Serum albumin and risk of cardiovascular events in primary and secondary prevention: a systematic review of observational studies and Bayesian meta-regression analysis

The predictive role of serum albumin (SA) has been evaluated in primary prevention studies. We want to assess the association of SA with the subsequent risk of cardiovascular events (CVE) in primary and secondary prevention studies. We performed a systematic review and Bayesian meta-regression analysis. Studies were identified by PubMed and EMBASE database using a combination of the following terms and MeSH terms: "serum albumin", "myocardial infarction, "cardiovascular events", "percutaneous coronary intervention" and "coronary restenosis". No time restriction of the research was applied. Two experienced physicians reviewed data on outcome measures and assessed the quality rating. The main outcomes were CVE including myocardial infarction, coronary heart disease, percutaneous coronary intervention and coronary restenosis. 15 studies of SA and CVE were identified involving 65,077 subjects with a mean age of 57.89 +/- 6.05 years and a mean follow-up of 9.4 (+/- 5.56) years. Subjects under SA cut-off of 3.8 g/dL had a combined hazard ratio (HR) for CVE of 2.16 [95% confidence interval (CI) 1.93-2.45]. An increased risk for CVE was also evident using SA as a continuous variable (HR = 1.89, 95% CI 1.5-2.39). Females and males had a similar risk for CVE (HR 2.46, 95% CI 1.92-3.16, and HR 1.46, 95% CI 1.27-1.69, respectively). We found a similar risk of CVE between primary and secondary prevention studies (HR 1.79, 95% CI 1.5-2.17, and HR 2.47, 95% CI 2.24-2.75, respectively). Low SA levels are associated with an increased risk of CVE, not only in subjects free from CVE, but also in patients who already experienced a CVE.