Arm length is associated with type 2 diabetes mellitus in Japanese-Americans

Abstract

Aims/hypothesis The aim of the study was to examine the association of type 2 diabetes mellitus with arm length as a marker for early life environment and development. Methods This was a cross-sectional analysis of 658 secondand third-generation Japanese-Americans (349 men and 309 women). Different armlength (total, upper and forearmlength) and leg length (total and lower leg length) measurements were performed. Type 2 diabetes was defined by the use of hypoglycaemic medication, fasting plasma glucose (FPG) =7 mmol/l or glucose at 2 h =11.1 mmol/l during an OGTT. Persons meeting the criteria for impaired glucose tolerance were excluded from these analyses (FPG <7 mmol/l and 2 h glucose during an OGGT <11.1 but =7.8 mmol/l). Multivariable logistic regression was used to estimate associations between prevalence of diabetes and limb length while adjusting for possible confounders. Results A total of 145 individuals had diabetes. On univariate analysis, arm and leg length were not associated with diabetes. After adjustment for age, sex, computed tomography-measured intra-abdominal fat area, height, weight, smoking status and family history of diabetes, total arm length and upper arm length were inversely related to diabetes (OR for a 1 SD increase 0.49, 95% CI 0.29, 0.84 for total arm length, and OR 0.56, 95% CI 0.36, 0.87 for upper arm length). Forearm length, height and leg length were not associated with diabetes after adjustment for confounding variables. Conclusions/interpretation Our findings of associations between arm lengths and prevalence of type 2 diabetes supports a role for factors that determine bone growth or their correlates in the development of this condition. © Springer-Verlag 2012.