Numerous published studies have suggested that there is association between heme oxygenase-1 (HO-1) gene polymorphisms and coronary heart disease (CHD) or restenosis (RS) after percutaneous coronary intervention (PCI). This study aimed to clarify this association using a meta-analysis method. We used a systematic search for studies on the association of HO-1gene polymorphisms with CHD or RS in PubMed, Web of Science, the Cochrane Library, Wanfang Data and CNKI (China National Knowledge Infrastructure). We used Stata 12.0 software to perform the meta-analyses. Twenty-three studies, involving 12,130 patients with CHD or RS and 14,181 controls, were included. A statistically significant association between the HO-1(GT)n repeat length polymorphism and CHD was observed under allelic (odds ratio (OR) = 0.929, 95% confidence interval (CI) = 0.881-0.978, p= 0.005), recessive (OR = 0.858, 95%CI = 0.780-0.945, p= 0.002), and co-dominant (OR = 0.843, 95%CI = 0.754-0.942, p= 0.003) models. Moreover, we also found a statistically significant association between the HO-1(GT)n repeat length polymorphism and RS under allelic (OR = 0.718, 95%CI = 0.541-0.953, p= 0.022) and co-dominant (OR = 0.522, 95%CI = 0.306-0.889, p=0.017) models. We found a significant association of the HO-1T(−413)A single-nucleotide polymorphism (SNP) with CHD under allelic (OR = 0.915, 95%CI = 0.842-0.995, p= 0.038), recessive (OR = 0.869, 95%CI = 0.760-0.994, p= 0.041), and co-dominant (OR = 0.792, 95%CI = 0.663-0.946, p=0.010) models. Our study indicates that both the HO-1(GT)n repeat length polymorphism and the T(−413)A SNP are associated with decreased risk of CHD. The (GT)n repeat length polymorphism was associated with RS following PCI.