Examining the Safety of Joint Injections In Patients on Warfarin
Many physicians who treat pain have been hesitant to perform invasive procedures on patients who are taking warfarin sodium (Coumadin) due to the potential risk of complications associated with the medications. However, the same patients who are on anticoagulants may also be the ones who would benefit the most from such procedures as joint injections and aspirations.
Warfarin sodium (Coumadin) is the most widely prescribed anticoagulant medication in North America.1 It is commonly used for the treatment and prevention of venous thrombosis, pulmonary embolism, and thromboembolic disorders. It is also used to reduce the risk of death, recurrent myocardial infarction (MI), and thromboembolic events after MI.2 However, according to the Physicians’ Desk Reference, there are a number of contraindications to using warfarin, including procedures with potential for uncontrollable bleeding (Table 1).3
Two of the most common procedures performed in the outpatient setting are joint injections and aspirations.4 Joint injections (intra-articular injections) are used in the diagnosis and treatment of a number of inflammatory and non-inflammatory joint conditions, such as rheumatoid arthritis, psoriatic arthritis, gout, tendonitis, bursitis, and osteoarthritis. Joint aspiration (arthrocentesis), on the other hand, uses a needle and syringe to collect synovial fluid from a joint capsule. It is commonly used in the diagnosis of gout, arthritis, and synovial infections and therapeutically in the management of joint effusions.4 These procedures are commonly used by multiple medical specialties and carry little risk of complications (Table 2).4-6
It has been suggested that there are increased risks of complications during these procedures in patients on warfarin sodium. However, there are currently very little data assessing the safety of joint injections/aspirations on patients taking warfarin sodium. Studies in the literature have assessed the safety of intramuscular and other injections for patients who are on warfarin sodium.7,8 Most of these studies detailing complications of taking warfarin sodium involved spontaneous hemarthrosis, which was typically both diagnosed and treated via joint aspiration.9-16
A review of three articles that describe studies assessing the safety of joint injection/aspiration for patients taking warfarin sodium found that bleeding complications occurred rarely. In a study conducted by the Mayo Clinic, Thumboo and O’Duffy did a prospective analysis of 32 joint or soft tissue aspirations and injections in patients taking warfarin sodium. Their study had no patient-reported complications of joint or soft tissue hemorrhage and they concluded that this procedure is low risk for hemorrhagic complications.17
In the second study, Salvati et al conducted a prospective analysis of 15 patients on oral anticoagulation (OAC) with a joint effusion. Of the 15 participants, 2 patients had a hemarthrosis seen as a complication of the arthrocentesis. It was noted that these 2 patients had supratherapeutic international normalized ratios (INRs) of 3.8 and 5.0, respectively. This article concluded that anticoagulation should not be an absolute contraindication to arthrocentesis.18
In the last study, Dunn and Turpie reviewed thromboembolic events and complications for patients taking OAC who needed to undergo surgery or invasive procedures. They found that major bleeding while receiving therapeutic OAC was rare for dental procedures (0.2% [4 of 2,014]), arthrocentesis (0.0% [0 of 32]), cataract surgery (0.0% [0 of 203]), and upper endoscopy or colonoscopy with or without biopsy (0.0% [0 of 111]).19
Many physicians hesitate to treat patients on OAC with invasive procedures. We conduced a retrospective study of patients who underwent joint injections/aspirations while concurrently on warfarin sodium, and compared the complication rates to those of a matched-control group who were not on OAC, in the hopes of determining whether such procedures are contraindicated in patients on anticoagulant therapy.
The study included 110 patients who received joint injections/aspiration procedures at the McConnell Spine, Sport, & Joint Center in Columbus, Ohio, between January 2006 and January 2009. The study population was divided into two groups: the study group (61 patients on warfarin, which included 2 men and 8 women who had undergone multiple injections) and the control group (18 men and 31 women). The age of the patients in the study group ranged from 42 to 79 years, with a mean age of 74 years. The age of the patients in the control group ranged from 24 to 88 years, with a mean age of 61 years.
All the patients who received a joint injection/aspiration while concurrently on warfarin sodium had to have their prothrombin time (PT)/INRs recorded within one month of the procedure.
As this was a retrospective study, patient data, including socio-demographics, were recorded and patient-specific injuries and/or complications from the procedures were assessed and charted via follow-up appointment or phone call. The INR was assessed by both chart review as well as accessing the OhioHealth Results Browser online.
Potential negative outcomes from the procedure were collected and could include a primary complication of hemarthrosis (bleeding complication), and additionally: infection (including septic arthritis), tendon rupture, post-injection flare, skin atrophy, fat atrophy, skin hypopigmentation, corticosteroid arthropathy, osteonecrosis, nerve damage, and excessive injection-site bleeding.
Results of the Study
None of the procedures in the study or control groups were associated with patient- or physician-reported joint and/or soft tissue hemorrhage. No incidences of post-procedural infection, tendon rupture, post-injection flare, skin atrophy, osteonecrosis, or nerve damage were reported in the study. One patient in the study group with a target INR reported “feeling ill” the night after the injection, which was not considered to fall under any of the procedural-related complications outlined in this study. One patient in the control group reported refractory pain after the procedure.
Non–procedural-related complications reported for the patients on OAC included eye hemorrhage, gum bleeding, easy bruising, and abdominal bleeding. Most of these complications notably occurred in patients within target and supratherapeutic INRs. Interestingly, the abdominal bleeding was only noted in patients with a subtherapeutic INR.
Areas injected in the study group included 52 knee injections, 30 of which had a therapeutic INR; 5 shoulder injections, 4 of which had a therapeutic INR; 2 hip/trochanteric bursa injections, all of which had a therapeutic INR; and 2 hand injections, all of which had a therapeutic INR. In the control group there were 24 knee injections, 17 shoulder injections, 9 hip/trochanteric bursa injections, and no hand injections (Figure).