Bidirectional temporal relationships between uric acid and insulin and their joint impact on incident diabetes

Abstract

Background and aimsThis study aims to examine the temporal relationship between uric acid (UA) and insulin and their joint impact on T2DM in middle-aged adults. Methods and resultsThe cohort consisted of 1351 non-diabetic adults who had serum UA and insulin measured twice at baseline and follow-up over 7.7 years on average, and incidence of T2DM in the outcome survey12.2 years later.After adjusting for covariates, the path coefficient from baseline UA to follow-up insulin was 0.082 (p < 0.001); the path from baseline insulin to follow-up UA was 0.060 (p = 0.030). In the mediation model with baseline UA as the predictor, total effect of baseline UA on incident T2DM was 0.089 (p = 0.016). The mediation effect through follow-up insulin on the UA-T2DM association was 28.1%. The direct effect of baseline UA on T2DM (0.064) became nonsignificant (p = 0.078). In the mediation model with baseline insulin as the predictor, total effect of baseline insulin on T2DM was 0.218 (p < 0.001). The mediation effect through follow-up UA on the insulin-T2DM association was 5.5%. The direct effect of baseline insulin on T2DM (0.206) remained significant (p < 0.001). The baseline hyperinsulinemia-follow-up hyperuricemia group showed the highest incidence rate of T2DM (27.9%). ConclusionsThe bidirectional temporal relationship suggests that UA and insulin influence each other in non-diabetic individuals, and the directionality plays pathogenic roles in the development of T2DM.