Background: Citrate, the currently preferred anticoagulant for continuous veno-venous hemofiltration (CVVH), may influence acid-base equilibrium.
Methods: The effect of 2 different citrate solutions on acid-base status was assessed according to the Stewart-Figge approach in two consecutive cohorts of critically ill adult patients. The first group received Prismocitrate 10/2 (PC10/2; 10 mmol citrate/L). The next group was treated with Prismocitrate 18/0 (PC18; 18 mmol citrate/L). Both groups received bicarbonate-buffered fluids in post-dilution.
Results: At similar citrate flow, the metabolic acidosis present at baseline in both groups was significantly attenuated in PC18 patients but persisted in PC10/2 patients after 24 h of treatment (median pH 7,42 vs 7,28; p = 0.0001). Acidosis in the PC10/2 group was associated with a decreased strong ion difference and an increased strong ion gap (respectively 43 vs. 51 mmol/L and 17 vs. 12 mmol/L, PC10/2 vs. PC18; both p = 0.001). Chloride flow was higher in PC10/2 than in PC18 subjects (25.9 vs 14.3 mmol/L blood; p < 0.05).
Conclusion: Correction of acidosis was blunted in patients who received 10 mmol citrate/L as regional anticoagulation during CVVH. This could be explained by differences in chloride flow between the applied citrate solutions inducing hyperchloremic acidosis.
Keywords: Acid-base balance; Acidosis; Alkalosis; Anticoagulation; Chloride; Citrate; Continuous veno-venous hemofiltration; Hyperchloremic acidosis; Stewart-Figge approach; Strong ion gap.