Effect of intra-articular corticosteroid injections for osteoarthritis on the subsequent use of pain medications: a UK CPRD cohort study.
Hawley S., Prats-Uribe A., Matharu GS., Delmestri A., Prieto-Alhambra D., Judge A., Whitehouse MR.
OBJECTIVES: To estimate the effect of intra-articular corticosteroid injection (IACI) for osteoarthritis on longer-term incidence of pain medications. METHODS: We conducted a cohort study of patients registered in the UK Clinical Practice Research Datalink (CPRD) GOLD primary care database with an incident diagnosis of knee, hip, hand, or shoulder osteoarthritis between 2005-2019. Exposure of interest was single or repeated use of IACI (analysed separately). Main outcome measures were five-year incidence of uncombined opioids, opioid-nonopioid analgesic combinations, oral corticosteroids, paracetamol, oral Non-Steroidal Anti-Inflammatory Drugs (NSAIDs), and topical NSAIDs. Instrumental Variable (IV) analysis was used given this methodology can account for strong and unmeasured confounding. Secondary analyses used propensity-score matching and Cox regression. RESULTS: Amongst 74 527 knee osteoarthritis patients, IACI use was associated with lower subsequent prescribing of most pain medications studied, including opioid-nonopioid analgesic combinations following single IACI (number needed to treat [NNT]=5 [5-6], p< 0.001) and uncombined opioids following repeat IACI (NNT = 12 [95% CI: 8-546], p= 0.049). Amongst 15 092 hand osteoarthritis patients, single IACI was associated with reduced use of opioid-nonopioid combinations, paracetamol, and oral NSAIDs. Secondary analyses confirmed lower incidence rates of opioid-nonopioid combinations after single IACI for knee (hazard ratio [HR] =0.88 [0.81-0.96]), hip (HR = 0.76 [0.62-0.92]), hand (HR = 0.77 [0.61-0.98]), or shoulder (HR = 0.72 [0.53-0.99]) osteoarthritis. CONCLUSIONS: IACI for knee or hand osteoarthritis showed lower incidence of several pain medications over the longer-term relative to no IACI use. Secondary findings suggest IACI may be effective in reducing longer-term use of opioid-nonopioid analgesic combinations for patients with knee, hip, hand, or shoulder osteoarthritis.