The impact of diabetes on ethnic disparities seen in kidney transplantation.

Graft failure rates following kidney transplant is disproportionately higher in African American (AA) renal transplant recipients. The aim of our study was to measure the impact of diabetes and other known confounding risk factors on this disparity. This was a long-term cohort study of adult kidney transplant recipients between 2000 and 2008 comparing AA transplant recipients to White recipients. 987 patients were included and patients were followed for up to 12 years post-transplant. Univariate analysis demonstrated AA recipients were more likely to have diabetes (35% vs 23%, P<.001), hypertension (97% vs 94%, P=.029), human leukocyte antigen mismatches (4 vs 3, P<.001), and receiving dialysis for a longer period prior to transplant (3.9 vs 2.0 yrs, P<.001). AA patients were also less likely to receive a living donor transplant (7% vs 31%, P<.001). Multivariable modeling established both AA ethnicity (HR 1.32 [95% CI 1.04-1.68]) and pre-existing diabetes (1.58 [95% CI 1.25-2.00]) as important predictors of graft failure. Diabetes was a significant modifier on the influence of AA ethnicity as a risk factor for graft loss (19% HR reduction); tight glycemic control, which was less common in AA recipients (35% vs 51%, P=.013), additionally attenuated the ethnic disparities seen in graft loss (28% risk reduction). In the final fully adjusted model, which included sociodemographic, immunologic, and cardiovascular risk factor as variables, the influence of AA ethnicity on graft failure was essentially nullified (HR 1.09 [.81-1.48]). In conclusion, AA ethnicity continues to be an important risk factor for graft loss, which can be significantly attenuated by controlling for pre-existing diabetes, glycemic control, and other transplant and cardiovascular variables.