Reduced‐dose direct oral anticoagulants (DOACs) in the extended treatment of venous thromboembolism: A systematic review and meta‐analysis—Response to Commentary

Abstract

Thank you to our colleagues Herold and Bauersachs for their commentary. The consideration for extended anticoagulation in patients with unprovoked deep vein thrombosis or pulmonary embolism is an important one and the goal of our review was to synthesize all available data in a systematic review and present the results as a meta‐analysis.1.Vasanthamohan L. Boonyawat K. Chai‐Adisaksopha C. Crowther M. Reduced‐dose direct oral anticoagulants in the extended treatment of venous thromboembolism: a systematic review and meta‐analysis.J Thromb Haemost. 2018; 16: 1288-1295Crossref PubMed Scopus (41) Google Scholar The commentary identifies an error in the data collected on the thrombotic events in the full‐dose and reduced‐dose anticoagulation cohort from the EINSTEIN CHOICE trial.2.Weitz J.I. Lensing A.W.A. Prins M.H. et al.Rivaroxaban or aspirin for extended treatment of venous thromboembolism.N Engl J Med. 2017; 376: 1211-1222Crossref PubMed Scopus (466) Google Scholar As mentioned by the authors, this error does lead to a change in the relative and absolute risk calculations for the comparisons of recurrent thrombotic events between groups. Specifically, in the comparison between the full‐dose and reduced‐dose groups, the relative risk changes to 0.85, with a 95% confidence interval (CI) of 0.51–1.42, and the absolute risk difference is two fewer events per 1000 (95% CI: −7 to +6 per 1000). As stated previously, there is no significant difference between the two groups with respect to recurrent thrombotic events. In the comparison between reduced‐dose anticoagulation and no anticoagulation, the relative risk changes to 0.22 (95% CI: 0.15–0.33) and absolute risk difference is 49 fewer events per 1000 (95% CI: −53 to −42 per 1000). Again, as aligned with the initial analysis, there remains a significantly reduced rate of recurrent thrombosis in the reduced‐dose anticoagulation group compared to the no anticoagulation group. The GRADE (Grading of Recommendations, Assessment, Development and Evaluations) summary of evidence table from our study was updated to reflect these changes and is shown below (Table 1). The relevant forest plots included in the article were also updated. Importantly, the underlying conclusions from the meta‐analysis are unchanged.TABLE 1GRADE summary of evidenceOutcomesParticipants (study types)Relative effect (95% CI)Absolute risk differenceQuality of evidence (GRADE)Reduced‐dose DOAC compared to full‐dose anticoagulationRecurrent VTEn = 3887 (2 RCTs)RR 0.85 (0.51–1.42)0.2% fewer (0.7% fewer to 0.6% more)ModerateMajor or clinically relevant nonmajor bleedingn = 3887 (2 RCTs)RR 0.74 (0.52–1.05)1.0% fewer (1.8% fewer to 0.2% more)ModerateReduced‐dose DOAC compared to no anticoagulationRecurrent VTEn = 3927 (2 RCTs)RR 0.22 (0.15–0.33)4.9% fewer (5.3% fewer to 4.2% fewer)HighMajor or clinically relevant nonmajor bleedingn = 3927 (2 RCTs)RR 1.19 (0.81–1.77)0.4% more (0.4% fewer to 1.8% more)ModerateAbbreviations: CI, confidence interval; DOAC, direct oral anticoagulant; GRADE, Grading of Recommendations, Assessment, Development and Evaluations; RCT, randomized controlled trials; RR, relative risk; VTE, venous thromboembolism. Open table in a new tab Abbreviations: CI, confidence interval; DOAC, direct oral anticoagulant; GRADE, Grading of Recommendations, Assessment, Development and Evaluations; RCT, randomized controlled trials; RR, relative risk; VTE, venous thromboembolism. We have submitted an erratum to our article to make the appropriate corrections. On behalf of the authors, we apologize for the errors and once again we thank the commenters for identifying this issue and helping to maintain accuracy in published literature. M. A. Crowther reports grants, personal fees, and participation in a drug safety monitoring board from Bayer, personal fees from Shionogi, personal fees from Alexion, grants from Leo pharma, personal fees from Pfizer, participation in a drug safety monitoring board for Daiichi, grants from Heart and Stroke Foundation, personal fees from Octapharma, personal fees from BMS Canada, and investments in Alynlam outside the scope of the submitted work. The other authors report no conflicts. LV reviewed the data, updated the figures and drafted the response. The other authors reviewed the manuscript prior to submission.