Highlights
- Lower bicarbonate replacement fluid (LBF, 22 mmol/l) significantly reduces the risk of pH and bicarbonate excursions compared to high bicarbonate fluid (HBF, 30 mmol/l) during CVVH.
- The use of HBF was associated with a 3.6-fold increased risk of bicarbonate levels falling outside the standard range.
- Importantly, the time to normalize pH in acidotic patients was not delayed by using the lower bicarbonate concentration.
- Regional citrate anticoagulation (RCA) acts as a significant endogenous source of bicarbonate, necessitating a reduction in exogenous bicarbonate to prevent metabolic overcorrection.
Introduction: The Challenge of Acid-Base Homeostasis in CRRT
Continuous renal replacement therapy (CRRT) is a cornerstone of management for critically ill patients with acute kidney injury (AKI). Among the various modalities, continuous veno-venous hemofiltration (CVVH) remains widely utilized. However, maintaining precise acid-base balance during these therapies is a complex clinical challenge. The introduction of regional citrate anticoagulation (RCA) has revolutionized the safety and circuit patency of CRRT, yet it introduces a significant metabolic variable: the conversion of citrate to bicarbonate.
In patients undergoing RCA, citrate is infused into the extracorporeal circuit to chelate calcium and inhibit coagulation. Once returned to the patient, citrate is metabolized, primarily in the liver, skeletal muscle, and kidneys, where one mole of citrate yields three moles of bicarbonate. Consequently, the choice of bicarbonate concentration in the replacement fluid must account for this endogenous production. Standard replacement fluids often contain bicarbonate concentrations between 30 and 35 mmol/l, which may predispose patients to metabolic alkalosis when combined with RCA. This study, published by Perschinka et al. in Intensive Care Medicine, provides a rigorous evaluation of whether a lower bicarbonate concentration (22 mmol/l) offers better physiological control than a standard higher concentration (30 mmol/l).
The Metabolic Role of Citrate: A Double-Edged Buffer
Understanding the results of this trial requires a deep dive into the pharmacology of citrate. When citrate is used for regional anticoagulation, it serves two purposes: preventing filter clotting and acting as a buffer precursor. In many ICU protocols, the ‘total’ bicarbonate load is the sum of the bicarbonate in the replacement/dialysate fluid and the bicarbonate generated from citrate metabolism. If the replacement fluid concentration is too high, the patient is at risk of metabolic alkalosis, which can lead to complications such as hypokalemia, ionized hypocalcemia, and a compensatory decrease in respiratory drive, potentially prolonging mechanical ventilation.
Conversely, there is a clinical fear that using a low-bicarbonate fluid might delay the correction of metabolic acidosis, a common feature of the critically ill population requiring CRRT. The trial by Perschinka et al. was specifically designed to address this tension between preventing excursions (overcorrection) and ensuring timely normalization of pH.
Study Design and Methodology
This study was a prospective, randomized, controlled, open-label, cross-over trial conducted at a single center. The cross-over design is particularly robust in this context as it allows each patient to serve as their own control, minimizing the impact of individual variations in citrate metabolism rates. A total of 88 patients requiring CVVH with RCA were enrolled.
Patients were randomized to receive either a low bicarbonate fluid (LBF, 22 mmol/l HCO3-) or a high bicarbonate fluid (HBF, 30 mmol/l HCO3-) for the first 48 hours. Following this first phase, patients were switched to the alternative fluid for a second 48-hour phase. The primary focus of the analysis was the ‘excursion rate’—the frequency with which pH and bicarbonate levels fell outside the standard clinical range (pH 7.35–7.45 and HCO3- 22–26 mmol/l). The researchers utilized generalized estimating equations (GEE) to estimate odds ratios for these excursions and employed Kaplan-Meier curves to assess the time to normalization for those starting with acidosis.
Key Findings: Superior Stability with Lower Bicarbonate Concentrations
The results of the trial strongly favor the use of lower bicarbonate concentrations in the setting of RCA. The data revealed that patients treated with HBF experienced significantly higher rates of both pH and bicarbonate excursions compared to those treated with LBF. Specifically, in the first phase, pH excursions occurred in 52% of the HBF group versus 41% of the LBF group. This trend was even more pronounced in the second phase (48% vs. 34%).
The statistical analysis using adjusted odds ratios (OR) highlighted the magnitude of this risk. The OR for at least one pH excursion was 1.78 (95% CI 1.12-2.82; p = 0.015) in favor of LBF. The results for bicarbonate levels were even more striking, with an OR of 3.60 (95% CI 2.16-5.99; p < 0.001). These findings suggest that the standard 30 mmol/l fluid significantly increases the likelihood of metabolic instability, primarily by driving patients toward metabolic alkalosis.
Normalization Rates: Dispelling the Fear of Under-Correction
A critical secondary endpoint was the time to normalization of acid-base status in patients who were acidotic at the start of the study. One might hypothesize that HBF would correct acidosis faster than LBF. However, the study found no significant difference between the two fluids. The p-values for time to pH normalization (p = 0.102) and bicarbonate normalization (p = 0.468) indicated that LBF is just as effective as HBF in bringing patients back to a physiological range, without the added risk of overshooting into alkalosis.
Clinical Implications and Expert Commentary
The clinical implications of this study are profound for ICU protocols. Most commercially available CRRT fluids are standardized, but this evidence suggests that ‘standard’ concentrations may be inappropriately high when RCA is utilized. The findings support a transition toward using replacement fluids with bicarbonate concentrations closer to 22 mmol/l in CVVH-RCA protocols.
Expert clinicians note that metabolic alkalosis is not a benign condition in the ICU. It shifts the oxyhemoglobin dissociation curve to the left, hindering oxygen delivery to tissues, and can induce cardiac arrhythmias through electrolyte shifts. By choosing LBF, clinicians can achieve a ‘gentler’ and more stable correction of acid-base status. The lack of difference in normalization time is perhaps the most reassuring finding, as it removes the primary clinical barrier to adopting lower-bicarbonate fluids.
Mechanistic Insights
The stability observed with LBF can be attributed to the ‘auto-regulating’ nature of citrate metabolism. When the exogenous bicarbonate load is lower, the body’s endogenous production from citrate becomes the primary driver of correction. This allows for a more controlled return to homeostasis. In contrast, the high exogenous load from HBF overrides these metabolic nuances, leading to the observed excursions.
Study Limitations and Generalizability
While the trial is well-designed, there are limitations to consider. As a single-center study, the results may reflect specific local practices regarding CVVH flow rates and citrate dosing. Furthermore, the open-label nature could theoretically introduce bias, although the primary endpoints (laboratory values) are objective and less susceptible to observer bias. The study focused on CVVH; while the principles likely apply to continuous veno-venous hemodialysis (CVVHD) or hemodiafiltration (CVVHDF), the specific bicarbonate kinetics in those modalities may differ slightly due to diffusive clearance.
Conclusion
The trial by Perschinka et al. provides high-quality evidence that a lower bicarbonate replacement fluid (22 mmol/l) is superior to a standard higher concentration (30 mmol/l) for maintaining acid-base balance during CVVH with regional citrate anticoagulation. LBF significantly reduces the incidence of pH and bicarbonate excursions while matching the efficacy of HBF in correcting metabolic acidosis. For critical care units looking to optimize their CRRT protocols, these findings suggest that ‘less is more’ when it comes to bicarbonate concentrations in the presence of citrate.
Funding and Clinical Trial Information
This study was registered at ClinicalTrials.gov under the identifier NCT04071171. The research was supported by institutional funding and conducted in accordance with the Declaration of Helsinki.
References
- Perschinka F, Köglberger P, Köhler A, et al. Comparison of two different bicarbonate replacement fluids during CVVH with RCA: a prospective, randomized, controlled trial. Intensive Care Med. 2025 Dec;51(12):2354-2366. doi: 10.1007/s00134-025-08175-7.
- Zarbock A, Küllmar M, Kindgen-Milles D, et al. Effect of Regional Citrate Anticoagulation vs Systemic Heparin Anticoagulation During Continuous Renal Replacement Therapy on Dialysis Filter Life: A Randomized Clinical Trial. JAMA. 2020;323(24):2482–2491.
- Schneider AG, Journois D, Rimmelé T. Complications of regional citrate anticoagulation: accumulation or overload? Critical Care. 2017;21(1):281.