Discontinuation of continuous renal replacement therapy: a post hoc analysis of a prospective multicenter observational study.

OBJECTIVES To describe current practice for the discontinuation of continuous renal replacement therapy in a multinational setting and to identify variables associated with successful discontinuation. The approach to discontinue continuous renal replacement therapy may affect patient outcomes. However, there is lack of information on how and under what conditions continuous renal replacement therapy is discontinued. DESIGN Post hoc analysis of a prospective observational study. SETTING Fifty-four intensive care units in 23 countries. PATIENTS Five hundred twenty-nine patients (52.6%) who survived initial therapy among 1006 patients treated with continuous renal replacement therapy. INTERVENTIONS None. MEASUREMENTS AND MAIN RESULTS Three hundred thirteen patients were removed successfully from continuous renal replacement therapy and did not require any renal replacement therapy for at least 7 days and were classified as the "success" group and the rest (216 patients) were classified as the "repeat-RRT" (renal replacement therapy) group. Patients in the "success" group had lower hospital mortality (28.5% vs. 42.7%, p < .0001) compared with patients in the "repeat-RRT" group. They also had lower creatinine and urea concentrations and a higher urine output at the time of stopping continuous renal replacement therapy. Multivariate logistic regression analysis for successful discontinuation of continuous renal replacement therapy identified urine output (during the 24 hrs before stopping continuous renal replacement therapy: odds ratio, 1.078 per 100 mL/day increase) and creatinine (odds ratio, 0.996 per [micro]mol/L increase) as significant predictors of successful cessation. The area under the receiver operating characteristic curve to predict successful discontinuation of continuous renal replacement therapy was 0.808 for urine output and 0.635 for creatinine. The predictive ability of urine output was negatively affected by the use of diuretics (area under the receiver operating characteristic curve, 0.671 with diuretics and 0.845 without diuretics). CONCLUSIONS We report on the current practice of discontinuing continuous renal replacement therapy in a multinational setting. Urine output at the time of initial cessation of continuous renal replacement therapy was the most important predictor of successful discontinuation, especially if occurring without the administration of diuretics.