Impact of Clinical Practice Guidelines on Use of Intra-Articular Hyaluronic Acid and Corticosteroid Injections for Knee Osteoarthritis.

BACKGROUND: The efficacy of corticosteroid and hyaluronic acid injections for knee osteoarthritis has been questioned. The purpose of this study was to determine the impact of the American Academy of Orthopaedic Surgeons (AAOS) clinical practice guidelines on the use of these injections in the United States and determine if utilization differed by provider specialty.

METHODS: Patients with knee osteoarthritis were identified within the Humana database from 2007 to 2015, and the percentage of patients receiving a knee injection relative to the number of patients having an encounter for knee osteoarthritis was calculated and was trended for the study period. The impact of each edition of the AAOS clinical practice guidelines on injection use was evaluated with segmented regression analysis. Injection trends were also analyzed relative to the specialty of the provider performing the injection.

RESULTS: Of 1,065,175 patients with knee osteoarthritis, 405,101 (38.0%) received a corticosteroid injection and 137,005 (12.9%) received a hyaluronic acid injection. The rate of increase in hyaluronic acid use, per 100 patients with knee osteoarthritis, decreased from 0.15 to 0.07 injection per quarter year (p = 0.02) after the first clinical practice guideline, and the increase changed to a decrease at a rate of -0.12 injection per quarter (p < 0.001) after the second clinical practice guideline. After the first clinical practice guideline, the rate of increase in utilization of corticosteroids, per 100 patients with knee osteoarthritis, significantly lessened to 0.12 injection per quarter (p < 0.001), and after the second clinical practice guideline, corticosteroid injection use plateaued (p = 0.72). The trend in use of hyaluronic acid injections by orthopaedic surgeons and pain specialists decreased with time following the second-edition clinical practice guideline but did not change for primary care physicians or nonoperative musculoskeletal providers.

CONCLUSIONS: Subtle but significant changes in hyaluronic acid and corticosteroid injections occurred following the publication of both clinical practice guidelines. Although the clinical practice guidelines did impact injection use, given the high costs of these injections and their questionable clinical efficacy, further interventions beyond publishing clinical practice guidelines are needed to encourage higher-value care for patients with knee osteoarthritis.