The effects of direct hemoperfusion using a polymyxin B-immobilized column in a pig model of severe Pseudomonas aeruginosa pneumonia

Abstract

Background: Hemoperfusion through a column containing polymyxin B-immobilized fiber (PMX-HP) is beneficial in abdominal sepsis. We assessed the effects of PMX-HP in a model of severe Pseudomonas aeruginosa pneumonia.

Methods: Eighteen pigs with severe P. aeruginosa pneumonia were mechanically ventilated for 76 h. Pigs were randomized to receive standard treatment with fluids and vasoactive drugs, or standard treatment with two 3-h PMX-HP sessions. Antibiotics against P. aeruginosa were never administered. We assessed endotoxemia through the endotoxin activity assay (EA). We measured the static lung elastance, ratio of arterial partial pressure per inspiratory fraction of oxygen (PaO2/FIO2), mean arterial pressure, cardiac output, systemic vascular resistance and inotropic score. Finally, every 24 h, we assessed complete blood count.

Results: In comparison with the control group, PMX-HP decreased percentage of circulating neutrophils from 47.4 ± 13.8 to 40.8 ± 11.5 % (p = 0.009). In a subgroup of animals with the worst hemodynamic impairment, EA in the control and PMX-HP groups was 0.50 ± 0.29 and 0.29 ± 0.14, respectively (p = 0.018). Additionally, in the control and PMX-HP groups, static lung elastance was 26.9 ± 8.7 and 25.3 ± 7.5 cm H2O/L (p = 0.558), PaO2/FIO2 was 347.3 ± 61.9 and 356.4 ± 84.0 mmHg (p = 0.118), mean arterial pressure was 81.2 ± 10.3 and 81.6 ± 13.1 mmHg (p = 0.960), cardiac output was 3.30 ± 1.11 and 3.28 ± 1.19 L/min (p = 0.535), systemic vascular resistance was 1982.6 ± 608.4 and 2011.8 ± 750.0 dyne/s/cm(-5) (p = 0.939), and inotropic score was 0.25 ± 0.10 and 0.26 ± 0.18 (p = 0.864).

Conclusions: In mechanically ventilated pigs with severe P. aeruginosa pneumonia, PMX-HP does not have any valuable clinical benefit, and studies are warranted to fully evaluate a potential role of PMX-HP in septic shock associated with severe pulmonary infections.

Keywords: Complications—septicemia; Infection—bacterial; Intensive care—pulmonary; Lung—endotoxemia.

Fig. 1

Differences in most significant hemodynamic parameters, between the control and treatment group. a Central venous pressure significantly differed between groups (p < 0.001) and changed over time (p < 0.001), without interactions between study groups and time of assessment. Similarly, pulmonary arterial pressure (b) significantly differed between groups (p = 0.006) and significantly changed over time (p < 0.001). There were no significant differences between study groups, no significant changes over time and no interactions between study groups and time of assessment in mean arterial pressure (c); cardiac output (d); systemic vascular resistance (e); and mixed venous oxygen saturation (f). *Polymyxin B-immobilized fiber hemoperfusion was carried out for 3 h

Fig. 2

Endotoxin activity in the control and treatment group during study time. Data are reported as mean and standard error. Polymyxin B-immobilized fiber hemoperfusion was carried out for 3 h, after 30 and 50 h. a Endotoxin activity in all subjects. Group effect, p = 0.224; study time effect, p = 0.882; group * study time effect, p = 0.572. b Post hoc analysis in only animals with a decrease in mean arterial pressure ≥10 %, within 24 h from bacterial challenge. Group effect, p = 0.018; study time effect, p = 0.473; group * study time effect, p = 0.830

Fig. 3

Differences in laboratory parameters and oxygenation parameters between the control and treatment group. a Total white blood cells count (a) (p = 0.031) differed between study groups, without significant changes over time, and no interactions between study groups and time of assessment; while percentage of neutrophils (b) differed between study groups (p = 0.009), significantly changed over time (p < 0.001), without interactions between study groups and time of assessment; similarly, percentage of lymphocytes (c) differed between study groups (p = 0.002), significantly changed over time (p < 0.001), without interactions between study groups and time of assessment; also, platelets count (d) differed between study groups (p < 0.001), significantly changed over time (p < 0.001), without interactions between study groups and time of assessment. Ratio of partial pressure arterial oxygen and fraction of inspired oxygen (PaO₂/FiO₂) (e) and pulmonary shunt (f) significantly changed over time (p < 0.001 and p = 0.018, respectively), in a similar fashion between groups, and without interactions between study groups and time of assessment. Of note, hemoperfusion through a column containing polymyxin B-immobilized fiber caused a significant reduction in circulating neutrophils and platelets and an increase in circulating lymphocytes