Predictors and outcomes of candiduria in renal transplant recipients

Background. The epidemiology of candiduria in renal transplantation is unknown. Methods. We performed a nested case-control study to evaluate the epidemiology of candiduria in renal transplant recipients at the University of Wisconsin (Madison) over an 8-year period. Results. Renal transplantations were performed on 1738 patients during this period, 192 of whom had 276 episodes of candiduria. Candida glabrata, which was recovered from 98 (51%) of 192 case patients, was the most common pathogen identified. Most case patients were asymptomatic. Independent predictors of candiduria were female sex (odds ratio [OR], 12.5; 95% confidence interval [CI], 6.7-23.0), intensive care unit admission (OR, 8.8; 95% Cl, 2.3-35.0), antibiotic use during the month before candiduria (OR, 3.8; 95% Cl, 1.7-8.3), presence of an indwelling bladder catheter (OR, 4.4; 95% Cl, 2.1-9.4), diabetes (OR, 2.2; 95% Cl, 1.3-3.9), neurogenic bladder (OR, 7.6; 95% Cl, 2.1-27), and malnutrition (OR, 2.4; 95% Cl, 1.3-4.4). Log-rank testing of Kaplan-Meier curves revealed that 60-day, 90-day, and cumulative survival rates were significantly different between case and control patients; there was no difference in the survival rate during the first 30 days after transplantation. A variety of regimens were used for treatment; 119 case patients (62%) underwent removal of the indwelling bladder catheter within I week after diagnosis of candiduria. Candiduria cleared in 148 case patients (77%). Treatment of candiduria was not associated with an improved survival rate. Conclusions. Candiduria occurs commonly in renal transplant recipients. Risk factors for candiduria in such persons are similar to those in hospitalized patients who have not received a transplant. Candiduria is associated with reduced survival rates among persons who have undergone renal transplantation; this is likely a marker for severity of illness. Treatment of asymptomatic candiduria in renal transplant recipients does not appear to result in improved outcome.