Sage Journals: Discover world-class research

Abstract

The present systematic review was conducted to assess the efficacy and safety of apixaban versus other anticoagulants, for the prevention of venous thromboembolism (VTE) following total hip replacement (THR) and total knee replacement (TKR) surgery. Electronic databases were interrogated to identify relevant randomized controlled trials. A series of direct/indirect comparisons and a network meta-analysis were conducted. Indirect comparisons found that the odds ratio of “all VTE and all-cause death” were significantly higher for dabigatran than for apixaban in patients with THR (odds ratio [OR], 2.51; 95% confidence interval [CI], 1.50-4.21) and TKR (OR, 1.72; 95% CI, 1.22-2.42). Rivaroxaban showed similar efficacy to apixaban in patients with THR and TKR (OR, 0.69; 95% CI, 0.38-1.25 and OR, 0.83; 95% CI, 0.57-1.19, respectively). No significant differences were observed in bleeding outcomes between treatments. The novel anticoagulants apixaban, rivaroxaban, and dabigatran demonstrated similar or improved efficacy and similar safety compared with current therapies in this indication.

Keywords

venous thromboembolism thrombosis prophylaxis factor Xa inhibitors anticoagulants deep venous thrombosis pulmonary embolism

Introduction

Venous thromboembolism (VTE) collectively describes the sometimes debilitating, painful, and potentially fatal conditions of deep vein thrombosis (DVT) and pulmonary embolism (PE). Deep vein thrombosis is the formation of a thrombus in a deep vein, most commonly of the lower limbs, and the potentially fatal PE is caused when such thrombi become dislodged and travel to the lungs.

Venous thromboembolism represents a considerable preventable morbidity and mortality burden to patients and is estimated to be responsible for 25 000 preventable deaths annually in the United Kingdom.¹ Risk factors for VTE include increasing age, obesity, cancer, and medical conditions such as cardiac and respiratory diseases, surgery, and inherited or acquired clotting tendency. High-risk surgical procedures can lead to VTE, and patients undergoing major orthopedic surgery, such as total hip replacement (THR) or total knee replacement (TKR) are in the highest risk category for VTE.² In the absence of anticoagulant prophylaxis, the estimated incidence of DVT following orthopedic surgery ranges from 40% to 60%, and the overall risk of fatal PE has been estimated to be between 0.2% and 0.3%.³

Thromboprophylaxis, both mechanical and pharmacological, is a current standard practice for the prevention of VTE in patients undergoing orthopedic surgery. Current treatment guidelines recommend the routine administration of a prophylactic anticoagulant for at least 10 days and up to 35 days following orthopedic surgery.^3,4 Currently available anticoagulant therapies include warfarin, fondaparinux, and the more commonly used low-molecular-weight heparins (LMWHs), such as enoxaparin, bemiparin, and tinzaparin. These agents have demonstrated efficacy but are associated with a number of limitations. There is a requirement for regular monitoring of warfarin and, in some regions, a requirement for the monitoring of platelet counts in those on LMWHs. Both LMWHs and fondaparinux require parenteral administration, and warfarin has a narrow therapeutic window which is difficult to attain.²

A number of new oral anticoagulants have been developed and are undergoing evaluation as thromboprophylaxis in orthopedic surgery. Rivaroxaban (a highly specific factor Xa inhibitor) and dabigatran etexilate (a direct thrombin inhibitor) have been assessed in large randomized controlled trials (RCTs) of THR and TKR surgery and have demonstrated similar or greater efficacy and similar safety compared with standard LMWH.^5

-10 Both rivaroxaban and dabigatran etexilate have been approved in over 70 countries for the prevention of VTE in patients having elective THR or TKR surgery.¹¹

Apixaban, recently approved in the European Union for the prevention of VTE following THR or TKR surgery, is a highly specific factor Xa inhibitor that is administered orally and does not require routine laboratory monitoring.¹² Clinical trials in patients who have undergone elective THR and TKR have demonstrated that apixaban has improved efficacy, when compared with enoxaparin, in reducing VTE and all-cause death, with a similar or lower risk of bleeding.^13,14

The aim of this systematic review and meta-analysis was to compare the efficacy and safety of apixaban versus other key comparators, for the prevention of VTE following elective TKR or THR surgery, via a series of direct and indirect comparisons and a network meta-analysis (NMA).

Methods

Systematic Review

Searches of MEDLINE, EMBASE, the Cochrane library, and CINAHL were conducted on July 8, 2010. Using Boolean operators, the searches combined terms (including MeSH headings as appropriate) for (1) VTE, (2) interventions, (3) clinical trial design, and (4) hip/knee replacement. The following conference proceedings were handsearched (2008-2010 inclusive):

American Society of Hematology

British Orthopaedic Association

International Congress on Thrombosis

International Society on Thrombosis and Haemostasis

American College of Chest Physicians

American College of Cardiology

American Academy of Orthopedic Surgeons

European Orthopaedic Research Society

European Federation of National Associations of Orthopaedics and Traumatology

European Hematology Society

British Society of Haematology.

Handsearching was also performed by reviewing the bibliographies of included studies and previously published systematic reviews identified through the database search. Prespecified inclusion and exclusion criteria for the systematic review are shown in Table 1 .

Table 1.

Summary of Inclusion and Exclusion Criteria for the Systematic Review

Study design	Only prospective, randomised controlled trials were included
Population	Adult patients (18 years or older) undergoing elective knee or hip replacement surgery. Patients were excluded if they were undergoing emergency knee or hip surgery, surgery for hip fracture repair, or any other types of surgery
Interventions	Studies were included if they evaluated ≥1 of the following treatments: Apixaban Low-molecular-weight heparins (enoxaparin, tinzaparin, and bemiparin) Fondaparinux Rivaroxaban Dabigatran Mechanical interventions and nursing care/physiotherapy were excluded.
Outcomes	Studies were included if they reported ≥1 of the following outcomes. Mortality (VTE-related and all-cause) Incidence of VTE Post-DVT complications including post-thrombotic syndrome Length of hospital stay Joint outcomes, including joint infection • Adverse events including bleeding events (intracranial bleeding, major bleeding and nonmajor, clinically relevant bleeding) • Health-related quality of life
Language of publication	Only publications in English language were included

Abbreviations: VTE, venous thromboembolism; DVT, deep vein thrombosis.

Identified studies were independently assessed by 2 reviewers in order to ascertain that they met the predefined inclusion/exclusion criteria and any discrepancies were resolved by a third party. Relevant information was extracted into a predefined document by a reviewer. A second reviewer checked the data extraction, and any inconsistencies were resolved through discussion.

Evaluation of Outcomes

The outcomes considered in the meta-analyses and corresponding definitions are summarized in Table 2 . This manuscript focuses on the primary efficacy outcomes of VTE composite and major VTE, and the primary safety outcomes of major bleeding, clinically relevant nonmajor (CRNM) bleeding, and any bleeding. Results from the analyses of the other outcomes are available as Supplement Data.

Table 2.

Outcomes of Interest for Meta-Analysis

Outcomes^a	Definition
VTE composite	Composite of DVT, PE, and fatal events, which includes studies reporting results for any of the following outcomes. Any DVT + symptomatic or fatal PE + any death Any DVT + nonfatal PE + any death Any DVT + symptomatic PE + any death Total VTE (any DVT or PE) + any death Any DVT + PE but only where no deaths are reported in the study
Major VTE (indirect comparisons only)	Composite of proximal DVT, nonfatal PE, and VTE-related death
Any DVT	Any DVT includes studies reporting results for any of the following Symptomatic DVT + asymptomatic DVT Distal + proximal DVT Total DVT, where type unspecified
Asymptomatic DVT	Reported Cases of DVT confirmed by venography or ultrasound with no prior clinical diagnosis of DVT Calculated Any DVT less symptomatic DVT
Symptomatic DVT (indirect comparisons only)	DVT where patients reported symptoms during the treatment period, which were subsequently confirmed by venography, ultrasound or other objective method
PE (indirect comparisons only)	Any PE event which includes studies reporting results for any of the following outcomes Symptomatic PE + asymptomatic PE Fatal + nonfatal PE Total PE, where type unspecified
Major bleeding	Major bleeding conforming to the main International Society on Thrombosis and Haemostasis criteria for major bleeding. Fatal bleeding and/or Symptomatic bleeding in a critical area or organ, such as intracranial, intraspinal, intraocular, retroperitoneal, intraarticular, or pericardial, or intramuscular with compartment syndrome, and/or Bleeding causing a fall in level of hemoglobin to 20 g/L (1.24 mmol/L) or more, or leading to transfusion of ≥2 units whole blood or red cells. Major bleeding (any definition).
CRNM bleeding	CRNM bleeding was defined slightly differently across all trials. Definitions included: multiple source bleeding, hematoma >25 cm², excessive wound hematoma, epistaxis >5 minutes, macroscopic hematuria (spontaneous, or lasting >24 hours if associated with an intervention), spontaneous rectal bleeding, gingival bleeding >5 minutes, hemoptysis, hematemesis, prolonged bleeding (>5 minutes) after venipuncture.
Any bleeding	Reported as a distinct end point or the sum of the reported major/minor/CRNM bleeding components.
Minor bleeding	Bleeds that did not meet the criteria for major bleed.

Abbreviations: CRNM, clinically relevant nonmajor; DVT, deep vein thrombosis; PE, pulmonary embolism; VTE, venous thromboembolism.

^aOutcomes shown in bold type are the focus of this report.

Statistical Methods for Meta-Analysis

It was decided a priori to conduct the following meta-analyses, where there were sufficient data for analysis. Both THR and TKR patient populations were analyzed separately.

A series of direct comparisons of head-to-head RCT data for the treatments of interest

A series of adjusted indirect comparisons to derive efficacy and safety effect sizes for apixaban versus the other anticoagulant treatments of interest, where head-to-head data were not available

An NMA allowed all evidence from both direct (head-to-head RCTs) and indirect (calculated from trials with common comparators) analyses to be evaluated. The NMA method allows the relative efficacy of several different treatments to be estimated by evaluating the entire network of RCT evidence.^15

–18

Treatments were grouped according to European licensed dose of launched products or doses in clinical development; other doses were included in the network if they contributed to the evidence for the treatment of interest. In addition, it was noted that the duration of treatment in some trials was longer than that indicated in Europe. Therefore, wherever feasible, the treatments of interest were separated into 3 prophylaxis groups: duration as European indication, extended prophylaxis where extension was less than 1 week, and extended prophylaxis where extension was 1 week or more. The reference treatment in the analyses was the treatment evaluated in the largest number of RCTs for the outcome measure considered; this was either enoxaparin 40 mg once daily (European licensed dose) or enoxaparin 30 mg twice daily (US licensed dose).

All the key outcomes of interest were measured using a dichotomous variable. The analyses were conducted on an intent-to-treat (ITT) basis for the outcomes of bleeding. However, since asymptomatic DVT can only be detected via an evaluable venogram, the evaluable population (those individuals with an evaluable venogram) was used in all analyses that included asymptomatic DVT (any DVT, asymptomatic DVT, major VTE, and VTE composite). The ITT analyses for these outcomes were conducted as sensitivity analyses (data not presented, available from the authors on request).

Wherever a direct meta-analysis was possible, this was conducted in Stata IC version 10.1 using the metan package SJ9_2: sbe24_3.^19,20 Results were expressed as odds ratios (ORs) and pooled using the DerSimonian and Laird random-effects method, which takes account of between-study variance. The estimate of heterogeneity was via the Mantel-Haenszel model.

The indirect comparisons between apixaban and other treatments of interest via a common comparator were made using the Bucher method²¹ and the pooled ORs produced from the direct meta-analysis. This method preserves randomization of treatments being compared indirectly.

The (indirect) OR between apixaban and the treatment of interest is given by

log (O R_{A ν B}) {= l o g (O R}_{A ν C}) - {l o g (O R}_{B ν C})

With standard error given by

S E [log (O R)_{A ν B}] = \sqrt{S E [log (O R_{A ν C})]^{2} + S E [log (O R_{B ν C})]^{2}}

The NMA methods are built on the principles of indirect comparisons and are based on relative treatment effects in order to preserve trial randomization and minimize bias. A random-effects NMA was conducted in WinBUGs version 1.4.1,^22
–24 which uses Bayesian Markov Chain Monte Carlo Gibbs sampling methods. Random-effect models allow the true treatment effect to vary between studies due to heterogeneity.

For the direct and indirect comparisons, 2 analyses were conducted for rivaroxaban 10 mg once daily in the THR patient population, one including and one excluding the RECORD 2 study. RECORD 2 compared rivaroxaban 10 mg once daily with enoxaparin 40 mg once daily; however, enoxaparin was administered for only 10 to 14 days compared with 31 to 39 days of rivaroxaban. The shorter period of enoxaparin treatment, half that recommended in the National Institute of Health and Clinical Excellence (NICE) VTE guideline,⁴ could lead to an overestimation of the treatment effect for rivaroxaban. Consequently, the 2 analyses were conducted in order to assess the variation in treatment effect contributed by this study.

Results

Systematic Review Results

A total of 1809 publications were identified through electronic searches and a further 4 through handsearching and manufacturer databases. Following assessment and exclusion of studies based on title, abstract, and full text, 40 records, reporting on 43 RCTs, were included in the final data set for the NMA (Figure 1 ).

Figure 1

Schematic representation for the systematic review of RCT evidence. Studies identified (i1) were initially assessed based on title and abstract. Articles not meeting the inclusion criteria were excluded (e1) and allocated a “reason code” to document the rationale for exclusion. Articles included after this stage (i2) were then assessed based on the full text; further articles were excluded (e2), yielding the final data set for inclusion (i3). CINAHL indicates cumulative index to nursing and allied health literature; mITT, modified intention-to-treat; RCT, randomized controlled trial.

Of the 43 studies identified in the final NMA data set, 25 did not compare the European licensed dose of the drug of interest with enoxaparin 40 mg once daily or enoxaparin 30 mg twice daily and were excluded from the direct/indirect meta-analyses.

Table 3 lists the studies identified by the systematic review, the interventions evaluated, and the outcomes evaluated in the TKR and THR patient populations. All the studies were included in the NMA, and highlighted studies were included in the indirect comparison.

Table 3.

Studies Identified by the Systematic Review and Considered for the NMA (All Studies) and Indirect Comparison (Shaded), Where Interventions and Outcomes Are Evaluated

Trial	Interventions	Composite of VTE Events (C)		Major VTE (Only Indirect Comparison)		Major Bleed		CRNM Bleed		Any Bleed
Trial	Interventions	Hip	Knee	Hip	Knee	Hip	Knee	Hip	Knee	Hip	Knee
ADVANCE-3¹³	Apixaban vs enoxaparin	X		X		X		X		X
APROPOS²⁵	Apixaban vs enoxaparin vs warfarin		X		X		X		X		X
ADVANCE-1²⁶	Apixaban vs enoxaparin		X		X		X		X		X
ADVANCE-2¹⁴	Apixaban vs enoxaparin		X		X		X		X		X
Bauer et al²⁷	Fondaparinux vs enoxaparin						X
BISTRO II ²⁸	Dabigatran vs enoxaparin	X	X			X	X	X	X	X	X
Eriksson et al²⁹	Dalteparin vs UFH					(C)
Eriksson et al³⁰	Desirudin vs enoxaparin	(E)				(C)
Eriksson et al³¹	Enoxaparin vs rivaroxaban	X				X		X		X
Fuji et al³²	Enoxaparin vs placebo	X				X		X		X
Fuji et al³²	Enoxaparin vs placebo		X				X		X		X
Fuji et al³³	Fondaparinux vs placebo	X				X				X(B)
Fuji et al³³	Fondaparinux vs placebo		X				X				X
Hull et al³⁴	Dalteparin vs warfarin					X(A)				X(A)
RENOVATE-II ³⁵	Dabigatran vs enoxaparin	X		X		X				(D)
Kakkar et al³⁶	Bemiparin vs UFH					(C)
Lassen et al³⁷	Enoxaparin vs fondaparinux	(D)				X				(D)
Lassen et al³⁸	Enoxaparin vs semuloparin		(E)				X				X
Lassen et al³⁹	Enoxaparin vs semuloparin		X				X		X		X
Leclerc et al⁴⁰	Enoxaparin vs placebo
Leclerc et al⁴¹	Enoxaparin vs warfarin						X				X
Levine et al⁴²	Enoxaparin vs heparin					X				X
Mouret et al⁴³	Enoxaparin vs semuloparin	X				X		X		X
Navarro-Quilis et al⁴⁴	Bemiparin vs enoxaparin		X		X		X				X
ODIXa-HIP Study ⁴⁵	Enoxaparin vs rivaroxaban	X				X		X		X
Pentathlon et al⁴⁶	Enoxaparin vs fondaparinux	X				X				X
Planes et al TRIAL 2 ⁴⁷	Enoxaparin
Planes et al TRIAL 3 ⁴⁷	Enoxaparin vs UFH					X(B)				X
Planes et al⁴⁸	Enoxaparin vs reviparin					X				X
Planes et al⁴⁹	Enoxaparin vs tinzaparin					X				X
RECORD 1 ⁵	Enoxaparin vs rivaroxaban	X		X		X		X		X
RECORD 2 ⁸	Enoxaparin vs rivaroxaban	X		X		X		X		X
RECORD 3 ⁹	Enoxaparin vs rivaroxaban		X		X		X		X		X
RECORD 4 ¹⁰	Enoxaparin vs rivaroxaban		X		X		X		X		X
REMOBILIZE ⁵⁰	Dabigatran vs enoxaparin		X		X		X		X		X
REMODEL ⁶	Dabigatran vs enoxaparin		X		X		X		X		X
RENOVATE ⁷	Dabigatran vs enoxaparin	X		X		X		X		X
Samama et al⁵¹	Enoxaparin vs placebo	X				X				X
Spiro et al⁵²	Enoxaparin					X				X
Torholm et al⁵³	Dalteparin vs placebo
Turpie⁵⁴	Enoxaparin vs placebo					X				X
Turpie et al⁵⁵	Enoxaparin vs fondaparinux	(D)				X				(D)
Turpie et al⁵⁶	Enoxaparin vs rivaroxaban		X				X		X		X

Abbreviations: CRNM, clinically relevant nonmajor; UFH, unfractionated heparin; VTE, venous thromboembolism; NMA, network meta-analysis.

(A), Although study reported relevant outcome, the treatment arms were not connected to the network, and therefore, the study was not included in the NMA; (B), Removed from NMA network due to poor fit/convergence that resulted from zero events being reported in one or more arms of the trial; (C), Using any of the following composite outcomes: VTE = any DVT + symptomatic nonfatal PE + any death; VTE = any DVT + nonfatal PE + any death; VTE = any DVT + symptomatic PE + any death; VTE = total VTE + any death; VTE = DVT + PE, where no deaths are reported in the study; (D), Data not used as composite VTE = DVT + PE, where deaths are reported in the study; (E), Other VTE composite.

The definition of major bleeding varied across studies (Table 4 ). Twenty-two studies reported a comparable major bleeding outcome in the THR patient population and 16 studies in the TKR patient population. All other studies did not report a major bleeding outcome. There was a slight variation in the definition of CRNM bleeding across studies (Table 5 ). Nine studies in the THR patient population and 11 in the TKR patient population reported a comparable CRNM bleeding outcome. Any bleeding was not reported as a distinct end point in all studies, therefore for these studies, the any bleeding end point used the reported major/minor/CRNM end points. If CRNM bleeding was reported as a distinct category of bleeding from minor bleeding, then major, minor, and CRNM bleeds were totaled, otherwise major and minor bleeds were totaled (a summary of the “any bleeding” definition is available as an online Supplement).

Table 4.

Definition of Major Bleeding Across Included Studies

Study	Indication	Fatal Bleeding	In a Critical Organ	Decrease in Hemoglobin ≥2 g/dL or Transfusion ≥2 Units	Requires Reintervention	Clinically Overt	Requiring Treatment Cessation	Other Definition/Inclusion	Location of bleeding
Study	Indication	Fatal Bleeding	In a Critical Organ	Decrease in Hemoglobin ≥2 g/dL or Transfusion ≥2 Units	Requires Reintervention	Clinically Overt	Requiring Treatment Cessation	Other Definition/Inclusion	Intracranial	Retroperitoneal	Intraocular	Intraspinal	Pericardial	Into Operated Joint	Intramuscular
ADVANCE-3 ¹³	THR	✓	✓	✓	✓	✓			✓	✓	✓	✓	✓	✓	✓
BISTRO II Trial ²⁸	THR	✓	✓	✓	✓	✓	✓		✓	✓	✓	✓
BISTRO II Trial ²⁸	TKR	✓	✓	✓	✓	✓	✓		✓	✓	✓	✓
Eriksson et al³¹	THR	✓	✓	✓	✓	✓	✓	✓_A	✓	✓	✓	✓
Fuji et al³²	THR	✓	✓	✓	✓				✓	✓	✓
Fuji et al³²	TKR	✓	✓	✓	✓				✓	✓	✓
RENOVATE-II ³⁵	THR	✓	✓	✓	✓	✓	✓		✓	✓	✓	✓
Lassen et al³⁷	THR	✓	✓	✓	✓	✓			✓	✓	✓	✓	✓
APROPOS ²⁵	TKR	✓	✓	✓	✓	✓	✓		✓	✓	✓	✓	✓	✓
ADVANCE-1 ²⁶	TKR	✓	✓	✓	✓	✓			✓	✓	✓	✓	✓	✓	✓
ADVANCE-2 ¹⁴	TKR	✓	✓	✓	✓	✓			✓	✓	✓	✓	✓	✓	✓
Navarro-Quilis et al⁴⁴	TKR	✓	✓	✓			✓		✓	✓
ODIXa-HIP Study ⁴⁵	THR	✓	✓	✓	✓	✓	✓		✓	✓	✓	✓	✓
Pentathlon et al⁴⁶	THR	✓	✓	✓	✓	✓			✓	✓	✓	✓	✓	✓
RECORD 1 ⁵	THR	✓	✓	✓	✓	✓			✓	✓	✓	✓	✓	✓
RECORD 3 ⁹	TKR	✓	✓	✓	✓	✓			✓	✓	✓	✓	✓		✓
RECORD 4 ¹⁰	TKR	✓	✓	✓	✓	✓			✓	✓	✓	✓	✓		✓
RECORD 2 ⁸	THR	✓	✓	✓	✓	✓			✓	✓	✓	✓	✓		✓
RE-MOBILIZE ⁵⁰	TKR	✓	✓	✓	✓	✓	✓		✓	✓	✓	✓
RE-MODEL ⁶	TKR	✓	✓	✓	✓	✓	✓		✓	✓	✓	✓
RENOVATE ⁷	THR	✓	✓	✓	✓	✓	✓		✓	✓	✓	✓
Samama et al⁵¹	THR	✓	✓	✓	✓	✓			✓	✓
Spiro et al⁵²	THR	✓	✓	✓		✓			✓	✓
Turpie et al⁵⁵	THR	✓	✓	✓	✓	✓			✓	✓	✓	✓	✓
Turpie et al⁵⁶	TKR	✓	✓	✓	✓	✓	✓		✓	✓	✓	✓	✓
Fuji et al³³	THR	✓	✓		✓	✓			✓	✓		✓
Fuji et al³³	TKR	✓	✓		✓	✓			✓	✓		✓
Lassen et al³⁸	TKR	✓	✓		✓	✓			✓	✓	✓	✓	✓	✓	✓
Bauer et al²⁷	TKR	✓	✓			✓	✓			✓	✓		✓
Hull et al³⁴	THR		✓	✓		✓			✓	✓	✓	✓	✓	✓
Leclerc et al⁴¹	TKR			✓		✓	✓	✓_B						✓
Levine et al⁴²	THR		✓	✓		✓			✓	✓
Planes et al⁴⁸	THR		✓	✓		✓	✓		✓	✓
Planes et al⁴⁹	THR		✓	✓		✓			✓	✓	✓
Turpie 1990 ⁵⁴	THR			✓		✓
Lassen, et al³⁹	TKR							✓_C
Mouret et al⁴³	THR							✓_C
Planes et al TRIAL 3 ⁴⁷	THR							✓_C
Studies not included in major bleeding NMA analysis
Kakkar, et al³⁶	THR						✓
Eriksson et al³⁰	THR							✓_D
Eriksson et al²⁹	THR							✓_E

Abbreviations: THR, total hip replacement; TKR, total knee replacement; NMA, network meta-analysis.

A, Starting ≥6 hours after surgery, or after the first dose of study medication (whichever came first), but not >2 days after the last administration of study drug; B, Interruption of physiotherapy for at least 24 hours; C, Not reported; D, Serious bleeding—perioperative transfusion of more than 5 U whole blood, red-cell concentrate, or both; transfusion of >7 U whole blood, red-cell concentrate, or both at any time; or transfusion of a total of >3500 mL of blood. E, excessive bleeding (>3000 mL).

Table 5.

Definition of CRNM Bleeding Across Included Studies

Study	Indication	Multiple Source Bleeding	hematoma >25 cm²	Excessive Wound Hematoma	Epistaxis >5 Minutes	Macroscopic Hemapturia (Spontaneous, or Lasting >24 hours If Associated With an Intervention)	spontaneous Rectal Bleeding	Gingival Bleeding >5 minutes	Hemoptysis	Hematemesis	Prolonged Bleeding (>5 minutes) After Venipuncture	Other Definition
ADVANCE-3 ¹³	THR			✓	✓	✓			✓
BISTRO II ²⁸	TKR		✓	✓	✓	✓	✓	✓
BISTRO II ²⁸	THR		✓	✓	✓	✓	✓	✓
Eriksson et al³¹	THR	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓
Fuji et al³²	THR				✓							A
Fuji et al³²	TKR				✓							A
APROPOS ²⁵	TKR											B
ADVANCE-1 ²⁶	TKR			✓	✓	✓			✓
ADVANCE-2 ¹⁴	TKR			✓	✓	✓			✓
ODIXa-HIP Study ⁴⁵	THR	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓
RECORD 4 ¹⁰	TKR	✓	✓	✓	✓	✓	✓	✓	✓	✓
RECORD 2 ⁸	THR	✓	✓	✓
RE-MOBILIZE ⁵⁰	TKR		✓	✓		✓	✓	✓
RE-MODEL ⁶	TKR		✓	✓	✓	✓	✓	✓
RENOVATE ⁷	THR		✓	✓	✓	✓	✓	✓
Turpie et al⁵⁶	TKR	✓	✓	✓	✓	✓
Lassen, et al³⁹	TKR											NR
Mouret et al⁴³	THR											NR
RECORD 1 ⁵	THR											NR
RECORD 3 ⁹	TKR											NR

Abbreviations: THR, total hip replacement; TKR, total knee replacement; NMA, network meta-analysis, CRNM, clinically relevant nonmajor; NR, not reported.

A, Defined as having ≥ 1 of the following features; epistaxis lasting >5 minutes or requiring intervention; ecchymosis or hematoma with a maximum size of 5 cm; hematuria not associated with urinary catheter trauma; gastrointestinal hemorrhage not related to intubation or a nasogastric tube; wound hematoma or hemorrhagic wound complications not associated with major hemorrhage; or subconjunctival hemorrhage requiring cessation of medication; B, Potentially significant nonovert bleeding was defined in the absence of clinically overt bleeding as either a decrease in hemoglobin of ≥ 4 g/dL (relative to postsurgery values) coupled with discontinuation of study medication due to a suspected bleeding event, or transfusion of ≥2 U whole blood product coupled with discontinuation of study medication due to a suspected bleeding event.

Data used in the meta-analyses are available as an online supplement.

Direct/Indirect Comparison Results

Results of the direct/indirect comparisons are summarized in Table 6 .

Table 6.

Results of the Direct and Indirect Comparisons by Outcome in the THR and TKR Populations

THR Population		TKR Population
VTE composite (evaluable population analysis)
Direct comparison		Direct comparison
Treatment vs enoxaparin 40 mg pooled	OR (95% CI)	Treatment vs enoxaparin 40 mg pooled	OR (95% CI)
Apixaban 2.5 mg bid	0.35 (0.224-0.546)	Apixaban 2.5 mg bid	0.55 (0.438-0.691)
Rivaroxaban 10 mg od	0.24 (0.16-0.36)	Bemiparin 3500 units	0.809 (0.514-1.272)
Rivaroxaban 10 mg od (excluding RECORD 2)	0.3 (0.17-0.5)	Rivaroxaban 10 mg od	0.455 (0.341-0.606)
Rivaroxaban 10 mg od (short duration)	0.352 (0.168-0.738)	Dabigatran etexilate 150 mg od	1.125 (0.876-1.444)
Dabigatran etexilate 150 mg od	1.309 (0.92-1.864)	Dabigatran etexilate 220 mg od	0.945 (0.733-1.22)
Dabigatran etexilate 220 mg od	0.879 (0.677-1.142)
Indirect comparison (via enoxaparin 40 mg pooled)		Indirect comparison (via enoxaparin 40 mg pooled)
Treatment vs apixaban 2.5 mg bid	OR (95% CI)	Treatment vs apixaban 2.5 mg bid	OR (95% CI)
Rivaroxaban 10 mg od	0.686 (0.375-1.253)	Bemiparin 3500 units	1.47 (0.886-2.44)
Rivaroxaban 10 mg od (excluding RECORD 2)	0.857 (0.416-1.765)	Rivaroxaban 10 mg od	0.827 (0.573-1.192)
Rivaroxaban 10 mg od (short duration)	1.006 (0.424-2.386)	Dabigatran etexilate 150 mg od	2.044 (1.459-2.865)
Dabigatran etexilate 150 mg od	3.74 (2.118-6.605)	Dabigatran etexilate 220 mg od	1.718 (1.221-2.417)
Dabigatran etexilate 220 mg od	2.511 (1.497-4.212)
Major VTE (evaluable population analysis)
Direct comparison		Direct comparison
Treatment vs enoxaparin 40 mg pooled	OR (95% CI)	Treatment vs enoxaparin 40 mg pooled	OR (95% CI)
Apixaban 2.5 mg bid	0.397 (0.19-0.828)	Apixaban 2.5 mg bid	0.496 (0.254-0.97)
Dabigatran etexilate 220 mg od	0.659 (0.45-0.964)	Bemiparin 3500 units	0.327 (0.087-1.231)
Rivaroxaban 10 mg od	0.118 (0.061-0.227)	Rivaroxaban 10 mg od	0.376 (0.174-0.813)
Rivaroxaban 10 mg od (excluding RECORD 2)	0.12 (0.04-0.34)	Dabigatran etexilate 150 mg od	1.08 (0.565-2.067)
Dabigatran etexilate 150 mg od	1.094 (0.687-1.743)	Dabigatran etexilate 220 mg od	0.722 (0.35-1.49)
Indirect comparison (via enoxaparin 40 mg pooled)		Indirect comparison (via enoxaparin 40 mg pooled)
Treatment vs apixaban 2.5 mg bid	OR (95% CI)	Treatment vs apixaban 2.5 mg bid	OR (95% CI)
Dabigatran etexilate 220 mg od	1.66 (0.724-3.806)	Bemiparin 3500 units	0.659 (0.149-2.911)
Rivaroxaban 10 mg od	0.297 (0.111-0.799)	Rivaroxaban 10 mg od	0.757 (0.272-2.105)
Rivaroxaban 10 mg od (excluding RECORD 2)	0.302 (0.081-1.135)	Dabigatran etexilate 150 mg od	2.177 (0.856-5.526)
Dabigatran etexilate 150 mg od	2.756 (1.153-6.587)	Dabigatran etexilate 220 mg od	1.456 (0.542-3.906)
Major bleed (ITT analysis)
Direct comparison		Direct comparison
Treatment vs enoxaparin 40 mg pooled	OR (95% CI)	Treatment vs enoxaparin 40 mg pooled	OR (95% CI)
Apixaban 2.5 mg bid	1.22 (0.65-2.28)	Apixaban 2.5 mg bid	0.64 (0.28-1.49)
Fondaparinux 2.5 mg od	1.49 (0.94-2.35)	Bemiparin 3500 units	1.01 (0.2-5.04)
Dabigatran etexilate 220 mg od	1.38 (0.83-2.27)	Rivaroxaban 10 mg od	1.19 (0.4-3.55)
Rivaroxaban 10 mg od (short duration)	0.36 (0.04-3.48)	Dabigatran etexilate 150 mg od	0.99 (0.39-2.5)
Tinzaparin 4500 units +Ext <1 week	0.49 (0.09-2.7)	Dabigatran etexilate 220 mg od	1.12 (0.45-2.78)
Rivaroxaban 10 mg od	2.29 (0.57-9.17)
Rivaroxaban 10 mg od (excluding RECORD 2)	3.02 (0.61-14.96)
Dabigatran etexilate 150 mg od	0.82 (0.41-1.64)
Indirect comparison (via enoxaparin 40 mg pooled)		Indirect comparison (via enoxaparin 40 mg pooled)
Treatment vs apixaban 2.5 mg bid	OR (95% CI)	Treatment vs apixaban 2.5 mg bid	OR (95% CI)
Fondaparinux 2.5 mg od	1.221 (0.56-2.665)	Bemiparin 3500 units	1.578 (0.256-9.722)
Dabigatran etexilate 220 mg od	1.131 (0.504-2.541)	Rivaroxaban 10 mg od	1.859 (0.473-7.304)
Rivaroxaban 10 mg od (short duration)	0.295 (0.03-2.901)	Dabigatran etexilate 150 mg od	1.547 (0.445-5.375)
Tinzaparin 4500 units +Ext <1 week	0.402 (0.066-2.449)	Dabigatran etexilate 220 mg od	1.75 (0.511-5.992)
Rivaroxaban 10 mg od	1.877 (0.408-8.639)
Rivaroxaban 10 mg od (excluding RECORD 2)	2.475 (0.444-13.81)
Dabigatran etexilate 150 mg od	0.672 (0.263-1.715)
CRNM bleed (ITT analysis)
Direct comparison		Direct comparison
Treatment vs enoxaparin 40 mg pooled	OR (95% CI)	Treatment vs enoxaparin 40 mg pooled	OR (95% CI)
Apixaban 2.5 mg bid	0.9 (0.69-1.17)	Apixaban 2.5 mg bid	0.75 (0.5-1.12)
Rivaroxaban 10 mg od (short duration)	0.65 (0.15-2.75)	Rivaroxaban 10 mg od	1.21 (0.72-2.01)
Rivaroxaban 10 mg od	1.22 (0.91-1.62)	Dabigatran etexilate 150 mg od	1.3 (0.84-2.02)
Rivaroxaban 10 mg od (excluding RECORD 2)	1.21 (0.84-1.75)	Dabigatran etexilate 220 mg od	1.09 (0.69-1.73)
Dabigatran etexilate 150 mg od	1.38 (0.91-2.09)
Dabigatran etexilate 220 mg od	1.21 (0.79-1.86)
Indirect comparison (via enoxaparin 40 mg pooled)		Indirect comparison (via enoxaparin 40 mg pooled)
Treatment vs apixaban 2.5 mg bid	OR (95% CI)	Treatment vs apixaban 2.5 mg bid	OR (95% CI)
Rivaroxaban 10 mg od (short duration)	0.722 (0.163-3.205)	Rivaroxaban 10 mg od	1.613 (0.835-3.117)
Rivaroxaban 10 mg od	1.356 (0.913-2.013)	Dabigatran etexilate 150 mg od	1.733 (0.955-3.145)
Rivaroxaban 10 mg od (excluding RECORD 2)	1.344 (0.856-2.112)	Dabigatran etexilate 220 mg od	1.453 (0.789-2.678)
Dabigatran etexilate 150 mg od	1.533 (0.936-2.513)
Dabigatran etexilate 220 mg od	1.344 (0.814-2.222)
Any bleed (ITT analysis)
Direct comparison		Direct comparison
Treatment vs enoxaparin 40 mg pooled	OR (95% CI)	Treatment vs enoxaparin 40 mg pooled	OR (95% CI)
Apixaban 2.5 mg bid	0.93 (0.79-1.09)	Apixaban 2.5 mg bid	0.81 (0.62-1.07)
Rivaroxaban 10 mg od (short duration)	0.68 (0.29-1.62)	Bemiparin 3500 units	1.01 (0.35-2.92)
Tinzaparin 4500 units + Ext <1 week	0.57 (0.29-1.1)	Rivaroxaban 10 mg od	1.02 (0.71-1.47)
Rivaroxaban 10 mg od	1.08 (0.89-1.32)	Dabigatran etexilate 150 mg od	1 (0.75-1.32)
Rivaroxaban 10 mg od (excluding RECORD 2)	1.02 (0.8-1.31)	Dabigatran etexilate 220 mg od	0.96 (0.72-1.27)
Dabigatran etexilate 150 mg od	1.07 (0.83-1.38)
Dabigatran etexilate 220 mg od	1.08 (0.84-1.39)
Indirect comparison (via enoxaparin 40 mg pooled)		Indirect comparison (via enoxaparin 40 mg pooled)
Treatment vs apixaban 2.5 mg bid	OR (95% CI)	Treatment vs apixaban 2.5 mg bid	OR (95% CI)
Rivaroxaban 10 mg od (short duration)	0.731 (0.307-1.741)	Bemiparin 3500 units	1.247 (0.418-3.72)
Tinzaparin 4500 units +Ext <1 week	0.613 (0.306-1.228)	Rivaroxaban 10 mg od	1.259 (0.803-1.975)
Rivaroxaban 10 mg od	1.161 (0.902-1.496)	Dabigatran etexilate 150 mg od	1.235 (0.834-1.828)
Rivaroxaban 10 mg od (excluding RECORD 2)	1.097 (0.819-1.47)	Dabigatran etexilate 220 mg od	1.185 (0.8-1.755)
Dabigatran etexilate 150 mg od	1.151 (0.851-1.556)
Dabigatran etexilate 220 mg od	1.161 (0.861-1.567)

Abbreviations: bid, twice daily; CI, confidence interval; Ext <1 week, length of treatment was longer than UK indication and extension was less than 7 days; Ext > 1 week, length of treatment was longer than UK indication and extension was 7 days or more; ITT, intention-to-treat; od, once daily; THR, total hip replacement; TKR, total knee replacement; VTE, venous thromboembolism; OR, odds ratio.

All VTE and All-Cause Death

Results are shown in Table 6 and Figure 2 . In direct comparisons, apixaban 2.5 mg twice daily was significantly better than enoxaparin 40 mg once daily for the prevention of the primary efficacy outcome “all VTE and all-cause death,” in both patients with THR (OR, 0.35; 95% confidence interval [CI], 0.224-0.546) and TKR (OR, 0.55; 95% CI, 0.438-0.691). Similarly, rivaroxaban 10 mg once daily was superior to enoxaparin for the prevention of all VTE and all-cause death for both THR surgery (OR, 0.24; 95% CI, 0.16-0.36) and TKR surgery (OR, 0.455; 95% CI, 0.341-0.606). Dabigatran etexilate 220 mg once daily showed similar efficacy to enoxaparin 40 mg once daily for this outcome for the THR (OR, 0.879; 95% CI, 0.677-1.142) and TKR (OR, 0.945; 95% CI, 0.733-1.22) patient populations.

Figure 2

Pooled estimates of the results of randomized controlled trials comparing the effects of apixaban, rivaroxaban, and dabigatran versus enoxaparin on venous thromboembolism (VTE) composite outcome for patients undergoing (A) total hip replacement and (B) total knee replacement.

The adjusted indirect comparison found that apixaban 2.5 mg twice daily was significantly better than dabigatran 220 mg once daily for the prevention of all VTE and all-cause death in patients with THR (OR dabigatran vs apixaban, 2.511; 95% CI, 1.497-4.212), and TKR (OR dabigatran vs apixaban, 1.718; 95% CI, 1.221-2.417). Rivaroxaban 10 mg once daily showed similar efficacy to apixaban 2.5 mg twice daily in both the patients with THR and TKR (OR, 0.686; 95% CI, 0.375-1.253 and OR, 0.827; 95% CI, 0.573-1.192, respectively).

Major VTE

Direct ORs (Table 6 ) suggest that apixaban 2.5 mg once daily (THR and TKR patient populations), dabigatran etexilate 220 mg once daily (THR patient population), and rivaroxaban 10 mg once daily (THR and TKR patient populations) are superior to enoxaparin 40 mg for the prevention of major VTE; and that bemiparin 3500 units (TKR patient population) has similar efficacy.

The adjusted indirect comparison suggested that rivaroxaban 10 mg once daily was superior to apixaban 2.5 mg once daily for the prevention of major VTE in the THR patient population (OR, 0.297; 95% CI, 0.111-0.799); however, on removal of the RECORD 2 study, this difference was not significant (OR, 0.302; 95% CI, 0.081-1.135). Apixaban 2.5 mg once daily was similar to dabigatran etexilate 220 mg once daily (THR and TKR patient populations), rivaroxaban 10 mg once daily (TKR patient population), and bemiparin 3500 units (TKR patient population) for this outcome.

Outcomes of Bleeding

For all outcomes of bleeding (major bleed, CRNM bleed, and any bleed), no significant differences between the treatments evaluated were observed in direct and indirect comparisons in either THR or TKR patient populations (Table 6 and Figure 3 ).

Figure 3

Pooled estimates of the results of randomized controlled trials comparing the effects of apixaban, rivaroxaban, and dabigatran versus enoxaparin on major bleeding for patients undergoing (A) total hip replacement and (B) total knee replacement.

Results of NMA

Results of the NMA are summarized in Table 7 . An example of the network of evidence for the VTE composite outcome in THR surgery is shown in Figure 4 . Due to the small number of studies reporting major VTE, an NMA was not possible for this outcome. In general, for all outcomes, there were few significant differences reported between the treatments. The wide credibility intervals were in contrast to the narrower CIs observed in the adjusted indirect comparisons for the same outcomes. The NMA results for apixaban 2.5 mg twice daily versus dabigatran 220 mg once daily for the VTE composite outcome were not in agreement with those observed in the adjusted indirect comparison. Furthermore, for the primary efficacy outcome, all VTE and all-cause death, the NMA results of apixaban 2.5 mg twice daily versus enoxaparin 40 mg once daily and rivaroxaban 10 mg once daily versus enoxaparin 40 mg once daily were inconsistent with the direct head-to-head RCT comparisons of these treatments.

Table 7.

Results From the NMA Base Model in the THR and TKR Populations

THR Indication		TKR Indication
VTE composite (evaluable population analysis)
Treatment vs enoxaparin 40 mg Ext >1 week	Median OR (CrI)	Treatment vs enoxaparin 30 mg bid	Median OR (CrI)
Enoxaparin 40 mg	0.625 (0.01-50.26)	Enoxaparin 40 mg pooled	1.563 (0.322-4.602)
Enoxaparin 40 mg +Ext <1 week	1.551 (0.084-30.21)	Rivaroxaban 10 mg od	0.686 (0.183-2.007)
Enoxaparin 30 mg bid	0.921 (0.014-111.5)	Fondaparinux 2.5 mg od+Ext <1 week	0.528 (0.032-5.122)
Rivaroxaban 10 mg od	0.29 (0.036-2.477)	Apixaban 2.5 mg bid	0.873 (0.278-1.942)
Fondaparinux 2.5 mg od	0.295 (0.005-28.25)	Dabigatran etexilate 150 mg od +Ext <1 week	1.506 (0.364-6.531)
Apixaban 2.5 mg bid	0.344 (0.038-3.114)	Dabigatran etexilate 220 mg od +Ext <1 week	1.339 (0.304-5.658)
Dabigatran etexilate 150 mg od	1.28 (0.176-10.14)	Dabigatran etexilate 150 mg od	1.761 (0.184-9.636)
Dabigatran etexilate 220 mg od	0.887 (0.199-4.11)	Dabigatran etexilate 220 mg od	1.478 (0.159-7.993)
Rivaroxaban 10 mg od (short duration)	0.22 (0.002-24.75)	Bemiparin 3500 units	1.244 (0.147-6.958)
Any major bleed (ITT analysis)
Treatment vs enoxaparin 40 mg +Ext >1 week	Median OR (CrI)	Treatment vs enoxaparin 30 mg bid	Median OR (CrI)
Enoxaparin 40 mg	0.006 (0.0000004-9.246)	Enoxaparin 40 mg pooled	0.853 (0.064-5.296)
Enoxaparin 40 mg +Ext <1 week	4.724 (0.108-474.7)	Rivaroxaban 10 mg od	1.655 (0.181-10.59)
Enoxaparin 30 mg bid	0.006 (0.0000005-9.036))	Fondaparinux 2.5 mg od+Ext 1 week	4.969 (0.01-1404)
Rivaroxaban 10 mg od	3.628 (0.533-40.78)	Apixaban 2.5 mg bid	0.497 (0.063-2.854)
Fondaparinux 2.5 mg od	0.011 (0.000001-17.67)	Dabigatran etexilate 150 mg od +Ext <1 week	0.389 (0.035-4.153)
Apixaban 2.5 mg bid	1.209 (0.241-6.092)	Dabigatran etexilate 220 mg od +Ext <1 week	0.399 (0.038-4.525)
Dabigatran etexilate 150 mg od	0.875 (0.194-4.197)	Dabigatran etexilate 150 mg od	0.849 (0.023-13.68)
Dabigatran etexilate 220 mg od	1.414 (0.456-4.612)	Dabigatran etexilate 220 mg od	0.99 (0.026-16.05)
Tinzaparin 4500 units +Ext <1 week	2.192 (0.022-337.3)	Bemiparin 3500 units	0.824 (0.018-20.96)
Rivaroxaban 10 mg od (short duration)	0.001 (0.0000001-4.217)	Fondaparinux 2.5 mg od	17.32 (0.987-729.8)
Any CRNM bleed (ITT analysis)
Treatment vs enoxaparin 40 mg +Ext >1 week	Median OR (CrI)	Treatment vs enoxaparin 30 mg bid	Median OR (CrI)
Rivaroxaban 10 mg od	1.217 (0.333-4.534)	Enoxaparin 40 mg pooled dose	0.94 (0.111-5.561)
Apixaban 2.5 mg bid	0.899 (0.15-5.608)	Rivaroxaban 10 mg od	1.227 (0.189-6.518)
Dabigatran etexilate 150 mg od	1.397 (0.216-9.106)	Apixaban 2.5 mg bid	0.663 (0.086-2.169)
Dabigatran etexilate 220 mg od	1.217 (0.187-7.637)	Dabigatran etexilate 150 mg od+Ext <1 week	1.083 (0.13-8.518)
Rivaroxaban 10 mg od (short duration)	NA	Dabigatran etexilate 220 mg od +Ext <1 week	1.12 (0.142-8.387)
		Dabigatran etexilate 150 mg od	1.259 (0.061-17.02)
		Dabigatran etexilate 220 mg od	1.012 (0.047-15.19)
Any bleed (ITT analysis)
Treatment vs enoxaparin 40 mg + Ext >1 week	Median OR (CrI)	Treatment vs enoxaparin 30 mg bid	Median OR (CrI)
Enoxaparin 40 mg other pooled^a	0.806 (0.071-9.088)	Enoxaparin 40 mg pooled	1.039 (0.592-2.116)
Rivaroxaban 10 mg od	1.01 (0.152-6.449)	Apixaban 2.5 mg bid	0.798 (0.464-1.341)
Apixaban 2.5 mg bid	0.92 (0.133-6.889)	Rivaroxaban 10 mg od	1.102 (0.621-2.101)
Dabigatran etexilate 150 mg od	1.081 (0.151-7.694)	Dabigatran etexilate 150 mg od +Ext <1 week	0.804 (0.357-1.913)
Dabigatran etexilate 220 mg od	1.082 (0.151-7.694)	Dabigatran etexilate 220 mg od +Ext <1 week	0.846 (0.374-1.967)
Enoxaparin 30 mg bid	0.81 (0.042-12.93)	Dabigatran etexilate 150 mg od	1.041 (0.443-2.889)
Tinzaparin 4500 units +Ext <1 week	0.439 (0.019-10.87)	Dabigatran etexilate 220 mg od	1.003 (0.419-2.805)
Rivaroxaban 10 mg od (short duration)	0.559 (0.026-12.84)	Bemiparin 3500 units	1.023 (0.259-4.6)
		Fondaparinux 2.5 mg od +Ext <1 week	0.881 (0.091-7.568)

Abbreviations: CrI, credible interval; the Bayesian equivalent of a confidence interval; Ext <1 week, length of treatment was longer than UK indication, extension was < 7 days; Ext >1 week, length of treatment was longer than UK indication, extension was 7 days or more; NA, not applicable; OR, odds ratio; ITT, intention-to-treat; od, once daily; THR, total hip replacement; TKR, total knee replacement; VTE, venous thromboembolism NMA, network meta-analysis; bid, twice daily; CRNM, clinically relevant nonmajor.

^aIn order to connect network, enoxaparin 40 mg od, where length of treatment as indication pooled with enoxaparin 40 mg od + Ext <1 week.

Figure 4

Network diagram for venous thromboembolism (VTE) composite outcome in the hip indication.

Discussion

In the absence of head-to-head RCT evidence for the new oral anticoagulants apixaban, rivaroxaban, and dabigatran, an adjusted indirect comparison with enoxaparin 40 mg once daily as the common comparator and an NMA were conducted to determine efficacy and safety effect sizes for these treatments.

The adjusted indirect comparison found that in both the THR and TKR patient populations, apixaban was superior to dabigatran etexilate and similar to rivaroxaban for the prevention of the primary efficacy outcome all VTE and all-cause death (Table 6 ). Apixaban was similar to dabigatran and rivaroxaban for the prevention of major VTE, in both THR and TKR patient populations.

The incidence of PE events was similar for apixaban, dabigatran etexilate, and rivaroxaban; however, the number of PE events across all the treatments was small (data shown in online Supplement Appendix).

The incidence of all bleeding events (minor, major, and CRNM) was comparable for apixaban, rivaroxaban, and dabigatran etexilate, in both THR and TKR patient populations. In the THR patient population, apixaban and fondaparinux had a similar incidence of major bleeding.

The results from the NMA and the adjusted indirect comparison were inconsistent for some treatment comparisons. Wide credibility intervals around treatment differences for some outcomes were observed in the NMA, which may have been due to the large number of trials contributing to the enoxaparin 40 mg once daily node within the NMA network. The NMA allows for more sources of uncertainty, which explains in part the wider credibility intervals. Some of the trials included in the NMA were of lower quality than the recent apixaban, rivaroxaban, dabigatran, and fondaparinux RCTs included in the main pairwise meta-analyses and adjusted indirect comparisons, with fewer study quality criteria reported. Many had fewer than 100 patients per treatment arm, and many compared enoxaparin 40 mg once daily against treatments not meeting the defined criteria of this review. All these factors could have contributed to a lack of precision and an increase in uncertainty in the relative treatment effects for enoxaparin 40 mg once daily observed in the NMA results, despite the apparent increase in power (ie, more eligible studies) afforded by the NMA study inclusion criteria.

In contrast, the adjusted indirect comparison approach, although restricting the number of studies for inclusion to those possessing a common comparator (ie, enoxaparin 40 mg once daily), may allow for more precision in the relative treatment effect estimates of interest. This is because included studies tended to report and fulfill more study quality criteria, have larger patient numbers (all in excess of 600 patients randomized per arm), and report similar outcome definitions and measures. However, it should be noted that some outcome definitions were inconsistent across included trials, particularly for the outcomes of bleeding (Tables 4 and 5).

The results of the direct and indirect comparisons are consistent with those observed in a recent meta-analysis and adjusted indirect comparison of rivaroxaban and dabigatran for the prevention of VTE.⁵⁷ The study found that for the prevention of VTE (for THR and TKR patient populations pooled), rivaroxaban was superior to enoxaparin (relative risk [RR], 0.56; 95% CI, 0.43-0.73, P < .0001), and dabigatran was not superior to enoxaparin (RR, 1.12; 95% CI, 0.97-1.29, P = .12),⁵⁷ thus in line with the conclusions of the current study. Adjusted indirect comparison showed that rivaroxaban was superior to dabigatran in preventing VTE; RR, 0.50 (95% CI, 0.37-0.68).⁵⁷

NMA and indirect comparison allow the efficacy of interventions to be compared in the absence of head-to-head evidence, which can be invaluable to health care providers, physicians, and patients for decision making when a number of new effective interventions are available. A limitation of meta-analysis is the underlying assumption that trials and outcomes are sufficiently similar to allow for data to be pooled. In particular, it assumes that all the studies relate to the same patient population and that any differences in study design, inclusion criteria, or baseline characteristics will not influence the relative efficacy of the various treatments.¹⁶ In the current analysis, methods for diagnosing DVT and PE and definitions of outcomes of bleeding may not have been comparable between trials (Tables 4 and 5). In addition, trials may have used different follow-up times or patient populations may have differed in the baseline risk of DVT.

In clinical trials, the new oral anticoagulants apixaban, rivaroxaban, and dabigatran etexilate have demonstrated similar or improved efficacy and similar safety compared with current therapies, such as LMWH, for the prevention of VTE in patients undergoing orthopedic surgery.^{5-10,13,14,25,26} The adjusted indirect comparison results reported here reflect the observations made in the clinical trials. Results of the adjusted indirect comparison also suggest that apixaban has an efficacy profile that is better than that of dabigatran etexilate and similar to that of rivaroxaban, while all 3 have a comparable safety profile. In addition to improved efficacy compared with current therapies, apixaban, dabigatran etexilate, and rivaroxaban have the advantage of oral administration and do not require routine laboratory monitoring.⁵⁸ Apixaban and rivaroxaban have a further benefit of not requiring dose adjustment for moderate or severe renal impairment or patient weight.⁵⁹

Considering practical and economic advantages, such as the ease of oral dosing and the substantial reduction in costs related to this, these new anticoagulants represent an appealing alternative to conventional thromboprophylaxis regimens in patients undergoing THR and TKR surgery and may improve patient compliance and standard of care.

Footnotes

Authors– Note

N.M. and D.R. are the employees of Pfizer Ltd. P.D. and T.S. are the employees of Bristol Myers Squibb. A.C., S.M., M.O. and A.S. were paid consultants to Bristol Myers Squibb and Pfizer Inc. in connection with conducting the meta-analysis and development of the manuscript.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: funded by Bristol Myers Squibb and Pfizer Inc.

References

House of Commons Health Committee. The Prevention of Venous Thromboembolism in Hospitalised Patients. 2nd Report of Session 2004-05. HC99. London, UK: The Stationary Office Limited; 2005.

Deitelzweig

McKean

Amin

Brotman

Jaffer

Spyropoulos

. Prevention of venous thromboembolism in the orthopedic surgery patient. Cleve Clin J Med. 2008;75(suppl 3):S27–S36.

Geerts

Bergqvist

Pineo

, et al. Prevention of venous thromboembolism: American college of chest physicians evidence-based clinical practice guidelines (8th Edition). Chest. 2008;133(6 suppl):381S–453S.

National Institute for Health and Clinical Excellence. NICE Clinical Guidance 92. Venous Thromboembolism: Reducing the Risk. Reducing the Risk of Venous Thromboembolism (Deep Vein Thrombosis and Pulmonary Embolism) in Patients Admitted to Hospital. London: National Institute for Health and Clinical Excellence; 2010.

Eriksson

Borris

Friedman

, et al. Rivaroxaban versus enoxaparin for thromboprophylaxis after hip arthroplasty. N Engl J Med. 2008;358(26):2765–2775.

Eriksson

Dahl

Rosencher

, et al. Oral dabigatran etexilate vs. subcutaneous enoxaparin for the prevention of venous thromboembolism after total knee replacement: the RE-MODEL randomized trial. J Thromb Haemost. 2007;5(11):2178–2185.

Eriksson

Dahl

Rosencher

, et al. Dabigatran etexilate versus enoxaparin for prevention of venous thromboembolism after total hip replacement: a randomised, double-blind, non-inferiority trial. Lancet. 2007;370(9591):949–956.

Kakkar

Brenner

Dahl

, et al. Extended duration rivaroxaban versus short-term enoxaparin for the prevention of venous thromboembolism after total hip arthroplasty: a double-blind, randomised controlled trial. Lancet. 2008;372(9632):31–39.

Lassen

Ageno

Borris

, et al. Rivaroxaban versus enoxaparin for thromboprophylaxis after total knee arthroplasty. N Engl J Med. 2008;358(26):2776–2786.

10.

Turpie

Lassen

Davidson

, et al. Rivaroxaban versus enoxaparin for thromboprophylaxis after total knee arthroplasty (RECORD 4): a randomised trial. Lancet. 2009;373(9676):1673–1680.

11.

Eriksson

Quinlan

Eikelboom

. Novel oral factor Xa and thrombin inhibitors in the management of thromboembolism. Annu Rev Med. 2011;62:41–57.

12.

Wong

Crain

Xin

, et al. Apixaban, an oral, direct and highly selective factor Xa inhibitor: in vitro, antithrombotic and antihemostatic studies. J Thromb Haemost. 2008;6(5):820–829.

13.

Lassen

Gallus

Raskob

Pineo

Chen

Ramirez

. Apixaban versus enoxaparin for thromboprophylaxis after hip replacement. N Engl J Med. 2010;363(26):2487–2498.

14.

Lassen

Raskob

Gallus

Pineo

Chen

Hornick

. Apixaban versus enoxaparin for thromboprophylaxis after knee replacement (ADVANCE-2): a randomised double-blind trial. Lancet. 2010;375(9717):807–815.

15.

Ades

Sculpher

Sutton

, et al. Bayesian methods for evidence synthesis in cost-effectiveness analysis. Pharmacoeconomics. 2006;24(1):1–19.

16.

Caldwell

Ades

Higgins

. Simultaneous comparison of multiple treatments: combining direct and indirect evidence. BMJ. 2005;331(7521):897–900.

17.

Jansen

Crawford

Bergman

Stam

. Bayesian meta-analysis of multiple treatment comparisons: an introduction to mixed treatment comparisons. Value Health. 2008;11(5):956–964.

18.

Ades

. Combination of direct and indirect evidence in mixed treatment comparisons. Stat Med. 2004;23(20):3105–3124.

19.

Harris

Bardburn

Deeks

Harbord

Altman

Sterne

. Metan: fixed-and random-effects meta-analysis. Stata J. 2008;8(1):3–28.

20.

Software updates (metan). Stata J. 2008;8(1):3–28.

21.

Bucher

Guyatt

Griffith

Walter

. The results of direct and indirect treatment comparisons in meta-analysis of randomized controlled trials. J Clin Epidemiol. 1997;50(6):683–691.

22.

Sutton

Abrams

. Bayesian methods in meta-analysis and evidence synthesis. Stat Methods Med Res. 2001;10(4):277–303.

23.

Spiegelhalter

Abrams

Myles

. Bayesian Approaches to Clinical Trials and Health Care Evaluation. Chicester, England: Wiley; 2004.

24.

The BUGS Project. http://www.mrc-bsu.cam.ac.uk/bugs/welcome.shtml. Accessed February 8, 2011.

25.

Lassen

Davidson

Gallus

Pineo

Ansell

Deitchman

. The efficacy and safety of apixaban, an oral, direct factor Xa inhibitor, as thromboprophylaxis in patients following total knee replacement. J Thromb Haemost. 2007;5(12):2368–2375.

26.

Lassen

Raskob

Gallus

Pineo

Chen

Portman

. Apixaban or enoxaparin for thromboprophylaxis after knee replacement. N Engl J Med. 2009;361(6):594–604.

27.

Bauer

Eriksson

Lassen

Turpie

. Fondaparinux compared with enoxaparin for the prevention of venous thromboembolism after elective major knee surgery. N Engl J Med. 2001;345(18):1305–1310.

28.

Eriksson

Dahl

Buller

, et al. A new oral direct thrombin inhibitor, dabigatran etexilate, compared with enoxaparin for prevention of thromboembolic events following total hip or knee replacement: the BISTRO II randomized trial. J Thromb Haemost. 2005;3(1):103–111.

29.

Eriksson

Kalebo

Anthymyr

Wadenvik

Tengborn

Risberg

Prevention of deep-vein thrombosis and pulmonary embolism after total hip replacement. Comparison of low-molecular-weight heparin and unfractionated heparin. J Bone Joint Surg Am. 1991;73(4):484–493.

30.

Eriksson

Wille-Jorgensen

Kalebo

, et al. A comparison of recombinant hirudin with a low-molecular-weight heparin to prevent thromboembolic complications after total hip replacement. N Engl J Med. 1997;337(19):1329–1335.

31.

Eriksson

Borris

Dahl

, et al. Oral, direct factor Xa inhibition with BAY 59-7939 for the prevention of venous thromboembolism after total hip replacement. J Thromb Haemost. 2006;4(1):121–128.

32.

Fuji

Ochi

Niwa

Fujita

. Prevention of postoperative venous thromboembolism in Japanese patients undergoing total hip or knee arthroplasty: two randomized, double-blind, placebo-controlled studies with three dosage regimens of enoxaparin. J Orthop Sci. 2008;13(5):442–451.

33.

Fuji

Fujita

Ochi

. Fondaparinux prevents venous thromboembolism after joint replacement surgery in Japanese patients. Int Orthop. 2008;32(4):443–451.

34.

Hull

Pineo

Francis

, et al. Low-molecular-weight heparin prophylaxis using dalteparin in close proximity to surgery vs warfarin in hip arthroplasty patients: a double-blind, randomized comparison. The North American Fragmin trial investigators. Arch Intern Med. 2000;160(14):2199–2207.

35.

Huo

Eriksson

Dahl

, et al. Oral Dabigatran Versus Enoxaparin for Thromboprophylaxis after Primary Total Hip Arthroplasty: The RENOVATE-II Randomised Trial (abstract). Presented at: European Hematology Association; 2010 October 22-24, 2010; Beirut, Lebanon.

36.

Kakkar

Howes

Sharma

Kadziola

Z. A

comparative double-blind, randomised trial of a new second generation LMWH (bemiparin) and UFH in the prevention of post-operative venous thromboembolism. The Bemiparin assessment group. Thromb Haemost. 2000;83(4):523–529.

37.

Lassen

Bauer

Eriksson

Turpie

. Postoperative fondaparinux versus preoperative enoxaparin for prevention of venous thromboembolism in elective hip-replacement surgery: a randomised double-blind comparison. Lancet. 2002;359(9319):1715–1720.

38.

Lassen

Dahl

Mismetti

Destree

Turpie

. AVE5026, a new hemisynthetic ultra-low-molecular-weight heparin for the prevention of venous thromboembolism in patients after total knee replacement surgery—TREK: a dose-ranging study. J Thromb Haemost. 2009;7(4):566–572.

39.

Lassen

Agnelli

Fisher

, et al. The Ultra-Low-Molecular-Weight Heparin (ULMWH) Semuloparin for Prevention of Venous Thromboembolism (VTE) After Elective Knee Replacement Surgery. 21st International Congress on Thrombosis; 2010; Milan, July 6-9, Italy.

40.

Leclerc

Geerts

Desjardins

, et al. Prevention of deep vein thrombosis after major knee surgery—a randomized, double-blind trial comparing a low molecular weight heparin fragment (enoxaparin) to placebo. Thromb Haemost. 1992;67(4):417–423.

41.

Leclerc

Geerts

Desjardins

, et al. Prevention of venous thromboembolism after knee arthroplasty. A randomized, double-blind trial comparing enoxaparin with warfarin. Ann Intern Med. 1996;124(7):619–626.

42.

Levine

Hirsh

Gent

, et al. Prevention of deep vein thrombosis after elective hip surgery. A randomized trial comparing low molecular weight heparin with standard unfractionated heparin. Ann Intern Med. 1991;114(7):545–551.

43.

Mouret

Agnelli

Fisher

, et al. The Ultra-Low-Molecular-Weight Heparin (ULMWH) Semuloparin for Prevention of Venous Thromboembolism (VTE) after Elective Hip Replacement Surgery (abstract). 21st International Congress on Thrombosis; 2010; Milan, July 6-9, Italy.

44.

Navarro-Quilis

Castellet

Rocha

Paz-Jimenez

Planes

. Efficacy and safety of bemiparin compared with enoxaparin in the prevention of venous thromboembolism after total knee arthroplasty: a randomized, double-blind clinical trial. J Thromb Haemost. 2003;1(3):425–432.

45.

Eriksson

Borris

Dahl

, et al. A once-daily, oral, direct Factor Xa inhibitor, rivaroxaban (BAY 59-7939), for thromboprophylaxis after total hip replacement. Circulation. 2006;114(22):2374–2381.

46.

Turpie

Gallus

Hoek

. A synthetic pentasaccharide for the prevention of deep-vein thrombosis after total hip replacement. N Engl J Med. 2001;344(9):619–625.

47.

Planes

Vochelle

Fagola

, et al. Once-daily dosing of enoxaparin (a low molecular weight heparin) in prevention of deep vein thrombosis after total hip replacement. Acta Chir Scand Suppl. 1990;556:108–115.

48.

Planes

Vochelle

Fagola

Bellaud

Comparison of two low-molecular-weight heparins for the prevention of postoperative venous thromboembolism after elective hip surgery. Reviparin Study Group. Blood Coagul Fibrinolysis. 1998;9(6):499–505.

49.

Planes

Samama

Lensing

, et al. Prevention of deep vein thrombosis after hip replacement—comparison between two low-molecular heparins, tinzaparin and enoxaparin. Thromb Haemost. 1999;81(1):22–25.

50.

Ginsberg

Davidson

Comp

, et al. Oral thrombin inhibitor dabigatran etexilate vs North American enoxaparin regimen for prevention of venous thromboembolism after knee arthroplasty surgery. J Arthroplasty. 2009;24(1):1–9.

51.

Samama

Clergue

Barre

Montefiore

Ill

Samii

Low molecular weight heparin associated with spinal anaesthesia and gradual compression stockings in total hip replacement surgery. Arar study group. Br J Anaesth. 1997;78(6):660–665.

52.

Spiro

Johnson

Christie

, et al. Efficacy and safety of enoxaparin to prevent deep venous thrombosis after hip replacement surgery. Enoxaparin clinical trial group. Ann Intern Med. 1994;121(2):81–89.

53.

Torholm

Broeng

Jorgensen

, et al. Thromboprophylaxis by low-molecular-weight heparin in elective hip surgery. A placebo controlled study. J Bone Joint Surg Br. 1991;73(3):434–438.

54.

Turpie

. Enoxaparin prophylaxis in elective hip surgery. Acta Chir Scand Suppl. 1990;556:103–107.

55.

Turpie

Bauer

Eriksson

Lassen

. Postoperative fondaparinux versus postoperative enoxaparin for prevention of venous thromboembolism after elective hip-replacement surgery: a randomised double-blind trial. Lancet. 2002;359(9319):1721–1726.

56.

Turpie

Fisher

Bauer

, et al. BAY 59-7939: an oral, direct factor Xa inhibitor for the prevention of venous thromboembolism in patients after total knee replacement. A phase II dose-ranging study. J Thromb Haemost. 2005;3(11):2479–2486.

57.

Loke

Kwok

. Dabigatran and rivaroxaban for prevention of venous thromboembolism—systematic review and adjusted indirect comparison. J Clin Pharm Ther. 2011;36(1):111–124.

58.

Selby

Geerts

. Prevention of venous thromboembolism: consensus, controversies, and challenges. Hematol Am Soc Hematol Educ Program. 2009:286–292.

59.

Squibb

B-M

. Summary of Product Characteristics (SmPC) (Draft). Apixaban. Data on File. 2010.

The Efficacy and Safety of Pharmacological Prophylaxis of Venous Thromboembolism Following Elective Knee or Hip Replacement

Abstract

Keywords

Introduction

Methods

Systematic Review

Evaluation of Outcomes

Statistical Methods for Meta-Analysis

Results

Systematic Review Results

Direct/Indirect Comparison Results

All VTE and All-Cause Death

Major VTE

Outcomes of Bleeding

Results of NMA

Discussion

Footnotes

Authors– Note

Declaration of Conflicting Interests

Funding

References