Abstract
Objective: To compare the effectiveness and safety profile of centrifugal and membrane plasma separation model in artificial liver therapy with a dual plasma molecular adsorption system (DPMAS). Method: A retrospective study was conducted. Data of inpatients with liver failure who were treated with DPMAS therapy in the Liver Disease Center of Nanfang Hospital, Southern Medical University, from October 2022 to June 2024 were included. Clinical data such as demographic characteristics, etiology, DPMAS treatment-related indicators (including plasma separation method, vascular access, frequency of treatment, treatment duration, type of anticoagulant drugs, and membrane rupture condition), and laboratory test indicators before and after DPMAS treatment were collected. Categorical variables were compared by the χ² test. Continuous variables were compared using a t-test or a non-parametric test between groups. Result: Data of 232 cases with liver failure who received artificial liver therapy with DPMAS were included. A total of 473 times DPMAS treatment was given. The average age was 50 years old, and males accounted for 82.3%. Centrifugal plasma separation was the initial DPMAS treatment in 176 (75.9%) cases, while membrane plasma separation was used in 56 cases (24.1%). The most common vascular access for DPMAS treatment was the internal jugular vein. The most commonly used anticoagulant was unfractionated heparin. The treatment duration of DPMAS was significantly higher with centrifugal separation than that with membrane separation (P<0.001). Hemoglobin levels (mean before and after treatment in the centrifugal: 112.8 g/L vs. 106.3 g/L, P<0.001; mean before and after treatment in the membrane group: 108.4 g/L vs. 103.3 g/L, P<0.001), red blood cell count (mean before and after treatment in the centrifugal group: 3.7×10(9)/L vs. 3.5×10(9)/L, P<0.001; mean before and after treatment in the membrane group: 3.5×10(9)/L vs. 3.3×10(9)/L, P<0.001) and platelet count (mean before and after treatment in the centrifugal group: 134.5×10(9)/L vs. 119.6×10(9)/L, P<0.001; mean before and after treatment in the membrane group: 120.7 ×10(9)/L vs. 97.3 ×10(9)/L, P<0.001) were slightly decreased following initial DPMAS treatment in both groups. The decrease in platelets was significantly lower in centrifugal separation than that in membrane separation (median: 10.4% vs. 17.0%; P=0.003). There was no statistically significant difference observed in the proportion of puncture site bleeding in terms of plasma separation-related adverse events between the two groups, but plasma separator membrane rupture occurred two times in the DPMAS treatment. Conclusion: Centrifugal and membrane separation, both with DPMAS therapy, can cause a slight decrease in hemoglobin, red blood cell count, and platelets in patients with liver failure. Membrane separation causes a larger drop in platelets than centrifugal plasma separation. The operational convenience of medical personnel, the risk of membrane rupture, the coagulation markers, the patient's vascular condition, and other factors should be comprehensively considered when choosing the plasma separation model.