Background
Pregnant women and their neonates represent 2 vulnerable populations with an interdependent immune system that are highly susceptible to viral infections. The immune response of pregnant women to severe acute respiratory syndrome coronavirus 2 and the interplay of how the maternal immune response affects the neonatal passive immunity have not been studied systematically.
Objective
We characterized the serologic response in pregnant women and studied how this serologic response correlates with the maternal clinical presentation and with the rate and level of passive immunity that the neonate received from the mother.
Study Design
Women who gave birth and who tested positive for immunoglobulin M or immunoglobulin G against severe acute respiratory syndrome coronavirus 2 using semiquantitative detection in a New York City hospital between March 22, 2020, and May 31, 2020, were included in this study. A retrospective chart review of the cases that met the inclusion criteria was conducted to determine the presence of coronavirus disease 2019 symptoms and the use of oxygen support. Serology levels were compared between the symptomatic and asymptomatic patients using a Welch 2 sample t test. Further chart review of the same patient cohort was conducted to identify the dates of self-reported onset of coronavirus disease 2019 symptoms and the timing of the peak immunoglobulin M and immunoglobulin G antibody levels after symptom onset was visualized using local polynomial regression smoothing on log 2 -scaled serologic values. To study the neonatal serology response, umbilical cord blood samples of the neonates born to the subset of serology positive pregnant women were tested for serologic antibody responses. The maternal antibody levels of serology positive vs the maternal antibody levels of serology negative neonates were compared using the Welch 2 sample t test. The relationship between the quantitative maternal and quantitative neonatal serologic data was studied using a Pearson correlation and linear regression. A multiple linear regression analysis was conducted using maternal symptoms, maternal serology levels, and maternal use of oxygen support to determine the predictors of neonatal immunoglobulin G levels.
Results
A total of 88 serology positive pregnant women were included in this study. The antibody levels were higher in symptomatic pregnant women than in asymptomatic pregnant women. Serology studies in 34 women with symptom onset data revealed that the maternal immunoglobulin M and immunoglobulin G levels peak around 15 and 30 days after the onset of coronavirus disease 2019 symptoms, respectively. Furthermore, studies of 50 neonates born to this subset of serology positive women showed that passive immunity in the form of immunoglobulin G is conferred in 78% of all neonates. The presence of passive immunity is dependent on the maternal antibody levels, and the levels of neonatal immunoglobulin G correlate with maternal immunoglobulin G levels. The maternal immunoglobulin G levels and maternal use of oxygen support were predictive of the neonatal immunoglobulin G levels.
Conclusion
We demonstrated that maternal serologies correlate with symptomatic maternal infection, and higher levels of maternal antibodies are associated with passive neonatal immunity. The maternal immunoglobulin G levels and maternal use of oxygen support, a marker of disease severity, predicted the neonatal immunoglobulin G levels. These data will further guide the screening for this uniquely linked population of mothers and their neonates and can aid in developing maternal vaccination strategies.
Introduction
As the novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) spread rapidly through New York City in March 2020—the global epicenter of the disease at that time—the obstetrical unit within a New York City hospital implemented universal testing of all women admitted to the labor and delivery unit to screen this uniquely vulnerable patient population. During this peak of the pandemic, 10% to 15% of all women admitted to labor and delivery units in the New York City area tested positive for SARS-CoV-2 using reverse transcription polymerase chain reaction (RT-PCR) testing. , An updated report from the Centers for Disease Control and Prevention in October 2020 stated that pregnant women with symptomatic coronavirus disease 2019 (COVID-19) infections were at an increased risk for intensive care unit admission, invasive ventilation, extracorporeal membrane oxygenation, and death. Additional prospective and retrospective studies have shown that pregnant women infected with SARS-CoV-2 are at an increased risk for other morbidities as well, including higher rates of cesarean delivery, increased postpartum complications (including fever, hypoxia, and hospital readmissions postdischarge) and placental pathology including fetal vascular malperfusion; however, it should be noted that the risk for premature delivery may still require further study.
Why was this study conducted?
Previous studies on the serologic response to severe acute respiratory syndrome coronavirus 2 viral infection have been focused on the general population but the timing and level of serologic response in pregnant women are not well characterized. The passive transmission of maternal antibodies to neonates have not been studied systematically at scale.
Key findings
Asymptomatic pregnant women mount a lower serologic response than symptomatic pregnant women. The timing of immunoglobulin M (IgM) and immunoglobulin G (IgG) antibody response levels peak at 15 days and 30 days after onset of coronavirus disease 2019 symptoms, respectively. The maternal IgG antibodies correlate positively with and predict the antibody levels of the neonates.
What does this add to what is known?
This study provides a comprehensive, semiquantitative analysis of the levels and timing of IgM and IgG antibodies in pregnant women. Mothers with higher antibody levels exhibit a higher likelihood of transferring antibodies to their neonates.
There has been a recent interest in serology testing as a means of detecting exposure to SARS-CoV-2, to limit disease spread, and to potentially predict outcomes. Studies have reported that nearly 100% of patients with confirmed SARS-CoV-2 will receive positive test results for immunoglobulin G (IgG) or immunoglobulin M (IgM) antibodies within 19 days of exposure, even after RT-PCR results revert to negative. Some data suggest that high levels of IgG are associated with more severe illness, whereas asymptomatic patients are more likely to convert to serology negative in the convalescent phase of infection. The protective nature of antibodies against SARS-CoV-2 against future infection is still unclear.
Pregnant women and their neonates have a unique interdependent immune system. The interactions between the maternal immune system and fetal placenta result in changes that alter both the innate and adaptive host responses to infections. For this reason, current studies on the serologic responses to SARS-CoV-2 may not be applicable to the pregnant population. Studies to date have demonstrated the rate of serology positive results in pregnant women, , but a detailed analysis of the timing and levels of the responses in these pregnant patients have not been well characterized.
Passive transfer of antibodies against SARS-CoV-2 has not been studied systematically beyond the demonstration of neonatal antibodies in a small number of cases, and it is not clear if these antibodies are protective against disease. There have been reports of transplacental transmission of SARS-CoV-2 and of symptomatic neonates who tested positive for SARS-CoV-2 after being born to mothers with positive PCR test results. However, in these cases, the serologic status of the mother was either negative or not reported. A small case series demonstrated the presence of antibodies in RT-PCR–negative, asymptomatic neonates born to symptomatic women. However, the rate of transfer of maternal antibodies from the mother to the neonate and if asymptomatic women can transfer antibodies to neonates have not been established yet. In this paper, we aimed to systematically explore the serologic responses of mothers and neonates in both symptomatic and asymptomatic cases.
Materials and Methods
Study population
A total of 88 pregnant women who tested positive for SARS-CoV-2–specific antibodies (serology positive) at a single institution in New York City were included in this study. Of the 88 women, 67 delivered between April 18, 2020, and May 31, 2020, and were identified to be serology positive using universal serology testing. An additional 21 of the 88 women were identified to be serology positive between March 22, 2020, and April 17, 2020, after undergoing testing because of a suspicion of SARS-CoV-2 infection or exposure. The neonates born to these mothers were also included in this study and underwent serology testing using umbilical cord blood.
Laboratory testing
Patients were tested for IgM and IgG antibodies against SARS-CoV-2 using the serum or plasma from peripheral blood (for the mothers) or the plasma from umbilical cord blood (for the neonates) using the clinical testing Pylon 3D platform (ET HealthCare, Palo Alto, CA). The Pylon 3D platform utilizes a fluorescence-based reporting system that allows for the semiquantitative detection of anti–SARS-CoV-2 IgG and IgM with a specificity of 98.8% and 99.4%, respectively. The antibody levels were expressed as log 2 (value) + 1.
Statistical analyses
To study the association between the symptoms and the serologic results, retrospective chart review was conducted to identify symptoms at the time of serology testing. Patients with any of the following COVID-19 symptoms, reported before or at the time of admission, were categorized as symptomatic: self-reported fever, cough, sore throat, rhinorrhea, shortness of breath, diarrhea, other gastrointestinal symptoms, myalgias, and loss of sense of taste or smell; , those without any of the listed symptoms were categorized as asymptomatic. The IgM and IgG values of symptomatic and asymptomatic patients were plotted as continuous variables and were expressed as the median (interquartile range) with error bars representing 95% confidence intervals. The serology antibody levels between the symptomatic and asymptomatic patients were compared using a Welch 2 sample t test. A P value of <.05 was considered statistically significant.
To study the time course of the antibody response at the cohort level, retrospective chart review was conducted to identify the dates of onset of the first COVID-19 symptoms. Of the 88 serology positive pregnant women, 34 had documentation of the specific dates of the onset of COVID-19 symptoms. The IgM and IgG antibody levels were correlated to the number of days elapsed from the date of COVID-19 symptom onset using local polynomial regression smoothing on log 2 -scaled serologic values.
We studied the relationship between the quantitative maternal and neonatal IgG levels using Pearson correlation and linear regression. To understand if the maternal antibody levels were different between the pregnant women who gave birth to serology positive neonates and those who gave birth to serology negative neonates, the maternal antibody levels were compared between those who delivered serology positive vs serology negative neonates using a Welch 2 sample t test.
Retrospective chart review identified 5 women who required oxygen support (3 women needed a nasal cannula and 2 women needed a nonrebreather) during their hospital course, which likely served as a marker of disease severity. A multiple linear regression analysis was conducted to predict neonatal IgG levels from maternal symptoms (present or absent), maternal IgG level, maternal IgM level, and maternal use of oxygen support.
Statistical analyses were performed using R (version 3.6.1 R Core Team, Austria), RStudio (version 1.1.463, RStudio Team, Boston, MA), and SPSS Statistics (version 1.0.0.1461, IBM Corp, Armonk, NY).
This study was approved by the institutional review board protocols 20-03021682 and 20-04021792. A waiver of consent was granted by the institutional review board.
Results
We identified 88 pregnant women who tested positive for SARS-CoV-2 IgM or IgG antibodies (10 women were IgM positive, 24 were IgM and IgG positive, and 54 women were IgG positive). A retrospective chart analysis of all 88 serology positive mothers showed that 42.0% (37/88) were symptomatic, whereas 58.0% (51/88) of serology positive patients were asymptomatic ( Figure 1 ). Both asymptomatic and symptomatic pregnant women mounted a detectable IgM ( Figure 1 , B) and IgG ( Figure 1 , A) response, however, the IgG levels were significantly higher in symptomatic mothers than in asymptomatic mothers ( P =.029) ( Figure 1 , A).
