Interference of maternal antibodies with the respective live-attenuated vaccines is well-known. 1 IgG antibodies are transferred from mother to fetus by an active transport mechanism starting at ~17 weeks of gestation. Cord blood antibody values are in balance with maternal titers at ~33 weeks of gestation and reach 1.5- to 2-fold higher values at term. 2 Therefore the magnitude of transplacentally acquired specific maternal antibodies should be lower in preterm newborns than in full term newborns. Two recent studies demonstrated lower antibody values against rubella and varicella-zoster virus (VZV) at birth in preterm infants than in full term infants, but no such data are available for measles and mumps. 3, 4
The goal of our study was to compare IgG antibody values against measles, mumps, rubella and VZV in full term and preterm infants and their mothers at birth and to assess the decay of passively acquired maternal antibodies during the infants’ first year of life.
Women who gave birth in either one of the two obstetric departments of the University Hospitals of Basel, Switzerland, between December 1999 and December 2000 were eligible for participation. Women were excluded if they had received immunoglobulin therapy at any time during pregnancy or infusion of blood or blood products within 3 months before delivery. Other exclusion criteria were chronic diseases with possible impairment of the immune system (i.e. HIV infection, malignancy, diabetes mellitus) or residency outside the Basel area.
Participants were categorized in two groups: Group A, women and their offspring born before 37 weeks of gestation (preterm); Group B, women and their infants born at or after 37 weeks of gestation (full term). For specific analyses Group A was subdivided into Groups A1 (24 to 27 weeks of gestation), A2 (28 to 31 weeks of gestation) and A3 (32 to 36 weeks of gestation).
Maternal age, nationality and personal history for measles, mumps, rubella and VZV were documented. Study enrollment was restricted to the first 70 women with term newborns, but it remained open for women with preterm newborns throughout the study period.
After informed consent had been obtained, aliquots of maternal blood specimens, which had been collected as part of the routine procedures immediately before or during delivery, were separated, and serum was kept at −20°C until analysis. A few milliliters of cord blood were collected directly after birth and were separated and stored as serum at −20°C. Further serum aliquots were collected from infants during their hospital stay whenever blood was taken for medical reasons.
After discharge from the hospital, a letter was sent to the responsible pediatrician to ask for an aliquot of serum in case blood would be taken from the infant during the first 12 months of life. In addition a blood sample was requested from each infant at 6 to 9 months of age. These blood specimens were taken either at the pediatrician’s office or in our outpatient clinic after a reminder had been mailed to the study families when the child had reached 6 months of age.
All analyses were performed in the virologic laboratory of the University Children’s Hospital, Basel, Switzerland. For quantitative determination of IgG antibody values against measles, mumps, rubella and VZV, commercially available enzyme-linked immunosorbent assays were used (measles: Enzygnost Anti-Masern-Virus/IgG; Dade-Behring, Düdingen, Switzerland; mumps: MUMPS-VIRUS ELISA IgG Antikörper-Test; RUWAG, Zürich, Switzerland; rubella: Platelia Rubella IgG; Bio-Rad Laboratories, Reinach, Switzerland; VZV: Enzygnost Anti-VZV/IgG; Dade-Behring). Paired maternal and children’s specimens were analyzed in parallel on the same plates, and positive and negative controls were included in all assays. Values above the cutoff values (according to instructions by the manufacturers: 10 IU/ml, 2.5 PEI/ml, 15 IU/ml and 100 mIU/ml for measles, mumps, rubella and VZV, respectively) were considered to be positive and were used for quantitative analyses.
Transplacental transfer ratios were calculated by dividing cord and maternal blood antibody values. 4 In twins and triplets cord blood values from the first child were used for calculations. Specimens from mothers with one or more negative antibody values were excluded from calculation of transplacental transfer ratios for the respective viruses. For infants with negative antibody values, 0.5 of the cutoff value of the respective test was used for calculations, provided that the value in the maternal serum specimen was positive.
χ2 tests and linear regression were used for statistical analysis. To analyze linear relationships of gestational age with antibody values of all mothers and of all children, titers were logarithmically transformed. A paired t test was used to compare antibody levels of cord blood specimens with levels of blood samples from the same newborns taken within the first 5 days of life. All statistical analyses were performed with STATA 7.0 (Stata Co.).
The study protocol was approved by the University’s ethics committee.
General characteristics of the study population.
We enrolled 152 mothers and 172 newborns including 18 pairs of twins (A1, 1; A2, 5; A3, 10; B, 2) and one set of triplets (A3). Age was known in 147 mothers, and the mean was 30 years (range, 19 to 41) with no significant differences between groups.
Antibody values in maternal blood specimens.
Blood samples from 96 (63%) mothers were available for analysis. We did not receive blood specimens from the remaining 56 mothers, because the short time between admission to hospital and birth was insufficient to obtain informed consent.
Positive maternal antibody values against measles, mumps, rubella and VZV were found in 92 (96%), 60 (62%), 88 (86%) and 94 (98%) subjects, respectively. Maternal geometric mean values (GMV) against measles, mumps, rubella and VZV were 2956 IU/ml [95% confidence interval (CI) 2328 to 3754], 6.5 PEI/ml (95% CI 5.3 to 8.0), 78 IU/ml (95% CI 66 to 92) and 798 mIU/ml (95% CI 660 to 966), respectively. They did not significantly vary among Groups A1, A2, A3 and B (data not shown).
Maternal age had no influence on the GMV against mumps, rubella and VZV (data not shown). In contrast GMV against measles increased significantly with maternal age with 1812 IU/ml (95% CI 1079 to 3043), 3232 IU/ml (2256 to 4630) and 4300 IU/ml (3021 to 6120) in women 19 to 28, 29 to 32 and 33 to 41 years of age, respectively (P < 0.001).
Antibody values in blood specimens from infants.
Cord blood specimens were available for analysis in 135 (78%) of the 172 newborns (n = 62 in full term infants and n = 73 in preterm infants). Of the full term infants 59 (95%) had positive antibody values against measles, 45 (73%) against mumps, 58 (94%) against rubella and 61 (98%) against VZV. Similarly of preterm infants 70 (96%) had positive antibody values against measles, 46 (63%) against mumps, 66 (90%) against rubella and 70 (96%) against VZV (all P > 0.05 when compared with full term infants). One infant (34 weeks of gestational age) had a negative antibody value against mumps, and the corresponding maternal blood specimen was just above the cutoff level (2.7 PEI/ml).
Antibody values in 59 cord blood specimens were compared with 59 venous blood specimens taken during the first 5 days of life from the same children. There were no significant differences in GMV against measles (4358 vs. 4341), mumps (5.5 vs. 5.1), rubella (88 vs. 111) and VZV (1324 vs. 1322).
Maternal and neonatal antibodies were highly correlated. There was an independent association between gestational age and cord blood antibody values for all four viruses (all P < 0.001). Geometric mean antibody values in the youngest gestational age group reached only 29 to 51% of the values of full term neonates: measles GMV increased from 1270 IU/ml (Group A1) to 4383 IU/ml (Group B); mumps GMV from 5.3 PEI/ml to 10.3 PEI/ml; rubella GMV from 47 IU/ml to 123 IU/ml; and VZV GMV from 307 mIU/ml to 1060 mIU/ml.
Transfer ratios of maternal and cord blood antibody values.
Transfer ratios (TR) of IgG antibody values against measles, mumps, rubella and VZV were calculated by dividing cord blood and maternal blood antibody values. There were 86 pairs of cord and maternal blood serum specimens available. For each of the 4 viruses, TR increased with gestational age of the newborn, and values >1 were reached by 32 to 36 weeks of gestation. The positive association of TR with gestational age was highly significant for all viruses (P < 0.001).
Antibody decay in infants during the first year of life.
As shown in Table 1, at the age of 1 to 3 months, 62, 17, 33 and 50% of preterm infants born before 28 weeks of gestation (Group A1) had measurable maternal antibodies against measles, mumps, rubella and VZV, respectively. In contrast all infants in Group A2 (born between 28 and 32 weeks of gestation) still had measurable antibodies against measles and VZV, and 25% had measurable antibodies against mumps and rubella. All children born between 32 and 36 weeks of gestation (Group A3) were positive for anti-measles antibodies, and 0, 29 and 43% were positive for mumps, rubella and VZV, respectively.
Between 6 and 12 months of age, most children (96%; 95% CI 89 to 99) had lost maternal antibodies against mumps. All infants from Groups A1 and A2 (95% CI 85 to 100) were also negative for measles, rubella and VZV. In contrast 19% (95% CI 10 to 33) and 10% (95% CI 3 to 21) of infants born after 32 weeks of gestation (Groups A3 and B), still had measurable IgG antibodies against measles and against rubella and VZV, respectively.
In this study >90% of mothers and their newborns had measurable antibodies against measles, rubella and VZV at birth irrespective of the infant’s gestational age. However, only 63% of mothers and 70% of newborns had measurable IgG values against mumps. These findings confirm similar observations from a recently published study by members of our group on the seroprevalence of measles, mumps and rubella antibodies in full term Swiss newborns. 5 It was previously shown that IgG seroconversion was only 38% in subjects vaccinated with the widely used Swiss Rubini mumps strain; this may explain our findings. 6 Maternal age was positively related to the magnitude of anti-measles IgG antibody values, but not to mumps, rubella and VZV. We believe that wild-type measles virus infection in older women, resulting in comparatively higher antibody values as opposed to vaccine-induced antibodies in younger women, is the most likely explanation for this observation. In contrast immunization against chickenpox has not been recommended in Switzerland, and similar antibody values independent of maternal age reflect previous natural infections. With respect to rubella, the former Swiss policy to immunize adolescent girls (since 1973) probably accounts for the findings. However, the rate of 14% of mothers seronegative for anti-rubella IgG serum antibodies is of concern, and better implementation of prepregnancy immunization against rubella is needed in Switzerland.
The prevalence of specific antibodies against measles, mumps, rubella and VZV in newborns reflected those of their mothers. This was independent of the children’s gestational age, indicating that transplacental transfer per se took place. However, gestational age significantly correlated with the magnitude of antibody transfer, as indicated by lower geometric mean ratios in preterm than in full term infants. This finding supports theoretical considerations based on the biology of transplacental IgG antibody transfer 2, 7 and similar clinical observations by other investigators. 3, 4, 8, 9 Van der Zwet et al. 8 demonstrated that maternal titers were the best predictive factor for neonatal titers of VZV, which is in line with our findings. In contrast to their analyses, our results also show that gestational age significantly influenced the level of neonatal anti-VZV IgG antibody values independent of values in maternal blood. This is indicated by lower values in preterm than in full term infants which we found not only for VZV but also for measles, mumps and rubella. Similarly in studies conducted in Israel by Linder et al., 3, 4 such a correlation was found for neutralizing rubella antibodies and for VZV antibodies in infants born beyond 28 weeks of gestation.
A significant number of infants born before 28 weeks of gestation lost maternal antibodies during the first 3 months of life, and between 6 and 12 months of age none of the infants born before 32 weeks of gestation still had measurable antibodies against measles, rubella and VZV. Of children born after 32 weeks of gestation, <20% still had measurable IgG antibodies against measles, rubella and VZV and none against mumps. Early loss of maternally derived serum antibodies puts preterm infants at increased risk for natural infection. For preterm infants the window of susceptibility may therefore be as wide as 6 to 9 months, which is especially of concern for measles. In this context it is tempting to speculate whether preterm infants should be immunized with MMR at an earlier age than full term infants. As studies in full term infants have shown, MMR given before 12 months of age will result in lower seroconversion rates compared with administration at 12 months of age or later. 10 However, specific T cell responses can be induced by early immunization despite reduced humoral responses. 11 Whether this is also true for preterm infants remains to be shown.
We thank the staff of the neonatal and obstetric departments of the University Hospitals of Basel for their support.
1. Markowitz LE, Albrecht P, Rhodes P, et al. Changing levels of measles antibody titers in women and children in the United States: impact on response to vaccination. Kaiser Permanente Measles Vaccine Trial Team. Pediatrics 1996;97:53–8.
2. Malek A, Sager R, Kuhn P, Nicolaides KH, Schneider H. Evolution of maternofetal transport of immunoglobulins during human pregnancy. Am J Reprod Immunol 1996;36:248–55.
3. Linder N, Sirota L, Aboudy Y, et al. Placental transfer of maternal rubella antibodies to full-term and preterm infants. Infection 1999;27:203–7.
4. Linder N, Waintraub I, Smetana Z, et al. Placental transfer and decay of varicella-zoster virus antibodies in preterm infants. J Pediatr 2000;137:85–9.
5. Desgrandchamps D, Schaad UB, Glaus J, Tusch G, Heininger U. Seroprevalence of IgG antibodies against measles, mumps and rubella in Swiss children during the first 16 months of life. Schweiz Med Wochenschr 2000;130:1479–86.
6. Schwarzer S, Reibel S, Lang AB, et al. Safety and characterization of the immune response engendered by two combined measles, mumps and rubella vaccines. Vaccine 1998;16:298–304.
7. Evans HE, Akpata SO, Glass L. Serum immunoglobulin levels in premature and full-term infants. Am J Clin Pathol 1971;56:416–8.
8. van der Zwet WC, Vandenbroucke-Grauls CM, van Elburg RM, Cranendonk A, Zaaijer HL. Neonatal antibody titers against varicella-zoster virus in relation to gestational age, birth weight, and maternal titer. Pediatrics 2002;109:79–85.
9. Mendez DB, Sinclair MB, Garcia S, Chou T, Meadow W. Transplacental immunity to varicella-zoster virus in extremely low birthweight infants. Am J Perinatol 1992;9:236–8.
10. Klinge J, Lugauer S, Korn K, Heininger U, Stehr K. Comparison of immunogenicity and reactogenicity of a measles, mumps and rubella (MMR) vaccine in German children vaccinated at 9–11, 12–14 or 15–17 months of age. Vaccine 2000;18:3134–40.
11. Gans H, Yasukawa L, Rinki M, et al. Immune responses to measles and mumps vaccination of infants at 6, 9, and 12 months. J Infect Dis 2001;184:817–26.