Thrombosis Research 134 (2014) 627–632
Contents lists available at ScienceDirect
Thrombosis Research journal homepage: www.elsevier.com/locate/thromres
Regular Article
Case fatality of bleeding and recurrent venous thromboembolism during, initial therapy with direct oral anticoagulants: A systematic review Cynthia Wu a,⁎, Ghazi S. Alotaibi a,b, Khalid Alsaleh b, M. Sean McMurtry a a b
Department of Medicine, University of Alberta, Edmonton, Alberta, Canada Department of Medicine, King Saud University, Riyadh, Saudi Arabia
a r t i c l e
i n f o
Article history: Received 22 April 2014 Received in revised form 30 June 2014 Accepted 1 July 2014 Available online 10 July 2014 Keywords: Direct oral anticoagulants Case fatality VTE recurrence Major bleeding Vitamin K antagonists Mortality
a b s t r a c t Introduction: The frequency and case fatality of venous thromboembolism (VTE) and major bleeding during the initial 3 months of therapy in those treated for symptomatic VTE with either direct oral anticoagulants (DOACs) or vitamin K antagonists (VKA) are important clinically relevant outcomes. We sought to measure it during the initial months of anticoagulation for symptomatic VTE. Material and Methods: We searched MEDLINE, EMBASE, and CENTRAL to identify studies that enrolled patients with acute symptomatic VTE treated with DOACs or VKA and reported data on bleeding, VTE recurrence and death. Studies were evaluated according to a priori inclusion criteria and critically appraised using established internal validity criteria. Single-proportion random-effects models were used to pool estimates. Results: Of the 2453 citations retrieved, 5 RCTs that enrolled 24,507 patients were included. The rate of major bleeding was 1.8 (95% CI: 1.3-2.5) and 3.1 (95% CI: 2.4-3.9) per 100 patient-years in DOAC and VKA arms, respectively. The rate of VTE recurrence was 3.7 (95% CI: 2.7-4.7) and 4.1 (95% CI: 3.0-5.4) per 100 patient-years of DOAC and VKA, respectively. The case fatality rate of bleeding was significantly higher in the VKA arms 10.4% (95% CI: 6.6-15.4) compared to DOACs 6.1% (95% CI: 2.7-11.7; p value for difference = 0.029) with no statistical difference between the case fatalities for recurrent VTE. The rate of death from either definite major bleeding or definite recurrent VTE was 0.27 (95% CI: 0.16-0.40) and 0.46 (95% CI: 0.32-0.63) per 100 patient-years for DOACs and VKAs respectively, resulting in a number needed to treat of 875 for DOACs to prevent one death. Conclusion: DOACs are attractive alternatives to VKAs for initial treatment of symptomatic VTE, with lower frequency and case fatality for major bleeding. However, the incremental safety benefit of DOACs over VKAs is small, with large numbers needed to treat. © 2014 Elsevier Ltd. All rights reserved.
Introduction Venous thromboembolism (VTE), including deep vein thrombosis (DVT) and pulmonary embolism (PE), is associated with a high rate of morbidity and mortality [1,2]. For decades, vitamin K antagonists (VKAs) have been the mainstay of VTE treatment and prevention [3]. Without adequate anticoagulant therapy, the risk of recurrent VTE is estimated to be 50% in the first 3 months following a new diagnosis [4,5]. In the absence of active bleeding, the risk of major bleeding is usually much smaller in this time period [6]. After 3 months, the risk of recurrent VTE drops off sharply and is estimated to be 3-5% per year ⁎ Corresponding author at: University of Alberta, 4-112 Clinical Sciences Building, 11350-83rd Avenue, Edmonton, AB, T6G 2G3. Tel.: +1 780 407 1584; fax: +1 780 407 2680. E-mail address: [email protected] (C. Wu).
http://dx.doi.org/10.1016/j.thromres.2014.07.001 0049-3848/© 2014 Elsevier Ltd. All rights reserved.
for events provoked by transient risk factors and 10% per year for idiopathic events [7,8]. Several direct oral anticoagulants (DOACs), including dabigatran, rivaroxaban, apixaban and edoxaban have emerged as alternatives to VKAs. Clinical trials have shown each of these agents to be at least as effective and safe as VKAs for the acute treatment of VTE and the secondary prevention of recurrent VTE [9–15]. The benefit of treating VTE with anticoagulant therapy must always be weighed against the risk of bleeding complications. The variable clinical course of recurrent VTE and bleeding complications make balancing the risks and benefits of anticoagulation difficult for clinicians and patients. Nevertheless, mortality is a serious and objective adverse outcome associated with both recurrent VTE and bleeding complications, and therefore a comparison of case fatality associated with recurrent VTE and with major bleeding may help judge the true benefit of anticoagulation. The purpose of this study was to look at patients with acute symptomatic
628
C. Wu et al. / Thrombosis Research 134 (2014) 627–632
VTE who were randomized to receiving DOACs compared to VKAs and estimate the rates and case-fatality of recurrent VTE and major bleeding during the initial months of therapy.
Material and Methods Studies Identification We conducted a systematic literature search to identify all randomized controlled trials (RCTs) that included adult patients (≥ 18 years old) treated with DOACs for acute symptomatic VTE treatment (i.e. not less than 3 months) based on a priori protocol available on request. Patients treated for other indications, including joint replacements, atrial fibrillation or acute coronary syndrome were excluded in order to minimize study heterogeneity when comparing safety and efficacy endpoints. The review was reported according to the PRISMA statement [16,17].
Literature Search We searched MEDLINE, Cochrane Central Register of Controlled Trials and EMBASE databases from inception through December 2013 to identify all published and unpublished literature, in any language, related to our topic. The search strategy included the MeSH terms venous thrombosis and pulmonary embolism. The following terms were also added to the search strategy: recurrent venous thromboembolism, new oral anticoagulants, dabigatran, apixaban, rivaroxaban, edoxaban, warfarin and acenocoumarol. We also screened major Hematology international conferences (American Society of Hematology, International Society of Thrombosis and Hemostasis, and European Hematology Association) for abstracts from their annual meetings from 2010-2013, registries of health technology assessments and clinical trials, and the reference list of retrieved studies for additional studies and unpublished data.
Study Selection We included studies that: I) were randomized controlled trials comparing therapeutic doses of dabigatran, apixaban, edoxaban or rivaroxaban to standard (heparin/VKAs) (target INR 2-3) for treatment of symptomatic VTE for at least 3 months; II) included study population with deep venous thromboembolism, pulmonary embolism or both; III) included a qualifying event (DVT or PE) that was confirmed objectively by the presence of intraluminal defect on venography or venous noncompressibility on duplex ultrasound for DVT [18], and the presence of intraluminal filling defect on pulmonary angiography or a high probability ventilation-perfusion scan for PE [19]; and IV) reported at least one of the following outcomes on anticoagulation: 1) recurrent events defined as symptomatic thromboembolic event that occurred after initiation of VTE treatment and diagnosed objectively by a new intraluminal filling defect on venography or a new non-compressible vein segment on duplex ultrasound for recurrent DVT or by a new intraluminal filling defect on pulmonary angiography or a new high probability ventilation-perfusion lung scan, 2) major bleeding events as defined by ISTH [20] or as per study definition, and 3) mortality caused by bleeding or venous thromboembolic event. Patients receiving additional treatment (e.g., thrombolysis, inferior vena filters) were excluded. Eligible studies were reviewed independently by two reviewers (CW, GA) to assess suitability for inclusion, and all included trials were reviewed to assess agreement in outcome event reporting. Studies considered relevant by one or both reviewers were retrieved and disagreements were resolved by discussion.
Outcome Measures and Data Extraction Our primary outcome measures were: 1) frequency of major bleeding; 2) frequency of fatal bleeding; 3) frequency of VTE recurrence; and 4) frequency of fatal VTE recurrence (definite and possible). Two reviewers independently abstracted the data describing baseline characteristics, treatment interventions and outcomes. Discrepancies were solved by discussion. Results of intention-to-treat analyses were collected if reported. Because the cause of death in cases with suspected VTE related death is not always objectively confirmed, we reported VTE deaths as definite or possible. Definite fatal recurrent VTE was defined as any VTE diagnosed postmortem, or a new intraluminal-filling defect detected on computed tomography, venography or ventilation perfusion scan, or a high clinical probability of fatal pulmonary embolism as adjudicated by the study investigators immediately before death. Possible VTE deaths included data reported as, “PE cannot be ruled out.” Quality Assessment Two reviewers independently assessed each study’s risk of bias using the 6 domains of the Cochrane Collaboration’s tool for assessing risk for in randomized trials [21] and disagreements were resolved by discussion. Data Synthesis and Analysis The results of individual studies were combined to determine the following outcomes: (1) the rate of major bleeding; (2) the rate of fatal bleeding; (3) the rate of VTE recurrence; (4) the rate of fatal VTE recurrence (definite and possible). We also calculated the case fatality rate of major bleeding, and recurrent VTE (both definite and possible). The case fatality rate of recurrent VTE was defined as the proportion of all recurrent VTE events (fatal and nonfatal) resulting in death. Fatal bleeding was defined as a major bleeding event directly leading to death or death was reported as “associated with bleeding”. The case fatality rate of a major bleeding event was defined as the proportion of all major bleeding events causing or associated with death [22]. All rates were expressed as events per 100 patient-years of anticoagulation to standardize for different follow-up durations across studies [23,24]. Proportions were transformed via the Freeman-Tukey double arcsine method [25,26] before pooling the case fatality rates. We then performed a DerSimonian–Laird random-effects model to pool the transformed rates [27]. After pooling the resulting estimates and their 95% CIs, limits were back-transformed to rates per 100 patient-years of follow-up. The pooled random effect incidence rate ratio (IRR) [28] was used to estimate of the effects of treatment with DOACs. The I2 statistic was used to estimate total variation among the pooled estimates across studies. An I2 value less than 25% was considered low-level heterogeneity, 25% to 50% was moderate-level, and greater than 50% was high-level [29]. Statistical analysis was done using StatsDirect statistical software (version 2.8.0). Results Study Selection and Patient Characteristics The literature search yielded 2453 studies (Fig. 1). Of these, 5 RCTs [9, 11,13–15] met the inclusion criteria and enrolled 12258 patients for acute VTE treatment in DOAC arms and 12249 patients for VKA arms. Trials evaluated three factor Xa inhibitors (rivaroxaban, n = 4150), (apixaban, n = 2691), (edoxaban, n = 4143) and one direct thrombin inhibitor (dabigatran, n = 1274). The majority of patients (N 80%) enrolled in these studies were treated for idiopathic (unprovoked) VTE events and were anticoagulated for a median time of 6.7 months (Table 1). The time from symptoms onset to randomization ranged
C. Wu et al. / Thrombosis Research 134 (2014) 627–632
4366 Potentially relevant references from databases screened. (Medline 1775, EMBASE 2287, CENTRAL 304)
629
Additional records identified through other sources (n =2)
Records after duplicates removed (n =2453)
Records screened (n =2453)
Records excluded (n =2426)
Full-text articles excluded (n =22): -Narrative reviews (n= 4) -Withdrawn drug (n=2) -Multiple publication (n = 1) -Non clinical outcomes (n = 3) -Studies for extended VTE prevention (n=9) - Dose ranging studies (n=3)
Articles retrieved and evaluated in full for inclusion (n =27)
Studies included in result synthesis (n =5)
Fig. 1. Search strategy and study selection as per PRISMA checklist.
from 2 to 11 days, during which all patients were treated with parenteral anticoagulation (e.g., LMWH, unfractionated heparin, or fondaparinux). Quality Assessment and Risk of Bias In all of the 5 included studies, the risk of bias was low. Open label design was implemented in three studies RCTs [11,13,14] but all outcome data were recorded with a routine structured technique and were adjudicated by an independent committee blinded to study drug allocation. Loss to follow-up rate was less than 5% in all included studies. Although all studies were funded by pharmaceutical industry, outcomes were assessed by independent steering committees. See summary of risk of bias in (Supplementary Figs. 1 and 2).
Synthesis of Results Outcomes for Acute VTE Management Overall analysis included 12258 patients who received 7206 patient-years of new oral anticoagulant therapy and 12249 patients who received 7134 patient-years of LMWH/VKA therapy. Table 2 summarizes the frequency of each outcome as reported by studies. Table 3 summarizes the pooled results by treatment type. Table 4 summarizes the comparison between the rates of bleeding and VTE recurrence. Major Bleeding. Major bleeding with DOACs occurred at a rate of 1.8 per 100 patient-years (95% CI: 1.3-2.5, I 2 = 67.6%) and fatal bleeding occurred at a rate of 0.14 per 100 patient years (95% CI: 0.07-0.24,
Table 1 Characteristics of studies enrolled patients for active treatment of acute VTE events. Study
Design
Treatments regimen
Duration, months
Patients
Efficacy outcome
Safety outcome
RE-COVER I, 2009
Randomized, double-blind
6
2539 patients with acute VTE
Recurrent VTE 2.4% dabigatran, 2.1% warfarin
Major bleeding: 1.6% dabigatran, 1.9% warfarin
EINSTEIN-PE, 2012
Randomized, open-label
3, 6 or 12
4832 patients with acute PE
Recurrent VTE 2.1% rivaroxaban, 1.8% enoxaparin/VKAs
Major bleeding 1.1% rivaroxaban, 2.2% enoxaparin/VKAs
EINSTEIN-DVT, 2010
Randomized, open-label
3, 6 or 12
3449 patients with acute DVT
Recurrent VTE 2.1% rivaroxaban, 3.0% enoxaparin/warfarin
Major bleeding 0.8% rivaroxaban, 1.2% VKAs
AMPLIFY, 2013
Randomized, double blind double-dummy
6
5395 Patients with acute DVT or PE
Hokusai-VTE, 2013
Randomized, double blind double-dummy
Dabigatran (150 mg BID) Warfarin Rivaroxaban (15 mg BID for 3 weeks, then 20 mg OD) VKAs Rivaroxaban (15 mg BID for 3 weeks, then 20 mg OD) VKAs Apixaban (10 mg BID for 7 days, then 5 mg BID) Warfarin Edoxaban (60 mg OD) Warfarin
12
8292 patients with acute DVT and/or PE
Recurrent VTE 2.3% apixaban, 2.7% enoxaparin/VKAs Recurrent VTE 3.2% edoxaban, 3.5% enoxaparin/VKA
Major bleeding 0.6% apixaban, 1.8% enoxaparin/VKAs Major bleeding 1.4% edoxaban, 1.6% enoxaparin/VKA
630
C. Wu et al. / Thrombosis Research 134 (2014) 627–632
Table 2 Outcome events per treatment in patients presenting with acute symptomatic VTE in included studies. Study, Year (Reference)
RE-COVER1 EINSTEIN-PE EINSTEIN-DVT AMPLIFY Hokusai-VTE
Study arm
Dabigatran VKA Rivaroxaban VKA Rivaroxaban VKA Apixaban VKA Edoxaban VKA
N
1274 1265 2419 2413 1731 1718 2691 2704 4143 4149
Mean on treatment duration (months)
6 6 3,6 or 12 3,6 or 12 3,6 or 12 3,6 or 12 6 6 3 to 12 3 to 12
MB
20 24 26 52 14 20 15 49 56 66
VTE recurrence
30 27 50 44 36 51 48 71 66 80
MB fatality rate
1 1 2 3 1 5 2 3 2 10
Recurrence fatality
Adjudication of Outcomes
P&D
D
1 3 11 7 4 6 12 16 26 26
1 3 2 1 1 0 1 2 4 3
Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
DOACs: new oral anticoagulants, MB: Major bleeding, CRNMB: clinically relevant non major bleeding, P&D: Possible and definite VTE recurrence, D: only definite VTE recurrent cases.
I2 = 0%), compared to 3.1 per 100 patient-years (95% CI 2.4-3.9, I 2 = 70.6%) and 0.33 per 100 patient-years (95% CI 0.21-0.48, I2 = 0%) for the VKA treatment. Out of 131 major bleeding episodes in patients receiving DOACs, 8 were fatal and this resulted in a case fatality rate for major bleeding of 6.1% (95% CI: 2.7-11.7). In patients treated with VKA, there were 211 incidents of major bleeding among which 22 resulted in death; the cases fatality for major bleeding in this group was 10.4% (95% CI: 6.6-15.4). Supplementary Figs. 3 and 4 shows the Forest plot of major bleeding proportion among patients randomized to DOACs and VKA, respectively, in included studies. VTE Recurrence. The VTE recurrence rate was 3.7 per 100 patient-years of DOACs therapy (95% CI: 2.7-4.7, I2 = 79.8%), definite fatal recurrent VTE occurred at a rate of 0.16 per 100 patient-years (95% CI: 0.08-0.26, I2 = 0%), and possible fatal VTE occurred at a rate of 0.72 per 100 patientsyears (95% CI: 0.48-1.00, I2 = 38%). In the VKA arms, the rate of VTE recurrence was 4.1 per 100 patient-years (95% CI: 3.0-5.4, I2 = 82.7%), the rate of definite fatal recurrent VTE cases was 0.14 per 100 patientyears (95% CI: 0.05-0.27, I2 = 32.2%), and the rate of possible fatal VTE recurrence was 0.82 per 100 patient-years (95% CI: 0.59-1.1, I2 = 24%). Of the 241 recurrent VTE in DOAC-treated patients, the case fatality rates in patients on DOACs were 3.7% (95% CI: 1.7-7.0) and 22.4% (95% CI: 17.3-28.2) for definite and possibly fatal events, respectively. For patients on VKAs, of the 273 cases of VTE recurrence, 9 of which lead to definite VTE related death and 58 were possibly fatal, the case fatality rate was 3.3% (95% CI: 1.5-6.2), and 21.2% (95% CI: 16.5-26.5), respectively. Supplementary Figs. 5 and 6 shows the Forest plot of
recurrent VTE proportion among patients randomized to DOACs and VKA, respectively, in included studies. Combined Major Bleeding and VTE Recurrence. The rates of composite fatality (death due to either major bleeding or definite recurrent VTE) were 0.27 per 100 patient years (95% CI 0.16-0.40, I2 = 0%) for DOACs and 0.46 per 100 patient years (95% CI 0.32-0.63, I2 = 0%) for VKAs. The rates of composite fatality (death due to either major bleeding or possible recurrent VTE) were 0.89 per 100 patient years (95% CI 0.65-1.1, I2 = 13.2%) for DOACs and 1.1 per 100 patient years (95% CI 0.86-1.4, I2 = 20.2%) for VKAs. Compared with VKAs, treatment with DOACs had lower incidence of major bleeding (IRR 0.59, 95% CI 0.420.87), lower incidence of fatal bleeding (IRR 0.39, 95% CI 0.17-0.9), and lower incidence of the composite fatality related to either major bleeding or definite recurrent VTE (IRR 0.55, 95% CI 0.30-0.99). There were no differences between treatment arms for recurrent VTE (IRR 0.87, 95% CI 0.73-1.0), recurrent fatal PE (IRR 0.91, 95% CI 0.33-2.5), or the composite of fatality related to either major bleeding or possible recurrent VTE (IRR 0.77, 95% CI 0.55-1.08). Discussion Our findings reveal that patients on DOACs had lower rates of major and fatal bleeding events (1.8 per 100 patient-years (95% CI 1.3-2.5) and 0.14 per 100 patient-years (95% CI 0.07-0.24), respectively, compared to 3.1 per 100 patient-years (95% CI 2.4-3.2), and 0.33 per 100 patientyears (95% CI 0.21-0.48), in patients treated with VKA therapy. This
Table 3 Outcomes according to treatment type. Outcomes
Acute DOACs
Acute VKA
Rate of MB, per 100 patient-years (95% CI) Rate of fatal bleeding, per 100 patient-years (95% CI) Bleeding CFR (95% CI) Rate of VTE recurrence, per 100 patient-years (95% CI) Rate of fatal VTE, per 100 patient-years (95% CI)
1.8 (1.3-2.5)
3.1 (2.4-3.9)
0.14 (0.07-0.24)
0.33 (0.21-0.48)
6.1% (2.7-11.7)
10.4% (6.6-15.4)
3.7 (2.7-4.7)
4.1 (3.0-5.4)
0.16 (0.08-0.26)
0.14 (0.05-0.27)
Definite Possible Recurrent VTE CFR (95% CI)
0.72 (0.48-1.00)
0.82 (0.59-1.1)
3.7% (1.7-7.0)
3.3% (1.5%-6.2%)
22.4% (17.3-28.2)
21.2% (16.5-26.5)
0.27 (0.16-0.40) 0.87 (0.65-1.1)
0.46 (0.32-0.63) 1.1 (0.86-1.42)
Definite Possible Rate of definite fatal events, per 100 patient-years (95% CI) Rate of definite and possible fatal events, per 100 patient-years (95% CI) DOACs: direct oral anticoagulants; VKA: vitamin K antagonists; MB: major bleeding; CFR: case fatality rate.
C. Wu et al. / Thrombosis Research 134 (2014) 627–632
631
Table 4 Comparison between the rate of bleeding and VTE recurrence in patients on DOACs vs VKA for acute VTE treatment. Incidence rate ratio
Major bleeding
DOACs vs. VKA 0.59 (0.42-0. 87, P = 0.007) (95% CI, P value)
Recurrent VTE
Fatal Bleeding
Fatal recurrent VTE (definite)
Fatal recurrent VTE (possible and definite)
Combined fatal bleeding and definite fatal VTE recurrence
Combined fatal bleeding and (definite and possible) fatal VTE recurrence
0.87 (0.73-1.0, P = 0.13)
0.39 (0.17-0.9, P = 0.029)
0.91 (0.33-2.5, P = 0.86)
0.93 (0.64-1.4, P = 0.70)
0.55 (0.30-0.99, P = 0.048)
0.77 (0.55-1.08, P = 0.13)
IRR: incidence rate ratio, DOACs: direct oral anticoagulants; VKA: vitamin K antagonists.
difference between DOAC and VKA therapy reached statistical significance in major and fatal bleeding and the use of DOACs was associated with a 41% and 61% relative reduction in the rate of major and fatal bleeding, respectively. The estimated case fatality rate for DOAC related major bleeding during acute VTE treatment was 6.1% (1 in 16 major bleeding events) compared to 10.4% bleeding case fatality in warfarin arms in the same studies and 11.3% from a previous study [7] (1 in 10 major bleeding events) during the acute management of VTE. These results illustrate a reduction in bleeding rates and associated fatality in patients on DOACs compared to warfarin for the acute treatment of VTE. It is important to note that DOAC therapy does not eliminate the incidence of major and fatal bleeding and that the overall rates of both in either treatment arms were small and the numbers needed to treat to prevent one case of major bleeding and fatal bleeding are 153 and 875, respectively. In contrast to major bleeding, the rates of VTE recurrence during the initial phase of acute VTE treatment were similar between the DOAC and VKA groups. 1 in 51 patients treated with DOACs had a recurrent VTE event as opposed to 1 in 45 of those treated with VKA, and the difference between these rates did not reach statistical significance. The rate of definite fatal VTE recurrence in both the DOAC and VKA arms was infrequent, 0.16 per 100 patient-years and 0.14 per 100 patientyears, leading to imprecise pooled estimates with large 95% confidence intervals. The case fatality rate of definite recurrent VTE case fatality for patients on DOACs was 3.7% compared to 3.3% in VKA arms in the same studies. These data suggest that the efficacy of all these medications for initial VTE therapy was comparable. Compared to previous data documenting the risk of recurrent VTE off anticoagulation in the acute phase of VTE therapy [4,5], this also confirms the known substantial decrease in recurrent VTE with appropriate anticoagulant use and reinforces the need for solid anticoagulation in this critical time frame. Importantly, we found that DOACs were associated with a lower risk for fatality definitely related to either a major bleed or a recurrent VTE, with IRR 0.55 (95% CI 0.30-0.99); initial treatment of acute VTE with DOACs may therefore be associated with lower risk for death than treatment with VKAs as well as lower risk for major bleeding. Though the numbers of events are small, and are drawn from selected patient populations recruited to randomized trials, the enhanced safety profile, non-inferior effectiveness of DOACs, and apparently lower risk for death definitely related to major bleeding or recurrent VTE may support a preference for a DOAC in the acute treatment of VTE in selected patients. However, the number needed to treat with DOACs to prevent 1 death from either major bleeding or VTE recurrence is high at 875. This study has limitations. First, some cases of fatal recurrent VTE were not objectively defined, therefore, we relied on “definite” and “possible” definitions as judged by the study investigators. Second, we note that for each DOAC there was only 1 phase III RCT for each studied indication, so results were not duplicated/confirmed and grouping all the DOACs together may not necessarily reflect each individual drug. However, we assumed the presence of a class effect on the main outcomes (e.g., stroke prevention in atrial fibrillation, major bleeding and recurrent VTE) as shown by several network meta-analyses [30,31].
Third, the pooled estimates of major bleeding and recurrent VTE had moderate to high heterogeneity. The most likely explanation for the heterogeneity is likely due to the variability in the follow-up periods. Pooled studies participated with different follow-up periods, which suggest that all studies were not estimating the same quantity. Fourth, publication bias could not be assessed using funnel plots due to the small number of included trials. To our knowledge, this is the first study to assess the rates and case fatality of major bleeding and recurrent VTE on DOACs and compare it to standard therapy in the active phase of VTE treatment. All reported events were independently adjudicated with a minimal loss to followup rate in all included studies. All included studies adopted the ISTH definition to quantify major bleeding and this allows for comparison between trials. One of the major potential limitations was the unavailability of any real world cohort data to give an estimate of case fatality outside clinical trials use of DOACs. Conclusion Owing to their comparable efficacy but lower risk of major bleeding, DOACs offer an attractive alternative to VKAs for acute treatment of symptomatic VTE. However, given the given the low incidence of adverse events and the high NNT, the absolute incremental benefit is limited. Funding This study was self-funded. Conflict of Interest Statement M.S. McMurtry and Wu C, are investigators for eTRIS, a randomized, open label, and multicenter study evaluating the efficacy and safety of edoxaban monotherapy versus low molecular weight heparin and warfarin in subjects with symptomatic deep vein thrombosis sponsored by Daiichi Sankyo, and an investigator for XALIA, a multicenter observational study of rivaroxaban as initial anticoagulation for venous thromboembolism. Dr Wu has served on the advisory board for LeoPharma. Acknowledgements M.S. McMurtry is supported by the Heart and Stroke Foundation of Canada. Alotaibi GS would like to thank the King Saud University for funding his graduate studies. Appendix A. Supplementary Data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.thromres.2014.07.001.
632
C. Wu et al. / Thrombosis Research 134 (2014) 627–632
References [1] Heit JA. The epidemiology of venous thromboembolism in the community. Arterioscler Thromb Vasc Biol 2008;28:370–2. [2] Kearon C. Epidemiology of venous thromboembolism. Semin Vasc Med 2001;1:7–26. [3] Kearon C, Kahn SR, Agnelli G, Goldhaber S, Raskob GE, Comerota AJ, et al. Antithrombotic therapy for venous thromboembolic disease: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition). Chest 2008;133:454S–545S. [4] Barritt DW, Jordan SC. Anticoagulant drugs in the treatment of pulmonary embolism. A controlled trial. Lancet 1960;1:1309–12. [5] Hull R, Delmore T, Genton E, Hirsh J, Gent M, Sackett D, et al. Warfarin sodium versus low-dose heparin in the long-term treatment of venous thrombosis. N Engl J Med 1979;301:855–8. [6] Hass B, Pooley J, Harrington AE, Clemens A, Feuring M. Treatment of venous thromboembolism - effects of different therapeutic strategies on bleeding and recurrence rates and considerations for future anticoagulant management. Thromb J 2012;10:24. [7] Carrier M, Le Gal G, Wells PS, Rodger MA. Systematic review: case-fatality rates of recurrent venous thromboembolism and major bleeding events among patients treated for venous thromboembolism. Ann Intern Med 2010;152:578–89. [8] Kearon C. Long-term management of patients after venous thromboembolism. Circulation 2004;110:I10–8. [9] Schulman S, Kearon C, Kakkar AK, Mismetti P, Schellong S, Eriksson H, et al. Dabigatran versus warfarin in the treatment of acute venous thromboembolism. N Engl J Med 2009;361:2342–52. [10] Schulman S, Kearon C, Kakkar AK, Schellong S, Eriksson H, Baanstra D, et al. Extended use of dabigatran, warfarin, or placebo in venous thromboembolism. N Engl J Med 2013;368:709–18. [11] Agnelli G, Buller HR, Cohen A, Curto M, Gallus AS, Johnson M, et al. Oral apixaban for the treatment of acute venous thromboembolism. N Engl J Med 2013;369:799–808. [12] Agnelli G, Buller HR, Cohen A, Curto M, Gallus AS, Johnson M, et al. Apixaban for extended treatment of venous thromboembolism. N Engl J Med 2013;368:699–708. [13] Einstein I, Bauersachs R, Berkowitz SD, Brenner B, Buller HR, Decousus H, et al. Oral rivaroxaban for symptomatic venous thromboembolism. N Engl J Med 2010;363:2499–510. [14] Einstein–Pe I, Büller HR, Prins MH, Lensin AW, Decousus H, Jacobson BF, et al. Oral rivaroxaban for the treatment of symptomatic pulmonary embolism. N Engl J Med 2012;366:1287–97. [15] Hokusai VTEI, Buller HR, Decousus H, Grosso MA, Mercuri M, Middeldorp S, et al. Edoxaban versus warfarin for the treatment of symptomatic venous thromboembolism. N Engl J Med 2013;369:1406–15.
[16] Canadian Agency for D, Technologies in H. Requirements for health technology assessment template for reports. Ottawa (ON): The Agency; 2008. [17] Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtzsche PC, Ioannidis JP, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: explanation and elaboration. BMJ 2009;339:10. [18] Bates SM, Jaeschke R, Stevens SM, Goodacre S, Wells PS, Stevenson MD, et al. Diagnosis of DVT: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012;141:e351S–418S. [19] Kearon C. Diagnosis of pulmonary embolism. CMAJ 2003;168:183–94. [20] Schulman S, Kearon C. Subcommittee on Control of Anticoagulation of the S, Standardization Committee of the International Society on T, Haemostasis. Definition of major bleeding in clinical investigations of antihemostatic medicinal products in non-surgical patients. J Thromb Haemost 2005;3:692–4. [21] Higgins JPT, Altman DG, Gøtzsche PC, Jüni P, Moher D, Oxman AD, et al. The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. BMJ 2011;343. [22] Douketis JD, Kearon C, Bates S, Duku EK, Ginsberg JS. Risk of fatal pulmonary embolism in patients with treated venous thromboembolism. JAMA 1998;279:458–62. [23] Koepsell TD. WNEmstooiNY. NY: Oxford University Press; 2003. [24] Kirkwood BR, Sterne JAC. Essential Medical Statistics. 2nd ed. Blackwell; 2003. [25] Miller J. The inverse of the Freeman-Turkey double arcsine transformation. Am Stat 1978;32:138. [26] Landis JR, Cooper MM, Kennedy T, Koch GG. A computer program for testing average partial association in three-way contingency tables (PARCAT). Comput Programs Biomed 1979;9:223–46. [27] DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials 1986;7:177–88. [28] Guevara JP, Berlin JA, Wolf FM. Meta-analytic methods for pooling rates when follow-up duration varies: a case study. BMC Med Res Methodol 2004;4:17. [29] Higgins JP, Thompson SG. Quantifying heterogeneity in a meta-analysis. Stat Med 2002;21:1539–58. [30] Castellucci LA, Cameron C, Le Gal G, Rodger MA, Coyle D, Wells PS, et al. Efficacy and safety outcomes of oral anticoagulants and antiplatelet drugs in the secondary prevention of venous thromboembolism: systematic review and network metaanalysis. BMJ 2013;347:f5133. [31] Weitz JI, Gross PL. New oral anticoagulants: which one should my patient use? Hematology/the Education Program of the American Society of Hematology American Society of Hematology Education Program, 2012; 2012. p. 536–40.
Contents lists available at ScienceDirect
Thrombosis Research journal homepage: www.elsevier.com/locate/thromres
Regular Article
Case fatality of bleeding and recurrent venous thromboembolism during, initial therapy with direct oral anticoagulants: A systematic review Cynthia Wu a,⁎, Ghazi S. Alotaibi a,b, Khalid Alsaleh b, M. Sean McMurtry a a b
Department of Medicine, University of Alberta, Edmonton, Alberta, Canada Department of Medicine, King Saud University, Riyadh, Saudi Arabia
a r t i c l e
i n f o
Article history: Received 22 April 2014 Received in revised form 30 June 2014 Accepted 1 July 2014 Available online 10 July 2014 Keywords: Direct oral anticoagulants Case fatality VTE recurrence Major bleeding Vitamin K antagonists Mortality
a b s t r a c t Introduction: The frequency and case fatality of venous thromboembolism (VTE) and major bleeding during the initial 3 months of therapy in those treated for symptomatic VTE with either direct oral anticoagulants (DOACs) or vitamin K antagonists (VKA) are important clinically relevant outcomes. We sought to measure it during the initial months of anticoagulation for symptomatic VTE. Material and Methods: We searched MEDLINE, EMBASE, and CENTRAL to identify studies that enrolled patients with acute symptomatic VTE treated with DOACs or VKA and reported data on bleeding, VTE recurrence and death. Studies were evaluated according to a priori inclusion criteria and critically appraised using established internal validity criteria. Single-proportion random-effects models were used to pool estimates. Results: Of the 2453 citations retrieved, 5 RCTs that enrolled 24,507 patients were included. The rate of major bleeding was 1.8 (95% CI: 1.3-2.5) and 3.1 (95% CI: 2.4-3.9) per 100 patient-years in DOAC and VKA arms, respectively. The rate of VTE recurrence was 3.7 (95% CI: 2.7-4.7) and 4.1 (95% CI: 3.0-5.4) per 100 patient-years of DOAC and VKA, respectively. The case fatality rate of bleeding was significantly higher in the VKA arms 10.4% (95% CI: 6.6-15.4) compared to DOACs 6.1% (95% CI: 2.7-11.7; p value for difference = 0.029) with no statistical difference between the case fatalities for recurrent VTE. The rate of death from either definite major bleeding or definite recurrent VTE was 0.27 (95% CI: 0.16-0.40) and 0.46 (95% CI: 0.32-0.63) per 100 patient-years for DOACs and VKAs respectively, resulting in a number needed to treat of 875 for DOACs to prevent one death. Conclusion: DOACs are attractive alternatives to VKAs for initial treatment of symptomatic VTE, with lower frequency and case fatality for major bleeding. However, the incremental safety benefit of DOACs over VKAs is small, with large numbers needed to treat. © 2014 Elsevier Ltd. All rights reserved.
Introduction Venous thromboembolism (VTE), including deep vein thrombosis (DVT) and pulmonary embolism (PE), is associated with a high rate of morbidity and mortality [1,2]. For decades, vitamin K antagonists (VKAs) have been the mainstay of VTE treatment and prevention [3]. Without adequate anticoagulant therapy, the risk of recurrent VTE is estimated to be 50% in the first 3 months following a new diagnosis [4,5]. In the absence of active bleeding, the risk of major bleeding is usually much smaller in this time period [6]. After 3 months, the risk of recurrent VTE drops off sharply and is estimated to be 3-5% per year ⁎ Corresponding author at: University of Alberta, 4-112 Clinical Sciences Building, 11350-83rd Avenue, Edmonton, AB, T6G 2G3. Tel.: +1 780 407 1584; fax: +1 780 407 2680. E-mail address: [email protected] (C. Wu).
http://dx.doi.org/10.1016/j.thromres.2014.07.001 0049-3848/© 2014 Elsevier Ltd. All rights reserved.
for events provoked by transient risk factors and 10% per year for idiopathic events [7,8]. Several direct oral anticoagulants (DOACs), including dabigatran, rivaroxaban, apixaban and edoxaban have emerged as alternatives to VKAs. Clinical trials have shown each of these agents to be at least as effective and safe as VKAs for the acute treatment of VTE and the secondary prevention of recurrent VTE [9–15]. The benefit of treating VTE with anticoagulant therapy must always be weighed against the risk of bleeding complications. The variable clinical course of recurrent VTE and bleeding complications make balancing the risks and benefits of anticoagulation difficult for clinicians and patients. Nevertheless, mortality is a serious and objective adverse outcome associated with both recurrent VTE and bleeding complications, and therefore a comparison of case fatality associated with recurrent VTE and with major bleeding may help judge the true benefit of anticoagulation. The purpose of this study was to look at patients with acute symptomatic
628
C. Wu et al. / Thrombosis Research 134 (2014) 627–632
VTE who were randomized to receiving DOACs compared to VKAs and estimate the rates and case-fatality of recurrent VTE and major bleeding during the initial months of therapy.
Material and Methods Studies Identification We conducted a systematic literature search to identify all randomized controlled trials (RCTs) that included adult patients (≥ 18 years old) treated with DOACs for acute symptomatic VTE treatment (i.e. not less than 3 months) based on a priori protocol available on request. Patients treated for other indications, including joint replacements, atrial fibrillation or acute coronary syndrome were excluded in order to minimize study heterogeneity when comparing safety and efficacy endpoints. The review was reported according to the PRISMA statement [16,17].
Literature Search We searched MEDLINE, Cochrane Central Register of Controlled Trials and EMBASE databases from inception through December 2013 to identify all published and unpublished literature, in any language, related to our topic. The search strategy included the MeSH terms venous thrombosis and pulmonary embolism. The following terms were also added to the search strategy: recurrent venous thromboembolism, new oral anticoagulants, dabigatran, apixaban, rivaroxaban, edoxaban, warfarin and acenocoumarol. We also screened major Hematology international conferences (American Society of Hematology, International Society of Thrombosis and Hemostasis, and European Hematology Association) for abstracts from their annual meetings from 2010-2013, registries of health technology assessments and clinical trials, and the reference list of retrieved studies for additional studies and unpublished data.
Study Selection We included studies that: I) were randomized controlled trials comparing therapeutic doses of dabigatran, apixaban, edoxaban or rivaroxaban to standard (heparin/VKAs) (target INR 2-3) for treatment of symptomatic VTE for at least 3 months; II) included study population with deep venous thromboembolism, pulmonary embolism or both; III) included a qualifying event (DVT or PE) that was confirmed objectively by the presence of intraluminal defect on venography or venous noncompressibility on duplex ultrasound for DVT [18], and the presence of intraluminal filling defect on pulmonary angiography or a high probability ventilation-perfusion scan for PE [19]; and IV) reported at least one of the following outcomes on anticoagulation: 1) recurrent events defined as symptomatic thromboembolic event that occurred after initiation of VTE treatment and diagnosed objectively by a new intraluminal filling defect on venography or a new non-compressible vein segment on duplex ultrasound for recurrent DVT or by a new intraluminal filling defect on pulmonary angiography or a new high probability ventilation-perfusion lung scan, 2) major bleeding events as defined by ISTH [20] or as per study definition, and 3) mortality caused by bleeding or venous thromboembolic event. Patients receiving additional treatment (e.g., thrombolysis, inferior vena filters) were excluded. Eligible studies were reviewed independently by two reviewers (CW, GA) to assess suitability for inclusion, and all included trials were reviewed to assess agreement in outcome event reporting. Studies considered relevant by one or both reviewers were retrieved and disagreements were resolved by discussion.
Outcome Measures and Data Extraction Our primary outcome measures were: 1) frequency of major bleeding; 2) frequency of fatal bleeding; 3) frequency of VTE recurrence; and 4) frequency of fatal VTE recurrence (definite and possible). Two reviewers independently abstracted the data describing baseline characteristics, treatment interventions and outcomes. Discrepancies were solved by discussion. Results of intention-to-treat analyses were collected if reported. Because the cause of death in cases with suspected VTE related death is not always objectively confirmed, we reported VTE deaths as definite or possible. Definite fatal recurrent VTE was defined as any VTE diagnosed postmortem, or a new intraluminal-filling defect detected on computed tomography, venography or ventilation perfusion scan, or a high clinical probability of fatal pulmonary embolism as adjudicated by the study investigators immediately before death. Possible VTE deaths included data reported as, “PE cannot be ruled out.” Quality Assessment Two reviewers independently assessed each study’s risk of bias using the 6 domains of the Cochrane Collaboration’s tool for assessing risk for in randomized trials [21] and disagreements were resolved by discussion. Data Synthesis and Analysis The results of individual studies were combined to determine the following outcomes: (1) the rate of major bleeding; (2) the rate of fatal bleeding; (3) the rate of VTE recurrence; (4) the rate of fatal VTE recurrence (definite and possible). We also calculated the case fatality rate of major bleeding, and recurrent VTE (both definite and possible). The case fatality rate of recurrent VTE was defined as the proportion of all recurrent VTE events (fatal and nonfatal) resulting in death. Fatal bleeding was defined as a major bleeding event directly leading to death or death was reported as “associated with bleeding”. The case fatality rate of a major bleeding event was defined as the proportion of all major bleeding events causing or associated with death [22]. All rates were expressed as events per 100 patient-years of anticoagulation to standardize for different follow-up durations across studies [23,24]. Proportions were transformed via the Freeman-Tukey double arcsine method [25,26] before pooling the case fatality rates. We then performed a DerSimonian–Laird random-effects model to pool the transformed rates [27]. After pooling the resulting estimates and their 95% CIs, limits were back-transformed to rates per 100 patient-years of follow-up. The pooled random effect incidence rate ratio (IRR) [28] was used to estimate of the effects of treatment with DOACs. The I2 statistic was used to estimate total variation among the pooled estimates across studies. An I2 value less than 25% was considered low-level heterogeneity, 25% to 50% was moderate-level, and greater than 50% was high-level [29]. Statistical analysis was done using StatsDirect statistical software (version 2.8.0). Results Study Selection and Patient Characteristics The literature search yielded 2453 studies (Fig. 1). Of these, 5 RCTs [9, 11,13–15] met the inclusion criteria and enrolled 12258 patients for acute VTE treatment in DOAC arms and 12249 patients for VKA arms. Trials evaluated three factor Xa inhibitors (rivaroxaban, n = 4150), (apixaban, n = 2691), (edoxaban, n = 4143) and one direct thrombin inhibitor (dabigatran, n = 1274). The majority of patients (N 80%) enrolled in these studies were treated for idiopathic (unprovoked) VTE events and were anticoagulated for a median time of 6.7 months (Table 1). The time from symptoms onset to randomization ranged
C. Wu et al. / Thrombosis Research 134 (2014) 627–632
4366 Potentially relevant references from databases screened. (Medline 1775, EMBASE 2287, CENTRAL 304)
629
Additional records identified through other sources (n =2)
Records after duplicates removed (n =2453)
Records screened (n =2453)
Records excluded (n =2426)
Full-text articles excluded (n =22): -Narrative reviews (n= 4) -Withdrawn drug (n=2) -Multiple publication (n = 1) -Non clinical outcomes (n = 3) -Studies for extended VTE prevention (n=9) - Dose ranging studies (n=3)
Articles retrieved and evaluated in full for inclusion (n =27)
Studies included in result synthesis (n =5)
Fig. 1. Search strategy and study selection as per PRISMA checklist.
from 2 to 11 days, during which all patients were treated with parenteral anticoagulation (e.g., LMWH, unfractionated heparin, or fondaparinux). Quality Assessment and Risk of Bias In all of the 5 included studies, the risk of bias was low. Open label design was implemented in three studies RCTs [11,13,14] but all outcome data were recorded with a routine structured technique and were adjudicated by an independent committee blinded to study drug allocation. Loss to follow-up rate was less than 5% in all included studies. Although all studies were funded by pharmaceutical industry, outcomes were assessed by independent steering committees. See summary of risk of bias in (Supplementary Figs. 1 and 2).
Synthesis of Results Outcomes for Acute VTE Management Overall analysis included 12258 patients who received 7206 patient-years of new oral anticoagulant therapy and 12249 patients who received 7134 patient-years of LMWH/VKA therapy. Table 2 summarizes the frequency of each outcome as reported by studies. Table 3 summarizes the pooled results by treatment type. Table 4 summarizes the comparison between the rates of bleeding and VTE recurrence. Major Bleeding. Major bleeding with DOACs occurred at a rate of 1.8 per 100 patient-years (95% CI: 1.3-2.5, I 2 = 67.6%) and fatal bleeding occurred at a rate of 0.14 per 100 patient years (95% CI: 0.07-0.24,
Table 1 Characteristics of studies enrolled patients for active treatment of acute VTE events. Study
Design
Treatments regimen
Duration, months
Patients
Efficacy outcome
Safety outcome
RE-COVER I, 2009
Randomized, double-blind
6
2539 patients with acute VTE
Recurrent VTE 2.4% dabigatran, 2.1% warfarin
Major bleeding: 1.6% dabigatran, 1.9% warfarin
EINSTEIN-PE, 2012
Randomized, open-label
3, 6 or 12
4832 patients with acute PE
Recurrent VTE 2.1% rivaroxaban, 1.8% enoxaparin/VKAs
Major bleeding 1.1% rivaroxaban, 2.2% enoxaparin/VKAs
EINSTEIN-DVT, 2010
Randomized, open-label
3, 6 or 12
3449 patients with acute DVT
Recurrent VTE 2.1% rivaroxaban, 3.0% enoxaparin/warfarin
Major bleeding 0.8% rivaroxaban, 1.2% VKAs
AMPLIFY, 2013
Randomized, double blind double-dummy
6
5395 Patients with acute DVT or PE
Hokusai-VTE, 2013
Randomized, double blind double-dummy
Dabigatran (150 mg BID) Warfarin Rivaroxaban (15 mg BID for 3 weeks, then 20 mg OD) VKAs Rivaroxaban (15 mg BID for 3 weeks, then 20 mg OD) VKAs Apixaban (10 mg BID for 7 days, then 5 mg BID) Warfarin Edoxaban (60 mg OD) Warfarin
12
8292 patients with acute DVT and/or PE
Recurrent VTE 2.3% apixaban, 2.7% enoxaparin/VKAs Recurrent VTE 3.2% edoxaban, 3.5% enoxaparin/VKA
Major bleeding 0.6% apixaban, 1.8% enoxaparin/VKAs Major bleeding 1.4% edoxaban, 1.6% enoxaparin/VKA
630
C. Wu et al. / Thrombosis Research 134 (2014) 627–632
Table 2 Outcome events per treatment in patients presenting with acute symptomatic VTE in included studies. Study, Year (Reference)
RE-COVER1 EINSTEIN-PE EINSTEIN-DVT AMPLIFY Hokusai-VTE
Study arm
Dabigatran VKA Rivaroxaban VKA Rivaroxaban VKA Apixaban VKA Edoxaban VKA
N
1274 1265 2419 2413 1731 1718 2691 2704 4143 4149
Mean on treatment duration (months)
6 6 3,6 or 12 3,6 or 12 3,6 or 12 3,6 or 12 6 6 3 to 12 3 to 12
MB
20 24 26 52 14 20 15 49 56 66
VTE recurrence
30 27 50 44 36 51 48 71 66 80
MB fatality rate
1 1 2 3 1 5 2 3 2 10
Recurrence fatality
Adjudication of Outcomes
P&D
D
1 3 11 7 4 6 12 16 26 26
1 3 2 1 1 0 1 2 4 3
Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
DOACs: new oral anticoagulants, MB: Major bleeding, CRNMB: clinically relevant non major bleeding, P&D: Possible and definite VTE recurrence, D: only definite VTE recurrent cases.
I2 = 0%), compared to 3.1 per 100 patient-years (95% CI 2.4-3.9, I 2 = 70.6%) and 0.33 per 100 patient-years (95% CI 0.21-0.48, I2 = 0%) for the VKA treatment. Out of 131 major bleeding episodes in patients receiving DOACs, 8 were fatal and this resulted in a case fatality rate for major bleeding of 6.1% (95% CI: 2.7-11.7). In patients treated with VKA, there were 211 incidents of major bleeding among which 22 resulted in death; the cases fatality for major bleeding in this group was 10.4% (95% CI: 6.6-15.4). Supplementary Figs. 3 and 4 shows the Forest plot of major bleeding proportion among patients randomized to DOACs and VKA, respectively, in included studies. VTE Recurrence. The VTE recurrence rate was 3.7 per 100 patient-years of DOACs therapy (95% CI: 2.7-4.7, I2 = 79.8%), definite fatal recurrent VTE occurred at a rate of 0.16 per 100 patient-years (95% CI: 0.08-0.26, I2 = 0%), and possible fatal VTE occurred at a rate of 0.72 per 100 patientsyears (95% CI: 0.48-1.00, I2 = 38%). In the VKA arms, the rate of VTE recurrence was 4.1 per 100 patient-years (95% CI: 3.0-5.4, I2 = 82.7%), the rate of definite fatal recurrent VTE cases was 0.14 per 100 patientyears (95% CI: 0.05-0.27, I2 = 32.2%), and the rate of possible fatal VTE recurrence was 0.82 per 100 patient-years (95% CI: 0.59-1.1, I2 = 24%). Of the 241 recurrent VTE in DOAC-treated patients, the case fatality rates in patients on DOACs were 3.7% (95% CI: 1.7-7.0) and 22.4% (95% CI: 17.3-28.2) for definite and possibly fatal events, respectively. For patients on VKAs, of the 273 cases of VTE recurrence, 9 of which lead to definite VTE related death and 58 were possibly fatal, the case fatality rate was 3.3% (95% CI: 1.5-6.2), and 21.2% (95% CI: 16.5-26.5), respectively. Supplementary Figs. 5 and 6 shows the Forest plot of
recurrent VTE proportion among patients randomized to DOACs and VKA, respectively, in included studies. Combined Major Bleeding and VTE Recurrence. The rates of composite fatality (death due to either major bleeding or definite recurrent VTE) were 0.27 per 100 patient years (95% CI 0.16-0.40, I2 = 0%) for DOACs and 0.46 per 100 patient years (95% CI 0.32-0.63, I2 = 0%) for VKAs. The rates of composite fatality (death due to either major bleeding or possible recurrent VTE) were 0.89 per 100 patient years (95% CI 0.65-1.1, I2 = 13.2%) for DOACs and 1.1 per 100 patient years (95% CI 0.86-1.4, I2 = 20.2%) for VKAs. Compared with VKAs, treatment with DOACs had lower incidence of major bleeding (IRR 0.59, 95% CI 0.420.87), lower incidence of fatal bleeding (IRR 0.39, 95% CI 0.17-0.9), and lower incidence of the composite fatality related to either major bleeding or definite recurrent VTE (IRR 0.55, 95% CI 0.30-0.99). There were no differences between treatment arms for recurrent VTE (IRR 0.87, 95% CI 0.73-1.0), recurrent fatal PE (IRR 0.91, 95% CI 0.33-2.5), or the composite of fatality related to either major bleeding or possible recurrent VTE (IRR 0.77, 95% CI 0.55-1.08). Discussion Our findings reveal that patients on DOACs had lower rates of major and fatal bleeding events (1.8 per 100 patient-years (95% CI 1.3-2.5) and 0.14 per 100 patient-years (95% CI 0.07-0.24), respectively, compared to 3.1 per 100 patient-years (95% CI 2.4-3.2), and 0.33 per 100 patientyears (95% CI 0.21-0.48), in patients treated with VKA therapy. This
Table 3 Outcomes according to treatment type. Outcomes
Acute DOACs
Acute VKA
Rate of MB, per 100 patient-years (95% CI) Rate of fatal bleeding, per 100 patient-years (95% CI) Bleeding CFR (95% CI) Rate of VTE recurrence, per 100 patient-years (95% CI) Rate of fatal VTE, per 100 patient-years (95% CI)
1.8 (1.3-2.5)
3.1 (2.4-3.9)
0.14 (0.07-0.24)
0.33 (0.21-0.48)
6.1% (2.7-11.7)
10.4% (6.6-15.4)
3.7 (2.7-4.7)
4.1 (3.0-5.4)
0.16 (0.08-0.26)
0.14 (0.05-0.27)
Definite Possible Recurrent VTE CFR (95% CI)
0.72 (0.48-1.00)
0.82 (0.59-1.1)
3.7% (1.7-7.0)
3.3% (1.5%-6.2%)
22.4% (17.3-28.2)
21.2% (16.5-26.5)
0.27 (0.16-0.40) 0.87 (0.65-1.1)
0.46 (0.32-0.63) 1.1 (0.86-1.42)
Definite Possible Rate of definite fatal events, per 100 patient-years (95% CI) Rate of definite and possible fatal events, per 100 patient-years (95% CI) DOACs: direct oral anticoagulants; VKA: vitamin K antagonists; MB: major bleeding; CFR: case fatality rate.
C. Wu et al. / Thrombosis Research 134 (2014) 627–632
631
Table 4 Comparison between the rate of bleeding and VTE recurrence in patients on DOACs vs VKA for acute VTE treatment. Incidence rate ratio
Major bleeding
DOACs vs. VKA 0.59 (0.42-0. 87, P = 0.007) (95% CI, P value)
Recurrent VTE
Fatal Bleeding
Fatal recurrent VTE (definite)
Fatal recurrent VTE (possible and definite)
Combined fatal bleeding and definite fatal VTE recurrence
Combined fatal bleeding and (definite and possible) fatal VTE recurrence
0.87 (0.73-1.0, P = 0.13)
0.39 (0.17-0.9, P = 0.029)
0.91 (0.33-2.5, P = 0.86)
0.93 (0.64-1.4, P = 0.70)
0.55 (0.30-0.99, P = 0.048)
0.77 (0.55-1.08, P = 0.13)
IRR: incidence rate ratio, DOACs: direct oral anticoagulants; VKA: vitamin K antagonists.
difference between DOAC and VKA therapy reached statistical significance in major and fatal bleeding and the use of DOACs was associated with a 41% and 61% relative reduction in the rate of major and fatal bleeding, respectively. The estimated case fatality rate for DOAC related major bleeding during acute VTE treatment was 6.1% (1 in 16 major bleeding events) compared to 10.4% bleeding case fatality in warfarin arms in the same studies and 11.3% from a previous study [7] (1 in 10 major bleeding events) during the acute management of VTE. These results illustrate a reduction in bleeding rates and associated fatality in patients on DOACs compared to warfarin for the acute treatment of VTE. It is important to note that DOAC therapy does not eliminate the incidence of major and fatal bleeding and that the overall rates of both in either treatment arms were small and the numbers needed to treat to prevent one case of major bleeding and fatal bleeding are 153 and 875, respectively. In contrast to major bleeding, the rates of VTE recurrence during the initial phase of acute VTE treatment were similar between the DOAC and VKA groups. 1 in 51 patients treated with DOACs had a recurrent VTE event as opposed to 1 in 45 of those treated with VKA, and the difference between these rates did not reach statistical significance. The rate of definite fatal VTE recurrence in both the DOAC and VKA arms was infrequent, 0.16 per 100 patient-years and 0.14 per 100 patientyears, leading to imprecise pooled estimates with large 95% confidence intervals. The case fatality rate of definite recurrent VTE case fatality for patients on DOACs was 3.7% compared to 3.3% in VKA arms in the same studies. These data suggest that the efficacy of all these medications for initial VTE therapy was comparable. Compared to previous data documenting the risk of recurrent VTE off anticoagulation in the acute phase of VTE therapy [4,5], this also confirms the known substantial decrease in recurrent VTE with appropriate anticoagulant use and reinforces the need for solid anticoagulation in this critical time frame. Importantly, we found that DOACs were associated with a lower risk for fatality definitely related to either a major bleed or a recurrent VTE, with IRR 0.55 (95% CI 0.30-0.99); initial treatment of acute VTE with DOACs may therefore be associated with lower risk for death than treatment with VKAs as well as lower risk for major bleeding. Though the numbers of events are small, and are drawn from selected patient populations recruited to randomized trials, the enhanced safety profile, non-inferior effectiveness of DOACs, and apparently lower risk for death definitely related to major bleeding or recurrent VTE may support a preference for a DOAC in the acute treatment of VTE in selected patients. However, the number needed to treat with DOACs to prevent 1 death from either major bleeding or VTE recurrence is high at 875. This study has limitations. First, some cases of fatal recurrent VTE were not objectively defined, therefore, we relied on “definite” and “possible” definitions as judged by the study investigators. Second, we note that for each DOAC there was only 1 phase III RCT for each studied indication, so results were not duplicated/confirmed and grouping all the DOACs together may not necessarily reflect each individual drug. However, we assumed the presence of a class effect on the main outcomes (e.g., stroke prevention in atrial fibrillation, major bleeding and recurrent VTE) as shown by several network meta-analyses [30,31].
Third, the pooled estimates of major bleeding and recurrent VTE had moderate to high heterogeneity. The most likely explanation for the heterogeneity is likely due to the variability in the follow-up periods. Pooled studies participated with different follow-up periods, which suggest that all studies were not estimating the same quantity. Fourth, publication bias could not be assessed using funnel plots due to the small number of included trials. To our knowledge, this is the first study to assess the rates and case fatality of major bleeding and recurrent VTE on DOACs and compare it to standard therapy in the active phase of VTE treatment. All reported events were independently adjudicated with a minimal loss to followup rate in all included studies. All included studies adopted the ISTH definition to quantify major bleeding and this allows for comparison between trials. One of the major potential limitations was the unavailability of any real world cohort data to give an estimate of case fatality outside clinical trials use of DOACs. Conclusion Owing to their comparable efficacy but lower risk of major bleeding, DOACs offer an attractive alternative to VKAs for acute treatment of symptomatic VTE. However, given the given the low incidence of adverse events and the high NNT, the absolute incremental benefit is limited. Funding This study was self-funded. Conflict of Interest Statement M.S. McMurtry and Wu C, are investigators for eTRIS, a randomized, open label, and multicenter study evaluating the efficacy and safety of edoxaban monotherapy versus low molecular weight heparin and warfarin in subjects with symptomatic deep vein thrombosis sponsored by Daiichi Sankyo, and an investigator for XALIA, a multicenter observational study of rivaroxaban as initial anticoagulation for venous thromboembolism. Dr Wu has served on the advisory board for LeoPharma. Acknowledgements M.S. McMurtry is supported by the Heart and Stroke Foundation of Canada. Alotaibi GS would like to thank the King Saud University for funding his graduate studies. Appendix A. Supplementary Data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.thromres.2014.07.001.
632
C. Wu et al. / Thrombosis Research 134 (2014) 627–632
References [1] Heit JA. The epidemiology of venous thromboembolism in the community. Arterioscler Thromb Vasc Biol 2008;28:370–2. [2] Kearon C. Epidemiology of venous thromboembolism. Semin Vasc Med 2001;1:7–26. [3] Kearon C, Kahn SR, Agnelli G, Goldhaber S, Raskob GE, Comerota AJ, et al. Antithrombotic therapy for venous thromboembolic disease: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition). Chest 2008;133:454S–545S. [4] Barritt DW, Jordan SC. Anticoagulant drugs in the treatment of pulmonary embolism. A controlled trial. Lancet 1960;1:1309–12. [5] Hull R, Delmore T, Genton E, Hirsh J, Gent M, Sackett D, et al. Warfarin sodium versus low-dose heparin in the long-term treatment of venous thrombosis. N Engl J Med 1979;301:855–8. [6] Hass B, Pooley J, Harrington AE, Clemens A, Feuring M. Treatment of venous thromboembolism - effects of different therapeutic strategies on bleeding and recurrence rates and considerations for future anticoagulant management. Thromb J 2012;10:24. [7] Carrier M, Le Gal G, Wells PS, Rodger MA. Systematic review: case-fatality rates of recurrent venous thromboembolism and major bleeding events among patients treated for venous thromboembolism. Ann Intern Med 2010;152:578–89. [8] Kearon C. Long-term management of patients after venous thromboembolism. Circulation 2004;110:I10–8. [9] Schulman S, Kearon C, Kakkar AK, Mismetti P, Schellong S, Eriksson H, et al. Dabigatran versus warfarin in the treatment of acute venous thromboembolism. N Engl J Med 2009;361:2342–52. [10] Schulman S, Kearon C, Kakkar AK, Schellong S, Eriksson H, Baanstra D, et al. Extended use of dabigatran, warfarin, or placebo in venous thromboembolism. N Engl J Med 2013;368:709–18. [11] Agnelli G, Buller HR, Cohen A, Curto M, Gallus AS, Johnson M, et al. Oral apixaban for the treatment of acute venous thromboembolism. N Engl J Med 2013;369:799–808. [12] Agnelli G, Buller HR, Cohen A, Curto M, Gallus AS, Johnson M, et al. Apixaban for extended treatment of venous thromboembolism. N Engl J Med 2013;368:699–708. [13] Einstein I, Bauersachs R, Berkowitz SD, Brenner B, Buller HR, Decousus H, et al. Oral rivaroxaban for symptomatic venous thromboembolism. N Engl J Med 2010;363:2499–510. [14] Einstein–Pe I, Büller HR, Prins MH, Lensin AW, Decousus H, Jacobson BF, et al. Oral rivaroxaban for the treatment of symptomatic pulmonary embolism. N Engl J Med 2012;366:1287–97. [15] Hokusai VTEI, Buller HR, Decousus H, Grosso MA, Mercuri M, Middeldorp S, et al. Edoxaban versus warfarin for the treatment of symptomatic venous thromboembolism. N Engl J Med 2013;369:1406–15.
[16] Canadian Agency for D, Technologies in H. Requirements for health technology assessment template for reports. Ottawa (ON): The Agency; 2008. [17] Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtzsche PC, Ioannidis JP, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: explanation and elaboration. BMJ 2009;339:10. [18] Bates SM, Jaeschke R, Stevens SM, Goodacre S, Wells PS, Stevenson MD, et al. Diagnosis of DVT: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012;141:e351S–418S. [19] Kearon C. Diagnosis of pulmonary embolism. CMAJ 2003;168:183–94. [20] Schulman S, Kearon C. Subcommittee on Control of Anticoagulation of the S, Standardization Committee of the International Society on T, Haemostasis. Definition of major bleeding in clinical investigations of antihemostatic medicinal products in non-surgical patients. J Thromb Haemost 2005;3:692–4. [21] Higgins JPT, Altman DG, Gøtzsche PC, Jüni P, Moher D, Oxman AD, et al. The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. BMJ 2011;343. [22] Douketis JD, Kearon C, Bates S, Duku EK, Ginsberg JS. Risk of fatal pulmonary embolism in patients with treated venous thromboembolism. JAMA 1998;279:458–62. [23] Koepsell TD. WNEmstooiNY. NY: Oxford University Press; 2003. [24] Kirkwood BR, Sterne JAC. Essential Medical Statistics. 2nd ed. Blackwell; 2003. [25] Miller J. The inverse of the Freeman-Turkey double arcsine transformation. Am Stat 1978;32:138. [26] Landis JR, Cooper MM, Kennedy T, Koch GG. A computer program for testing average partial association in three-way contingency tables (PARCAT). Comput Programs Biomed 1979;9:223–46. [27] DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials 1986;7:177–88. [28] Guevara JP, Berlin JA, Wolf FM. Meta-analytic methods for pooling rates when follow-up duration varies: a case study. BMC Med Res Methodol 2004;4:17. [29] Higgins JP, Thompson SG. Quantifying heterogeneity in a meta-analysis. Stat Med 2002;21:1539–58. [30] Castellucci LA, Cameron C, Le Gal G, Rodger MA, Coyle D, Wells PS, et al. Efficacy and safety outcomes of oral anticoagulants and antiplatelet drugs in the secondary prevention of venous thromboembolism: systematic review and network metaanalysis. BMJ 2013;347:f5133. [31] Weitz JI, Gross PL. New oral anticoagulants: which one should my patient use? Hematology/the Education Program of the American Society of Hematology American Society of Hematology Education Program, 2012; 2012. p. 536–40.
