Association of Maternal SARS-CoV-2 Infection in Pregnancy With Neonatal Outcomes

Question  What are the outcomes in newborn infants of mothers testing positive for SARS-CoV-2 in pregnancy?

Findings  In this nationwide, prospective cohort study that included 88 159 infants from Sweden, SARS-CoV-2 infection in pregnancy was significantly associated with higher risk of any neonatal respiratory disorder (2.8% vs 2.0%; odds ratio, 1.42) and some other neonatal morbidities, but not neonatal mortality (0.30% vs 0.12%; odds ratio, 2.55).

Meaning  Maternal SARS-CoV-2 infection in pregnancy was significantly associated with small increases in the absolute risk of respiratory disorders and some other neonatal morbidities.

Importance  The outcomes of newborn infants of women testing positive for SARS-CoV-2 in pregnancy is unclear.

Objective  To evaluate neonatal outcomes in relation to maternal SARS-CoV-2 test positivity in pregnancy.

Design, Setting, and Participants  Nationwide, prospective cohort study based on linkage of the Swedish Pregnancy Register, the Neonatal Quality Register, and the Register for Communicable Diseases. Ninety-two percent of all live births in Sweden between March 11, 2020, and January 31, 2021, were investigated for neonatal outcomes by March 8, 2021. Infants with malformations were excluded. Infants of women who tested positive for SARS-CoV-2 were matched, directly and using propensity scores, on maternal characteristics with up to 4 comparator infants.

Exposures  Maternal test positivity for SARS-CoV-2 in pregnancy.

Main Outcomes and Measures  In-hospital mortality; neonatal resuscitation; admission for neonatal care; respiratory, circulatory, neurologic, infectious, gastrointestinal, metabolic, and hematologic disorders and their treatments; length of hospital stay; breastfeeding; and infant test positivity for SARS-CoV-2.

Results  Of 88 159 infants (49.0% girls), 2323 (2.6%) were delivered by mothers who tested positive for SARS-CoV-2. The mean gestational age of infants of SARS-CoV-2–positive mothers was 39.2 (SD, 2.2) weeks vs 39.6 (SD, 1.8) weeks for comparator infants, and the proportions of preterm infants (gestational age <37 weeks) were 205/2323 (8.8%) among infants of SARS-CoV-2–positive mothers and 4719/85 836 (5.5%) among comparator infants. After matching on maternal characteristics, maternal SARS-CoV-2 test positivity was significantly associated with admission for neonatal care (11.7% vs 8.4%; odds ratio [OR], 1.47; 95% CI, 1.26-1.70) and with neonatal morbidities such as respiratory distress syndrome (1.2% vs 0.5%; OR, 2.40; 95% CI, 1.50-3.84), any neonatal respiratory disorder (2.8% vs 2.0%; OR, 1.42; 95% CI, 1.07-1.90), and hyperbilirubinemia (3.6% vs 2.5%; OR, 1.47; 95% CI, 1.13-1.90). Mortality (0.30% vs 0.12%; OR, 2.55; 95% CI, 0.99-6.57), breastfeeding rates at discharge (94.4% vs 95.1%; OR, 0.84; 95% CI, 0.67-1.05), and length of stay in neonatal care (median, 6 days in both groups; difference, 0 days; 95% CI, −2 to 7 days) did not differ significantly between the groups. Twenty-one infants (0.90%) of SARS-CoV-2–positive mothers tested positive for SARS-CoV-2 in the neonatal period; 12 did not have neonatal morbidity, 9 had diagnoses with unclear relation to SARS-CoV-2, and none had congenital pneumonia.

Conclusions and Relevance  In a nationwide cohort of infants in Sweden, maternal SARS-CoV-2 infection in pregnancy was significantly associated with small increases in some neonatal morbidities. Given the small numbers of events for many of the outcomes and the large number of statistical comparisons, the findings should be interpreted as exploratory.

Since the start of the COVID-19 pandemic, there has been concern about how to protect vulnerable persons from SARS-CoV-2. In the past, newborn infants experienced increased mortality in pandemics,¹ and infections are the most common cause of mortality in children younger than 5 years. These reasons have most likely been important drivers for guidelines—especially in countries affected early by COVID-19—recommending mother-infant separation at birth in case the mother tested positive for SARS-CoV-2 on delivery.^2-4 In other guidelines, nonseparation of mother-infant dyads and breastfeeding have been promoted.^2-4 While recommendations on management of newborn infants of mothers with COVID-19 vary, they are all hampered by uncertainty and associated with potentially severe adverse effects from interfering too much or too little with mother-infant interactions after birth.^2,4,5

Systematic reviews have begun to shed light on maternal and pregnancy outcomes, but knowledge on neonatal outcomes following maternal COVID-19 in pregnancy is still based on case series and center experiences.^3,6-12 Because of potential selection biases, lack of denominators, lack of contemporary comparators, and limited and often unspecific symptoms and poorly defined outcome measures reported, there are insufficient data to draw solid conclusions about neonatal complications after maternal COVID-19 in pregnancy.^9,13 A comprehensive assessment would help to build a more robust evidence base for clinical practice. To test the hypothesis that maternal SARS-CoV-2 positivity in pregnancy may be associated with adverse neonatal outcomes, a Swedish nationwide, prospective cohort study was conducted.

All pregnant women in antenatal care and all parents of infants admitted for neonatal care received information regarding processing of personal data by the Swedish national quality registers, with a right to opt out and have all personal information removed. The study was approved by the Swedish Ethical Review Authority, which waived patient informed consent.

In Sweden, participation in the prenatal care program is almost universal, and prenatal, delivery, and neonatal care is tax funded. The Swedish COVID-19 strategy has been described elsewhere.¹⁴ The first pregnant woman testing positive for SARS-CoV-2 in Sweden gave birth on March 11, 2020. A national recommendation on perinatal management of SARS-CoV-2 was issued on March 17. As in other countries, the rigor of development and the evidence base of this guideline was of low quality.² After an update on April 5, the main features of the guideline for pregnant women testing positive for SARS-CoV-2 and their infants were (1) mode of delivery according to standard obstetric assessment; (2) nonseparation of mothers and infants unless any was unwell; (3) all infants to be tested (polymerase chain reaction [PCR] on nasopharyngeal swab) within 12 to 24 hours after birth, and for infants admitted for neonatal care, PCR tests repeated at 48 and 96 hours; (4) breastfeeding and expressed breast milk allowed given strict hygiene routines; (5) antiviral drugs not recommended to be given to infants; and (6) standard criteria for hospital discharge. Primary separation at birth of SARS-CoV-2–positive mothers and their infants was indicated only if an infant needed resuscitation followed by neonatal admission or if a condition diagnosed antenatally was known to necessitate admission to a neonatal unit (eg, preterm birth). Secondary separation was done if an infant developed a need for neonatal care after birth. SARS-CoV-2–positive mothers were not allowed in neonatal units, and if an infant was admitted after having had postnatal contact with the mother, the infant was considered incubated with SARS-CoV-2 and isolated until the infant had a negative SARS-CoV-2 test result.

On February 1, 2020, SARS-CoV-2 was included in the Swedish Communicable Diseases Act, making it mandatory to report all laboratory-confirmed PCR cases within 24 hours to the Public Health Agency’s national register for communicable diseases, the Swedish Register for Communicable Diseases (SmiNet). The National Board of Health and Welfare also issued instructions to assign SARS-CoV-2–positive patients with the International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10) code U07.1 followed by a code for manifestation (eg, ICD-10 code P23 for neonatal pneumonia).¹⁴ All ICD codes were captured in population-based health and quality registers.

The study cohort consisted of all live-born infants delivered by women captured in the Swedish Pregnancy Register between March 11, 2020, and January 31, 2021, and investigated for neonatal outcomes by March 8, 2021. Infants with malformations as defined by a Q code in ICD-10 were excluded, with the exception of patent ductus arteriosus in infants younger than 37 weeks of gestational age (eTable 1 in the Supplement).

The inception of this study was on March 25, 2020. Using the unique personal identity numbers for mothers and infants, data were linked between the Swedish Pregnancy Register, the Swedish Neonatal Quality Register, and SmiNet. All 3 registers prospectively extracted standardized and predefined data on a daily basis. The Swedish Pregnancy Register provided information on each individual pregnancy and birth outcomes for 92% of all pregnant women in Sweden; incompleteness is attributed to 4 hospitals not reporting to the register.¹⁵ The Swedish Neonatal Quality Register provided detailed information on procedures and diagnoses for all infants admitted for neonatal care.¹⁶ SmiNet included dates of first positive result on a PCR test for SARS-CoV-2 in mothers and infants.

The exposure was SARS-CoV-2 test positivity in mothers from conception to 1 week after birth. In the beginning of the study period, only pregnant women admitted for hospital care with symptoms of COVID-19 were tested. In June 2020, a general testing strategy was implemented including outpatient testing and contact tracing.¹⁴ Some but not all hospitals tested all women admitted for delivery.

Maternal, pregnancy, and birth characteristics were collected for descriptive purposes and to create cohorts matched on maternal characteristics for associations between SARS-CoV-2 in pregnancy and neonatal outcomes.^8-11,17-19 Small for gestational age was defined as a birth-weight z score more than 2 SDs below the mean in a Swedish reference for normal fetal growth.²⁰

Because near-term infants with gestational ages of 35 weeks and 0 days to 36 weeks and 6 days were not routinely admitted for neonatal care, gestational age was categorized into very preterm (gestational age less than 32 weeks 0 days); moderately preterm (gestational age of 32 weeks 0 days to 34 weeks 6 days); near-term or term (gestational age of 35 weeks 0 days to 41 weeks 6 days); and postterm (gestational age of 42 weeks 0 days or more).

This study was exploratory in design, and no primary outcome was predefined. Some of the neonatal outcomes described herein have previously been described in association with maternal SARS-CoV-2 in pregnancy^{6-10,12,13,18,21-26} whereas others have not. Altogether, the neonatal outcomes, as defined herein or in eTable 1 in the Supplement, included (1) neonatal resuscitation (assisted ventilation [ie, mask ventilation or use of continuous positive airway pressure {CPAP}] or intubation at birth); (2) admission for neonatal care; (3) neurologic conditions (hypoxic-ischemic encephalopathy grade 2-3,²⁷ neonatal convulsions, and severe brain injury in very preterm infants, defined as intraventricular hemorrhage grade 3-4²⁸ or cystic periventricular leukomalacia²⁹); (4) respiratory disorders (respiratory distress syndrome, transient tachypnea of the newborn, and meconium aspiration syndrome) and their treatments (CPAP, mechanical ventilation, surfactant administration, postnatal corticosteroids for severe lung disease, and need for supplemental oxygen at 28 days of postnatal age); (5) circulatory problems (persistent pulmonary hypertension of the newborn and patent ductus arteriosus [treated pharmacologically or by surgery]); (6) infections (neonatal pneumonia and blood culture–proven sepsis) and antibiotic therapy; (7) gastrointestinal disease (necrotizing enterocolitis in very preterm infants, surgically treated); (8) metabolic and hematologic problems (hypoglycemia [defined as plasma glucose level <2.6 mmol/L {<47 mg/dL} more than 3 hours after birth], treated hyperbilirubinemia, and blood transfusion); (9) in-hospital mortality; (10) length of hospital stay; and (11) breastfeeding at discharge. A composite outcome of respiratory distress syndrome, transient tachypnea of the newborn, meconium aspiration syndrome, or neonatal pneumonia was also defined as any respiratory disorder. In addition, infant test positivity for SARS-CoV-2 in the early neonatal (0-6 days), late neonatal (7-28 days), and postneonatal periods were included as outcomes.

To assess any association between SARS-CoV-2 in pregnancy and neonatal outcomes, each exposed infant was matched with up to 4 comparator infants of mothers without a positive SARS-CoV-2 test result on parity, multiple pregnancy, health care region, and a propensity score (estimated using logistic regression) that included maternal age, early pregnancy body mass index, educational level, country of birth, smoking status, living with partner, and prepregnancy comorbidity (a composite including hypertension, cardiovascular disease, diabetes, lung disease, and kidney disease). In the propensity score model, the dependent variable was maternal SARS-CoV-2 test result. In the propensity score matching, the nearest neighbor was sampled, allowing a maximum caliper width of 0.2 of the pooled standard deviation of the logit of the propensity score. For variables included in the matching algorithm (maternal age, parity, early pregnancy body mass index, educational level, birth country, and smoking), missing information was replaced using a missing indicator category.

Standardized differences between the 2 groups were calculated before and after matching.³⁰ Logistic regression was used to estimate odds ratios (ORs) and 95% confidence intervals conditioned on each matching set.

To provide a deeper understanding of mediating pathways among maternal exposure to SARS-CoV-2, pregnancy and birth events, and neonatal outcomes, a mediation analysis was performed. The exposure was maternal SARS-CoV-2 in pregnancy, and the relations between confounding and mediator variables were identified, as shown in Figure 1. Three potential mediators were tested: preterm birth (gestational age <37 weeks), mode of delivery (cesarean delivery), and Apgar score less than 4 at 5 minutes. The neonatal outcome, chosen on the basis of the most prevalent symptoms and conditions reported in newborn infants with positive SARS-CoV-2 test results,¹² was any respiratory disorder.

Data were analyzed using SAS software, version 9.4 (SAS Institute Inc). Mediation analyses were performed using CAUSALMED procedures, and median differences were estimated using quantile regression with QUANTREG procedures in SAS. A standardized mean difference of less than −0.10 or more than 0.10 was considered to indicate significant group differences or imbalance in matching. For 95% CIs, those that did not include 1 for ORs and those that did not include 0 for risk or median differences were considered statistically significant. A sensitivity analysis with a narrowed exposure window from conception to birth was also performed. Because of the potential for type I error due to multiple comparisons, the findings of this study should be interpreted as exploratory.

A total of 2905 infants (3.2%) were excluded because of malformations. Among 88 159 newborn infants included (49.0% girls), 2323 (2.6%) were delivered by 2286 SARS-CoV-2–positive mothers. Among SARS-CoV-2–positive mothers, the median time from a positive test result to delivery was 36 (interquartile range, 5-85) days. Among the 2323 infants, 642 (28%) had mothers testing positive at the time of delivery (ie, from 2 weeks before to 2 days after delivery), and 68 infants (2.9%) had mothers who tested positive within 1 week after delivery.

Maternal age, parity, prepregnancy comorbidity, maternal education, proportions living with a partner, smoking in pregnancy, multiple pregnancy, gestational diabetes, type of delivery onset, and mode of delivery were not significantly different between SARS-CoV-2–positive mothers and mothers without a positive SARS-CoV-2 test result. In contrast, non-Nordic country of birth, high body mass index, antenatal corticosteroid treatment for preterm birth, and delivery at a hospital with a neonatal intensive care unit were significantly more common in SARS-CoV-2–positive mothers than in mothers without a positive SARS-CoV-2 test result. The distribution of health care region also differed significantly between the 2 groups. Infant sex and proportions with Apgar scores less than 7 or less than 4 at 5 minutes were not significantly different in the 2 groups. Gestational age and birth weight were significantly lower in infants born to SARS-CoV-2–positive mothers than in comparator infants (Table).

After matching, the groups were well balanced on matching factors except for significantly higher proportions of exposed infants having antenatal corticosteroid treatment for preterm birth, being born preterm, and being delivered at university hospitals than in the comparator group (Table).

In the unmatched cohort, 7 (0.30%) of 2323 infants of SARS-CoV-2–positive mothers died and 96 (0.11%) of 85 836 infants of mothers without a positive SARS-CoV-2 test result died. The clinical characteristics of the 7 SARS-CoV-2–exposed infants who died are presented in eTable 2 in the Supplement. In addition, 65 (2.8%) of the 2323 infants of SARS-CoV-2–positive mothers and 1518 (1.8%) of the 85 836 infants of mothers without a positive SARS-CoV-2 test result had a diagnosis of a noninfectious respiratory disorder, and 3 (0.13%) of the SARS-CoV-2–exposed infants vs 85 (0.10%) of the control infants had a neonatal diagnosis of pneumonia, none of which was attributed to neonatal COVID-19 (Figure 2 and Figure 3).

After matching infants by maternal characteristics, some neonatal outcomes (procedures and morbidities) were significantly more common in infants of SARS-CoV-2–positive women (n = 2323) than in infants of comparator women (n = 9275): assisted ventilation at birth (with mask or CPAP; 6.4% vs 5.1%; OR, 1.28; 95% CI, 1.05-1.55), intubation at birth (0.6% vs 0.3%; OR, 1.90; 95% CI, 1.002-3.59), admission for neonatal care (11.7% vs 8.4%; OR, 1.47; 95% CI, 1.26-1.70), respiratory distress syndrome (1.2% vs 0.5%; OR, 2.40; 95% CI, 1.50-3.84), use of CPAP (4.9% vs 3.8%; OR, 1.32; 95% CI, 1.06-1.64), mechanical ventilation (1.6% vs 0.5%; OR, 3.51; 95% CI, 1.85-6.65), any respiratory disorder (2.8% vs 2.0%; OR, 1.42; 95% CI, 1.07-1.90), persistent pulmonary hypertension (0.3% vs 0.1%; OR, 3.00; 95% CI, 1.04-8.65), antibiotic therapy (2.8% vs 2.0%; OR, 1.36; 95% CI, 1.02-1.82), and hyperbilirubinemia (3.6% vs 2.5%; OR, 1.47; 95% CI, 1.13-1.90). Mortality (0.30% vs 0.12%; OR, 2.55; 95% CI, 0.99-6.57), breastfeeding rates at discharge (94.4% vs 95.1%; OR, 0.84; 95% CI, 0.67-1.05), and length of stay in neonatal care (median, 6 days in both groups; difference, 0 days; 95% CI, −2 to 7 days) did not differ significantly between the groups. Hypoxic-ischemic encephalopathy, transient tachypnea, meconium aspiration, pneumonia, sepsis, and hypoglycemia also did not differ significantly between the 2 groups (Figure 2 and Figure 3).

In mediation analysis, preterm birth (gestational age <37 weeks; 205 [8.8%] of 2323 SARS-CoV-2–exposed infants; 518 [5.6%] of 9275 comparator infants matched on maternal characteristics) was identified to mediate 89.3% of the estimated association between maternal SARS-CoV-2 test positivity and any neonatal respiratory disorder. Adding cesarean delivery and an Apgar score less than 4 at 5 minutes as potential mediators diluted the mediation of preterm birth. There was no statistically significant direct (infectious) association between maternal SARS-CoV-2 test positivity and any neonatal respiratory disorder (eTable 3 in the Supplement).

Twenty-one (0.90%) of 2323 infants of SARS-CoV-2–positive mothers tested positive for SARS-CoV-2 in the neonatal period. Of these, 12 did not have any neonatal morbidity, 9 had diagnoses with unclear relation to SARS-CoV-2, and none had congenital pneumonia. The proportion of infants testing positive for SARS-CoV-2 among infants born to mothers who tested positive at delivery was 17 (2.7%) of 642. In the comparator group of infants of mothers without a positive SARS-CoV-2 test result in pregnancy up to 1 week after birth, 279 (0.33%) of 85 836 infants tested positive for SARS-CoV-2, 230 of them (82%) after the neonatal period (0-28 days after birth). Characteristics of all infants testing positive for SARS-CoV-2 in the neonatal period are presented in eTable 4 in the Supplement.

The sensitivity analysis restricting the exposure window for test positivity among pregnant women from conception to delivery did not alter the results and interpretation in any significant way (eTable 5 in the Supplement).

In a nationwide cohort of infants in Sweden, maternal SARS-CoV-2 test positivity was significantly associated with increased risks of some neonatal morbidities. Preterm birth was identified as a mediator of neonatal respiratory disorders.

Previously, viral RNA has been detected in the placenta³² and fetal membranes,³³ but intrauterine transmission of SARS-CoV-2 from mother to fetus seems to be rare.¹² Postnatal viral transmission has been described to occur.^{12,13,21,26,34-36} One case series reported most infants of SARS-CoV-2–positive mothers to be asymptomatic, but a smaller proportion experienced neonatal morbidity, and 6 of 450 died.¹³ However, because none of the fatal cases were SARS-CoV-2 positive, death due to other causes could not be excluded.

A review of 77 studies mainly from China, the US, and Europe reported that 2 (0.5%) of 427 infants of COVID-19–positive mothers died, which was a higher mortality rate than in historical controls (0.1%).¹⁰ The difference in mortality rates found herein was smaller, and no deceased infant was considered to have had COVID-19 infection as currently defined by the World Health Organization.³⁷ In addition, considering that maternal SARS-CoV-2 test positivity in pregnancy occurred 18 to 74 days before delivery for 5 of the 7 deceased infants and that none of the mothers had been admitted for intensive care during pregnancy or at delivery, a potentially causal relationship (direct or indirect) between SARS-CoV-2 and neonatal mortality appears to be very weak.

In 176 cases of neonatal SARS-CoV-2 from different parts of the world, selected according to diagnostic criteria based on PCR tests,³⁸ various but nonspecific symptoms were reported in 55%.¹² In a subgroup analysis, mother-infant co-care was reported to increase the risk of neonatal infection.¹² Increased admission rates for neonatal care among infants born to mothers infected with SARS-CoV-2 have also been reported previously in the United Kingdom,²⁶ from Turkey,³ and in systematic reviews of case series from China, the US, and elsewhere.^9,10,12,13 The present study confirms an increased admission rate in infants of SARS-CoV-2–positive women. Given that all preterm infants less than 35 weeks of gestational age in both groups were admitted for neonatal care (except for infants who died after birth but before neonatal admission), the higher admission rates in infants of SARS-CoV-2–positive women were confined to near-term or term infants. Based on the findings presented herein, the excess admissions in near-term or term infants of SARS-CoV-2–positive women most likely reflect widened indications for observation and isolation in neonatal units,¹⁸ maternal illness after delivery, or recommendations in guidelines.³

This study could neither confirm nor rule out an association between maternal SARS-CoV-2 test positivity and increased fetal distress during delivery or birth asphyxia.^8,39 Point estimates for ORs regarding asphyxia-related outcomes in near-term and term infants, such as hypoxic-ischemic encephalopathy grade 2 to 3 and convulsions, were above 1. This may indicate an increased vulnerability in some SARS-CoV-2–exposed fetuses and infants to asphyxia-related complications, possibly mediated by placental insufficiency or maternal fever superimposed on a hypoxic-ischemic event during delivery.⁴⁰ There was no information on maternal temperature during delivery in this study. To detect a statistically significant difference (with an OR of 1.7 as estimated herein and with a statistical power of 80%) between SARS-CoV-2–exposed and unexposed infants in the rare outcome of hypoxic-ischemic encephalopathy grade 2 or 3, approximately 33 000 infants in the exposed group would be needed. This means sampling for more than 13 years in a country such as Sweden, provided that the rates of SARS-CoV-2 in pregnancy would stay the same as in March 2020–January 2021, which is highly unlikely.

Preterm birth has been reported to be a major risk among pregnant women who test positive for SARS-CoV-2.^{3,7,8,11,26,39,41} This study found support for preterm birth being an important mediator of the association between maternal SARS-CoV-2 and increased neonatal respiratory morbidity, whereas no statistically significant direct, infectious association with maternal SARS-CoV-2 could be demonstrated. The mediator of preterm birth may explain other associations between maternal SARS-CoV-2 and neonatal outcomes as well, such as more frequent need for assisted ventilation (at birth and later), a higher proportion of infants with hyperbilirubinemia, and more frequent use of antibiotics among SARS-CoV-2–exposed infants.

The follow-up time for SARS-CoV-2 test positivity among participants in this study was at least 1 month. All SARS-CoV-2–exposed infants underwent SARS-CoV-2 PCR testing at least once (those admitted for neonatal care were tested 3 times), and a majority tested negative for SARS-CoV-2 in the neonatal period. While a low transmission rate is in accordance with previous reports,^3,8,21,26 interpretations vary. Given that mothers and their infants were kept together, the results presented herein suggest that the risk of viral transmission from mothers to their newborns and older infants is low, and should it occur, infants are not severely affected. These findings suggest that routine use of interventions such as mother-infant separation¹² and stopping breastfeeding may not be necessary. An additional message of this study is that SARS-CoV-2 test positivity in mothers did not prolong hospital stay for families.

Strengths of this study include its population-based, nationwide design including a large number of infants and the direct and propensity score matching procedure followed by formal mediation analyses to minimize bias and confounding. Matching and mediation analyses were based on maternal and perinatal characteristics known to affect clinically important outcomes such as admission rates for neonatal care, neonatal morbidity and mortality, and breastfeeding. A wide range of neonatal outcomes that have been discussed as a matter of concern in connection to SARS-CoV-2 were explored.¹² The follow-up time for SARS-CoV-2 test positivity in infants reached at most 9 months. In addition, the possibility of daily data capture online in all 3 national registers and using unique personal identity numbers for linkage enabled a minimum of delay in reporting this study’s results.

This study had a number of limitations. First, numbers for some neonatal outcomes were small and the magnitude of the absolute risks was low. Second, imbalanced testing between health care regions and hospitals as well as regional variations may have occurred, attributed to clusters of COVID-19 infection. However, all analyses adjusted for health care region. Third, some mothers of matched comparators could have been infected since COVID-19 infection in pregnancy may be asymptomatic.¹¹ Furthermore, women with near-term or term infants admitted for neonatal care could have been more likely to be tested, which may lead to an overestimation of risks associated with SARS-CoV-2 test positivity in near-term and term infants. Fourth, although surveillance bias could not be excluded, there were no recommendations in the Swedish guidelines for COVID-19 in pregnancy to test women with pregnancy complications differently than other pregnant women, ie, on suspicion of COVID-19 infection. Fifth, there was no information on severity of illness among SARS-CoV-2–positive women before or during delivery. Sixth, all newborn infants of SARS-CoV-2–positive mothers had 1 to 3 SARS-CoV-2 PCR tests after birth, but testing in the unexposed population was performed only in infants with suspected symptomatic infection, introducing detection bias. Seventh, neonatal tests on the day of birth may have been contaminated by maternal viral DNA, resulting in falsely high test positivity among SARS-CoV-2–exposed infants. On the other hand, some SARS-CoV-2–positive women may have become negative by the time of delivery, suggesting that their infants were unexposed to postnatal viral transmission. Eighth, although test positivity could be traced throughout infancy, clinical outcomes were limited to the neonatal period. Ninth, given population differences and that general precautionary measures and specific recommendations for SARS-CoV-2–positive mothers and their infants are country-specific,^2,4 external validity of the findings may be limited.

In a nationwide cohort of infants in Sweden, maternal SARS-CoV-2 infection in pregnancy was significantly associated with small increases in some neonatal morbidities. Given the small numbers of events for many of the outcomes and the large number of statistical comparisons, the findings should be interpreted as exploratory.

Corresponding Author: Mikael Norman, MD, Division of Pediatrics, Department of Clinical Science, Intervention and Technology, Novum, Blickagången 6A, Karolinska Institutet, SE-141 57 Stockholm, Sweden (mikael.norman@ki.se).

Correction: This article was corrected September 14, 2021, to change 1.6% to 2.6% in the first sentence in the Results section of the abstract and in the second sentence in the Results section of the text.

Published Online: April 29, 2021. doi:10.1001/jama.2021.5775

Accepted for Publication: March 30, 2021.

Author Contributions: Dr Söderling had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Norman, Söderling, Hervius Askling, Aronsson, Byström, Jonsson, Sengpiel, Stephansson.

Acquisition, analysis, or interpretation of data: Norman, Navér, Söderling, Ahlberg, Hervius Askling, Aronsson, Byström, Ludvigsson, Håkansson, Stephansson.

Drafting of the manuscript: Norman.

Critical revision of the manuscript for important intellectual content: Navér, Söderling, Ahlberg, Hervius Askling, Aronsson, Byström, Jonsson, Sengpiel, Ludvigsson, Håkansson, Stephansson.

Statistical analysis: Norman, Söderling, Ahlberg.

Obtained funding: Norman, Stephansson.

Administrative, technical, or material support: Norman, Stephansson.

Supervision: Norman.

Conflict of Interest Disclosures: Dr Ludvigsson reported coordinating a study on behalf of the Swedish IBD Quality Register (SWIBREG), which has received funding from Janssen. No other disclosures were reported.

Funding/Support: This study was supported by grants from the Swedish Society of Medicine (2020-937944) and NordForsk (105545), by grants from a regional agreement on clinical research between Region Stockholm and Karolinska Institutet (ALF2020-0443), and by the Childhood Foundation of the Swedish Order of Freemasons. The Swedish Neonatal Quality Register and the Swedish Pregnancy Register are funded by the Swedish government (Ministry of Health and Social Affairs) and the body of Regional Health Care Providers (County Councils).

Role of the Funder/Sponsor: The funding bodies had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; or decision to submit the manuscript for publication.

Additional Contributions: We thank all obstetric and pediatric departments in Sweden for contributing and sharing data to the quality registers.