Resting End-Tidal Carbon Dioxide Predicts Respiratory Complications in Patients Undergoing Thoracic Surgical Procedures

Background Ventilatory efficiency (V˙E/V˙CO₂ slope [minute ventilation to carbon dioxide output slope]) has been shown to predict morbidity and mortality in lung resection candidates. Patients with increased V˙E/V˙CO₂ during exercise also exhibit an increased V˙E/V˙CO₂ ratio and a decreased end-tidal CO₂ at rest. This study hypothesized that ventilatory values at rest predict respiratory complications and death in patients undergoing thoracic surgical procedures. Methods Inclusion criteria for this retrospective, multicenter study were thoracotomy and cardiopulmonary exercise testing as part of routine preoperative assessment. Respiratory complications were assessed from the medical records (from the hospital stay or from the first 30 postoperative days). For comparisons, Student's t test or the Mann-Whitney U test was used. Logistic regression and receiver operating characteristic analyses were performed for evaluation of measurements associated with respiratory complications. Data are summarized as mean ± SD; p <0.05 is considered significant. Results Seventy-six subjects were studied. Postoperatively, respiratory complications developed in 56 (74%) patients. Patients with postoperative respiratory complications had significantly lower resting tidal volume (0.8 ± 0.3 vs 0.9 ± 0.3L; p = 0.03), lower rest end-tidal CO₂ (28.1 ± 4.3vs 31.5 ± 4.2 mm Hg; p < 0.01), higher resting V˙E/V˙CO₂ ratio (45.1 ± 7.1 vs 41.0 ± 6.4; p = 0.02), and higher V˙E/V˙CO₂ slope (34.9 ± 6.4 vs 31.2 ± 4.3; p = 0.01). Logistic regression (age and sex adjusted) showed resting end-tidal CO₂ to be the best predictor of respiratory complications (odds ratio: 1.21; 95% confidence interval: 1.06 to 1.39; area under the curve: 0.77; p = 0.01). Conclusions Resting end-tidal CO₂ may identify patients at increased risk for postoperative respiratory complications of thoracic surgical procedures.