Are shots or squats utilized more often in the 4 years prior to knee replacement surgery?

Abstract

Purpose: Exercise therapy is strongly recommended as a core treatment for knee osteoarthritis, though recent studies show an increase in utilization of medications including opioids over the past 10 years in the general population. The recommendation for corticosteroid injections is conditional or neutral in most guidelines, but their use is common in clinical practice. Some clinicians advocate the use of the latter immediately prior to exercise in order to improve exercise tolerance. Little is known about utilization of either of these interventions in the Military Health System for knee osteoarthritis. The purpose of this study was to compare utilization rates, to include order and timing, of two common interventions, corticosteroid injection and exercise therapy, delivered 4 years prior to knee replacement surgery. We also examined the relationship between order and timing of each intervention with the fill of an opioid prescription. Methods: We identified all eligible TRICARE beneficiaries that underwent a knee replacement surgical procedure in either a military or civilian hospital, where TRICARE was the payer, for a 3-year period from 1 January 2016 through 31 December 2018. We used the Military Health System Data Repository which is a single-payer data repository capturing all medical care at the person-level, whether it occurs in a military or civilian hospital all around the world. We identified all persons with a current procedural terminology (CPT) code for a knee replacement surgical procedure. We also abstracted all CPT codes that reflected CSI and therapeutic exercise. We ensured that these procedure codes were present with knee-related International Classification of Disease (ICD) 9th and 10th revision codes. We also utilized the Pharmacy Data Transaction Service (PDTS) with MDR to identify all opioid prescription fills using the American Hospital Formulary Service therapeutic class codes 280808 and 280812. We identified all healthcare utilization in any setting (military or civilian clinic) for the full 4-year period prior to the knee replacement procedure (1 January 2013 through 30 December 2018). We reported means, medians, and frequency counts for each variable, and chi-square with 95% confidence intervals for odds ratio calculations. Results: Out of 54,422 unique persons that underwent a total knee replacement surgery during the surveillance period, 16,999 (31.2%) received at least 1 CSI (mean [SD] = 4.6 [4.5]), while 9668 (17.8%) received at least 1 ET session (mean [SD] = 10.0 [13.4]). Only 7379 had both CSI and ET (13.6%), but in the majority of cases (72.5%) there was more than 60 days between the CSI and the ET session. The first CSI occurred a mean of 729 days (2 years) before surgery, whereas the first ET session occurred a mean of 598 days (1.6 years) before surgery. In patients that received both CSI and ET, receiving a CSI before receiving ET was associated with higher odds of also filling at least 1 opioid prescription (OR = 1.46; 95CI 1.32 to 1.61). The results were not different based on whether care occurred in a military or civilian hospital setting. Conclusions: In this cohort across a large health system, CSI was utilized as an intervention at a rate almost double that of ET for individuals with knee osteoarthritis prior to joint replacement. When patients did receive both interventions, 60.2% of the time the CSI was administered 60 or more days before the ET, perhaps outside the window of optimal relief from the injection. These Results challenge claims of guideline-adherent care occurring within the Military Health System. Future research is necessary to better understand clinical decision-making that is guiding the use of these interventions within this health system.