Perioperative Management of Novel Anticoagulants in Patients Undergoing Arthroplasty

1University of California San Francisco, San Francisco, CA
2University of California, San Francisco, San Francisco, CA

Meeting: Hospital Medicine 2014, March 24-27, Las Vegas, Nev.

Abstract number: 96


The use of novel oral anticoagulants (NOACs) such as dabigatran and rivaroxaban has been rapidly increasing. However, there is limited experience with the periprocedural management of these agents, and uncertainty about the safety of NOACs in the setting of high risk procedures (such as neuraxial anesthesia) or renal insufficiency. The objective of our study was to describe the perioperative management and outcomes of patients taking chronic NOAC therapy who were undergoing elective arthroplasty.


We identified all patients who were taking chronic NOAC therapy (dabigatran, rivaroxaban, or apixaban) prior to undergoing elective knee or hip arthroplasty at a tertiary care medical center between 7/2010 to 6/2013. We selected matched controls based on procedure type and date, age, and sex. Electronic medical records were reviewed to obtain data on demographics, clinical comorbidities, periprocedural anticoagulation management, hospital course, and 60 day outcomes.


Among 2406 patients undergoing arthroplasty during the time period, 25 patients were taking NOACs prior to surgery (19 on dabigatran and 6 on rivaroxaban, all for atrial fibrillation). We identified 50 matched control patients not on anticoagulants. The mean age of the cohort was 74.5 and 60% were women. In the NOAC treated group, the perioperative stroke risk by CHADS2 score was “low risk, CHADS2 0‐2” in 18 (72%), “intermediate risk, CHADS2 3‐4” in 6 (24%), and “high risk, CHADS2 5‐6” in 1 (4%). NOAC anticoagulation was stopped ≥4 days prior to surgery in 20% of patients, 3 days in 52%, and ≤ 2 days in 16%; stroke risk was not associated with how long the NOAC was held. Twenty‐four percent of NOAC patients had impaired renal function (estimated glomerular filtration rate of <50 ml/min). Three patients (50%) of NOAC patients with renal insufficiency had anticoagulation discontinued for shorter than the recommended periods given their renal function. Bridging anticoagulation with enoxaparin was used in 1 patient in the NOAC group. Average hospital length of stay was 3.5 days (range 2‐11 days) in NOAC group vs. 3.04 (2‐6 days) in non‐anticoagulated patients, p=0.11. Post‐operatively, NOACs were restarted within: 1‐2 days (52%), 3‐5 days (32%), and 5 or more days (16%). Control patients were all given post‐operative enoxaparin at 40mg daily. A total of 3 patients in the NOAC group received ≥ 2 units blood transfusions post‐operatively, compared to 8 in the control group (p=0.27). There were no differences in the number of venous thromboembolism events, infections, hematomas, readmissions or ED visits between the two groups (p>0.05). All patients received neuraxial anesthesia either through spinal or peripheral nerve catheters and there were no documented epidural hematomas.


We observed variation in the perioperative management of NOACs in arthroplasty patients, particularly in when patients were instructed to stop their medication and when full anticoagulation was restarted after surgery. In this pilot study of perioperative NOAC outcomes, patients taking chronic NOACs and undergoing major surgery did not have more bleeding or thromboembolism than control patients not on NOACs. Standardization of practice regarding the perioperative management of anticoagulants may be helpful to minimize potential complications associated with NOAC use around the time of surgery.

To cite this abstract:

Magan Y, Fang M. Perioperative Management of Novel Anticoagulants in Patients Undergoing Arthroplasty. Abstract published at Hospital Medicine 2014, March 24-27, Las Vegas, Nev. Abstract 96. Journal of Hospital Medicine. 2014; 9 (suppl 2). Accessed April 7, 2020.

« Back to Hospital Medicine 2014, March 24-27, Las Vegas, Nev.