This meta-analysis aimed to explore the effectiveness of partial spleen arterial embolization (PSAE) in the treatment of hypersplenism secondary to hepatic cirrhosis. PubMed, Embase, Cochrane Library, Wan Fang, CNKI, Vip, and CBM databases were searched for randomized controlled trials, cohort studies, and case-control studies that compared PSAE with splenectomy in the treatment of hypersplenism secondary to hepatic cirrhosis from their inception to July 25, 2015. Statistical analysis was conducted in Cochrane Network RevMan v5.3. Primary outcomes included the pre- and postoperative platelet and leukocyte counts and hemoglobin levels. Secondary outcomes were operative time, intraoperative volume of bleeding, and length of hospital stay. Mean and standard deviation were obtained from each study and then pooled using fixed- or random-effects models to calculate the mean difference. Ten original studies investigating 737 patients were included. Both the PSAE group and the splenectomy group yielded higher postoperative platelet and leukocyte counts and hemoglobin levels than the preoperative. The difference of platelet and leukocyte counts and hemoglobin levels between postoperative and preoperative levels in the PSAE group was smaller than that in the splenectomy group. Besides, compared with the splenectomy group, the PSAE group exhibited shorter operative time, less intraoperative bleeding, and shorter length of stay. PSAE is a mini-invasive therapy, which can be applied to treat hypersplenism secondary to hepatic cirrhosis effectively, particularly for patients with a poor overall condition. However, further high-quality studies should be conducted because this meta-analysis is limited by the quality of studies and the large statistical heterogeneity.