Background: Hyperuricemia is frequently observed in kidney transplant recipients and it has been implicated in poor allograft function and survival. However, this association remains controversial since time-varying confounders such as kidney function have not been properly addressed in prior studies.

Methods: We examined 1,169 kidney transplant recipients who were transplanted between 1 Jan 2000 and 31 Dec 2010 (followed until 31 Dec 2011). Hyperuricemia was defined as serum uric acid (UA) > 416 Μmol/L (> 7.0 mg/dL) in males and > 357 Μmol/L (> 6.0 mg/dL) in females. UA and kidney function (CKD-EPI formula) were measured after kidney transplantation at 1-month, 3-months, and every 3-months thereafter. Other variables were collected at baseline. The primary outcome was time to total graft failure (i.e., graft loss or death). The association of UA and total graft failure was assessed in a marginal structural Cox model accounting for kidney function as a time-varying confounder.

Results: When UA was considered a time-fixed exposure at 1-month, it was associated with an increased risk of total graft failure regardless of how it was measured, i.e., continuous or binary (see Table). UA as a time-varying exposure showed an increased risk of total graft failure when measured as a continuous (but not binary) variable. Finally, simultaneously accounting for the time-varying effects of UA and kidney function showed a modestly protective association of UA when measured as a continuous (but not binary) variable.

Conclusions: UA was generally associated with an increased risk of total graft failure when considered as a time-fixed or time-varying exposure. However, accounting for the time-varying effects of kidney function reduced this association to null or modestly protective. The latter may be an indicator of nutritional status, especially within the range of normouricemia, but this is speculative and requires further study.

« Back to 2013 American Transplant Congress