Blood perfusion with polymyxin B immobilized columns in patients with COVID-19 requiring oxygen therapy

Abstract

Extracorporeal blood purification with polymyxin B immobilized fiber column direct hemoperfusion (PMX-DHP), is reported to be effective in treating COVID-19 pneumonitis with oxygen demand. This multicenter prospective study evaluated the efficacy and safety of PMX-DHP in oxygen-requiring patients with COVID-19 admitted between September 28, 2020, and March 31, 2022. The primary endpoint was the percentage of clinical improvement 15 days after treatment. The secondary endpoint was the percentage of worsened disease status. Data from the COVID-19 patient registry were used for the synthetic control group. The improvement rate on Day 15 did not differ between PMX-treated patients and controls; however, the deterioration rate was 0.38 times lower in the PMX-treated group, and the death rates on Day 29 were 0 and 11.1% in the PMX-treated and control groups, respectively. The PMX group showed a 0.73 times higher likelihood for reduced intensive care demand, as 16.7% of PMX-treated patients and 22.8% of controls worsened. After treatment blood oxygenation improved, urinary β2-microglobulin and liver-type fatty acid-binding protein showed significant decreases, and IL-6 decreased once during treatment but did not persist. In this study, PMX treatment effectively prevented the worsening of COVID-19 pathology, accompanied by improved oxygenation. PMX treatment to remove activated cells may effectively improve patient outcomes.

Figure 1

Flow diagram of the study. A total of 21 patients received PMX-DHP therapy in this study; 3 patients were excluded for reasons related to matching in the control group. In the control group, 1853 patients were selected from the COVIREGI-JP database, of which 407 were used for propensity score estimation. Finally, 76 individuals were selected as synthetic controls. COVIREGI-JP, COVID-19 Registry Japan; PMX-DHP, polymyxin B immobilized fiber column direct hemoperfusion.

Figure 2

Kaplan–Meier estimate. Mortality on Day 15 was 0% in the PMX-treated group and 7.8% in the control group; mortality on Day 29 was 0% in the PMX-treated group and 11.1% in the control group.

Figure 3

Clinical effectiveness in the PMX-treated group. Variations in various parameters before, during, and after PMX treatment for FAS. Horizontal black bars indicate the first quartile, median, and third quartile at each time point. Red diamonds indicate mean values at each time point. (A) P/F ratio, (B) serum IL-6, (C) urinary β2-microbloburin, (D) urinary L-FABP. L-FABP, liver-type fatty acid-binding protein; P/F ratio, PaO₂/FiO₂ ratio.