The Prediction of Fissure Integrity by Quantitative Computed Tomography Analysis

Abstract

Abstract Background Incomplete interlobar fissure may increase the difficulty of thoracoscopic lobectomy. Herein, we compared the accuracy of visual versus quantitative analysis to predict fissure integrity in lung cancer patients undergoing thoracoscopic lobectomy and evaluated the effects of fissure integrity on surgical outcome. Methods This was a single-center retrospective study including consecutive patients undergoing VATS (video-assisted thoracoscopic surgery) lobectomy for lung cancer. The target interlobar fissures were classified as complete or incomplete by visual and quantitative analysis. Using the intraoperative finding as the reference method, the diagnostic accuracy of the two methods to define fissure completeness (dependent variable) was calculated and statistically compared. Yet, we evaluated differences in postoperative outcomes between patients with complete and incomplete fissure integrity. Results A total of 93 patients were included in the study; 33/93 (36%) presented complete fissure. Visual and quantitative analyses correctly identified complete fissure in 19/33 (57%) and 29/33 (88%) patients, respectively, and incomplete fissure in 56/60 (93%) and 58/60 (96%) patients, respectively. Quantitative analysis had better diagnostic accuracy than visual analysis (81 vs. 93%; p = 0.01). Patients with incomplete fissure compared with those with complete fissure had a higher conversion rate (6 vs. 13%; p = 0.43), higher persistent air leak rate (0/33 vs. 14/60; p = 0.03), and longer hospitalization (12.6 ± 3.8 vs. 7.1 ± 2.4 days; p = 0.01). Conclusions Quantitative analysis accurately predicted the fissures' integrity; it may be useful for selecting suitable cases for thoracoscopic lobectomy especially for surgeons with limited minimally invasive experience.