Individual and combined effects of chemical and mechanical power on postoperative pulmonary complications: a secondary analysis of the REPEAT study

Abstract

Introduction: Intra-operative supplemental oxygen and mechanical ventilation expose the lungs to potentially injurious energy. This can be quantified as 'chemical power' and 'mechanical power', respectively. In this study, we sought to determine if intra-operative chemical and mechanical power, individually and/or in combination, are associated with postoperative pulmonary complications.

Methods: Using an individual patient data analysis of three randomised clinical trials of intra-operative ventilation, we summarised intra-operative chemical and mechanical power using time-weighted averages. We evaluated the association between intra-operative chemical and mechanical power and a collapsed composite of postoperative pulmonary complications using multivariable logistic regression to estimate the odds ratios related to the effect of 1 J.min-1 increase in chemical or mechanical power with adjustment for demographic and intra-operative characteristics. We also included an interaction term to assess for potential synergistic effects of chemical and mechanical power on postoperative pulmonary complications.

Results: Of 3837 patients recruited to three individual trials, 2492 with full datasets were included in the analysis. Intra-operative time-weighted average (SD) chemical power was 10.2 (3.9) J.min-1 and mechanical power was 10.5 (4.4) J.min-1. An increase of 1 J.min-1 in chemical power was associated with 8% higher odds of postoperative pulmonary complications (OR 1.08, 95%CI 1.05-1.10, p < 0.001), while the same increase in mechanical power raised odds by 5% (OR 1.05, 95%CI 1.02-1.08, p = 0.003). We did not find evidence of a significant interaction between chemical and mechanical power (p = 0.40), suggestive of an additive rather than synergistic effect on postoperative pulmonary complications.

Discussion: Both chemical and mechanical power are independently associated with postoperative pulmonary complications. Further work is required to determine causality.