Introduction

Venous thromboembolism (VTE) occurs in up to 25% of patients undergoing total hip (THA) or knee arthroplasty (TKA) without the use of prophylactic anticoagulation. Low molecular weight heparins (LMWH) are the standard agents for preventing VTE in this setting. In recent years, rivaroxaban, apixaban and dabigatran have been approved for this indication and, although results from randomized trials suggest that they are non-inferior and potentially superior to LMWH, information regarding outcomes in routine use is lacking.

Objectives

To evaluate the safety and efficacy of rivaroxaban for the prevention of VTE in patients undergoing THR or TKR in routine practice.

Methods

We conducted a population-based retrospective cohort study using linked healthcare databases in Ontario, Canada, including information on hospital discharge, emergency room visits, medication use, demographics and physician billing. In Ontario older patients have universal drug coverage and thus we included patients aged 66 years or older who received an outpatient prescription for a LMWH, (including dalteparin, tinzaparin and enoxaparin) or rivaroxaban after discharge from THR or TKR between 2002 and 2012 across 121 hospitals. Patients were excluded if they had other indications for anticoagulation. Primary efficacy and safety outcomes in the 30 days after surgery were the occurrence of an Emergency Room visit or hospitalization with a VTE (either deep vein thrombosis or pulmonary embolism) or a hospitalization with non traumatic major hemorrhage, respectively. Secondary outcomes included the previous 2 endpoints at 90 days as well as hospitalization for digestive system endoscopy (a proxy for gastrointestinal hemorrhage) and all cause mortality, both at 30 and 90 days after surgery. Unadjusted and adjusted odds ratios with 95% confidence intervals (CI) were obtained using logistic regression and reported as relative risks (RR) (appropriate given the incidence observed).

Results

The cohort included 24,321 patients and there was no significant difference on over 35 baseline characteristics between the LMWH (n=11,471) and rivaroxaban (n=12,850) groups. The median age for both groups was 73 years, 14,366 patients were women (59.1%) and 8,612 patients (35.4%) underwent THR. Anticoagulants were prescribed for a median of 14 days after discharge (interquartile range 10 to 21). The main results are shown in the table. Results were consistent in multiple additional analyses accounting for years rivaroxaban was approved in provincial formulary, adjusting for potential confounders, secular trends, individual LMWH, prescriber characteristics, and for subgroup analyses examining THR and TKR separately.

Table

Clinical outcomes in patients undergoing THR or TKR receiving rivaroxaban or LMWH.

30-Days Outcomes90-Days Outcomes*
UnadjustedAdjustedUnadjustedAdjusted
n/N
(%)
RR95% CIP ValueRR95% CIP Valuen/N
(%)
RR95% CIP ValueRR95% CIP Value
Primary Outcomes 
Emergency room visit or hospitalization with VTE 
LMWH 93 / 11471 (0.81%) Ref.  Ref.  134 / 11136 (1.20%) Ref.  Ref.  
Rivaroxaban 61 / 12850 (0.47%) 0.58 (0.42, 0.81) 0.001 0.58 (0.42, 0.80) 0.001 84 / 11823 (0.71%) 0.59 (0.45, 0.77) <0.001 0.59 (0.43, 0.81) 0.001 
Hospitalization with non-traumatic major hemorrhage 
LMWH 23 / 11471 (0.20%) Ref.  Ref.  30 / 11136 (0.27%) Ref.  Ref.  
Rivaroxaban 23 / 12850 (0.18%) 0.89 (0.50, 1.59) 0.700 0.88 (0.49, 1.58) 0.675 28 / 11823 (0.24%) 0.88 (0.53, 1.47) 0.623 0.63 (0.34, 1.16) 0.138 
Secondary Outcomes 
Hospitalization with digestive system endoscopy (a proxy for gastrointestinal bleeding) 
LMWH 26 / 11471 (0.23%) Ref.  Ref.  42 / 11136 (0.38%) Ref.  Ref.  
Rivaroxaban 31 / 12850 (0.24%) 1.06 (0.63, 1.79) 0.814 1.05 (0.62, 1.76) 0.869 47 / 11823 (0.40%) 1.05 (0.69, 1.60) 0.804 0.76 (0.47, 1.23) 0.270 
All cause mortality 
LMWH 6 / 11471 (0.05%) ― ― ― ― ― ― 25 / 11136 (0.22%) Ref.  Ref.  
Rivaroxaban ≤5 / 12850 (0.02%) ― ― 16 / 11823 (0.14%) 0.60 (0.32, 1.13) 0.114 0.52 (0.27, 1.02) 0.058 
30-Days Outcomes90-Days Outcomes*
UnadjustedAdjustedUnadjustedAdjusted
n/N
(%)
RR95% CIP ValueRR95% CIP Valuen/N
(%)
RR95% CIP ValueRR95% CIP Value
Primary Outcomes 
Emergency room visit or hospitalization with VTE 
LMWH 93 / 11471 (0.81%) Ref.  Ref.  134 / 11136 (1.20%) Ref.  Ref.  
Rivaroxaban 61 / 12850 (0.47%) 0.58 (0.42, 0.81) 0.001 0.58 (0.42, 0.80) 0.001 84 / 11823 (0.71%) 0.59 (0.45, 0.77) <0.001 0.59 (0.43, 0.81) 0.001 
Hospitalization with non-traumatic major hemorrhage 
LMWH 23 / 11471 (0.20%) Ref.  Ref.  30 / 11136 (0.27%) Ref.  Ref.  
Rivaroxaban 23 / 12850 (0.18%) 0.89 (0.50, 1.59) 0.700 0.88 (0.49, 1.58) 0.675 28 / 11823 (0.24%) 0.88 (0.53, 1.47) 0.623 0.63 (0.34, 1.16) 0.138 
Secondary Outcomes 
Hospitalization with digestive system endoscopy (a proxy for gastrointestinal bleeding) 
LMWH 26 / 11471 (0.23%) Ref.  Ref.  42 / 11136 (0.38%) Ref.  Ref.  
Rivaroxaban 31 / 12850 (0.24%) 1.06 (0.63, 1.79) 0.814 1.05 (0.62, 1.76) 0.869 47 / 11823 (0.40%) 1.05 (0.69, 1.60) 0.804 0.76 (0.47, 1.23) 0.270 
All cause mortality 
LMWH 6 / 11471 (0.05%) ― ― ― ― ― ― 25 / 11136 (0.22%) Ref.  Ref.  
Rivaroxaban ≤5 / 12850 (0.02%) ― ― 16 / 11823 (0.14%) 0.60 (0.32, 1.13) 0.114 0.52 (0.27, 1.02) 0.058 

n, number of events; N, number of patients; RR, relative risk; CI confidence interval; LMWH, low molecular weight heparin; Ref. Reference.

*

All 90-Days outcomes are secondary.

Conclusions

In this routine practice population-based study, the use of rivaroxaban compared to LMWH was associated with a lower risk of VTE without an increase in bleeding events.

Financial Support

Canadian Institutes of Health Research; ICES Western Scholars program.

Disclosures:

Lazo-Langner:Pfizer: Honoraria; Leo Pharma: Honoraria; Boehringer Ingelheim: Honoraria.

Author notes

*

Asterisk with author names denotes non-ASH members.

This icon denotes a clinically relevant abstract

Sign in via your Institution