Could Thoracoscore predict postoperative mortality in patients undergoing pneumonectomy?

Thoracoscore is incorporated in the new British Thoracic Society and National Institute of Health and clinical Excellence guidelines to evaluate the operative mortality risk of patients undergoing thoracic surgery. This study examines the accuracy of Thoracoscore in predicting postoperative mortality in patients undergoing pneumonectomy. All patients who underwent pneumonectomy from January 1998 to March 2008 were included. Thoracoscore was calculated based on the following variables: age, sex, American Society of Anaesthesiologists' class, performance status classification, dyspnoea score, priority of surgery, procedure class, Diagnosis group and comorbidities score. Two hundred and forty-three patients with a mean age of 63 +/- 9 years were included and 81% were male. The predicted postoperative mortality based on Thoracoscore was 8 +/- 2.6% (95% confidence interval (CI) 4.56-11.43), while actual in-hospital mortality was 4.5% (11/243) (95% CI 1.87-7.12). 54% (6/11) of in-hospital mortality was of those who were > 70 years old and 73% (8/11) of patients who died in hospital were male. Nine of 11 (82%) patients had pneumonectomy for malignancy. Thoracoscore was divided into four risk groups: low (0-3), moderate (3.1-5), high (5.1-8) and very high (> 8). It underestimated mortality in low-risk group while overestimated in high-risk groups. The 30-day, 1-year, 2-year and 3-year observed mortalities were 5.3, 29, 43 and 55%, respectively. Although advanced age, the male sex and malignancy proved to be strong predictors of in-hospital mortality in our study, Thoracoscore failed to predict accurate risk of in-hospital mortality in pneumonectomy patients in this study. Further studies are required to validate the Thoracoscore in different subgroups of thoracic surgery.