In patients with acute kidney injury (AKI) undergoing continuous renal replacement therapy (CRRT), appropriate ultrafiltration management is important. Multi-frequency Bioelectrical Impedance Analysis (MF-BIA) is a noninvasive tool for assessing body fluid status. Previous study used the total body water-to-height squared (TBW/HT²) index; however, this parameter has limitations in reflecting extracellular fluid distribution. This study evaluated the clinical utility of an MF-BIA–based fluid overload (FO) index.

This retrospective multicenter study was conducted using the VENUS trial cohort, which included AKI patients undergoing CRRT at eight tertiary hospitals in Korea. Patients were included if they had MF-BIA measurements within 48 hours of CRRT initiation, and patients with implausible or erroneous BIA measurements were excluded. MF-BIA measurements were obtained on CRRT Days 0, 1, 2, and 7 using InBody S10. The primary outcome was 28-day mortality. ROC/AUROC analyses were used to evaluate prediction of 28-day mortality, and optimal cutoffs were determined using the Youden index. Kaplan–Meier analysis and log-rank tests were performed using the optimal cutoffs. Cox proportional hazards models were used to calculate hazard ratios (HRs) and 95% confidence intervals (CIs).

A total of 485 patients were included in the analysis. No significant differences were observed in age, body weight, TBW, or TBW/HT² between survivors and non-survivors. On Day 2, extracellular water (ECW), ECW ratio, and FO were significantly higher in non-survivors. FO consistently showed higher AUROC than TBW/HT² at all time points. Both indices showed their best predictive performance on Day 2. Day 2 FO showed the highest AUROC (0.667 vs. 0.563 for TBW/HT²). The optimal Day 2 cutoffs were 1.79 L for FO and 15.29 L/m² for TBW/HT². In survival analysis, patients were classified into high-FO (FO ≥ 1.79 L, n=118) and low-FO (FO < 1.79 L, n=160) groups using the optimal Day 2 FO cutoff. Patients with higher FO had an increased risk of 28-day mortality, with a hazard ratio of 2.52 (95% CI 1.75–3.63, p<0.001).

FO showed better predictive performance for 28-day mortality than TBW/HT² in severe AKI patients undergoing CRRT, particularly on Day 2. MF-BIA-based FO may support early fluid assessment and risk stratification in CRRT patients.