Comparative pain reduction of oral non-steroidal anti-inflammatory drugs and opioids for knee osteoarthritis: systematic analytic review

Abstract

Objective Summarize the comparative effectiveness of oral non-steroidal anti-inflammatory drugs (NSAIDs) and opioids in reducing knee osteoarthritis (OA) pain. Methods Two reviewers independently screened reports of randomized controlled trials, published in English between 1982 and 2015, evaluating oral NSAIDs or opioids for knee OA. Included studies were at least eight weeks duration, conducted in Western Europe, the Americas, New Zealand, or Australia, and reported baseline and follow-up pain using the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Pain subscale (0–100, 100-worst). Effectiveness was evaluated as reduction in pain, accounting for study dropout and heterogeneity. Results 27 treatment arms (9 celecoxib, 4 non-selective NSAIDs [diclofenac, naproxen, piroxicam], 11 less potent opioids [tramadol], and 3 potent opioids [hydromorphone, oxycodone]) from 17 studies were included. NSAID and opioid studies reported similar baseline demographics and efficacy withdrawal rates; NSAID studies reported lower baseline pain and toxicity withdrawal rates. Accounting for efficacy-related withdrawals, all drug classes were associated with similar pain reductions (NSAIDs: −18; less potent opioids: −18; potent opioids: −19). Meta-regression did not reveal differential effectiveness by drug class but found that study cohorts with a higher proportion of male subjects and worse mean baseline pain had greater pain reduction. Similarly, results of the network meta-analysis did not find a significant difference in WOMAC Pain reduction for the three analgesic classes. Conclusion NSAIDs and opioids offer similar pain relief in OA patients. These data could help clinicians and patients discuss likely benefits of alternative analgesics.