Abstract Background Acute kidney injury (AKI) is a common complication in critically ill patients with decompensated cirrhosis who often require kidney replacement therapy (KRT) but prognostic uncertainty remains in critically ill ICU patients, particularly regarding short- and medium- term outcomes after KRT initiation. Methods We conducted a retrospective, single center cohort analysis of critically ill patients with cirrhosis and AKI requiring KRT in the medical ICU of a tertiary care liver transplant center. We gathered data pertaining to CLIF-C ACLF and MELD-Na on the day of ICU admission. Survival outcomes of interest included days to death from the start of KRT, and at one- and six-month interval. Results 131 patients met the inclusion criteria of which 21 were listed for liver transplantation at time of ICU admission. 86.3% of patients were diagnosed with acute tubular necrosis (ATN) and the remaining with HRS-AKI. 69.5% of patients were prescribed Continuous Veno Venous Hemofiltration (CVVH) as initial KRT modality. Mean clinical severity scores were MELD-Na 33.1 (SD 8.3) and CLIF-C ACLF score 63.4 (SD 10.3). 21.4% and 15.3% of the 131 patients included, were alive at 1 and 6 months after ICU admission. Median survival time from KRT initiation in ICU was 5 days (IQR 3-12.5 days). Survival did not differ significantly by AKI etiology or transplant listing status although this is limited by small sample size and dynamic nature of transplant listing status. In unadjusted analyses, survival differed by initial KRT modality, with higher mortality among patients initiated on CVVH; however, in multivariable Cox regression, only CLIF-C ACLF score HR 1.03 (1.00- 1.05), p = 0.04, platelets HR 0.997 (0.994–0.999), p = 0.02 and INR HR 1.22 (1.00-1.49), p = 0.05 were significantly associated with survival time. Conclusions AKI requiring KRT in critically ill patients with cirrhosis is associated with limited survival. Mortality was primarily associated with overall severity of acute-on-chronic liver failure, as reflected by CLIF-C ACLF score, rather than MELD-Na. No clear differences were observed by AKI etiology or transplant listing status. Although unadjusted survival differed by initial KRT modality, this likely reflects confounding by illness severity rather than a modality-specific effect. Clinical trial number Not applicable.
Kerns et al. (Thu,) studied this question.