Comparison of two different bicarbonate replacement fluids during CVVH with RCA: a prospective, randomized, controlled trial

Abstract

Purpose: This study aimed to compare a high (HBF) and low (LBF) bicarbonate (HCO3−) replacement fluid over 96 h after continuous veno-venous hemofiltration (CVVH) initiation using regional citrate anticoagulation to determine which fluid achieved better acid–base balance. Methods: In this prospective, randomized, controlled, open-label, cross-over, single-center trial, patients were randomized to initially receive either LBF (22 mmol/l HCO3−) or HBF (30 mmol/l HCO3−) for 48 h (first phase), followed by changing to the other fluid for a further 48 h (second phase). Intracorporeal pH and HCO3− changes were analyzed by calculating the rates of levels outside the standard values (excursion rates). Generalized estimating equations were performed to estimate the odds ratios for at least one pH/HCO3− excursion. Time to normalization within the first phase for pH and HCO3− was assessed using Kaplan–Meier curves and log-rank tests. Results: A total of 88 patients were included. The rates for at least one pH and HCO3− excursion were higher in patients treated with HBF compared to LBF (pH: first phase 52% vs. 41%; second phase 48% vs. 34%; HCO3−: first phase 68% vs. 43%; second phase 73% vs. 43%). Adjusted odds ratios were 1.78 (95% CI 1.12–2.82; p = 0.015) for pH and 3.60 (95% CI 2.16–5.99; p < 0.001) for HCO3−. The time to normalization of pH in acidotic patients and of HCO3− in patients with low HCO3− did not differ significantly between HBF and LBF (p = 0.102, p = 0.468). Conclusion: LBF showed significantly lower rates of pH and HCO3− excursions compared to HBF and no significant differences to the pH normalization in acidotic patients were observed. ClinicalTrials.gov-identifier: NCT04071171.