Predicting of Postoperative Pulmonary Complications Following Esophageal Cancer Surgery Using Univariate and Multiplelogistic Regression Analysis

Kazusada Shirao

The First Department of Surgery, Kagoshima University School of Medicine

In order to predict postoperative pulmonary complications, I studied 47 patients with esophageal cancer who underwent esophagectomy between October 1987 and March 1991, using univariate analysis and multiple logistic regression analysis. Pulmonary complications occurred in 15 patients. Age, vital capacity, forced expiratory volume and serum transferrin were significantly different between the cases with and without pulmonary complications. As a result of multiple logistic regression analysis, vital capacity, percentage of ideal body weight and serum prealbumin were selected as factors of predicting postoperative pulmonary complications. When the decision level was fixed at 0.3, I had an overall prediction rate of 83% (39/47 cases) between October 1987 and March 1991 by retrospective study, and 73% (30/41 cases) between April 1991 and October 1993 by prospective study. It is concluded that predicting pulmonary complications is useful for perioperative management of esophageal cancer surgery

Key words
esophageal cancer, multiple logistic regression analysis, pulmonary complication

Reprint requests
Kazusada Shirao First Department of Surgery, Kagoshima University School of Medicine
8-35-1 Sakuragaoka, Kagoshima, 890 JAPAN

Accepted
May 17, 1995

To read the PDF file you will need Abobe Reader installed on your computer.