|| Checking for direct PDF access through Ovid
Identify clinical variables associated with mortality in patients with sepsis-associated acute kidney injury (SA-AKI) receiving continuous renal replacement therapy (CRRT) and examine timing of initiation of CRRT in reference to those variables identified.Retrospective study conducted at two tertiary care hospitals including 939 septic shock patients with SA-AKI who received CRRT in the intensive care unit (ICU). Cox regression models were used to identify variables associated with 90-day mortality. Timing of CRRT initiation was assessed in relationship to significant clinical variables identified.Overall 90-day mortality was 62.9%. Variables prior to CRRT associated with 90-day mortality included: age (aHR, 1.02; 95%CI, 1.01–1.02, p < 000.1), APS-III score (1.01, 1.0–1.0, p < 0.048), days from hospital admission to CRRT initiation (1.01, 1.0–1.0, p < 0.01), blood urea nitrogen (1.01, 1.0–1.0, p < 0.04), medical admission (1.76, 1.5–2.1, p < 0.0001), creatinine (0.99, 0.9–1.0, p < 0.001), and urine output (0.77, 0.6–0.9, p = 0.049). In patients with advanced SA-AKI at ICU admission receiving CRRT within the first 5 days (n = 433), urine output during the 24 h prior to CRRT initiation was a strong predictor of survival (2.6, 1.6–4.3, p < 0.001).In patients with SA-AKI, survival is lower when CRRT is started in the setting of low urine output.Severity of SA-AKI at CRRT demonstrates a complex relationship to 90-day mortality.Higher creatinine prior to CRRT did not predict worst outcome.Lower urine output prior to CRRT was associated with greater mortality.Longer time from ICU admission to CRRT did not predict worst outcome.