Prognostic Impact of Early Versus Late Initiation of Renal Replacement Therapy Based on Early Warning Algorithm in Critical Care Patients With Acute Kidney Injury.

Abstract

The aim of our study is to evaluate the impact of early vs. late initiation of continuous renal replacement therapy (CRRT), defined by clinical information system (CIS) software using an early warning algorithm based on acute kidney injury network (AKIN) stages, on survival outcome of critically ill intensive care unit (ICU) patients with acute kidney injury (AKI). Of 1144 patients (mean [SD] age: 61.3 [17.9] years, 57.7% were males) hospitalized in ICU over a 2-year-period from January 2016 to December 2017, a total of 272 patients who had developed AKI requiring CRRT were included in this retrospective cross-sectional study. Data on patient demographics (age, gender), reason for ICU hospitalization, AKIN stage, Sequential Organ Failure Assessment (SOFA) score, Acute Physiology and Chronic Health Evaluation II (APACHE II) score, indications for CRRT, and time of CRRT initiation with respect to AKIN early warning algorithm were retrieved from hospital records and the CIS software database. Survivorship status was assessed based on total, in-hospital and 90-day post-discharge mortality rates and analyzed with respect to CRRT onset before vs. after AKIN alarm. CRRT was initiated before the AKIN alarm in 41(15.0%) patients, and after the AKIN alarm in 231(85.0%) patients involving treatment within 0-24 h of alarm in 146 (63.2%) patients and within 24-120 h of alarm in 85 (36.8%) patients. Mortality occurred in 175 (64.3%) patients involving 25 (61.0%) out of 41 patients who received CRRT before AKIN alarm and 150 (64.9%) out of 231 patients who received CRRT after AKIN alarm. Mortality rate was significantly higher in those who received CRRT 24-120 h vs. 0-24 h after the AKIN alarm (82.4% vs. 54.8%, P < 0.001). Pre- and post-CRRT SOFA scores were significantly lower in patients who received CRRT 0-24 h vs. 24-120 h after the AKIN alarm (P = 0.009 and P = 0.004, respectively), while pre-CRRT APACHE II scores were significantly lower in patients who received CRRT before vs. after the AKIN alarm (P = 0.008). In conclusion, our findings indicate the potential role of using AKIN stage-based early warning system in guiding time to start CRRT and improved survival in critically ill patients with AKI, provided that the CRRT was initiated within the early (first 24 h) of the alarming AKIN Stage II-III events. Future well-designed clinical trials addressing early vs. late initiation of CRRT in critical care patients with AKI are needed to find and answer to the ongoing controversy and help clinicians in refining their indications for starting CRRT.