Journal Article Annotations
2017, 4th Quarter
Annotation by Jacquelin Esque, MD
Type of study: Cohort study
The finding: This was a cohort study evaluating risks of congenital malformations in pregnant women exposed to stimulants in the first trimester. Authors looked at data from US Medicaid beneficiaries (1,813,894 pregnancies) and conducted validation analyses using data from the Nordic Health registries (2,560,069 pregnancies). The authors reported a 28% increase in risk of cardiac malformations associated with intrauterine exposure to methylphenidate when looking at pooled data from the US and Nordic countries (RR 1.28 (95% CI, 1.00-1.64). The authors did not find a significantly increased risk for all congenital malformations with methamphetamine use and did not find any significantly increased risks for congenital malformations or specifically cardiac malformations for amphetamine use.
Strength and weaknesses: Strengths include that this study evaluated data from 6 different countries. Exposure and outcome data was independently and prospectively collected preventing recall bias. Surveillance bias was thought to be limited as noted that there was no increased risk for malformations in infants exposed to amphetamines. Authors accounted for possibility of nonadherence to prescribed stimulants by increasing specificity of the exposure definition, requiring medication to be dispensed to women twice during the time window and noted that this did not change findings.
Limitations include that both data sets looked only at live births, which may result in underestimation of relative risk due to selection bias. There was also the possibility that residual confounding by factors associated with ADHD severity could have contributed to the finding of increase risk of malformations in the methylphenidate group. It is additionally noted that when looking at the central finding that methylphenidate was associated with increased risk for cardiovascular malformations in the pooled data, the value for z is 1.93 which is less than the 1.96 formally required for statistical significance and the lower limit of the confidence interval was 1.0, indicating borderline statistical significance.
Relevance: Recently there has been an increase in stimulant medication use in adults for ADHD symptoms. Previous data regarding these medications in early pregnancy is limited. Although the statistical and clinical significance of the small increase in risk of cardiac malformations in infants exposed to methylphenidate is questionable, this study provides information that can be useful when discussing risks and benefits of different treatments for ADHD in pregnant women and young women of reproductive age.