nearly one-half of maternal records (43.3% [39.7–46.8]) had insufficient information to define COVID-19 illness severity, 16.8% (13.7–19.9) had an asymptomatic infection, and illness was mild in 29.7% (26.2–33.3), moderate-severe in 8.6% (7.2–10.1) and critical in 1.6% (1.1–2.0). The proportion of asymptomatic illness was higher, and the proportions of other illness categories were lower among those with infections occurring ≤14 days before delivery (Table 1).

Most infants were born term (≥37 weeks) (85.6% [82.4–88.7]), singleton (96.0% [94.2–97.8]), and had no birth defect diagnosis through 6 months of age (96.7% [95.8–97.7]) (Table 1). There were no differences in these birth and infant characteristics by timing of maternal infection relative to delivery. Among infants with incomplete follow-up MRA, maternal age, timing of maternal infection, sex, and gestational age at birth were similar, and a higher proportion of pregnant people were Asian, had no or unreported insurance, and had unknown illness severity (data not shown).

Among SARS-CoV-2-exposed infants reported by jurisdictions with complete electronic laboratory records for infants (n = 6601), 0.6% (0.5–0.7) tested positive, 10.2% (9.2–11.3) tested negative, and 89.1% (88.0–90.3) were not known to be tested in the late postnatal period (15 days to 6 months of age). Results during the late postnatal period did not vary by timing of maternal infection (Table 2). The incidence of SARS-CoV-2 infection was 0.4% (0.3–0.5) in the early postnatal period (≤14 days of age); all early postnatal infant infections occurred among those exposed to maternal SARS-CoV-2 infections at ≤14 days before delivery (1.9% [1.4–2.3]). From birth through 6 months, 1.0% (0.8–1.1) tested positive, 19.1% (17.5–20.6) tested negative, and 80.0% (78.4–81.6) were not known to be tested, with differences in the incidence of infant SARS-CoV-2 infection by timing of maternal infection. Age at infant’s first positive test result differed by timing of maternal infection. Survival curves reveal that most early infant infections occurred among those with maternal infections ≤14 days before delivery; no infant infections occurred during the first month of life among those with maternal infection >14 days before delivery and ≥ 20 weeks’ gestation, and a similar pattern was observed among the 3 groups thereafter (Fig 2).

In a sensitivity analysis limiting data to infants known to be tested with positive or negative results (n = 1692), the percent positivity in the late postnatal period was 5.6% (4.6–6.5) overall, 5.8% (4.0–7.6) when maternal infection occurred <20 weeks’ gestation and >14 days before delivery, 4.7% (3.5–5.8) when ≥20 weeks’ gestation and >14 days before delivery, and 7.3% (5.7–9.0) when ≤14 days before delivery. Percent positivity in the early postnatal period was 3.7% (2.8–4.7). Percent positivity from birth through 6 months of age was 4.8% (4.1–5.5) (data not shown).

Among infants with completed MRA (n = 3967), the prevalence of breastmilk feeding initiation among SARS-CoV-2-exposed infants was 85.8% (82.8–88.7) overall and breastmilk feeding declined at subsequent well-child visit timepoints: newborn visit (78.6% [74.5–82.6]), 2 months (59.0% [55.3–62.7]), and 6 months (41.5% [38.0–45.0]). When examining by timing of maternal infection, breastmilk feeding initiation was significantly lower when maternal infection occurred ≤14 days before delivery (77.6% [72.5–82.6]) compared with earlier maternal infections (88.3% [84.7–92.0]), but there were no differences in breastmilk feeding at subsequent visits (Table 3).

Most newborn well-child visits were in person (98.9% [98.5–99.4]). Jaundice requiring phototherapy after birth hospitalization was observed in 5.6% (4.3–6.9), with no differences by timing of maternal infection (Table 3).

The incidence of all-cause hospitalization through 6 months of age was 4.1% (2.0–6.2) and did not differ by timing of maternal infection (Table 3). The most common reasons for hospitalization were acute respiratory illness excluding COVID-19 (12%), jaundice (11%), and feeding issues (3%). COVID-19 was listed as a reason for hospitalization for 1 infant (0.2% of hospitalizations), born to a person infected >14 days before delivery. Of 1199 infants with vital records information, 17 deaths were observed. All-cause mortality from birth through 6 months of age was 0.4% (0.3–0.6), with a higher incidence among those with maternal infection ≤14 days before delivery (1.0% [0.4–1.6]) compared with maternal infection that occurred at >14 days (0.3% [0.1–0.5]) (Table 3). No deaths occurred among infants with positive SARS-CoV-2 test results nor was COVID-19 a listed cause of death (data not shown).

Our study, representative of multiple United States jurisdictions, describes the preliminary 6 month outcomes of infants born to people with SARS-CoV-2 infections in 2020, which are largely reassuring. The incidence of SARS-CoV-2 infections in the early postnatal period was higher when maternal infection occurred in the 14 days preceding delivery, and most infant infections occurred between 15 days and 6 months of age. Breastmilk feeding initiation was lower when maternal infection occurred in the 14 days preceding delivery and all-cause mortality was higher.

We found 1.0% of SARS-CoV-2-exposed infants tested positive for SARS-CoV-2 infection from birth to 6 months of age; 0.4% in the early postnatal period and 0.6% in the late postnatal period. We report an incidence of early postnatal infections among SARS-CoV-2-exposed infants similar to that reported by authors of other studies from North America.¹⁸  Similar to previous reports on SARS-CoV-2 infections among those tested during the birth hospitalization,⁶  infant infections were more common when maternal infection occurred in the 14 days preceding delivery, but SARS-CoV-2 infections occurred beyond 14 days of life with no significant differences by timing of maternal infection. This may reflect ongoing transmission in the community and within the dyad’s household.

In a previous analysis using data from 20 jurisdictions participating in SET-NET, Olsen et al found that 13% of neonates born to people with SARS-CoV-2 infection during pregnancy were known to be PCR-tested during the birth hospitalization and, of those, 3.6% tested positive.⁶  In that analysis, the proportion known to be tested increased among those with maternal infection close to delivery, with 80% of infants tested at the birth hospitalization when maternal infection was ≤14 days before delivery. Here, we used data from 10 jurisdictions with 6-month infant follow-up data. When maternal infection occurred ≤14 days before delivery, infant testing was more common during the early than the late postnatal period (47% vs 8%). A possible explanation for testing differences between these 2 time periods is that neonatal testing is recommended for all newborns born to people with COVID-19, whereas late postnatal testing was likely more commonly conducted for infants with symptoms or other exposures.

The prevalence of breastmilk feeding initiation was similar among SARS-CoV-2-exposed infants (82.3%) compared with prepandemic estimates in the general population (83.9%). However, the prevalence of breastmilk feeding at newborn, 2-month, and 6-month well-child visits was lower than historical estimates.¹⁹  In line with previous reports among SARS-CoV-2-exposed infants,²⁰  breastmilk feeding initiation was lower when the pregnant person tested positive for SARS-CoV-2 infection in the 14 days preceding delivery. However, at the newborn and subsequent well-child visits, those with earlier maternal infections were fed breastmilk in a similar proportion. Additional breastmilk feeding support for new parents who have experienced a SARS-CoV-2 infection during pregnancy may be needed.

The incidence of all-cause mortality, all-cause hospitalization, and most common reasons for hospitalization from birth through 6 months of age was similar to historical estimates.^21,22  There were no deaths attributed to COVID-19, and unrecognized SARS-CoV-2 infections resulting in infant death were unlikely given the population and time period examined. The reason for the observed higher mortality among infants born to people who tested positive ≤14 days before delivery is unclear and may be due to SARS-CoV-2-related factors in utero, during delivery, or in the neonatal period.

Although the COVID-19 pandemic expanded access to telehealth services,²³  SARS-CoV-2-exposed infants were seen in person for their newborn visits, as was recommended.¹³  The proportion with jaundice requiring phototherapy after the birth hospitalization was 5.6%, higher than estimates reported for infants not exposed to SARS-CoV-2.²⁴  The reasons for the higher proportion with jaundice requiring phototherapy after birth hospitalization among SARS-CoV-2 exposed infants are unknown and may require further study.

The present report is strengthened by longitudinal, representative data from geographically diverse jurisdictions. Nonetheless, some limitations need to be considered. The timing of the first positive test result is an imprecise measure of the timing of maternal infection. There are also known testing and reporting biases during the early postnatal period by timing of maternal infection, maternal characteristics, and birth outcome⁶  and potential testing bias among infants in the late postnatal period by symptom status. This analysis is limited to infants with PCR-confirmed infection and may underreport asymptomatic and nonmedically attended infections. Jurisdiction-wide records were searched making it reasonable to assume infants without reported results did not have laboratory testing performed; however, one cannot conclude whether untested infants were positive or negative for SARS-CoV-2. Several analyses relied on MRA, which was not completed for all infants selected because of delays in reporting. It is unclear how the addition of these records might change the findings. Liveborn infants with MRA not yet reported had a similar distribution of sex and gestational age at birth and were born to mothers with similar age and timing of infection but with some differences in race/ethnicity and illness severity.

Data collection through SET-NET is ongoing, with plans to include infants born to pregnant people with SARS-CoV-2 infections occurring in 2021. The current analysis included infants born to people with SARS-CoV-2 during pregnancy before the circulation of more recent variants (eg, οmicron) and when COVID-19 vaccines were not yet widely available. Increasing evidence reveals that COVID-19 vaccination during pregnancy is safe^25–27  and effective²⁸  and can also help protect infants born to those vaccinated during pregnancy.^29,30  COVID-19 vaccination is recommended for people who are pregnant, breastmilk feeding, trying to get pregnant, or might become pregnant.³¹ 

In these preliminary findings, we found generally low morbidity among infants born to people with SARS-CoV-2 infection and a low incidence of postnatal PCR-confirmed SARS-CoV-2 infection. Despite the lack of an association with poor infant health outcomes, COVID-19 may pose a larger threat to infants through more immediate impacts of severe disease experienced by pregnant people and preterm birth.³²  Given a lack of vaccination and treatment options available to infants, strategies to prevent SARS-CoV-2 infection throughout pregnancy and postpartum should be implemented to reduce SARS-CoV-2 infections among infants as well as their gestational parents.

We thank the staff supporting the SET-NET work, including those at participating health departments, for data collection and reporting. We thank the following persons for their contributions to this project: Miguel Valencia-Prado, MD, Puerto Rico Department of Health, Tyler Faulkner and Erika Fuchs, PhD, MPH, Nebraska Department of Health and Human Services, Mayra Trujillo, MS, Kansas Department of Health and Environment, Amber Palmer, MPH, Nadia Thomas, MS, RN, New York State Department of Health, Kirsten Short, MPH, Tahani Hamdan, MPH, Zakariya Mahmod, MPH, Syed F Imam, BDS, MPH, Dieula A. Casimyr, MPH, Ikram Shai Cheref, MPH, Elizabeth Herrera, BS, Houston Health Department, Shannon Baack, BS, Ona Loper, MPH, Minnesota Department of Health, Tiffany Chen, MPH, Washington State Department of Health. Finally, we thank the COVID-19 Pregnancy and Infant Linked Outcomes Team, the Epidemiology and Surveillance Task Force.