ObjectiveTo explore the correlation of pretreatment systemic immune inflammation index (SII) with prognosis in esophageal cancer patients.MethodsWe searched the PubMed, Web of Science, Embase, Cochrane Library, China National Knowledge Infrastructure, VIP, Chinese Biology Medicine, and Wanfang databases to identify eligible studies evaluating the relation between pretreatment SII and prognosis in patients with esophageal cancer from establishment of databases to December 2018. SII was defined as the absolute neutrophil count multiplied by the absolute platelet count divided by the absolute lymphocyte count. The primary endpoint was overall survival (OS), and the secondary endpoints were cancer-specific survival and disease-free survival. The Stata 12.0 software was applied for the meta-analysis, and the hazard ratio (HR) and 95% confidence interval (CI) were assessed.ResultsA total of six retrospective studies involving 2 376 esophageal cancer patients were included and all patients were from China or Japan. The results revealed that elevated pretreatment SII was significantly associated with poor OS in esophageal cancer [HR=1.50, 95%CI (1.15, 1.95), P=0.002]. Subgroup analyses of OS indicated that SII had a high prognostic value in patients who received surgery [HR=1.54, 95%CI (1.14, 2.08), P=0.005] and were diagnosed as esophageal squamous cell carcinoma [HR=1.50, 95%CI (1.11, 2.02), P=0.007]; however, no significant relation was observed between SII and prognosis in esophageal cancer patients who were treated with radiotherapy [HR=1.318, 95%CI (0.611, 2.841), P=0.482]. Furthermore, compared with neutrophil to lymphocyte ratio and platelet to lymphocyte ratio, SII showed a higher predictive value for the prognosis of esophageal cancer.ConclusionsPretreatment SII may serve as an independent risk factor for prognosis of Chinese and Japanese esophageal cancer patients, especially patients who were treated with surgery and with esophageal squamous cell carcinoma. However, more prospective studies with big samples from other countries or regions are still needed to verify our findings.