OBJECTIVE To clarify the association between premature coronary heart disease of patients ≤55 years and high-density lipoprotein cholesterol (HDL-C) plasma levels. METHODS Searches were performed between 2002 and 2013 using PubMed, Web of Science, Science Direct, and Google Scholar. The original articles selected published premature coronary heart disease diagnosed by World Health Organization criteria or via angiograph both in males and females ≤55 years and with plasma HDL-C levels in both the case and control groups. The 'related articles' function and manual searches of the related references was used to broaden the search. The Newcastle-Ottawa Quality Assessment Scale was used to assess the quality of papers. Standard mean difference with 95% confidence interval was used as a measure of the association between HDL and premature coronary heart disease, after pooling data across trials in a random effect model. Sensitivity and subgroup analyses were used to explore sources of heterogeneity and the effect of potential confounders. The influences of publication year, sample size and district were assessed by meta-regression. STATA (version 11.0) was used to conduct all statistical analyses. RESULTS Only 13 case-control studies met the criteria, which included 1775 patients and 1989 controls. The Newcastle-Ottawa Quality Assessment Scale score was about 5-7. A strong association was identified between HDL-C and premature coronary heart disease. The premature coronary heart disease patients had lower levels of HDL-C compared with the controls: standard mean difference = -0.48, 95% confidence interval = -0.71 to -0.26, p < 0.001, pheterogeneity < 0.001. By meta-regression and subgroup analysis, we found publication year might be the source of heterogeneity, but not the main reason for heterogeneity. After removing the heterogeneity of outlier studies, the significant association between low HDL-C levels and premature coronary heart disease was still retained. CONCLUSIONS Low plasma HDL-C levels are positively associated with premature coronary heart disease in patients ≤55 years. As only small sample size case-control studies were found to focus on this age group in the last 10 years, additional population-based cohort studies with large samples are necessary.