Thus we used the term ‘sRIFLE’ in our manuscript to distinguish from the original RIFLE. Therefore, observations accrued here might not be extrapolated to patients with AKI elsewhere. Further multicenter randomized clinical trials are warranted to confirm our findings.ConclusionsLD defined by RIFLE-I or RIFLE-F of ‘simplified’ RIFLE classification is an independent predictor for in-hospital mortality in the current study. Our findings support earlier initiation of RRT, and also underscore the importance of predicting prognoses of patients with AKI by using RIFLE classification.Key messages? AKI is a common problem in critically ill patients, and postoperative AKI is one of the most serious complications in surgical patients.? The RIFLE classification was proposed to standardize AKI study, and it’s predictive value for patient outcome was supported by many studies.? Late initiation of RRT defined by RIFLE-I or RIFLE-F is an independent predictor for in-hospital mortality in the current study. Our findings support early initiation of RRT, and also underscore the importance of predicting prognoses of patients with AKI by using RIFLE classification.AbbreviationsAKI: acute kidney injury; APACHE II: Acute Physiology and Chronic Health Evaluation II; BUN: blood urea nitrogen; CI: confidence interval; CKD: chronic kidney disease; CVP: central venous pressure; ED: early dialysis; GCS: Glascow Coma Scale; GFR: glomerular filtration rate; GI: gastrointestinal; HR: hazard ratio; ICU: intensive care unit; LD: late dialysis; MDRD: Modification of Diet in Renal Disease; RR: relative risk; RRT: renal replacement therapy; sCr: serum creatinine; sK+:serum K; SOFA: Sequential Organ Failure Assessment.Competing interestsThe authors declare that they have no competing interests.Authors’ contributionsCCS, VCW, and CCK have made substantial contributions to conception and design, and drafted the manuscript. WYL, DMH, SLL and PRT were involved in acquisition and interpretation of data. YFL, GHY, and CHW participated in the sequence alignment and drafted the manuscript. FCH, NKC, and THL participated in the design of the study and performed the statistical analysis. TWK, YCY, and YMC participated in its design and coordination and helped to draft the manuscript. MTL, AC, WJK, and KDW revised the manuscript critically for important intellectual content, and have given final approval of the version to be published. All authors read and approved the final manuscriptAcknowledgementsThis study was financially supported by the Improving Dialysis Quality Research Funds, Ta-Tung Kidney Foundation, and Taiwan National Science Council (grant NSC 98-2314-B-002-108-MY2).