Inappropriate inclusion of multiple studies in a meta-analysis of randomized controlled trials of atorvastatin loading prior to percutaneous coronary intervention for acute coronary syndrome