Objective: To analyze the extubation indications of tracheotomy patients with severe neuropathy by Meta-analysis in order to determine the effective indication parameters for successful extubation. Methods: The literatures in databases including China National Knowledge Infrastructure (CNKI), Wanfang, VIP, Web of Science, PubMed and Cochrane Library were retrieved from their establishment to April 2019. The literatures were case-control studies, cohort studies, randomized controlled trials (RCTs) or surveys related to indication parameters for successful extubation in patients with severe neuropathy. Two researchers identified and extracted literatures and data independently. The quality of literatures was assessed by the Newcastle-Ottawa scale (NOS). Meta-analysis was performed by RevMan 5.3 software. The stability of the results were evaluated by assessing the statistical models (the fixed effects model or the random effects model) and literatures quality (inferior or superior), and by sensitivity analysis. The publication bias of literatures was assessed by funnel plot. Results: Eleven studies involving 1 357 participants were enrolled, and the rate of successful extubation was 46.7%-97.5%. Nine studies defined successful extubation as no need of re-intubation, and other two did not explain. All studies were high quality research, with NOS score of 6-8. Meta-analysis showed that the correlation between the level of consciousness [Glasgow coma score (GCS) ≥ 8 vs. < 8: odds ratio (OR) = 3.34, 95% confidence interval (95%CI) was 2.22-5.03, P < 0.001], the amount of tracheal secretions (less vs. more: OR = 13.07, 95%CI was 5.64-30.32, P < 0.001), cough reflex (with vs. without: OR = 14.33, 95%CI was 6.36-32.28, P < 0.001), swallowing function (good vs. bad: adjusted OR = 18.56, 95%CI was 8.16-42.21, P < 0.001) and successful extubation was statistically significant, and the correlation between the pulmonary infection (with vs. without: adjusted OR = 1.94, 95%CI was 0.87-4.35, P = 0.11), oxygen saturation (≥ 0.95 vs. < 0.95: OR = 2.34, 95%CI was 1.11-4.91, P = 0.12), tolerance of tube plugging (good vs. bad: OR = 2.12, 95%CI was 0.67-6.71, P = 0.20), method of tube drawing (gradually vs. abruptly: OR = 0.99, 95%CI was 0.95-1.04, P = 0.93) and successful extubation was not statistically significant. Sensitivity analysis showed that the results were stable. Funnel plot showed that the studies distributed in both sides of the funnel symmetrically, indicating the publication bias of literatures was small. Conclusion The effective indication of extubation is consciousness level (GCS score ≥ 8), less secretion, cough reflex and good swallowing function.