Association of long-term use of non-steroidal anti-inflammatory drugs with knee osteoarthritis: a prospective multi-cohort study over 4-to-5 years

Salis Z, Sainsbury A · Scientific Reports

Long-term NSAID users had higher odds of pain worsening (OR 2.04), disability (OR 2.21), and total knee replacement (OR 3.13). On radiographic deterioration the main adjusted analysis found higher odds of a worse Kellgren-Lawrence grade (OR 1.43, 95% CI 1.15–1.77), and the propensity-matched sensitivity analysis the abstract leads with did not reach significance (OR 1.25, 95% CI 0.94–1.65).

Population

4,197 participants (8,394 knees) from the OAI, MOST and CHECK cohorts followed 4–5 years; 435 long-term users against 3,762 non-users

Intervention

Long-term NSAID use, reported at baseline and every follow-up visit — prescribed NSAIDs including COX-2 inhibitors in OAI and MOST, and self-reported ibuprofen, diclofenac, naproxen, celecoxib or rofecoxib in CHECK (observed)

Comparator

Non-use

Limitations

Severe confounding by indication — users were heavier, more often women and had worse baseline symptoms and grades; symptom outcomes especially vulnerable; NSAID types pooled; over-the-counter painkillers and opioids not adjusted for; continuous use between visits assumed. No competing interests declared; the cohorts are NIH-funded with Merck, Novartis, GlaxoSmithKline and Pfizer (OAI) and the Dutch Arthritis Foundation (CHECK).