Associations of pretransplant diabetes mellitus, new-onset diabetes after transplant, and acute rejection with transplant outcomes: an analysis of the Organ Procurement and Transplant Network/United Network for Organ Sharing (OPTN/UNOS) database

Abstract

Background: Diabetes and acute rejection are major contributors to morbidity and mortality in kidney transplant recipients. Immunosuppressive medications decrease acute rejection, but increase the frequency of new-onset diabetes after transplant. Our objective was to investigate the joint associations of diabetes (pretransplant diabetes and new-onset diabetes after transplant) and acute rejection with transplant outcomes in a recent transplant cohort.

Study design: Historical cohort study.

Setting & participants: 37,448 recipients (age ≥ 18 years; 2004-2007) surviving with a functioning transplant for longer than 1 year were identified in the Organ Procurement and Transplant Network/United Network for Organ Sharing (OPTN/UNOS) database as of May 22, 2009.

Predictors: Recipients were stratified into 6 mutually exclusive groups according to status of diabetes and acute rejection at 1 year: group 1, neither (reference; n = 20,964); group 2, new-onset diabetes alone (n = 2,140); group 3, pretransplant diabetes alone (n = 10,730); group 4, acute rejection alone (n = 2,282); group 5, new-onset diabetes and acute rejection (n = 361); and group 6, pretransplant diabetes and acute rejection (n = 1,061). Analyses were adjusted for other recipient, donor, and transplant characteristics. OUTCOMES MEASUREMENTS: Multivariate Cox regression analysis of time to transplant failure (overall and death censored) and mortality (all-cause and cardiovascular).

Results: Median follow-up after 1 year was 548 days (25th-75th percentiles, 334-752 days). During this time, there were 3,047 outcomes of overall transplant failure. New-onset diabetes alone (group 2) was not associated significantly with any study outcomes. Groups 3-6 were associated with higher overall transplant failure risk. However, only groups 4-6 were associated with higher death-censored transplant failure risk. Group 3, 4, and 6 were associated with higher all-cause mortality risk, whereas only groups 3 and 6 were associated with higher cardiovascular mortality risk.

Limitations: Potential information bias with exposure, covariable, or outcome misclassification; relatively short follow-up.

Conclusions: Pretransplant diabetes is the major predictor of all-cause and cardiovascular mortality, and acute rejection during the first year is the major predictor of death-censored transplant failure in kidney recipients surviving with a functioning transplant for at least 1 year. The influence of new-onset diabetes on long-term outcomes needs further observation.