Associations of angiotensin converting enzyme gene (ACE) insertion/deletion (I/D) polymorphism with type 2 diabetes (T2D) have been inconsistent with both positive, null and negative results. We thereby performed a meta-analysis from all English-published reports to examine ACE I/D polymorphism in association with T2D risk. Case-control studies were identified from MEDLINE, EMBASE and Web of Science as of Dec 10, 2009. A total of 14 studies with 1985 patients with T2D and 4602 controls were finally identified. Random-effects model was applied irrespective of between-study heterogeneity. Study quality was assessed in duplicate. Compared with ACE I allele, presence of D allele conferred a significant increased risk for T2D (OR=1.33; 95% CI, 1.10-1.61; p=0.003). This trend was potentiated after comparing homozygotes of D allele with I allele with a 90% increased risk (p=0.0008). Carriers of D allele had a moderate increased risk for T2D compared with the II genotype carriers (OR=1.34; 95% CI, 1.04-1.72; p=0.02), whereas under recessive model this effect was significantly enhanced (OR=1.73; 95% CI, 1.26-2.38; p=0.0008). Subgroup analyses indicated significant association for population-based study design only, as well as among populations from Africa and Europe ancestries rather than from Asia ancestry. No publication bias was observed using the fail-safe number at the level of 0.05. Our results demonstrated that ACE D allele was significantly associated with an increased risk of T2D, and this effect appeared to be additive. Moreover, this association was more prominent for population-based studies and among Africans and Caucasians.