Preoperative prediction of unresectability in malignant pleural mesothelioma

Diffuse chest wall invasion (DCWI) is a common finding in patients undergoing intended resection for malignant pleural mesothelioma. We sought to determine the incidence and preoperative predictors of this finding, and to test our anecdotal impression that contraction of the ipsilateral hemithorax is associated with DCWI. This was a single-institution retrospective study of 170 patients undergoing intended macroscopic complete resection for malignant pleural mesothelioma from 2014-2018. A novel metric of thoracic cage volume was calculated by preoperative chest computed tomography. Univariable analyses were performed to determine associations of preoperative variables with DCWI. Macroscopic complete resection was achieved by pleurectomy/decortication in 104 patients (61%) and by extrapleural pneumonectomy in 39 patients (23%). Unresectable disease was discovered at thoracotomy in 27 (16%) of patients; 24 (14%) by DCWI and 3 (2%) by intrathoracic organ invasion. In univariable analysis, decreased ipsilateral thoracic cage volume demonstrated the strongest association with unresectability by DCWI (P = .009) with >5% decrease in thoracic cage volume representing the optimal cutoff (P = .014; area under the curve, 0.67). Other preoperative variables associated with DCWI included preoperative chest pain requiring opioids (P = .028), prior pleurodesis (P = .036), decreased forced vital capacity (P = .023), decreased ipsilateral lung perfusion by ventilation/perfusion lung scan (P = .007), and magnetic resonance imaging findings of chest wall invasion (P = .035). Preoperative identification of DCWI will avoid unnecessary thoracotomy and accelerate initiation of nonsurgical therapy in malignant pleural mesothelioma. Our data suggest that contraction of thoracic cage volume has merit in predicting malignant pleural mesothelioma unresectability and should be validated in prospective studies.