Soft tissue aspirations and injections may be used diagnostically and therapeutically in a variety of rheumatic conditions. The patients who require these procedures often are older and may be taking warfarin for the treatment of atrial fibrillation, valvular heart disease or cerebrovascular disease. The risk of traumatic hemorrhage or hemarthrosis following a soft tissue or joint injection in these patients is unknown. Thumboo and O'Duffy performed a prospective study at the Mayo Clinic to evaluate the frequency of joint or soft tissue bleeding in patients receiving warfarin therapy who required aspiration or injection of the soft tissue or joints.

The primary inclusion criteria for the trial included the use of a stable dosage of warfarin and an International Normalized Ratio (INR) of less than 4.5. Patients were excluded if the INR was more than 4.5, or if they had overlying cellulitis or were receiving concomitant heparin therapy.

The injections or aspirations were performed by a staff rheumatologist, a rheumatology fellow or an internal medicine resident under supervision. The medications used were either betamethasone or methylprednisolone, with 1 percent lidocaine typically given before aspiration procedures. The patients were instructed to avoid any strenuous activity or movement of the affected joint for 24 to 48 hours after the procedure. One of the two authors contacted all patients by telephone four weeks after the procedure to obtain information about warmth, swelling or bruising at the injection site. The patients were also asked about follow-up treatment for any complications related to the procedure. Patients who received corticosteroid injections were also asked to rate their pain as worse, slightly better, much better or fully resolved.

Thirty-two aspirations or injections (15 joint and 17 soft tissue procedures) were performed in 25 patients. Members of the study group had a median age of 74 years, and the median INR was 2.6. The INR assays were performed from zero to 129 days before the procedure, but the median time was only 1.5 days. There were no patient-reported joint or soft tissue hemorrhages. Based on a statistical rule known as the “rule of threes” (if no events of a specific type occur in a study group of X individuals, it is 95 percent certain that the event occurs no more than three times per X), the risk of significant soft tissue or joint hemorrhage in this study group can be calculated as less than 10 percent. Resolution or marked improvement in pain was reported by 74 percent of patients who received corticosteroid injections, and specific diagnostic information was obtained in 53 percent of patients who had joint aspirations.

The authors conclude from their study that joint or soft tissue aspirations and injections in patients who are undergoing anticoagulation with warfarin are associated with a low risk of hemorrhage. In addition, these procedures often provide therapeutic or diagnostic benefits.

editor's note: This is a small but clinically useful study. I believe most physicians would feel comfortable performing a soft tissue or joint injection in a patient taking warfarin provided a recent and therapeutic INR was available. If there is uncertainty, one could wait several hours until a prothrombin time could be obtained before administering a corticosteroid injection. On the other hand, if more urgent diagnostic information is required, as in a case of a possible septic joint, knowing that an INR less than 4.5 appears to be safe is reassuring. There would, however, still be a theoretic concern about bleeding in patients receiving concomitant therapy with aspirin or other nonsteroidal anti-inflammatory drugs.—j.t.k.