To ascertain the ability of adrenomedullin (ADM) and proadrenomedullin (proADM) to predict mortality in sepsis patients. A systematic literature search was made of the PubMed, EMBASE, Cochrane and China National Knowledge Infrastructure (CNKI) databases before May 2017, supplemented by manual searches of references. A meta-analysis of high-quality clinical studies was subsequently performed to assess the association between ADM/proADM and mortality risk among patients with sepsis. Thirteen studies involving 2556 patients were included in the study. Two reviewers independently identified articles, extracted data, assessed quality and cross-checked the results. The predictive values of ADM and proADM referred to mortality were assessed by relative risk (RR). The overall diagnostic accuracy of ADM and proADM in application to sepsis was pooled according to a bivariate model. Publication bias was assessed using Deek's funnel plot asymmetry test. Elevated ADM or proADM levels were associated with increased mortality (pooled RR=3.31; 95%CI 2.31-4.75). Subgroup analyses indicated the pooled RRs were 3.12 (95%CI 1.75-5.56) and 3.43 (95%CI 2.21-5.31) for ADM and proADM, respectively. The pooled sensitivity and specificity were 0.72 (95%CI 0.64-0.78) and 0.77 (95%CI 0.69-0.83), respectively. The overall area under the summary receiver operating characteristic (SROC) curve was 0.80 (95%CI 0.77-0.84). Publication bias was not statistically significant. Both ADM and proADM might serve as useful markers for predicting the prognosis of sepsis.