Risk factors for citrate accumulation in patients with liver failure undergoing continuous renal replacement therapy with regional citrate anticoagulation

Objective To investigate the risk factors of citrate accumulation in patients with liver failure treated with regional citrate anticoagulated continuous renal replacement therapy (RCA-CRRT). Methods The clinical data of liver failure patients with RCA-CRRT admitted to department of intensive care unit (ICU) of Nantong Third People's Hospital from January 2017 to June 2020 were retrospectively analyzed. The selected patients were divided into citrate accumulation group and control group according to whether there was citrate accumulation (serum total calcium/free calcium ratio ≥ 2.4) during CRRT. The age, acute physiology and chronic health evaluation II (APACHE II), mean arterial pressure (MAP), norepinephrine (NE) dose, blood lactic acid (Lac) concentration, liver function status, citrate dose, filter time and prognosis of the patients were compared between the two groups. Unconditional Logistic regression was used to analyze the risk factors for citrate accumulation. Results Among 48 patients with RCA-CRRT and liver failure, 20 patients had citrate accumulation (accumulation group), and a total of 96 CRRTs were performed; the remaining 28 patients did not have citrate accumulation (control group), a total of 106 CRRTs were performed. There were no significant differences in age and APACHE II score between the two groups. Compared with the control group, the MAP in the accumulation group was lower [mmHg (1 mmHg = 0.133 kPa): 66.9±13.6 vs. 86.4±8.3, P = 0.032], and the dosage of NE (μg/min: 16.3±8.4 vs. 5.9±2.8, P = 0.015) and lactic acid level (mmol/L: 4.89±1.45 vs. 2.98±0.87, P = 0.004) were higher, the damage of liver function was more serious [total bilirubin (TBil, μmol/L): 220.4±45.2 vs. 163.4±43.8, P = 0.012; Child-Pugh score: 12.0±2.5 vs. 8.8±1.4, P = 0.029; model for end-stage liver disease (MELD) score: 31.30±8.22 vs. 21.78±6.40, P = 0.041], hourly citric acid dosage (mmol/h: 27.4±6.9 vs. 19.3±4.9, P = 0.032) and total citric acid dosage (mmol: 3 393±809 vs. 1 819±502, P = 0.039) were higher. Although there were no significant differences in the length of ICU stay, total length of hospitalization stay and cost of hospitalization between the two groups, the 28-day mortality of the accumulation group was higher than that of the control group (60.0% vs. 28.6%, P = 0.039). Unconditional Logistic regression analysis showed that MAP [odds ratio (OR) = 2.901, 95% confidence interval (95%CI) was 0.921-19.493, P = 0.019], NE dosage (OR = 2.098, 95%CI was 1.923-12.342, P = 0.002), Lac level (OR = 5.201, 95%CI was 3.211-9.433, P = 0.012), Child-Pugh score (OR = 1.843, 95%CI was 0.437-7.420, P = 0.018), MELD score (OR = 3.012, 95%CI was 0.384-12.843, P = 0.031), hourly citric acid dosage (OR = 4.254, 95%CI was 1.734-11.839, P = 0.011) and total citric acid dosage (OR = 4.109, 95%CI was 1.283-18.343, P = 0.001) were risk factors for citrate accumulation. Conclusions In patients with tissue hypoperfusion and severe liver function damage, citrate anticoagulation should be avoided or the dosage of citric acid should be reduced, in order to avoid citrate accumulation.