Abstract
We evaluated the capacity to predict severe respiratory complications (SRCs) following upper abdominal surgery (UAS) by using the results of a respiratory questionnaire and preoperative pulmonary function tests. Lung volumes, flows and transfer factor of the lung for carbon monoxide (TL,CO,sb) were assessed in 361 consecutive adult patients (248 males and 113 females). SRCs were diagnosed 24 h after UAS by clinical examination and chest radiography. Univariate and stepwise multiple logistic regression analyses were performed to estimate the odds ratio (OR) and 95% confidence interval (95% CI) of each single input variable, and to determine which indices best predicted outcome. These patients had a 1% mortality rate and 14% incidence of SRCs, with a male:female ratio of 0.86. The best predictors for SRCs by multiple analysis were: preoperative current hypersecretion of mucus (OR=133; p<0.0001); an increase in residual volume (RV) (OR=3.11; p=0.01); and, to a lesser extent, low percentage of predicted values both of forced expiratory volume in one second (FEV1 % pred) and TL,CO,sb. The algorithm thus obtained (logit theta) was extremely sensitive (84%), specific (99%), and accurate (95%) for preoperative prediction of SRCs. We have found that preoperative current hypersecretion of mucus and pulmonary hyperinflation, and to a lesser extent percentage predicted values both of forced expiratory volume in one second and transfer factor of the lung for carbon monoxide, have a significant predictive capacity for severe respiratory complications following upper abdominal surgery.
