Maternal respiratory syncytial virus antibody titers: season and children matter

The seasonal variation in concentrations of maternal serum antibodies to respiratory syncytial virus fusion protein was investigated with the use of 761 maternal sera at delivery of term infants during a 12-month period. Mean antibody titers were significantly higher between April and September and significantly higher in women who already had children at home. Respiratory syncytial virus (RSV) is a significant cause of infant morbidity caused by lower respiratory tract infection. Increasing the RSV-specific maternal serum antibody concentrations near the time of delivery has been proposed as an option to confer some protection for newborn infants especially in the first 6 months of life. 1 This concept is based on evidence that transplacental transfer of maternal antibody to RSV is protective against severe disease in infants. 2–4 Of the proteins that are likely to be a more specific correlate of protection is antibody to RSV fusion (F) protein, which is conserved in both A and B subtypes of RSV. This protein induces IgG1 antibody production, and its presence in large titers before infection has correlated with decreased risk and severity of disease in children. Purified F protein shows promise as a candidate RSV vaccine, and its use as a maternal immunogen has been reported. 5 The purpose of this study was to determine the seasonal variation in titers of antibody to the F protein of RSV in women at the time of delivery of term infants. Methods. The Ethics Review Board of the Ottawa Hospital approved the protocol. The hospital is a university-affiliated tertiary care hospital serving low and high risk obstetrics populations. A convenience sample was selected among women who had delivered full term (≥37 weeks gestation) healthy infants in the previous 48 h. The research nurse visited the obstetric unit three times a week and approached women who could communicate in English or French, were willing to hear about the study and were in good general health. Women were excluded if they were immunocompromised, had HIV infection, had any chronic or pulmonary illness or were unwilling to participate. After obtaining informed consent we administered a questionnaire and collected blood by venipuncture. Data collected included maternal age, smoking history, number of children living at home and history of upper (“cold”) or lower (pneumonia) respiratory tract infections in the 3 months before delivery. Using a birth registry database for the entire city, we compared characteristics of women in our sample with all women in the city who gave birth in the year 2000. Serum specimens were centrifuged, separated and frozen at −70°C. IgG titers to purified RSV F of the A subgroup of RSV (Long strain) were determined by enzyme immunoassay. Wyeth-Lederle Vaccines and Pediatrics supplied the purified RSV fusion protein. The RSV proteins were affinity-purified according to published methods and coated to 96-well enzyme immunoassay plates. 6 Serial 2-fold dilutions of serum were incubated in the wells followed by addition of alkaline phosphatase-conjugated goat anti-human IgG followed by substrate. Tests were done in duplicate. The antibody titers are reported as reciprocal log2 titers. The number of pediatric RSV confirmed (direct fluorescent antigen detection and/or culture of RSV) hospital admissions with home addresses within the city were obtained by discharge diagnosis of RSV infection [International Classification of Diseases, 9th revision, Clinical Modification code (ICD-9-CM) 079.6] or RSV bronchiolitis (ICD-9-CM 466.11) or RSV pneumonia (ICD-9-CM 480.1) from hospital medical records. Descriptive statistics were calculated for the participants. Summary statistics were expressed as the means ± sd. Using chi square tests we compared demographic data of our cohort with the total number of women from Ottawa who gave birth to full term healthy infants in 2000. Mean antibody concentrations by season and month were compared with an analysis of variance using Tukey’s test for multiple comparisons. The difference in antibody titers between those who had children at home was assessed with the t test. Correlation coefficients were assessed among smoking status, recent maternal respiratory tract symptoms and level of maternal antibody present. Multivariate regression analysis was performed to explore the relation between antibody titers and seasons, number of children living at home and history of upper or lower respiratory tract infections in the 3 months before delivery. Results. Blood was collected from 762 women from September 30, 1997 to September 29, 1998. Serum from one patient was missing, leaving 761 women with complete data sets and sera available for analysis. This sample represents 28.6% of the 2661 deliveries (preterm and full term) at the hospital during the study period. All women were previously healthy and delivered healthy term infants. There were 78.1% of Caucasian origin; the remainder was equally divided among eastern European, African, Caribbean, Asian and other origins. The mean age of the mothers was 30 years (sd 5.1). The percentage of mothers who had no other children at home was 39.6% (301 of 761). For those who had children, the mean number of children at home was 1.48 (sd 0.75). The study population was similar in age to the cohort of 7463 women in the city who had term births in the year 2000. However, 40% of women in our sample gave birth to their first child and 15.4% were smokers compared with 44.7% (P = 0.008) and 7.8% (P < 0.001) in the entire city cohort in the year 2000. The mean antibody titer for the RSV fusion protein F was 13.94 log2 with a minimum value of 4.17 log2 and a maximum value of 19.04 log2. Antibody titers from April to June and July to September were significantly higher than in January to March (P < 0.001 for both). Figure 1 shows the mean RSV F antibody titers and the number of children with RSV-confirmed hospital admissions in the Ottawa region by month.Fig. 1: Mean monthly RSV antibody titer (――) and number of children with confirmed RSV hospitalized in the Ottawa region from September 1997 to August 1998 (░⃞).The mean antibody level of women without children at home (13.70 ± 1.89) was lower (P = 0.006) than that of mothers who had at least one child at home (14.10 ± 1.99). A maternal history of respiratory tract infections, bronchitis or pneumonia in the preceding 3 months or smoking did not correlate with antibody titers. According to the linear regression model, each additional child at home increased (P = 0.019) the maternal mean antibody titers by 0.18 (95% confidence interval, 0.03, 0.33) compared with women who did not have children at home. No other variable included in the model was associated with variation in antibody titers. Discussion. There is a strong association between the presence of higher concentrations of maternal neutralizing antibodies to RSV and protection against severe RSV disease in young infants. 2–4, 7 Titers of RSV IgG antibody were previously shown to be higher during May to August. 8 Our data specifically illustrate the seasonal variation in specific antibody titers to the F protein, a more specific and immunogenic surface glycoprotein of RSV. Antibody concentrations are significantly higher during the second and third quarters of the year than in the last and first quarters. Assuming that most infections occur during the months of January to March, maternal antibody also seems to persist for 6 months after infection and would be waning before the next RSV season. The mean antibody titer in this cohort was quite high, indicating that pregnant women respond well to natural RSV infection as an antigenic stimulus. Titers of 12.6 log2 in cord blood of term infants have been previously documented indicating a close correlation between our results and theirs. 9 Although the titer of antibody that is “protective” is not precisely defined, Hall et al. 10 determined a risk of reinfection of 75% if the level of F antibody was <9.64 log2 and 25% if it was >12.64 log2 in adults. Accordingly the titers we detected in the second and third quarters would appear to be protective if adequate placental transfer occurred. Higher titers in women who have other children at home may be due to frequent viral spread from children to adults resulting in subclinical infections. 1112 The limitations of the study include its cross-sectional design and convenience sample that might limit the generalizability of the results. The titers that we found support the concept that high titers are found naturally in pregnant women at term, especially in the second and third quarters of the year. The seasonal variation of disease and number of children at home could affect RSV specific antibody titers and should be taken into consideration when evaluating for efficacy in foreseeable maternal immunization programs against RSV. Acknowledgments. We acknowledge the data provided to us by the Perinatal Partnership Program of Eastern and Southern Ontario. The study was funded by the Children’s Hospital of Eastern Ontario Research Institute.