Objective To evaluate the efficacy of the subtotal splenectomy versus total splenectomy with gastroesophageal devascularization for patients of hepatic cirrhosis and portal hypertension. Methods We searched the Cochrane Library (Issue 2, 2008), MEDLINE (1966 to August, 2008), EMbase (1966 to August, 2008), the China Biological Medicine Database (1978 to August, 2008), Chinese Sci-tech Periodical Full-text Database (1989 to August, 2008) and Chinese Periodical Full-text Database (1994 to August, 2008), as well as hand-searched several related journals and conference proceedings for the randomized controlled trials involving the comparison of the efficacy of the subtotal splenectomy with the total splenectomy for the patients of the hepatic cirrhosis and portal hypertension. Results Three studies involving 136 patients were identified. The results of two studies indicated that both of the subtotal splenectomy increased less the peripheral platelet count and there was a statistically significant difference between the two groups [WMD= – 39.27, 95%CI (– 62.57, – 15.97)]. Two studies indicated that the serum tuftsin level was increased significantly after the subtotal splenectomy [WMD= 165.28, 95%CI (159.36, 171.21)]. One study indicated that both of the subtotal splenectomy and total splenectomy with gastroesophageal devascularization increased the peripheral white blood cell count and there was statistical difference between the two groups [WMD= – 0.93, 95%CI (– 1.52, – 0.34)]. There was no statistical difference in serum IgA level between the two groups. One study indicated the average fever time after the total splenectomy was longer than the subtotal splenectomy; there was statistical differences in 5 years survival rate between the two groups (Plt;0.05). Conclusions After the subtotal splenectomy the hypersplenism of patients suffered from hepatic cirrhosis and portal hypertension was corrected obviously and the blood cells were increased smoothly so that high blood viscosity was prevented and occurrence of the thrombotic diseases was decreased. And the immune system reserve functions of the patients with hepatic cirrhosis and portal hypertension were maintained. More randomized controlled trials, with large sample sizes, may lead to more accurate results.