Lung cancer surgery in octogenarians: a meta-analysis of predictors for postoperative complications

Background: Recently, several reports had focused on the feasibility of lung cancer surgery in octogenarians, describing encouraging short and long-term survival. However, the postoperative complication rate remains high. Some papers analyzed the role of potential predictors of post-operative complications, showing discordant results. We aimed at systematically assessing the evidence on risk factors for post-operative complications in octogenarians undergoing lung cancer surgery. Methods: Literature search was performed through PubMed and MEDLINE using the terms “lung cancer”, “non-small cell lung cancer” or “lung cancer surgery” combined with “octogenarian” or “80 years”. Last search was performed on 03/2021. Inclusion criteria were: clinical studies involving octogenarians undergoing lung cancer surgery between 1990 and 2020; studies explicitly appraising risk factors for postoperative complications and providing adjusted risk effect estimates. Studies that did not explicitly investigate predictors of postoperative complications and papers published in other language than English were not included. Risk of bias was assessed using the Quality in Prognostic Factor Studies (QUIPS) tool. More than 20 potential predictors were screened, finally limiting the inferential analysis to the six most frequently reported: performance status (PS), forced expiratory volume in 1 second (FEV1), type of resection, surgical approach, history of tobacco abuse and male gender. Adjusted P values from individual studies were pooled with the Fisher method. A pooled 2-tailed P value ≤0.05 was considered statistically significant. Results: A final set of 13 cohort studies, including 2,596 patients, was considered. Effect estimates were reported for PS by 6 studies (46%), for FEV1 by 10 (77%), for type of resection by 11 (85%), for surgical approach by 4 (31%), for history of tobacco abuse by 6 (46%), and for male gender by 8 (62%). After inferential analysis with the Fisher method, a significant association with post-operative complications was found for PS (P=0.004), major resections (P=0.002) and open approach (P=0.007); male gender showed only a borderline trend (P=0.053). Preoperative FEV1 (P=0.109) and history of tobacco abuse (P=0.399) were not associated with increased postoperative complications. Conclusions: Lung cancer surgery in octogenarian patients represent a viable and useful treatment. According to our analysis, patients should be carefully selected considering that a good preoperative PS is associated with a lower rate of postoperative complications. Similarly, thoracoscopic approach and sublobar resection should be preferred in octogenarian patients if technically and oncologically sound. Interestingly, impaired lung function was not significantly associated to post-operative complications.