Uric acid and the risk of graft failure in kidney transplant recipients: a re-assessment.

The association of hyperuricemia with kidney allograft outcomes remains controversial. We studied this problem in 1170 kidney transplants from January 2000 to December 2010. The primary endpoint was total graft failure (i.e. graft loss or death). Conventional, time-dependent and marginal structural Cox proportional hazards models were fitted, the latter accounting for kidney function as a time-varying confounder affected by prior uric acid levels. Uric acid level was associated with an increased risk of total graft failure in time-fixed and time-varying models (HR 1.02 [95% CI: 1.003-1.04] and HR 1.02 [95% CI: 1.01-1.03], respectively, for every 10 µmol/L increase in uric acid). In contrast, the marginal structural model showed a modestly protective effect (HR 0.90 [95% CI: 0.85-0.94] for every 10 µmol/L increase in uric acid). Similar results were observed for death-censored graft failure and death with graft function. In summary, the absence of a deleterious association between elevated uric acid and graft outcome after accounting for graft function as a time-varying confounder suggests that uric acid is not an independent risk factor for graft failure. The modestly protective effect of uric acid may be an indicator of nutritional status but further study is warranted.