Abstract
Low birthweight is a risk factor for type 2 diabetes. We hypothesised that differential associations between birthweight and clinical characteristics in persons with and without type 2 diabetes may provide novel insights into the role of birthweight in type 2 diabetes and its progression. We analysed UK Biobank data from 9,442 persons with and 254,446 without type 2 diabetes. Associations between birthweight, clinical traits, and genetic predisposition were assessed using adjusted linear and logistic regression, comparing the lowest and highest 25% of birthweight to the middle 50%. Each kg increase in birthweight was associated with higher BMI, waist, and hip circumference, with stronger effects in persons with versus without type 2 diabetes (BMI: 0.74 [0.58, 0.90] vs. 0.21 [0.18, 0.24] kg/m2; waist: 2.15 [1.78, 2.52] vs. 1.04 [0.98, 1.09] cm; hip: 1.65 [1.33, 1.97] vs. 1.04 [1.04, 1.09] cm). Family history of diabetes was associated with higher birthweight regardless of diabetes status, albeit with a twofold higher effect estimate in type 2 diabetes. Low birthweight was further associated with prior myocardial infarction regardless of type 2 diabetes status (OR 1.33 [95% CI 1.11, 1.60] for type 2 diabetes; 1.23 [95% CI 1.13, 1.33] without), and hypertension (OR 1.25 [1.23, 1.28] and stroke 1.24 [1.14, 1.34]) only among persons without type 2 diabetes. Differential associations between birthweight and cardiometabolic traits in persons with and without type 2 diabetes illuminate potential causal inferences reflecting the roles of pre- and postnatal environmental versus genetic aetiologies and disease mechanisms.