Several molecular epidemiological studies have been conducted to examine the association between glutathione S-transferase mu-1 (GSTM1) and glutathione S-transferase theta-1 (GSTT1) null polymorphisms and childhood acute leukemia; however, the conclusions remain controversial. We performed an extensive meta-analysis on 26 published case-control studies with a total of 3252 cases and 5024 controls. Crude odds ratios (ORs) with 95% confidence interval were used to assess the strength of association between childhood acute leukemia risk and polymorphisms of GSTM1 and GSTT1. With respect to GSTM1 polymorphism, significantly increased risk of childhood acute leukemia was observed in the overall analysis (OR = 1.30; 95%CI, 1.11-1.51). Furthermore, a stratification analysis showed that the risk of GSTM1 polymorphism are associated with childhood acute leukemia in group of Asians (OR = 1.94; 95%CI, 1.53-2.46), Blacks (OR = 1.76; 95%CI, 1.07-2.91), ALL (OR = 1.33; 95%CI, 1.13-1.58), '< 100 cases and <100 controls' (OR = 1.79; 95%CI, 1.21-2.64), '≥ 100 cases and ≥ 100 controls' (OR = 1.25; 95%CI, 1.02-1.52), and population-based control source (OR = 1.40; 95%CI, 1.15-1.69). With respect to GSTT1 polymorphism, significant association with childhood acute leukemia risk was only found in subgroup of Asian. This meta-analysis supports that GSTM1 null polymorphism is capable of causing childhood acute leukemia susceptibility.