The Effects of Preserving the Diaphragm on Early and Late Outcome of Lung-Sparing Radical Surgery for Malignant Pleural Mesothelioma

Background: The accepted aim of radical surgery for malignant pleural mesothelioma (MPM) is the achievement of macroscopic complete resection (MCR) whilst reducing perioperative morbidity by preserving normal tissue. Whilst preservation of the lung by pleurectomy/decortication (PD) has become widely utilised, there remains debate regarding the management of the diaphragm. Muscle-sparing complete excision of the diaphragmatic pleura is technically challenging; thus, surgeons may proceed to extended PD with phrenectomy and possible increased morbidity or to preserve the diaphragmatic pleura at the expense of MCR with potential survival deficit. We aimed to evaluate the effects of an intentional change in protocol to diaphragm-sparing PD whilst maintaining MCR as the treatment of choice for MPM. Methods: In a series of 136 patients (111M:25F, median age 68(63-73) years) undergoing radical surgery for MPM, we identified 28 patients (22M:6F, median age 67(60-71) years) in whom MCR was achieved without phrenectomy (PD group). We compared their perioperative outcomes and survival with a historical control group of 18 patients (18M:0F, median age 69(57-78) years) in whom MCR had been achieved with phrenectomy (EPD group) but in whom there was no histological evidence of diaphragm muscle invasion and who, in retrospect, could have undergone muscle-sparing MCR if this procedure had been attempted. Results: There was no significant intergroup difference in demographics or tumour cell type; the majority of both groups were found to be epithelial (PD 85.7%, EPD 77.8%). The EPD group was found to be more locally advanced (T3 55.56%) than the PD group (T1 46.43%) (p = 0.03). All the following parameters were significantly reduced after PD compared to EPD: operative time (188 vs. 220 min, p = 0.007); duration of air leak (5 vs. 10 days, p = 0.001), duration of inotrope (p = 0.009) and post-operative hospital stay (8 vs. 13 days, p = 0.034). There were no significant differences (p = 0.123) in overall survival (OS) between the two groups, but the median survival in the PD group had not been reached at a median follow up of 33.9 (24.2-46) months. Conclusions: A surgical strategy of attempting to spare the diaphragm whilst still achieving MCR wherever possible is justified by improved perioperative outcomes without compromising OS.