MENINGOCOCCAL VACCINE IN PREGNANCY: AN ASSESSMENT OF INFANT RISK

Abstract

Disease caused by Neisseria meningitidis occurs in the United States population with an incidence of 0.6 to 1.3/100 000 depending on serotype (∼2600 cases annually). Case fatality is 12% with 45% of annually reported cases caused by Group C organisms. Outbreaks of Serogroup C meningococcal disease have occurred with increasing frequency in the United States in the recent past.1 Asymptomatic carriage in the general population at any given time is from 5 to 10%.2 Group C meningococcus is a serotype for which a polysaccharide vaccine is efficacious in persons older than 2 years3 and is considered a useful tool for outbreak control. There have been case reports of serious maternal4, 5 and neonatal5 illness with N. meningitidis in pregnancy although the disease is rare in the perinatal period. There are several reports on the use of polysaccharide vaccines in pregnancy.6-12 The number of pregnant vaccinees reported is small, but the aggregate positive experience with several different polysaccharide vaccines has led to increasing comfort with use of these vaccines in pregnancy. Reported experience of meningococcal vaccine use in pregnancy includes 109 women who received North American quadrivalent vaccine at 30 to 38 weeks of gestation6, 7 and 21 who received a Brazilianmanufactured meningococcal vaccine whose gestation at time of immunization was not reported.8 No birth defects were present in the infants born of these vaccinated mothers. We report our experience with 34 women who received meningococcal vaccine, predominantly in the second and third trimesters, during a community-wide outbreak of Group C meningococcal disease. We evaluated their infants for the presence of birth defects and for evidence of developmental, neurologic or other abnormalities during a 2-year period following a mass immunization campaign in northwestern Wyoming. Methods. A mass immunization with meningococcal vaccine was prompted by the occurrence of four cases of Group C meningococcal disease, including one fatality, during a 9-month period in Teton County, WY, with a population of ∼13 000 permanent residents. Only two documented cases of meningococcal disease had been reported in the preceding 23 years in this county. Three of the 4 cases occurred in individuals between the ages of 13 and 25 years. County and State Public Health Officials made the decision to immunize the estimated 7000 members of the community between the ages of 2 and 30 years with Menomune® (quadrivalent meningococcal vaccine with antigens of N. meningitidis Groups A, C, Y and W135 manufactured by Connaught Laboratories). All physicians providing perinatal care in the county were asked to consider offering meningococcal vaccine to their pregnant patients after individual assessments of those patients. All of these physicians concluded that the benefit of the vaccine outweighed any potential risk and subsequently kept records of all pregnant women immunized during the mass immunization campaign conducted throughout the county in October and November of 1994. We reviewed, retrospectively, the medical records of all immunized pregnant women and of their delivered infants 2 years after the vaccine campaign. All diagnoses noted in hospital or clinic records occurring in the first 30 days of life were compared with expected frequencies of the same diagnoses published in the medical literature. We did not include suspected diagnoses that were not borne out by clinical investigation, such as suspected sepsis or suspected hip dislocation. Because our population was racially homogeneous (Caucasian), rates for Caucasian populations were used for comparison where race has been noted to be a factor in association with occurrence of an event. Maternal data collected on standardized forms included demographic information, pregnancy risk factors and complications, medications, alcohol and tobacco use, number and onset of prenatal visits and pregnancy outcome. Infant data collected on similar forms included demographic information, Apgar scores, birth weights, physical abnormalities at birth and any complications at birth or during the hospital stay. Infant charts were also reviewed for longitudinal data on growth and development, adequacy of routine preventive (well child) care immunizations and any conditions that might result from in utero exposure to meningococcal vaccine. All clinical conditions, growth or developmental abnormalities were tabulated for evidence of potential teratogenic effects apparent at birth or effects appearing during the first 2 years of life. Differences in observed and expected frequencies of abnormalities were compared by the Mantel-Haenzsel chi square test or the Fisher exact test as appropriate. Results. A total of 5994 doses of vaccine were given to residents of the county, 4932 (70%) of which were among the estimated 7000 individuals belonging to the target age group of 2 to 30 years. In the 3 years after the meningococcal immunization campaign, there have been no further cases of meningococcal disease reported in the county. Thirty-four pregnant women ages 20 to 38 years (mean, 26.7 years) were immunized with meningococcal vaccine between October 4 and November 2, 1994. During the period of expected delivery for women immunized (October 5, 1994, and June 30, 1995), 201 live births were documented. Because no attempt was made to knowingly immunize women in the first trimester and assuming that those pregnancies would be roughly one-third of the total, there were ∼135 pregnant women who could have been immunized. We therefore estimate that 34 of 135 (25.2%) of those in the second or third trimester during the 1-month campaign were immunized. The 34 women who received quadrivalent meningococcal vaccine during pregnancy had 34 singleton deliveries, all of whom survived and were followed during the study period. All mother-infant pairs were Caucasian. The average length of follow-up for the infants was 13.2 months (range, 1 to 24 months) with 29 of 34 (85.3%) being followed ≥6 months and 27 of 34 (79.4%) ≥12 months. The trimester of immunization was first for 4 (11.8%), second for 17 (50.0%) and third for 13 (38.2%) mothers. For 32 of 34 (94.1%) and 31 of 34 (91.2%) infants, well-child assessment and immunizations, respectively, were up to date. None of the 34 infants had significant growth or developmental delays in the follow-up period. Unusual birth events and their expected rate of occurrence are shown in Table 1. Eight of these abnormalities occurred at a frequency that was not significantly different than the expected frequency reported in the literature. Two abnormalities occurred with significantly lower frequency than reported in the literature, cardiac murmur and physiologic jaundice. Only one child had an abnormality that occurred at greater than unexpected frequency. This child, whose mother had received vaccine at 33 weeks of gestation, was born with anomalies consistent with the oromandibular limb hypogenesis spectrum (Charlie M syndrome).24 The genetics/dysmorphology evaluation at the University of Utah included multiple follow-up visits for 2 years and consultation with Dr. R. J. Gorlin who first described the syndrome and who concluded that this child best fit the "Charlie M" variant of the spectrum. Charlie M syndrome is not presently identified as a genetic condition and occurs sporadically with no known cause. There are no reports of possible cause by teratogens.25TABLE 1: Unusual birth events Discussion. This study found no teratogenicity of polysaccharide quadrivalent meningococcal vaccine when given to pregnant women. The number of women immunized in the first trimester was only four, limiting the power of this observation. Teratogenicity has not been noted in prior studies of polysaccharide vaccine use in pregnancy.6-12 All but one of these studies used vaccine in the third trimester. The one exception was similar to the situation we describe, in that it resulted from a mass immunization during a meningococcal outbreak. Unfortunately these authors did not report the gestational age at immunization, and it is not clear that there was complete ascertainment of the pregnant vaccinees.8 We identified one child with birth defects consistent with a defined syndrome that does not have a reported cause. It seems unlikely that meningococcal vaccine would be responsible for this syndrome. In our case the vaccine was given well into the third trimester. The type of anomalies seen in this patient very likely would have occurred in the first 4 weeks of the first trimester.26 There has not been vaccine or other potential teratogen-associated exposure reported occurring with any of the rare oromandibular limb hypogenesis spectrum variants. We attach no significance to the lesser occurrence of cardiac murmur and physiologic jaundice observed in our patients. We think that we have complete ascertainment of immunized pregnancies that were recognized. Some women may have been immunized at Public Health clinics before their pregnancies were recognized and in the absence of specific concerns may not have reported the immunization to their physicians. This seems unlikely because of the prominent media attention to the outbreak and the widespread participation of the public in the immunization campaign. The choice for immunization was left to the woman and her primary obstetric provider leaving no record of meningococcal vaccine refusals or the total number of pregnant women offered vaccine. Given these limitations we estimate that 25% of eligible pregnant women recognized the risks and elected immunization. Prior studies of meningococcal and other polysaccharide vaccines used in pregnancy have focused on maternal and infant antibody response rather than teratogenic effects.6-12 These studies have noted that polysaccharide vaccine use in pregnancy yields lower cord blood:maternal blood ratios of antibody than seen with other antigens given during pregnancy such as tetanus, measles, mumps and rubella.7 This has raised the question of induced immune tolerance when exposing a neonate or fetus to polysaccharide antigen. It is known, however, that factors such as maternal malaria infection affect the ratio of antibody transfer across the placenta.7 In addition there is a correlation between lower transplacental transfer to the fetus and the length of time between maternal immunization and delivery of the neonate.12 Studies in the neonate with the use of Haemophilus influenzae type b (Hib) conjugate vaccines have shed some light on this phenomenon. Granoff et al.27 found enhancement of response to immunization at 2, 4 and 6 months of age with Hib conjugate vaccines (PRP-T and HbOC) when neonates were immunized first with diphtheria-tetanus vaccine at one month of age. Lieberman et al., however, found that immunization with diphtheria-tetanus vaccine at birth resulted in lower Hib and diphtheria antibody levels in response to PRP-T, HbOC and diphtheria-tetanus-pertussis given at 2, 4 and 6 months of age.28 Alaskan infants immunized with the Hib conjugate vaccine PRP-OMP at birth and given subsequent doses, had lower Hib antibody concentrations than their counterparts who received their first dose of PRP-OMP at 2 months of age.29 These data suggest that induction of immune tolerance in the infant is related to timing of immunization whether the first dose of vaccine is given to the mother or to the infant. Unfortunately there is no such direct evidence of immune tolerance induction by neonatal immunization with the currently licensed meningococcal vaccine. It would be difficult to study given the generally poor response of infants to this polysaccharide. Our study made no attempt to assess the immune response of the infants of immunized mothers. We looked directly at the question of teratogenicity and developmental or growth abnormalities as a possible effect of meningococcal vaccination in pregnancy. We found no evidence of subtle adverse effects in this cohort. These children were evaluated with a high level of consistency and timeliness. The quality of follow-up is a consequence of the highly developed health care system in this county and the special attention paid to this particular cohort of children. As with all studies published to date, the small numbers make an accurate assessment difficult. In addition the "expected" frequencies are those derived from reported literature and have the disadvantage of not having come from a direct comparison group. Nonetheless the aggregate of information on the use of these types of vaccines in pregnancy suggests safety of use in pregnancy. This is particularly important with meningococcal disease given the potential seriousness of the disease when acquired in pregnancy. Because of the potential value of polysaccharide vaccines in pregnancy and the nagging concerns about the safety of their use in this circumstance, some thought should be given to collecting national data on the use of these and other vaccines in pregnancy to create a data set large enough to resolve this issue in a definitive fashion. This would facilitate timely use of these vaccines in pregnancy in future disease outbreaks or for other public health purposes such as prevention of neonatal group B streptococcal disease. Acknowledgments. We acknowledge the help and diligence in identifying and following the patients by Roger Brecheen, M.D., Robert "Buzz" Bricca, M.D., Anne Fenn, M.D., Thomas Pockat, M.D., Richard Sugden, M.D., Elizabeth Ridgway, M.D., Teton County Public Health Nursing and Wyoming Immunization Program. We thank Connaught Laboratories for use of their Menommune® vaccination data form as a template for questionnaire development and Jay Wenger, M.D., of the Centers for Disease Control and Prevention for arranging electropherotype testing of the available case isolates and providing excellent advice and assistance on the management of the outbreak. G. William Letson, M.D., M.S. James R. Little, M.D. Jayne Ottman, R.N., M.S. Gayle L. Miller, D.V.M., M.P.H. Wyoming Department of Health; Cheyenne, WY (GWL, JO, GLM) Jackson Pediatrics; Jackson, WY (JRL)