Impact of adherence to statins on cardiovascular adverse events in patients with coronary artery disease: a meta-analysis

Abstract

To evaluate the impact of different adherence mode to statins on cardiovascular adverse events in patients with coronary artery disease (CAD). Electronic searches, including PubMed, Scopus, Ovid MEDLINE, Ovid EMBASE, Ovid EBM Reviews CENTRAL, CINAHL, The Cochrane Library, Ovid PsycInfo, Wanfang data, CNKI and Science & Technology Magazine Online, were performed and all related literatures of all languages were retrieval till to March 1, 2015. The full text was obtained through manual retrieval, inter-library loan and document delivery service, or by contacting the author directly. According to inclusion and exclusion criteria, data was extracted dependently by two raters. The high adherence to statin was use defined by the ratio of statins cover time and the total time (proportion of days covered, PDC≥80%). Data were analyzed quantitatively using RevMan 5.1. Then implement subgroup analysis was made according to different statin adherence and classification of clinical outcomes. The impact of adherence to statin on cardiovascular events (all-cause mortality, non-fatal myocardial infarction, hospitalization due to unstable angina pectoris, heart insufficiency attack) in CAD patients was evaluated. Present analysis enrolled eight relevant retrospective and observational studies. Because there were only few literatures describing the impact of statin adherence on clinical outcomes, we also included literatures with low adherence group (4 studies in PDC<80%, 2 studies in PDC<40% and 2 studies in PDC<20%). High adherence group includes 189 556 cases; low adherence group includes 11 384 cases. Compare with low adherence group, cardiovascular events rate reduced by 32% in high adherence group (OR=0.68, 95%CI 0.58-0.80, P<0.001). Subgroup analysis with 4 literatures with PDC≥80% or<80% showed that the cardiovascular events prominently decreased in high adherence group compared to low adherence group (OR=0.63, 95%CI 0.53-0.76, P<0.001). According to 5 literatures with all-cause mortality parameter, we also found a borderline decrease in all-cause mortality in high adherence group compared to low adherence group ( OR=0.67, 95%CI 0.44-1.02, P=0.06), while non-fatal cardiovascular events were significantly reduced by 18% in high adherence group (OR=0.82, 95%CI 0.77-0.87, P<0.001). High adherence to statins is related to significantly lower cardiovascular events in CAD patients.