Abstract
This observational study explores whether rubella serostatus, which is routinely assessed during pregnancy, can serve as a proxy for measles serostatus in parturient persons.
Measles cases worldwide increased 79% during the first 2 months of 2022 compared with 2021, and experts warn of increasing outbreaks of measles.1 Newborns and infants are at particular risk for severe complications and death from measles and rely on maternally derived, placentally transferred antibodies for protection until routine measles-mumps-rubella (MMR) vaccination; the vaccine is administered starting at 12 months of age in the US.2 There are limited data on contemporary measles serostatus among parturient persons, and it is unclear whether rubella serostatus, which is routinely assessed during pregnancy, can serve as a proxy for measles serostatus.
Methods
This observational study was conducted with biobank serum samples collected from patients at hospital admission for childbirth from April 26, 2021, to October 24, 2021, at 2 academic birth hospitals in Philadelphia, Pennsylvania, serving a diverse population. The institutional review board at the University of Pennsylvania approved this study as having minimal risk and with a waiver of consent. Registry data included qualitative results from routine prenatal rubella IgG measurements collected up to 270 days before delivery.
Qualitative measles IgG testing was performed with biobank serum samples using a chemiluminescent immunoassay (Diasorin). Equivocal IgG results were considered seronegative per guidance from the US Centers for Disease Control and Prevention.3 The MMR administration dates were abstracted from medical records, but were not routinely available. Based on preliminary rubella serological data, we estimated 5% to be seronegative for measles and determined a priori that a random sample of 500 patients would allow for measles immunity estimation with an SE of 0.01. We oversampled by 10% to allow for unavailable serum samples. The proportions of patients seronegative for rubella and measles were estimated with 95% CIs.
Demographic and clinical characteristics were compared between patients seropositive and seronegative for measles using standard descriptive statistics (Mann-Whitney test, χ2 test, or the Fisher exact test as appropriate). The Cohen κ coefficient with a 95% CI was computed to determine agreement between measles and rubella serostatus. A 2-sided P<.05 was considered statistically significant. Analyses were performed using SAS version 9.4 (SAS Institute Inc).
Results
Of 4886 patients who gave birth during the study period, 550 were randomly selected. Of the 550 patients, 535 (97.3%) had a rubella result and 513 (93.3%) had an available serum sample for measles testing; 500 (90.9%) had both. Rubella seronegativity was documented in 43 of 535 patients (8.0% [95% CI, 5.7%-10.3%]; 10 of 43 were equivocal) and measles seronegativity was documented in 103 of 513 patients (20.1% [95% CI, 16.6%-23.5%]; 13 of 103 were equivocal). Of the 42 patients who were seronegative for rubella (among those with both rubella and measles results), 14 (33.3% [95% CI, 19.1%-47.6%]) were seronegative for measles.
Characteristics of the patients who were seronegative vs seropositive for measles were not significantly different except for rubella status (Table 1). There were no significant differences between patients who were seropositive vs seronegative for measles during prepregnancy in MMR documentation (31.0% vs 25.2%, respectively) or in timing between last MMR vaccine dose and delivery. Among the 500 patients with both measles and rubella results, agreement between serostatus was poor (κ, 0.089 [95% CI, −0.002 to 0.181]) and the concordant proportion was 77.2% (95% CI, 73.5% to 80.9%; Table 2).
Table 1. Demographic and Clinical Characteristics by Seropositive and Seronegative Status for Measles (n = 513).
| Seropositive for measles (n = 410)a | Seronegative for measles (n = 103)a | P valueb | |
|---|---|---|---|
| Maternal characteristics | |||
| Age, median (IQR), y | 32 (27-35) | 31 (26-35) | .37 |
| Parity, No. (%) | |||
| 0 | 191 (46.6) | 47 (45.6) | .86 |
| ≥1 | 219 (53.4) | 56 (54.4) | |
| Insurance status, No. (%)c | |||
| Private | 224 (54.6) | 57 (55.3) | .15 |
| Public | 185 (45.1) | 44 (42.7) | |
| Other or unknown | 1 (0.2) | 2 (1.9) | |
| Race and ethnicity, No. (%)d | |||
| Hispanic or Latino | 38 (9.3) | 11 (10.7) | .27 |
| Non-Hispanic Asian | 31 (7.6) | 7 (6.8) | |
| Non-Hispanic Black | 175 (42.7) | 34 (33.0) | |
| Non-Hispanic White | 147 (35.9) | 48 (46.6) | |
| Other or unknowne | 19 (4.6) | 3 (2.9) | |
| Prepregnancy BMI, No. (%)f | |||
| <18.5 | 12 (3.0) | 3 (3.0) | .62 |
| 18.5-<25.0 | 183 (45.1) | 39 (39.0) | |
| 25.0-<30.0 (overweight) | 102 (25.1) | 25 (25.0) | |
| ≥30.0 (obese) | 109 (26.8) | 33 (33.0) | |
| Maternal history, No. (%) | |||
| Diabetesg | 21 (5.1) | 6 (5.8) | .78 |
| Hypertensive disorderg | 77 (18.8) | 18 (17.5) | .76 |
| Cesarean delivery | 125 (30.5) | 42 (40.8) | .05 |
| Preterm delivery <37 weeks’ gestation | 45 (11.0) | 13 (12.6) | .64 |
| Multiple gestation | 10 (0.2) | 3 (0.8) | .73 |
| Received ≥1 documented MMR vaccine dose before deliveryh | 127 (31.0) | 26 (25.2) | .26 |
| Received >1 documented MMR vaccine dose before delivery | 109 (26.6) | 21 (20.4) | .20 |
| Duration between most recent prepregnancy MMR vaccine dose and delivery, median (IQR), di | 7309 (5493-8535) | 6369 (2564-8352) | .36 |
| Duration between rubella serology testing and measles serology testing, median (IQR), dj | 189 (166-204) | 197 (172-205) | .95 |
| Seropositive for rubellaj | 372 (93.0) | 86 (86.0) | .02 |
| Seronegative for rubellaj | 28 (7.0) | 14 (14.0) | |
| Documented MMR vaccine dose after delivery | 11 (2.7) | 7 (6.8) | .07 |
| Duration between postdelivery MMR vaccine dose and delivery, median (IQR), dk | 2 (2-3) | 2 (2-3) | .64 |
| Infant characteristics | |||
| Gestational age at delivery, median (IQR), wkl | 39.0 (37.9-39.9) | 39.4 (38.4-40.3) | .76 |
| Female sex, No. (%)l,m | 204 (48.7) | 53 (50.0) | .86 |
Abbreviations: BMI, body mass index; MMR, measles-mumps-rubella.
Patients with missing data for a variable were excluded from the statistical comparison for that variable.
The differences in maternal age, gestational age at delivery, duration between rubella serology and measles serology, duration between most recent prepregnancy MMR vaccine dose and delivery, and duration between postdelivery MMR vaccine dose and delivery were tested using the Mann-Whitney test. The differences in all other characteristics were tested using the χ2 test or the Fisher exact test as appropriate.
Categorization was determined via manual review by a health system physician financial advisor.
Self-reported and included because previous studies have reported racial and ethnic disparities in measles immunity among pregnant patients.
American Indian, Pacific Islander, multiple races, and missing race.
Calculated as weight in kilograms divided by height in meters squared. Data were missing for 4 patients who were seropositive for measles and 3 who were seronegative.
Diagnoses were based on delivery admission International Statistical Classification of Diseases and Related Health Problems, Tenth Revision diagnosis codes for diabetes (O24, E08-E13, and Z79.4) and hypertension (O10, O11, O13-O16, I10-I13, and I15).
Prepregnancy MMR data outside the health system were not routinely documented or available.
Based on data from 153 patients who had at least 1 prepregnancy MMR vaccine dose administered that was documented and dated; of which, 127 were seropositive for measles and 26 were seronegative.
Data were missing for 10 patients who were seropositive for measles and 3 who were seronegative.
Based on data from 18 patients who had at least 1 postdelivery MMR administration documented and dated; of which, 11 were seropositive for measles and 7 were seronegative.
Based on data from 526 infants (500 singletons and 26 twins).
Sex was unknown for 1 infant born to a patient seropositive for measles.
Table 2. Concordance Between Measles and Rubella Serostatusa.
| No. (%) | |||
|---|---|---|---|
| Seropositive for measles | Seronegative for measles | Total | |
| Rubella status | |||
| Seropositive | 372 (74.4) | 86 (17.2) | 458 (91.6) |
| Seronegative | 28 (5.6) | 14 (2.8) | 42 (8.4) |
| Total | 400 (80.0) | 100 (20.0) | 500 (100.0) |
Agreement between serostatus was poor (κ, 0.089 [95% CI, −0.002 to 0.181]) and the concordant proportion was 77.2% (95% CI, 73.5% to 80.9%).
Discussion
In this study conducted at 2 hospitals in Philadelphia, 1 in 5 parturient patients were seronegative for measles, suggesting that 1 in 5 newborns lack maternally derived, placentally transferred measles antibodies. Furthermore, rubella serostatus was a poor surrogate for measles serostatus. These findings underscore the need for a more detailed understanding of how maternal measles serostatus could be used to optimize infant protection during measles outbreaks.
Infants born to seropositive mothers may not benefit from early vaccination during outbreaks given decreased immunogenicity in the presence of maternal measles antibodies. Infants born to seronegative mothers, however, may benefit from supplemental early vaccination during measles outbreaks.4 Postpartum vaccination of patients seronegative for measles could also promote infant immunity via placental antibody transfer during subsequent pregnancies. These data expand on findings from studies of measles serostatus in other urban pregnant cohorts and are less prone to selection bias.5,6
This study has limitations. Serological immunity may differ from clinical immunity because patients who are seronegative may maintain immunity via an anamnestic antibody response. Only 2 centers from 1 urban location were included. Prepregnancy MMR administration data were incomplete and likely underestimated the proportion of vaccinated patients. The sample size may have precluded identification of significant demographic differences between seronegative and seropositive patients.
Section Editors: Jody W. Zylke, MD, Deputy Editor; Kristin Walter, MD, Senior Editor.
Data sharing statement
References
- 1.World Health Organization . UNICEF and WHO warn of perfect storm of conditions for measles outbreaks affecting children. Published April 27, 2022. Accessed July 26, 2022. https://www.who.int/news/item/27-04-2022-unicef-and-who-warn-of--perfect-storm--of-conditions-for-measles-outbreaks--affecting-children
- 2.Rasmussen SA, Jamieson DJ. What obstetric health care providers need to know about measles and pregnancy. Obstet Gynecol. 2015;126(1):163-170. doi: 10.1097/AOG.0000000000000903 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.US Centers for Disease Control and Prevention . Measles serology. Accessed December 13, 2022. https://www.cdc.gov/measles/lab-tools/serology.html
- 4.Principi N, Esposito S. Early vaccination: a provisional measure to prevent measles in infants. Lancet Infect Dis. 2019;19(11):1157-1158. doi: 10.1016/S1473-3099(19)30520-1 [DOI] [PubMed] [Google Scholar]
- 5.Chu Lam MT, Schmidt-Beuchat E, Geduldig E, et al. What is the prevalence of measles immunity among pregnant women? Am J Perinatol. 2021;38(01):016-022. doi: 10.1055/s-0040-1713817 [DOI] [PubMed] [Google Scholar]
- 6.Kassir E, Holliman K, Negi M, et al. Risk factors for measles nonimmunity in rubella-immune pregnant patients. Am J Perinatol. Published online May 31, 2022. doi: 10.1055/a-1799-5714 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data sharing statement
