Background: Mortality of critically ill acute kidney injury (AKI) initiated on renal replacement therapy (RRT) is very high. Identifying modifiable and unmodifiable clinical variables at dialysis start that are associated with hospital survival can help, not only in prognostication, but also in clinical triaging. Methods: A retrospective observational study was done on AKI-D patients who were initiated on RRT in medical and surgical intensive care units (ICU)s of a high-acuity academic medical center from January 2010 through December 2015. We excluded patients with suspected poisoning, End Stage Renal Disease, Stage 5 CKD not on dialysis or AKI-D patients initiated on RRT outside of ICU setting. Primary outcome was in-hospital mortality. Results: Of the 416 critically ill AKI-D patients admitted to the medical (38%), surgical (40.9%) and cardiac (21.1%) ICUs, with nearly 75% on artificial organ support, mean age 62.1 +/-14.8 years, mean SOFA scores of 11.8+/-4.3, dialysis was initiated using continuous RRT in 261(62.7%) and intermittent hemodialysis in 155 (37.3%) patients. Incidence of survival to hospital discharge was 47.6%. Using multivariable logistic regression with stepwise backward elimination, a prognostic model was created which included the variables age, CKD, COPD, admission and within 24 hour of start SOFAscore, refractory hyperkalemia and uremic encephalopathy as dialysis indications, BUN > 100mg/dl, serum creatinine, serum lactate, serum albumin, CRRT as initial modality, severe volume overload and abdominal surgery. The model exhibited good calibration (goodness of fit test, p =0.8315) and excellent discrimination (optimism-corrected c-statistic 0.9287). Conclusions: In this single-center, diverse, critically-ill AKI-D population, a novel prognostic model which combined widely used ICU scores, clinical and biochemical data at dialysis start as well as dialysis indication and modality, robustly predicted short-term survival. External validation is needed to prove generalizability of study findings.