On the non-linear association between serum uric acid levels and all-cause mortality rate in patients with type 2 diabetes mellitus

Background and aims High levels of serum uric acid (SUA) are associated with increased mortality risk in the general population. Contrasting results are available in people with diabetes. The aim of our study was to investigate the association and its functional form between SUA and all cause-mortality in patients with type 2 diabetes mellitus (T2DM). Methods We studied three cohorts of patients with T2DM: Gargano Mortality Study, Foggia Mortality Study, Pisa Mortality Study. All-cause mortality rate was the end point of this study. Results The most reliable relationship between SUA levels and all-cause mortality rate was quadratic, with such model being well approximated by SUA tertiles. Both tertiles 1 and 3 were at higher risk of mortality as compared to tertile 2: Hazard Ratio (HR) [95% Confidence Interval (CI)] = 1.34 (1.07–1.68) and 1.61 (1.29–1.99), respectively. In the pseudo-sample, created from the real pooled sample, the best relationship between SUA and all-cause mortality rate was quadratic. In a tree-based Recursive Partitioning and Regression Tree analysis two subgroups at increased risk of mortality were identified, namely those with SUA levels ≥7.28 mg/dl and with SUA levels <4.16 mg/dl as compared to patients with intermediate SUA levels (i.e. 4.16–7.28), thus providing further evidence on the J-shaped relationship between SUA levels and mortality rate. Conclusions SUA was not linearly associated with all-cause mortality rate in patients with T2DM. For clinical and public health purposes such association is J-shaped.