Background and Aims: As one of the most common malignant tumors, liver cancer has long been a human health concern. Hepatectomy is the preferred treatment of liver cancer, but the high postoperative recurrence rate and short survival time seriously influence the surgical efficacy. With the deepening of studies in tumor-associated inflammation, a series of systemic inflammatory indicators, including platelet to lymphocyte ratio (PLR), have been gradually proposed, and are considered to be markers that can be used to predict the prognosis of patients with malignant tumors. In recent years, studies have found that preoperative PLR can be used as an evaluation index to predict the prognosis of patients after hepatectomy, but the results of various studies are controversial. This study was designated to evaluate the relationship between preoperative PLR and prognosis of patients after hepatectomy by Meta-analysis, so as to provide reference for clinical treatment of liver cancer. 
Methods: The relevant studies concerning PLR and hepatocellular carcinoma published as of March 11, 2020 were searched in PubMed, Embase, Cochrane Library, Web of Science, CNKI, Wanfang Database and SinoMed databases. After literature screening and data extraction, Meta-analysis was performed by Stata 12.0 software, using hazard ratio (HR) and its 95% confidence interval (CI) as effect indicators.
Results: A total of 26 studies with 12 288 patients were finally included. The results of Meta-analysis showed that the shortened overall survival time (OS) after hepatectomy was related to high preoperative PLR (HR=1.03, 95% CI=1.01–1.04, P<0.001), and the increased preoperative PLR could predict poor disease-free survival (DFS) or recurrence free survival time (RFS) after operation (HR=1.05, 95% CI=1.02–1.07, P<0.001). Subgroup analysis showed that preoperative PLR could predict OS shortening in patients with BCLC-0/A disease (HR=1.47, 95% CI=1.17–1.80, P<0.05), but was irrelevant to DFS/RFS (HR=1.16, 95%CI=0.91–1.48, P=0.227); in patients receiving postoperative transarterial chemoembolization, local ablation therapy and other anti-cancer therapy, the poor OS was also related to the preoperative PLR (HR=1.07, 95% CI=1.030–1.109, P<0.001). Further exploring the range of effective cut-off value of PLR found that preoperative PLR was irrelevant to the shortened OS or DFS/RFS (HR=1.12, 95% CI=0.88–1.41, P=0.365; HR=1.26, 95% CI=0.93–1.72, P=0.135) when the cut-off value was less than 100. Heterogeneity analysis and publication bias test found that the heterogeneity was due to different tumor stages, different cut-off values of PLR and there was a significant publication bias among the included studies (Egger's test: P>|t|=0.000). After 11 studies were added, the results still steadily showed that high preoperative PLR was associated with poor OS.
Conclusion: Preoperative PLR is a biological index to predict the poor prognosis of patients undergoing hepatectomy, and the cut-off value for effective prediction should be more than 100. The above conclusions need to be further verified by high-quality multicenter prospective studies in the future, so that PLR can be better used in clinical practice.