Adiponectin, TNF-α and inflammatory cytokines and risk of type 2 diabetes: A systematic review and meta-analysis

Aims: There has been growing evidence that adiponectin, tumor necrosis factor-alpha (TNF-alpha) and inflammatory cytokines involved in insulin resistance and may be attractive candidates for assessing risk of the incident type 2 diabetes (T2DM). A systematic review and meta-analysis of prospective studies was conducted to assess the associations of levels of serum adiponectin, TNF-alpha and inflammatory markers (Interleukin-1 beta (IL-1(3), Interleukin-6 (IL-6), Interleukin-18 (IL-18), C-reactive protein (CRP)) with risk of T2DM. Materials/methods: We searched PubMed, ISI Web of Knowledge, EMBASE, and Cochrane Library databases up until February 1, 2016 for eligible studies which were matched to search subjects. Either fixed-effects or random-effects models were used to estimate the summary risk incorporated between study variations. Results: 19 studies comprising a total of 39,136 participants and 7924 cases were included in the meta analysis. Our findings showed that an obvious association of elevated CRP levels with T2DM risk (relative risk [RR] 1.48 [95% CI 1.26-1.71]), with the absence of publication bias. For IL-6, the meta-analysis involved 16 cohorts with a total of 24,929 participants and 4751 cases. Using data from all trials, a strong positive correlation (1.32 [1.14, 1.51]) was observed between basal plasma IL-6 and T2DM, whereas relatively lower relation between TNF-alpha (1.16 [0.87, 1.45]), IL-18 (1.45 [1.16, 1.73]), IL-1 beta (0.87, [0.59, 1.15]) and independently increased risk to occurrence of T2DM. Conversely, we also found that the level of adiponectin decreased significantly in patients with T2DM. Sensitivity analyses further supported the associations. Conclusions: This meta-analysis indicates that T2DM risk as whole was strongly associated with elevated levels of inflammatory cytokines (IL-1 beta, IL-6, IL-18, CRP), TNE-alpha and low levels of adiponectin. Despite an overall detectable association in the meta-analysis, considerable heterogeneity existed between studies. Further work is needed, it seems clear that a complex interplay of inflammation and the development of DM. Moreover, these biomarkers are predictors of T2DM subjects and should take more attention to measure levels of these as well as to target therapy/interventions. (C) 2016 Elsevier Ltd. All rights reserved.