Rivaroxaban, an oral anti-Xa inhibitor, was approved in 2008 in Canada for prevention of venous thromboembolism (VTE) post elective total hip replacement (THR). Studies completed in elective THR patients showed superiority or noninferiority of rivaroxaban over injectable low molecular weight heparin (LMWH) for prevention of VTE, with no difference in trial defined major bleeding. Many centres use rivaroxaban for VTE prevention post THR as standard of care. Our centre introduced rivaroxaban for THR patients in May 2009. As there is limited data on real world experience with this agent, beyond bleeding and VTE events, we evaluated the change in VTE prophylaxis for this indication.


A retrospective cohort study was completed comparing the change in use of a LMWH, enoxaparin, to rivaroxaban for VTE prophylaxis for elective THR patients ≥ 19 years not requiring therapeutic anticoagulation. The setting was a regional health authority servicing a population of 290 000, and the timeframe was 18 months before and 18 months after the introduction of rivaroxaban, allowing a 3 month period in between for incorporation of rivaroxaban into routine care. Patients with THR completed pre introduction of rivaroxaban were in the enoxaparin (E) group and patients with THR completed post introduction were in either the enoxaparin and rivaroxaban (E+R, given sequentially) group, or the rivaroxaban (R) group. Data were abstracted from computerized and paper records using a standardized form. Data collected included demographics, medical and medication history, risk factors for bleeding and VTE, type of anesthesia, VTE prophylaxis details, post-discharge community health (CH) nursing visits and hospital visits (emergency room (ER) and inpatient admissions), and major bleeding or VTE events. Hospital visits and VTE were assessed up to 30 days, and major bleeding up to 2 days, after the last presumed dose of medication. Primary outcomes were differences between E, E+R, and R in hospital visits, as well as referrals to, and reasons for, post discharge CH nursing visits.


There were 246 THR procedures in E, 43 in E+R, and 190 in R. Baseline characteristics were similar between the three groups. For E, E+R, R, 52%, 49% and 53% were female with a mean age of 64, 62, 64 years and a mean BMI of 31, 32, 31 kg/m2. There was no difference in surgery duration (mean 92, 97, 91 min for E, E+R, R), or number of patients transfused (19, 16, 14% for E, E+R, R). More patients (74%) in E+R received regional anesthesia exclusively, compared to E (57%) and R (52%), p=0.01.

There was no difference in post-surgery time to VTE prophylaxis initiation (mean of 22, 22, 23 hours for E, E+R, R). There was a difference in the length of stay (mean 7.1, 8.4, 6.3 days for E, E+R, R, p=0.005), as well as total duration of prophylaxis (mean 10.4, 16.1, 14.5 days for E, E+R, R, p<0.0001). Patients in the E+R sequentially received a mean of 1.7 days of enoxaparin, then 14.2 days of rivaroxaban.

Post discharge, there was no difference in the number of patients with at least one ER visit (7.3, 9.3, 6.3% for E, E+R, R, p=0.775) or hospital admission (3.3, 9.3, 2.6%, p=0.096). Most patients had a referral for CH nursing visits (94, 91, 95% for E, E+R, R, p=0.253). Most referrals requested dressing changes (81, 77, 82% for E, E+R, R, p=0.720) and removal of staples (91, 96, 91%, for E, E+R, R, p=0.671). The mean number of days post discharge for staple removal was 4.1, 4.3, 5.1 days for E, E+R, R, p<0.0001. The most common frequency of dressing changes requested was every 2 days (74, 63, 60% for E, E+R, R, p=0.009). Medication administration was requested for patients in E only (84, 0, 0% for E, E+R, R, p<0.0001), with a frequency of once daily (98%). Symptomatic VTE occurred in 1.6, 2.3, and 2.1% of patients for E, E+R, and R. There were 5 major bleeding events in E (2 extra-surgical site bleeds, 3 unexpected surgical site bleeds), and 0 events in both E+R and R.


This study provides useful data, beyond VTE and bleeding events, on the impact of a change in VTE prophylaxis post elective THR. No statistically significant difference was observed in post discharge hospital visits. Most patients had post discharge CH nursing visits requested; the overall requested visit frequency was reduced for all patients receiving rivaroxaban. Differences in type of anesthesia, length of stay, and duration of prophylaxis were noted for E+R. Differences between this group and R should be further explored.


Grewal:Bayer: Honoraria.

Author notes


Asterisk with author names denotes non-ASH members.