Value of the oxygenation index during 1-lung ventilation for predicting respiratory complications after thoracic surgery

Purpose: Thoracic surgery requiring 1-lung ventilation (OLV) is associated with significant postoperative pulmonary complications (PPCs) due to patient-and surgery-specific risk factors. The aim of this study is to determine whether the oxygenation index (OI) (fraction of inspired oxygen x mean airway pressure/partial pressure of oxygen) during OLV predicts the development of PPC. Materials and methods: The Society of Thoracic Surgeons General Thoracic Surgery Database at the University of Virginia was used to identify all thoracic cases in 2012 to 2014. All subjects requiring OLV and in whom a blood gas was available for analysis were included. Univariate and multivariate analyses were performed to determine whether the OI predicted the development of PPC. Results: A total of 296 subjects had OLV with a calculable OI during the study period. Composite PPC were identified in 97 (33%) of subjects. In the PPC group, mortality at 30 days was 7.2%, whereas no deaths were identified in the non-PPC group. In both univariate and multivariate analyses, OI was not predictive of PPC analysis (odds ratio [OR], 1.0; 95% confidence interval [CI], 0.9-1.1; P = .8). The only predictors of mortality were forced expiratory volume (OR, 0.98; 95% CI, 0.96-0.99; P = . 03) and diffusion capacity of carbon monoxide (OR, 0.98; 95% CI, 0.96-0.99; P = .03). Conclusion: The calculated OI during OLV is not predictive of PPC in our sample. Additional studies in a larger cohort of patients are needed to identify intraoperative ventilation parameters during OLV that can predict the risk of PPC. (C) 2016 Elsevier Inc. All rights reserved.