Using a large, international cohort, we sought to determine the effect of initial technique of renal replacement therapy (RRT) on the outcome of acute renal failure (ARF) in the intensive care unit (ICU). We enrolled 1218 patients treated with continuous RRT (CRRT) or intermittent RRT (IRRT) for ARF in 54 ICUs in 23 countries. We obtained demographic, biochemical and clinical data and followed patients to either death or hospital discharge. Information was analyzed to assess the independent impact of treatment choice on survival and renal recovery. Patients treated first with CRRT (N=1006, 82.6%) required vasopressor drugs and mechanical ventilation more frequently compared to those receiving IRRT (N=212, 17.4%), (p<0.0001). Unadjusted hospital survival was lower (35.8% vs. 51.9%, p<0.0001). However, unadjusted dialysis-independence at hospital discharge was higher after CRRT (85.5% vs. 66.2%, p<0.0001). Multivariable logistic regression showed that choice of CRRT was not an independent predictor of hospital survival or dialysis-free hospital survival. However, the choice of CRRT was a predictor of dialysis independence at hospital discharge among survivors (OR: 3.333, 95% CI: 1.845 - 6.024, p<0.0001). Further adjustment using a propensity score did not significantly change these results. We conclude that worldwide, the choice of CRRT as initial therapy is not a predictor of hospital survival or dialysis-free hospital survival but is an independent predictor of renal recovery among survivors. © Wichtig Editore, 2007.