[ad_1]
Study site, population, and study design
The COVID-19 Outcomes in Mother-Infant Pairs (COMP) study is a longitudinal cohort study of pregnant patients who had SARS CoV-2 during gestation and their infants7,13. Pregnant individuals, ≥16 years old or older, with confirmed SARS-CoV-2 infection by nasopharyngeal reverse transcription polymerase chain reaction (RT-PCR), antigen (Ag) or serology during gestation were eligible for enrollment, regardless of preexisting conditions. Participants were primarily recruited by the Department of Obstetrics at the University of California, Los Angeles (UCLA) from 15 April 2020 to 31 August 2022. Beginning in April 2020, all women admitted to UCLA labor and delivery were screened for SARS-CoV-2 by nasopharyngeal swab. Two-hundred and twenty-one pregnant individuals, aged 16 to 56 years old, and 227 SARS-CoV-2 exposed fetuses were enrolled in our study. This resulted in 199 live births following in utero exposure to COVID-19. Maternal-infant pairs were followed longitudinally until the infants reached 6 months of age. The UCLA Medical Center comprises of multiple teaching hospitals, including tertiary and quaternary referral centers, and services the Los Angeles region in Southern California.
Informed consent for participation was obtained for all participants prior to enrollment. If a participant was incapable to provide consent (i.e., due to an acute hospitalization or intubated), consent was provided by a surrogate decision maker and the participant was re-consented once they regained capacity. Information was obtained directly from participants using RedCap survey software and from clinical chart review using EPIC. Our study was approved by the UCLA Internal Review Board (IRB).
Sampling, variables, and definitions
Our primary outcome was neonatal RD, which was defined as infants with at least two of the following: respiratory rate of 60 breaths per minute, retractions, nasal flaring, or central cyanosis. Infants were considered premature if they were born at a gestational age less than 37 weeks. Maternal COVID-19 severity was determined by NIH classification14. Briefly, maternal critical illness describes patients with respiratory failure (requiring mechanical ventilation) or signs of multiple organ failure; severe illness was defined as patients with oxygen saturation (SpO2) < 94% on room air, a ratio of arterial partial pressure of oxygen to fraction of inspired oxygen (PaO2/FiO2) < 300 mmHg, a respiratory rate >30 breaths/min, or lung infiltrates >50%; moderate illness was defined as individuals with evidence of lower respiratory disease on clinical assessment or imaging with a SpO2 ≥ 94% but did not require supplemental oxygen; mild illness describes symptomatic patients without shortness of breath, dyspnea or abnormal chest imaging; and asymptomatic individuals showed no symptoms14. Women were considered vaccinated if they received at least one dose of an mRNA COVID-19 vaccine prior to infection. Women who received the Janssen/Johnson & Johnson COVID-19 during pregnancy were not included (n = 1).
Maternal race and ethnicity were operationalized into three categories (Black, Hispanic, and Latina; Asian, Mixed-Race, and Other; or White) based on self-reported racial identity. We acknowledge that race is a social construct and our categorizations may not adequately reflect an individual’s identity. However, we included race in our univariate analysis given the history of systemic racism that has contributed to poor maternal outcomes among black women in the United States43. None of the mothers in our study self-identified as non-binary or transgender, therefore we have used gendered language in our text to refer to pregnant persons.
Statistical analysis and model construction
We compared the demographics of infants born with and without RD using one-way t-tests. We considered variables related to infant characteristics (sex, delivery method, prematurity, low birth weight), maternal predictors (maternal age, ethnicity, preexisting medical conditions), pregnancy complications (e.g., preeclampsia, gestational hypertension, chorioamnionitis, etc.), and COVID-19 predictors (maternal vaccination, trimester of infection, severity, symptoms, treatment, viral variant). The Fisher exact statistical test was used to obtain p-values. We did not adjust for multiple comparisons in our bivariate analyses because it was exploratory. Next, we conducted logistic regression analyses on neonatal RD, prioritizing variables that were significant in the previous t-tests and based on clinical suspicion of intermediate variables and effect modifiers. The following variables were used as predictors in our univariate analysis: maternal ethnicity, trimester of infection, COVID-19 severity, maternal vaccination status, and binary prematurity. We selected variables to include in our final model using a backwards selection and WALDs test. As recommended by ref. l44, we initially included all variables in a model and eliminated each variable above the p-value threshold of 0.25. This process was repeated until the remaining variables had p values below the threshold. We considered potential collinearity or lack of independence among predictor variables using chi-squared tests for independence. Similarly, our final multivariable model did not include maternal COVID-19 severity—despite evidence of significance in univariate regression models—because of the known association between COVID-19 severity and vaccination status. Including both COVID-19 severity and vaccination status would have likely attenuated the strength of our findings. In order to evaluate whether prematurity mediates the effect of maternal COVID-19 on neonatal respiratory distress, we utilized the Valeri and Vanderweele mediation analysis45. Approximately 18% of the risk of neonatal respiratory distress is mediated by prematurity, although not statistically significant (p = 0.3). Our final model for neonatal RD included binary maternal vaccination status and infant prematurity as predictors. Finally, we conducted a chart and imaging review to better characterize features of infant RD born to women infected with COVID-19 during pregnancy.
We performed a post-hoc power analysis in which we calculated the power achieved in our study based upon the false positive rate, the effect size of maternal COVID-19 vaccination prior to infection in protecting against adverse perinatal outcomes, and the number of pregnant individuals in our cohort. To carry out the calculation, we needed to supply a value for the effect size that was independent of our cohort. According to a systematic review and meta-analysis, rates of adverse perinatal outcomes were 15% higher in unvaccinated pregnant individuals than among those who were vaccinated46. Therefore, our post-hoc power analysis assumed that the rate of such outcomes would be at least 15% higher among pregnant women who were not vaccinated. The null hypothesis was that the rates of adverse outcomes would be the same among vaccinated and unvaccinated pregnant people. In the present cohort of 227 pregnancies, approximately one-third of the participants were vaccinated before COVID-19 infection.
We carried out a post-hoc power analysis using G*Power 3.1.9.447. Our calculations assumed a 5% false-positive rate and used a two-tailed test of the difference in proportions between the vaccinated and unvaccinated groups. The results indicated that our analysis had 90% power, in other words, the chance of rejecting the null hypothesis if it was false was 90%. This rate is reasonably high and we believe it provides support for our conclusions about this cohort.
Data was analyzed using R language48. Statistical analysis was conducted using a combination of epiDisplay package49, tableone package49, stats base package, and aod package50.
Proteomic profiling analysis
The present data is reanalyzed from a dataset previously published by our group7. We conducted a proteomic reanalysis to explore potential associations between respiratory distress and canonical pathways possibly associated with SARS CoV-2 in a subset of infants for whom proteomics was performed. We analyzed a subset of 45 SEU infants born in the first year of the pandemic (April 2020 to March 2021) matched to seven control infants born to unexposed healthy women at the pandemic onset, for a total of 52 infants (Fig. 1). Controls were from a convenience sample of healthy mothers who did not have SARS-CoV-2 exposure and for whom infant specimens were available. Infants were matched based on gestational age. This analysis utilizes peripheral infant blood specimens collected between 24 and 48 h of life. This timeframe was selected to coincide with routine bilirubin checks in order to minimize blood draws. The SEU infant cohort was clustered according to RD outcome and gestational age, resulting in four groups: no RD term infants (n = 27), no RD preterm infants (n = 3), RD term infants (n = 4), and RD preterm infants (n = 11). Significant differentially expressed proteins between healthy and the four COVID-19-exposed infant groups were determined by two-tailed Mann Whitney U test using the R base package, t.test, considering fold-change ≥2 and FDR-adjusted p value < 0.05. Enrichment analysis was conducted using the online platform Enricher51. Comparisons between multiple groups using 1-way ANOVA with uncorrected Fisher’s test in GraphPad Prism v9.4.0. Network and pathways analyses were performed using QIAGEN Ingenuity Pathway Analysis (IPA) v01-19-007,42.
Ethics and inclusion statement
This research was conducted locally in Los Angeles, California and was approved by the UCLA Internal Review Board. The authors come from diverse socioeconomic backgrounds and have expertise in a variety of disciplines across medicine and public health. This manuscript cites prior studies from across the United States, including Los Angeles, with an emphasis on sources about maternal and child health. Our results are locally relevant with the goal of providing physicians and patients more information about the benefits of COVID-19 vaccination. Furthermore, it can help guide future research and understandings of the long-term sequelae of SARS-CoV-2.
The roles and responsibilities of the authors were agreed upon prior to conducting the analysis. This project and future projects have been designed with special attention to continue to train medical students, medical residents and fellows, graduate students, and post doctorial researchers.
The data collection process and analysis methods did not result in increased stigmatization, incrimination, discrimination, or other personal risks to our study participants. This study did not cause increased health, safety, or security risks to the researchers or participants. No animal welfare, environmental protection, or biorisk-related regulations were violated by our study. Our study did not involve the transfer of biological materials, cultural artefacts, or associated traditional knowledge.
Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
[ad_2]
Source link