Résumé
Prior meta-analyses have suggested a protective link between smoking and knee osteoarthritis (KOA), but they relied on aggregate data, potentially obscuring the true relationship. To address this limitation, we conducted an Individual Participant Data (IPD) meta-analysis using data from three large cohorts: the Osteoarthritis Initiative (OAI), the Multicenter Osteoarthritis Study (MOST), and the Cohort Hip and Cohort Knee (CHECK) study. Participants from 16 centers in the USA and Netherlands were categorized as current, former, or never smokers. We assessed six outcomes, three related to structural changes over 4-5 years of follow-up, and three related to changes in KOA symptoms over 2-2.5 years, 5 years, and 7-8 years of follow-up. First, the incidence of radiographic KOA was evaluated in 10,072 knees, defined as having a Kellgren-Lawrence grade ≥ 2 ('radiographic KOA') at follow-up but not at baseline. Second, the progression of radiographic KOA was evaluated in 5274 knees, defined as an increase in Kellgren-Lawrence grade between baseline and follow-up in knees that had radiographic KOA at baseline. Third, the incidence of symptomatic KOA was evaluated in 12,910 knees, defined as having radiographic KOA in addition to symptoms at follow-up but not at baseline. Fourth, fifth, and sixth, we investigated changes between baseline and all follow-ups in scores for the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) subscales of pain, disability, and stiffness (in 2640 knees). There were no differences between smoking groups in any of these six outcomes. Our study, leveraging data from three large cohorts and the advantages of IPD, finds no evidence that smoking offers any protection against KOA, refuting the notion that smoking may benefit joint health.