Developmental origins of cardiometabolic health outcomes in twins: a systematic review and meta-analysis

Abstract

Background and aims: Studies of twins can reduce confounding and provide additional evidence about the causes of disease, due to within-pair matching for measured and unmeasured factors. Although findings from twin studies are typically applicable to the general population, few studies have taken full advantage of the twin design to explore the developmental origins of cardiometabolic health outcomes. We aimed to systematically review the evidence from twin studies and generate pooled estimates for the effects of early-life risk factors on later-life cardiometabolic health. Methods and results: An initial search was conducted in March 2018, with 55 studies of twins included in the review. Risk of bias was assessed using the Newcastle–Ottawa Scale, and eligible studies were included in a meta-analysis, where pooled estimates were calculated. Twenty-six studies analysed twins as individuals, and found that higher birthweight was associated with lower SBP (β = −2.02 mmHg, 95%CI: −3.07, −0.97), higher BMI (β = 0.52 kg/m 2, 95%CI: 0.20, 0.84) and lower total cholesterol (β = −0.07 mmol/L, 95%CI: −0.11, −0.04). However, no associations were reported in studies which adjusted for gestational age. Few of the included studies separated their analyses into within-pair and between-pair associations. Conclusions: Early-life risk factors were associated with cardiometabolic health outcomes in twin studies. However, many estimates from studies in this review were likely to have been confounded by gestational age, and few fully exploited the twin design to assess the developmental origins of cardiometabolic health outcomes.