Efficacy and safety of dabigatran etexilate for the prevention of venous thromboembolism following total hip or knee arthroplasty - A meta-analysis
Sorrel E. Wolowacz 1; Neil S. Roskell 1; Jonathan M. Plumb 2; Joseph A. Caprini 3; Bengt I. Eriksson4
1 RTI Health Solutions, Williams House, Manchester University Science Park, Lloyd Street North, Manchester, UK; 2 Boehringer Ingelheim GmbH, Binger Strasse, Ingelheim am Rhein, Germany; 3 Robert R. McCormick School of Engineering and Applied Sciences, Department of Surgery, Evanston Northwestern Healthcare, Evanston IL and Northwestern University Feinberg School of Medicine, Chicago, Illinois, USA; 4 Professor of Orthopaedics, Surgical Sciences Gothenburg University, Sahlgrenska University Hospital/ÖSTRA, Gothenburg, Sweden
Meta-analysis, thrombosis, anticoagulants, prophylaxis, Embolism
Dabigatran etexilate has been investigated in three phase III trials for the prevention of venous thromboembolism (VTE). Health technology assessment agencies increasingly require meta-analyses of all relevant evidence for an intervention, if appropriate. The objective of this study was to perform a metaanalysis of efficacy and safety data for the recommended dose of dabigatran etexilate, 220 mg once daily (od), for VTE prophylaxis after total knee arthroplasty (TKA) and total hip arthroplasty (THA), and discuss the appropriateness of combining the data. Risk ratios (RR) for VTE and bleed end-points were estimated using fixed and random effects meta-analysis. Analyses were performed combining RE-MODEL and RE-NOVATE, which compared dabigatran etexilate with enoxaparin 40 mg od after TKA and THA, respectively, and also including RE-MOBILIZE, which compared dabigatran etexilate with enoxaparin 30 mg twice daily after TKA. Tests for statistical heterogeneity were performed using the Chi-squared statistic. No significant differences were detected between dabigatran etexilate and enoxaparin in any of the end-points analysed, either in the two trial analysis (all p>0.15), or when all three trials were combined ( all p>0.30). RRs (random effects) for the composite end-point total VTE and all-cause mortality were 0.95 [95% confidence intervals 0.82 – 1.10] and 1.05 [0.87 – 1.26] in the two and three trial analyses, respectively. Meta-analysis of RE-MODEL and RE-NOVATE supported the conclusions of the individual trials that dabigatran etexilate is non-inferior to enoxaparin 40 mg od, with a similar safety profile. Meta-analysis of all three trials found no significant differences between treatments in any of the end-points analysed. Heterogeneity between the trials cannot be ruled out.