To determine the incidence of fatal pulmonary events after extrapleural pneumonectomy and hemithoracic intensity-modulated radiotherapy (IMRT) for malignant pleural mesothelioma.

We retrospectively reviewed the records of 63 consecutive patients with malignant pleural mesothelioma who underwent extrapleural pneumonectomy and IMRT at the University of Texas M. D. Anderson Cancer Center. The endpoints studied were pulmonary-related death (PRD) and non-cancer-related death within 6 months of IMRT.

Of the 63 patients, 23 (37%) had died within 6 months of IMRT (10 of recurrent cancer, 6 of pulmonary causes [pneumonia in 4 and pneumonitis in 2], and 7 of other noncancer causes [pulmonary embolus in 2, sepsis after bronchopleural fistula in 1, and cause unknown but without pulmonary symptoms or recurrent disease in 4]). On univariate analysis, the factors that predicted for PRD were a lower preoperative ejection fraction (p = 0.021), absolute volume of lung spared at 10 Gy (p = 0.025), percentage of lung volume receiving >or=20 Gy (V(20); p = 0.002), and mean lung dose (p = 0.013). On multivariate analysis, only V20 was predictive of PRD (p = 0.017; odds ratio, 1.50; 95% confidence interval, 1.08-2.08) or non-cancer-related death (p = 0.033; odds ratio, 1.21; 95% confidence interval, 1.02-1.45).

The results of our study have shown that fatal pulmonary toxicities were associated with radiation to the contralateral lung. V20 was the only independent determinant for risk of PRD or non-cancer-related death. The mean V20 of the non-PRD patients was considerably lower than that accepted during standard thoracic radiotherapy, implying that the V20 should be kept as low as possible after extrapleural pneumonectomy.