Leukocyte filtration ameliorates the inflammatory response in patients with mild to moderate lung dysfunction.

Background: Chronic obstructive pulmonary disease is a risk factor for postoperative lung injury. Contradictory results have been published about leukocyte filtration (LF) because of the heterogeneity of patients and interventions, type of LF, and comorbidities. Methods: Sixty patients with mild moderate chronic obstructive pulmonary disease (forced expiratory volume in 1 second 40% to 80%) undergoing aortic valve surgery were randomly assigned to receive systemic arterial and cardioplegic LF during cardiopulmonary bypass (group L, 30 patients) or standard cardiopulmonary bypass (group S). Perioperative interleukin-6, interleukin-8, and tumor necrosis factor-alpha were sampled at different time points. The PaO2/inspired oxygen fraction (FiO2) and alveoloarterial oxygen gradient (AaDO2) were measured preoperatively, at intensive care unit arrival, and at 24, 48, and 72 hours postoperatively; lung compliance was measured after intubation, at intensive care unit arrival, and at 4 and 8 hours postoperatively; and radiographic lung injury score was determined preoperatively and at 24, 48 and 72 hours. Length of intubation, intensive care unit stay, hospital stay, need for noninvasive positive-pressure ventilation, acute lung injury, and pneumonia were recorded. Repeated-measures analysis of variance assessed group, time, and group by-time interactions. Results: Preoperative and intraoperative data were comparable. Proinflammatory cytokine leakage was reduced by LF. Group L showed shorter intubation time (median 9.5 hours versus group S, 15.0 hours; p = 0.0001), and intensive care unit length of stay (median 19.0 hours versus group S, 24.5; p = 0.0001), lower need for noninvasive positive-pressure ventilation (5 of 30, 16.7%, versus 12 of 30, 40%; p = 0.042). The AaDO2, PaO2/FiO2, lung compliance, and radiographic lung injury score worsened early postoperatively, followed by progressive improvements (time p ≤ 0.001 for all). Such decline of AaDO2, PaO 2/FiO2, lung compliance, and radiographic lung injury score was significantly attenuated by LF (group by-time p = 0.0001 for AaDO 2, PaO2/FiO2, and lung compliance; p = 0.004 for radiographic lung injury score). Conclusions: Arterial plus cardioplegic LF significantly reduced proinflammatory cytokine release after cardiopulmonary bypass, thus ameliorating postoperative indexes of lung function and overall respiratory outcome.