Association between circulating adipocytokine concentrations and microvascular complications in patients with type 2 diabetes mellitus: A systematic review and meta-analysis of controlled cross-sectional studies

Abstract

Background The adipocytokines leptin and adiponectin have been variously associated with diabetic microvascular complications. No comprehensive clinical data exist examining the association between adipocytokines and the presence of these complications. Methods This is a systematic review of cross-sectional studies comparing circulating adipocytokines in patients with type 2 diabetes mellitus (T2DM), with and without microvascular complications. Studies were retrieved from MEDLINE, EMBASE, Scopus and Cochrane databases. Study quality was evaluated using a modified Newcastle-Ottawa Scale. Meta-analysis was performed using an inverse-variance model, providing standardised mean differences (SMD) and 95% confidence intervals (CI). Heterogeneity was determined by I2 statistic. Results Amongst 554 identified studies, 28 were included in the review. Study quality range was 3.5-9 (maximum 11). Higher leptin levels were associated with microalbuminuria (SMD = 0.41; 95% CI = 0.14-0.67; n = 901; p = 0.0003), macroalbuminuria (SMD = 0.68; 95% CI = 0.30-1.06; n = 406; p = 0.0004), and neuropathy (SMD = 0.26; 95% CI = 0.07-0.44; n = 609; p = 0.008). Higher adiponectin levels were associated with microalbuminuria (SMD = 0.55; 95% CI = 0.29-0.81, n = 274; p < 0.001), macroalbuminuria (SMD = 1.37; 95% CI = 0.78-1.97, n = 246; p < 0.00001), neuropathy (SMD = 0.25; 95% CI = 0.14-0.36; n = 1516; p < 0.00001), and retinopathy (SMD = 0.38; 95% CI = 0.25-0.51; n = 1306; p < 0.00001). Meta-regression suggested no influence of body mass index and duration of diabetes on effect size, and a weak trend in terms of age on effect size. Discussion Our meta-analysis suggests leptin and adiponectin levels are higher in T2DM patients with microvascular complications. Studies were limited by cross-sectional design. Large prospective analyses are required to validate these findings.