Evidence of a causal relationship between high serum adiponectin levels and increased cardiovascular mortality rate in patients with type 2 diabetes

Abstract

Background Despite its beneficial role on insulin resistance and atherosclerosis, adiponectin has been repeatedly reported as an independent positive predictor of cardiovascular mortality.

Methods A Mendelian randomization approach was used, in order to evaluate whether such counterintuitive association recognizes a cause-effect relationship. To this purpose, single nucleotide polymorphism rs822354 in the ADIPOQ locus which has been previously associated with serum adiponectin at genome-wide level, was used as an instrument variable. Our investigation was carried out in the Gargano Heart Study-prospective design, comprising 356 patients with type 2 diabetes, in whom both total and high molecular weight (HMW) adiponectin were measured and cardiovascular mortality was recorded (mean follow-up = 5.4 ± 2.5 years; 58 events/1922 person-year).

Results The A allele of rs822354 was associated with both total and HMW adiponectin [β (SE) = 0.10 (0.042), p = 0.014 and 0.17 (0.06), p = 0.003; respectively]. In a Poisson model comprising age, sex, smoking habits, BMI, HbA1c, total cholesterol, HDL-cholesterol, triglycerides, insulin therapy and hypertension, both rs822354 (IRR = 1.94, 95 % CI 1.23–3.07; p = 0.005), as well as the genetic equivalent of total adiponectin change (IRR = 1.07, 95 % CI 1.02–1.12; p = 0.003) were significantly associated with cardiovascular mortality. The observed genetic effect was significantly greater than that exerted by the genetic equivalent change of serum adiponectin (p for IRR heterogeneity = 0.012). In the above-mentioned adjusted model, very similar results were obtained when HMW, rather than total, adiponectin was used as the exposure variable of interest.

Conclusions Our data suggest that the paradoxical association between high serum adiponectin levels and increased cardiovascular mortality rate is based on a cause-effect relationship, thus pointing to an unexpected deleterious role of adiponectin action/metabolism on atherosclerotic processes.