A risk model to predict 2-year survival after video-assisted thoracoscopic surgery lobectomy for non-small-cell lung cancer. (13th November 2019)

• Record Type: Journal Article
• Authors: Tan, Hui Xian; Drake, Benjamin Cooper; Chaudhuri, Nilanjan; Kefaloyannis, Manos; Milton, Richard; Papagiannopoulos, Kostas; Tcherveniakov, Peter; Brunelli, Alessandro
• Is Part Of: European journal of cardio-thoracic surgery. Volume 57:Number 4(2020)
• Page Start: 781
• Page End: 787
• Publication Date: 2019-11-13
• Subjects: Non-small-cell lung cancer -- Lung surgery -- Risk score -- Survival -- Video-assisted thoracoscopic surgery; Heart -- Surgery -- Periodicals; Chest -- Surgery -- Periodicals; 617.54
• DOI: 10.1093/ejcts/ezz304
• Languages: English
• ISSNs: 1010-7940

Abstract

Abstract: : OBJECTIVES: We sought to identify the risk factors associated with mortality post-video-assisted thoracoscopic surgery (VATS) lobectomy over a 2-year period. METHODS: Analysis was performed using a sample from an institutionally maintained database. All lobectomies for non-small-cell lung cancer from April 2014 to March 2018 started with VATS approach and with a complete follow-up were included ( n = 732). Several clinical variables were screened using the Cox univariate analysis for their association with 2-year survival. Those with a P -value <0.1 were included in a Cox proportional hazard model. RESULTS: After multivariable analysis, the following variables showed significant association with 2-year survival: age >75 [hazard ratio (HR) 1.527, P = 0.043], carbon monoxide lung diffusion capacity <70 (HR 1.474, P = 0.061), body mass index (BMI) <18.5 (HR 2.628, P = 0.012), American Society of Anesthesiologist Physical Status >2 (HR 1.518, P = 0.047), performance status >1 (HR 1.822, P = 0.032) and male gender (HR 2.700, P < 0.001). A score of 2 was assigned to the male gender and BMI <18.5, with all other variables assigned a score of 1. Each patient was scored and placed into their risk class. A Kaplan–Meier estimate for 2-year survival was calculated for each class. These were collapsed into the following 3 classes of risk based on their similar 2-year survival: class A (score 0) 97%, 95% CI 88–99, class B (score 1–3) 84%, 95% CI 80–88, class C (score > 3) 66%, 95% CI 57–74. CONCLUSION: Our scoring system can serve as an adjunct to a clinician's experience in risk-stratifying patients during multidisciplinary tumour board discussion and the shared decision-making process.