Pandemic Influenza Vaccine and the Fetus
Pasternak B, Svanström H, Mølgaard-Nielsen D, et al
JAMA. 2012;308:165-174
The 2009 influenza A (H1N1) pandemic was associated with an increase in morbidity and mortality in pregnant women, prompting heavy targeting of these women for vaccination. Recognizing the need for more definitive fetal safety data related to this product, Pasternak and colleagues examined the relationship of maternal H1N1 vaccination and fetal outcome. Their primary goal was to determine whether H1N1 vaccination was associated with an increased risk for major birth defects, preterm birth, or fetal growth restriction in infants born to mothers vaccinated during pregnancy compared with infants who were not exposed to vaccination. This cohort study included more than 53,000 Danish infants born alive from November 2009 through September 2010. All data were obtained through nationalized health information systems. Multiple gestation infants were excluded, along with infants with any known chromosomal, genetic, or other birth defects and infants who experienced perinatal or congenital viral infections that might influence the outcomes.
Pregnant women with chronic medical problems were generally vaccinated during the first trimester, whereas women without chronic medical problems were vaccinated in the second or third trimester. Infants exposed in the first trimester (covering the period of embryogenesis) were specifically assessed for diagnosis of a major birth defect before 1 year of age. The annual frequency of major birth defects among live births in Denmark was approximately 2.4%. The frequency of birth before 37 weeks and birth in the lowest 10th percentile of gestational age-specific birth weight was evaluated for all infants.
Propensity scores were used to adjust for maternal confounders, including maternal age, place of birth, living location, parity, maternal smoke exposure and tobacco use, and body mass index, as well as other familial conditions that might be associated with the infant outcomes of interest. The cohort included 53,432 live-born infants, 13.1% of whom (6989) were exposed to influenza vaccine through maternal vaccination. The median gestational age at which the children were exposed to the vaccine fell during the second trimester. As expected, women who received influenza vaccine during the first trimester were older, had a previous history of adverse pregnancy outcomes, and had medical comorbid conditions.
In analyses matched by propensity scores, the frequency of major birth defects among infants exposed to vaccine in the first trimester was 5.5% compared with 4.5% among nonexposed infants, with a prevalence odds ratio of 1.12 (95% confidence interval, 0.60-2.45), a difference that was not statistically significant. In a similar pattern, the frequency of preterm birth was 9.4% in the infants exposed to the vaccine during the first trimester compared with 7.3% in the nonexposed infants, but this difference did not reach statistical significance. When looking at second- or third-trimester exposure, the rates of preterm birth were identical at 4.6% in both groups, and no increase in the frequency of low birth weight was seen.
Pasternak and colleagues found no significant associations between H1N1 vaccine exposure and the outcomes of interest, including major birth defects. They emphasized that the mothers who received first-trimester influenza vaccination were very different as a result of high-risk pregnancy conditions. These data provide reassurance for the safety of the 2009 H1N1 vaccine during pregnancy.
Pasternak and colleagues speculated that this safety information may also apply to other influenza vaccines, emphasizing the relatively small pool of data available for assessing fetal risk for influenza vaccination exposure. An accompanying editorial reviews other recent data, including data from Canada that demonstrated a similar lack of concerning outcomes in infants born of mothers who received influenza vaccination during pregnancy. Moreover, US data have not revealed an association with poor infant outcomes. These findings are reassuring, but they suggest that additional studies are needed to try to overcome the bias present in cohort and observational studies. In a study of pregnant women with H1N1, 22% of the women required intensive care and 8% died, so not vaccinating pregnant women likely poses more potential harm to the fetus than vaccine exposure.
Abstract
Risk of Adverse Fetal Outcomes Following Administration of a Pandemic Influenza A (H1N1) Vaccine During Pregnancy
Pasternak B, Svanström H, Mølgaard-Nielsen D, et al
JAMA. 2012;308:165-174
Safety of Maternal Pandemic Influenza Vaccination
The 2009 influenza A (H1N1) pandemic was associated with an increase in morbidity and mortality in pregnant women, prompting heavy targeting of these women for vaccination. Recognizing the need for more definitive fetal safety data related to this product, Pasternak and colleagues examined the relationship of maternal H1N1 vaccination and fetal outcome. Their primary goal was to determine whether H1N1 vaccination was associated with an increased risk for major birth defects, preterm birth, or fetal growth restriction in infants born to mothers vaccinated during pregnancy compared with infants who were not exposed to vaccination. This cohort study included more than 53,000 Danish infants born alive from November 2009 through September 2010. All data were obtained through nationalized health information systems. Multiple gestation infants were excluded, along with infants with any known chromosomal, genetic, or other birth defects and infants who experienced perinatal or congenital viral infections that might influence the outcomes.
Pregnant women with chronic medical problems were generally vaccinated during the first trimester, whereas women without chronic medical problems were vaccinated in the second or third trimester. Infants exposed in the first trimester (covering the period of embryogenesis) were specifically assessed for diagnosis of a major birth defect before 1 year of age. The annual frequency of major birth defects among live births in Denmark was approximately 2.4%. The frequency of birth before 37 weeks and birth in the lowest 10th percentile of gestational age-specific birth weight was evaluated for all infants.
Propensity scores were used to adjust for maternal confounders, including maternal age, place of birth, living location, parity, maternal smoke exposure and tobacco use, and body mass index, as well as other familial conditions that might be associated with the infant outcomes of interest. The cohort included 53,432 live-born infants, 13.1% of whom (6989) were exposed to influenza vaccine through maternal vaccination. The median gestational age at which the children were exposed to the vaccine fell during the second trimester. As expected, women who received influenza vaccine during the first trimester were older, had a previous history of adverse pregnancy outcomes, and had medical comorbid conditions.
In analyses matched by propensity scores, the frequency of major birth defects among infants exposed to vaccine in the first trimester was 5.5% compared with 4.5% among nonexposed infants, with a prevalence odds ratio of 1.12 (95% confidence interval, 0.60-2.45), a difference that was not statistically significant. In a similar pattern, the frequency of preterm birth was 9.4% in the infants exposed to the vaccine during the first trimester compared with 7.3% in the nonexposed infants, but this difference did not reach statistical significance. When looking at second- or third-trimester exposure, the rates of preterm birth were identical at 4.6% in both groups, and no increase in the frequency of low birth weight was seen.
Pasternak and colleagues found no significant associations between H1N1 vaccine exposure and the outcomes of interest, including major birth defects. They emphasized that the mothers who received first-trimester influenza vaccination were very different as a result of high-risk pregnancy conditions. These data provide reassurance for the safety of the 2009 H1N1 vaccine during pregnancy.
Viewpoint
Pasternak and colleagues speculated that this safety information may also apply to other influenza vaccines, emphasizing the relatively small pool of data available for assessing fetal risk for influenza vaccination exposure. An accompanying editorial reviews other recent data, including data from Canada that demonstrated a similar lack of concerning outcomes in infants born of mothers who received influenza vaccination during pregnancy. Moreover, US data have not revealed an association with poor infant outcomes. These findings are reassuring, but they suggest that additional studies are needed to try to overcome the bias present in cohort and observational studies. In a study of pregnant women with H1N1, 22% of the women required intensive care and 8% died, so not vaccinating pregnant women likely poses more potential harm to the fetus than vaccine exposure.
Abstract