You are here
A Pilot Study Comparing Hospital Readmission Rates In Patients Receiving Rivaroxaban or Enoxaparin After Orthopedic Surgery
Venous thromboembolism (VTE), consisting of deep vein thrombosis (DVT) and pulmonary embolism (PE), is a potential complication of total hip arthroplasty (THA) or total knee arthroplasty (TKA). The intravascular damage resulting from surgery may activate the intrinsic clotting cascade and increase the risk of clot formation. In the absence of prophylaxis, 40% to 60% of patients develop DVT within seven to 14 days after hip or knee surgery.1 Several oral and injectable anticoagulants are available to prevent clot formation in this susceptible population, including fondaparinux (Arixtra, GlaxoSmithKline), apixaban, rivaroxaban (Xarelto, Janssen Pharmaceuticals), low-dose unfractionated heparin, warfarin, aspirin, and dabigatran.2 The American College of Chest Physicians (ACCP) recommends low-molecular-weight heparin as the standard of care for VTE prophylaxis after THA or TKA.2 In 2008, prospective, randomized studies showed that rivaroxaban had a safety profile similar to that of enoxaparin (Lovenox, Sanofi-Aventis), but a superior efficacy profile.3,4
The Centers for Medicare and Medicaid Services (CMS) Readmission Reduction Program originally measured 30-day, all-cause readmissions after hospitalization for acute myocardial infarction, heart failure, or pneumonia.5 In 2015, the program added patients with chronic obstructive pulmonary disease and those undergoing an elective THA or TKA. The CMS will reduce the reimbursement to a hospital if the annual number of readmissions for these disorders is higher than the national average.
Our institution currently does not have a preferred anticoagulant for VTE prophylaxis after THA or TKA. We set out to determine whether the use of rivaroxaban would reduce the 30-day all-cause readmission rate after THA or TKA compared with enoxaparin. We evaluated the use of rivaroxaban and enoxaparin and the associated readmission rates at our institution, with the goal of identifying the preferred anticoagulant.
This pilot study was a single-center, retrospective, non-randomized chart review comparing the 30-day all-cause readmission rates for patients who received either rivaroxaban or enoxaparin for VTE prophylaxis after THA or TKA at Hillcrest Hospital—a Cleveland Clinic facility in Mayfield Heights, Ohio. Our institutional review board approved the study in August 2014. We identified evaluable electronic medical records through a quality management database. We placed each patient into one of two treatment groups, rivaroxaban or enoxaparin, depending on the agent used for VTE prophylaxis after surgery. We collected the data listed in
Adult patients (18 years of age and older) who underwent THA or TKA between May 2013 and May 2014 were eligible for evaluation. We included all patients who received one or more prophylactic doses of oral rivaroxaban (10 mg once daily) or subcutaneous enoxaparin (30 mg twice daily or 40 mg once daily). Patients were excluded from the study if they had a history of heparin-induced thrombocytopenia, a history of allergy associated with heparin or rivaroxaban, a hypercoaguable condition, or creatinine clearance of less than 30 mL/min. They were also excluded if they required anticoagulation before admission, were discharged more than 30 days after surgery, or died before discharge.
The study’s primary endpoint was the 30-day all-cause readmission rate. The secondary endpoint was the 30-day readmission rate due to VTE or a major or minor bleeding event, which had to be diagnosed within 48 hours of readmission. A major bleeding event was defined as a bleed in a critical organ, a bleed requiring reoperation, or an extrasurgical-site bleed that was clinically overt and associated with a decrease in hemoglobin of at least 2 g/dL or with the transfusion of two or more units of blood.3 All other bleeding events were considered to be minor.
We analyzed the primary and secondary outcomes using Fisher’s exact test, and used descriptive statistics to evaluate data related to the patients’ baseline demographic characteristics.
A total of 543 patients underwent THA or TKA between May 2013 and May 2014. We reviewed 405 patient charts and included 240 patients (27 in the enoxaparin group and 213 in the rivaroxaban group). We did not review the charts of all 543 patients because low rates of enoxaparin use had been found during an interim analysis. The reasons for exclusion are listed in
The primary endpoint (the 30-day all-cause hospital re admission rate) was reached in eight of 213 patients in the rivaroxaban group and in one of 27 patients in the enoxaparin group (3.76% versus 3.70%, respectively) (
The secondary endpoint of the 30-day readmission rate due to VTE or a bleeding event was reached in two of 213 patients (0.9%) in the rivaroxaban group. One patient experienced a major bleed with an infected hematoma requiring surgical evacuation. This patient had received one prophylactic dose of rivaroxaban and had developed a PE prior to discharge, and was therefore on full-dose anticoagulation with rivaroxaban upon readmission. The patient still met the study’s inclusion criteria since he received one prophylactic dose after surgery and did not use full-dose anticoagulation before his initial admission. The second patient experienced a minor bleed with a hematoma around the incision site.
The average postoperative length of stay was one day longer in the enoxaparin group than in the rivaroxaban group (5.0 days versus 4.1 days, respectively), which may be an area for further investigation in terms of potential cost-savings to the hospital. The average times to the administration of the first postoperative dose of VTE prophylaxis were 18.4 hours for enoxaparin and 18.9 hours for rivaroxaban.
Four patients developed a DVT or a PE before being discharged from the hospital. These events occurred between two and seven days after surgery when the patients were initiated on full-dose anticoagulation. Three of the four patients were female; all underwent TKA; and three received rivaroxaban. Their weights ranged from 95 kg to 109 kg. Among the four patients, the times to the administration of the first dose of VTE prophylaxis were 19.0, 20.5, 21.0, and 32.0 hours, respectively.
Eight of the nine readmitted patients were female, with average ages of 81.0 years in the enoxaparin group and 78.4 years in the rivaroxaban group (
Several patients presented to the emergency room (ER) and were never readmitted to the hospital. Three patients in the enoxaparin group (11.1%), 13 patients in the rivaroxaban group (6.1%), and six patients in the excluded group (3.7%) presented to the ER. In both the rivaroxaban and enoxaparin groups, a higher number of patients presented to the ER than were readmitted to the hospital. This did not include patients who came through the ER prior to admission. Six of the eight readmitted patients in the rivaroxaban group initially came through the ER, and two patients were directly admitted from their physicians’ offices. The one readmitted patient in the enoxaparin group initially presented to the ER. The group of excluded patients had a higher readmission rate (9.3%) and a lower ER visit rate. Seven of the 13 patients treated with rivaroxaban had a chief complaint related to their orthopedic surgeries: five experienced lower-extremity swelling; one had knee pain; and one developed cellulitis. The average time to ER presentation was 5.3 days for those patients. Two of three patients in the enoxaparin group presented to the ER for lower-extremity swelling negative for DVT, with an average time to ER presentation of five days. Twelve of the 16 patients presenting to the ER who received rivaroxaban or enoxaparin were not readmitted and did not present to the ER at a later date. Two of these 16 patients did present to the ER and were readmitted to the hospital two months after surgery; one underwent THA repair and the other had DVT. The other two patients presented to the ER at least one year later with shortness of breath and chest pain, both of which were unrelated to their surgeries and were not due to PE.
In a large randomized trial, oral rivaroxaban (10 mg/day) was significantly more effective than subcutaneous enoxaparin (40 mg/day) for DVT prophylaxis after THA, with similar safety outcomes.3 The study’s primary endpoint was a composite of DVT, nonfatal PE, or death from any cause for up to 36 days after the initiation of treatment. Rivaroxaban was noninferior to enoxaparin in the per-protocol group (absolute risk reduction [ARR], 2.5%) and was superior to enoxaparin in the modified intent-to-treat group (ARR, 2.6%).The primary endpoint was reached in 18 of 1,595 patients (1.1%) in the rivaroxaban group and in 58 of 1,558 patients (3.7%) in the enoxaparin group. The rates of major bleeding were 0.3% for rivaroxaban and 0.1% for enoxaparin (P = 0.18).
Similar results were demonstrated in another large randomized comparison of rivaroxaban and enoxaparin as DVT prophylaxis after TKA.4 Again, the primary outcome was a composite endpoint of DVT, nonfatal PE, or death from any cause within 13 to 17 days after initiation of treatment. Rivaroxaban was found to be noninferior to enoxaparin in the per-protocol group (ARR, 9.2%) and superior to enoxaparin in the modified intention-to-treat group (ARR, −9.2%). The primary endpoint was reached in 79 of 824 rivaroxaban-treated patients (9.6%) and in 166 of 878 enoxaparin-treated patients (18.9%). The major bleeding rates were 0.6% for rivaroxaban and 0.5% for enoxaparin (P = 0.77).
Based on the results of these two studies, the ACCP has identified rivaroxaban as an effective alternative to enoxaparin for DVT prophylaxis following THA or TKA.2
In our pilot study, we did not find a substantial difference in the 30-day all-cause readmission rate between rivaroxaban and enoxaparin (3.76% versus 3.70%, respectively). The total 30-day, all-cause, unadjusted readmission rate from May 2013 to May 2014 was 6%, which was calculated from both included and excluded patients. The 3.76% readmission rate in the rivaroxaban group demonstrated that this drug was not the leading cause of readmissions. Only one patient was readmitted in the enoxaparin group, making an analysis difficult. We did not identify other studies that looked at readmission rates for rivaroxaban versus enoxaparin and therefore cannot compare our results with the previous literature. The major bleeding rate was 0.5% in the rivaroxaban group, which was similar to the major bleeding rates reported in the RECORD trials (0.3% to 0.6%).3,4 We used the same definition of major bleeding, but our bleeding rate may have been underestimated because we reviewed only patients who had been readmitted to Cleveland Clinic-affiliated hospitals. In addition, patients were not included in our readmission analysis if they experienced a major bleeding event prior to discharge.
Our institution does not have a preferred agent for VTE prophylaxis after orthopedic surgery. Rather, we found that the agent of choice in this setting depended on physician preferences and not on the type of surgery being performed. Rivaroxaban and enoxaparin appear to be similar in terms of low readmission rates and favorable safety profiles, and both may be considered as first-line options for VTE prophylaxis. We prefer rivaroxaban over enoxaparin, however, because it has the benefit of oral administration with a cost similar to that of enoxaparin. Enoxaparin also had an average length of stay that was one day longer than that of rivaroxaban. The latter finding should be explored further as it suggests a cost-saving benefit with rivaroxaban.
We did not anticipate the extensive use of aspirin as the sole pharmacological agent for VTE prophylaxis. The surgeons involved in our study explained their preference for aspirin by saying that it was less expensive than either rivaroxaban or enoxaparin, and that it offered an oral alternative to enoxaparin injections. The readmission rate for aspirin was 10 of 127 patients (7.9%)—twice that of rivaroxaban and enoxaparin combined (3.8%). Six of the 10 readmissions were related to surgery: four wound infections, one hematoma, and one case of intractable knee pain. The average time to readmission for patients who received aspirin was 8.6 days.
There is no clinical evidence to support the use of aspirin over anticoagulants for VTE prophylaxis after orthopedic surgery. Because of the lack of head-to-head comparison trials and the use of endpoints with questionable clinical significance (e.g., asymptomatic DVT versus symptomatic DVT or PE), the American Academy of Orthopaedic Surgeons declined to recommend either for or against specific VTE prophylactic strategies.7 In 2008, the ACCP recommended against the use of aspirin (grade 1A), but their guidelines now include a grade 2C recommendation for aspirin over no VTE prophylaxis, based on the results of a study with a flawed design.2,8 The investigators in that trial excluded patients who were at high risk for VTE or a bleeding episode; did not clearly define the primary endpoint; and did not control the use of compression devices and additional VTE prophylactic agents.9 Another study found no difference between aspirin and enoxaparin when used in combination with compression devices.10 Additional studies are needed to identify patients who would benefit from VTE prophylaxis with aspirin rather than anticoagulants.
As mentioned previously, a major limitation of our study was our inability to identify patients who were readmitted to hospitals outside of the Cleveland Clinic health system. Therefore, we may have underestimated the readmission rates. Moreover, individual surgeons’ preferences and their surgical techniques may have confounded the readmission rates for rivaroxaban and enoxaparin because the use of these medications was largely surgeon dependent.
Another study limitation was the unanticipated use of aspirin, which prevented us from meeting the target sample size necessary for calculating statistical significance. Moreover, our findings may have been influenced by selection bias resulting from surgeons’ comfort with older agents over rivaroxaban.
Finally, the nature of a retrospective study limits the interpretation of results because of the inherent lack of control and jurisdiction over patient adherence, patient preference, and patient satisfaction.
In this retrospective pilot study, we did not find a substantial difference in the 30-day all-cause readmission rates between rivaroxaban and enoxaparin. Both agents may be considered as appropriate first-line options for VTE prophylaxis. We recommend rivaroxaban over enoxaparin, however, because rivaroxaban is administered orally and may improve compliance. Larger studies should be conducted to explore the use of aspirin in this patient population, to review the most appropriate time to administer the first dose of VTE prophylaxis, and to determine whether rivaroxaban can reduce a patient’s length of stay compared with enoxaparin.
Figure and Tables
CrCl = creatinine clearance; HIT = heparin-induced thrombocytopenia;
IPC = intermittent pneumatic compression
IPC = intermittent pneumatic compression; VTE = venous thromboembolism.
Demographic and Clinical Characteristics of Patient Cohorts
|Female (no. [%])||137 (64.3)||19 (70.4)|
|Age in years, mean (range)||68.0 (40–92)||65.6 (45–87)|
|Height in inches, mean (range)||65.9 (53–78)||65.9 (60–74)|
|Weight in kg, mean (range)||85.9 (46.5–155.6)||85.8 (42–169)|
|Serum creatinine, mean (SD), range||0.8 mg/dL (± 0.2), 0.4–1.9||0.8 mg/dL (± 0.3), 0.5–1.7|
|Creatinine clearance, mean (SD), range||77.7 mL/min (± 30.1), 30.0–173.6||79.4 mL/min (± 30.5), 33.5–139.4|
|Documented renal function (no. [%])||163 (76.5)||22 (81.5)|
|History of smoking (no. [%])||121 (56.8)||16 (59.3)|
|History of VTE (no. [%])||21 (9.9)||7 (25.9)|
|History or active malignancy (no. [%])||39 (18.3)||4 (14.8)|
|Type of surgery (no. [%])|
|Total hip arthroplasty||126 (59.2)||17 (63)|
|Total knee arthroplasty||87 (40.8)||10 (37)|
|Indication for surgery (no. [%])|
|Osteoarthritis||194 (91.1)||18 (66.7)|
|Degenerative joint disease||12 (5.6)||5 (18.5)|
|Joint necrosis||4 (1.9)||1 (3.7)|
|Trauma||3 (1.4)||3 (11.1)|
|Postoperative length of stay in days, mean (SD), range||4.1 (± 1.6), 3–14||5.0 (± 3), 3–14|
|Time to first dose, mean (SD), range||18.9 hours (± 6.5), 5.5–56.5||18.4 hours (± 5.9), 5–27|
|IPC (no. [%])||213 (100)||27 (100)|
|Concomitant antiplatelet (no. [%])||71 (33.3)||8 (29.6)|
|Aspirin 81 mg daily||60 (28.2)||4 (14.8)|
|Aspirin 81 mg twice daily||2 (0.9)||0 (0)|
|Aspirin 325 mg daily||9 (4.2)||2 (7.4)|
|Aspirin 650 mg daily||0 (0)||2 (7.4)|
|Topical tranexamic acid (no. [%])||165 (77.5)||20 (74.1)|
|0 g||48 (22.5)||7 (25.9)|
|1 g||18 (8.5)||2 (7.4)|
|2 g||147 (69)||18 (66.7)|
IPC = intermittent pneumatic compression; SD = standard deviation; VTE = venous thromboembolism.
Incidence of Primary Endpoint
|Rivaroxaban||10 mg daily||213||8 (3.76)|
|Enoxaparin||40 mg daily or 30 mg twice daily||27||1 (3.70)|
Demographic and Clinical Characteristics of Readmitted Patients
|Female (no. [%])||7 (87.5)||1 (100)|
|Age in years, mean (range)||78.4 (54–85)||81.0|
|Height in inches, mean (range)||64.4 (60–70)||65.0|
|Weight in kg, mean (range)||89.0 (53.1–127)||60.3|
|Serum creatinine, mean (SD), range||0.8 mg/dL (± 0.2), 0.5–1.1||1.1 mg/dL|
|Creatinine clearance, mean (SD), range||73.9 mL/min (± 32.7), 34.8–125.7||34.8 mL/min|
|Documented renal function (no. [%])||7 (87.5)||1 (100)|
|History of smoking (no. [%])||5 (62.5)||0|
|History of VTE (no. [%])||4 (50)||0|
|History or active malignancy (no. [%])||1 (12.5)||0|
|Type of surgery (no. [%])|
|Total hip arthroplasty||5 (62.5)||0|
|Total knee arthroplasty||3 (37.5)||1 (100)|
|Indication for surgery (no. [%])|
|Degenerative joint disease||1 (12.5)||1 (100)|
|Joint necrosis||1 (12.5)||0|
|Postoperative length of stay in days, mean (SD), range||4.6 (± 1.7), 3–7||6|
|Time to first dose, mean (SD), range||19.4 hours (± 6.6), 8.5–32||7 hours|
|IPC (no. [%])||8 (100)||1 (100)|
|Concomitant antiplatelet (no. [%])|
|Aspirin 81 mg daily||3 (37.5)||0|
|Aspirin 325 mg daily||0 (0)||1 (100)|
|Topical tranexamic acid (no. [%])|
|0 g||1 (12.5)||0|
|1 g||3 (37.5)||1 (100)|
|2 g||4 (50)||0|
|Time to readmission in days, average (range)||12 (3–25)||10|
|Reason for readmission (no. [%])|
|Related to surgery||4 (50)||0|
|Unrelated to surgery||4 (50)||1 (100)|
IPC = intermittent pneumatic compression; SD = standard deviation; VTE = venous thromboembolism.
- Geerts WH, Pineo GF, Heit JA, et al. Prevention of venous thromboembolism: the seventh ACCP conference on antithrombotic and thrombolytic therapy. Chest 2004;126;(suppl):338s–400s.
- Falck-Ytter Y, Francis C, Johanson N, et al. Prevention of VTE in orthopedic surgery patients: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012;141:e278s–e325s.
- Eriksson B, Borris L, Friedman R, et al. Rivaroxaban versus enoxaparin for thromboprophylaxis after hip arthroplasty. N Engl J Med 2008;358:2765–2775.
- Lassen M, Ageno W, Borris L, et al. Rivaroxaban versus enoxaparin for thromboprophylaxis after total knee arthroplasty. N Engl J Med 2008;358:2776–2786.
- Centers for Medicare and Medicaid Services. Readmissions reduction program (HRRP). Available at: www.cms.gov/medicare/medicare-fee-for-service-payment/acuteinpatientpps/readmissions-reduction-program.html. Accessed October 17, 2014
- The Joint Commission. Surgical Care Improvement Project. Available at: www.jointcommission.org/surgical_care_improvement_project. Accessed October 17, 2014
- Stewart DW, Freshour JE. Aspirin for the prophylaxis of venous thromboembolic events in orthopedic surgery patients: a comparison of the AAOS and ACCP guidelines with review of the evidence. Ann Pharmacother 2013;47:63–74.
- Geerts WH, Bergqvist D, Pineo GF, et al. Prevention of venous thromboembolism: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines, 8th edition. Chest 2008;133;(suppl):381s–453s.
- Rodgers A, MacMohon S. Prevention of pulmonary embolism and deep vein thrombosis with low dose aspirin: pulmonary embolism prevention (PEP) trial. Lancet 2000;355:1295–1302.
- Westrich GH, Bottner F. VenaFlow plus Lovenox vs VenaFlow plus aspirin for thromboembolic disease prophylaxis in total knee arthroplasty. J Arthroplasty 2006;21:139–143.