Objective: The objective of this study was to investigate the association between periodontitis (PD) and etanercept (ETN) discontinuation in anti-tumor necrosis factor (anti-TNF)-naive patients with rheumatoid arthritis (RA). Methods: This retrospective nationwide population-based cohort study identified 3359 anti-TNF-naive patients with RA (age at diagnosis ≥16 years) in whom ETN treatment was initiated using administrative data. We identified PD exposure within 5 years before ETN initiation and during ETN treatment. Cox proportional hazard models were used to assess ETN discontinuation risk associated with PD within 5 years before ETN initiation, shown as hazard ratios with 95% confidence intervals (CIs). Stratified analyses were performed on the basis of PD during ETN treatment to avoid violating the Cox regression assumptions. Results: Patients with PD history during the 5 years before ENT initiation had a higher risk of ETN discontinuation compared with those without such history; the hazard ratios of ETN discontinuation were 1.27 (95% CI, 1.01-1.60) and 1.17 (95% CI, 1.06-1.30) among patients with and without PD during ETN treatment, respectively. Other risk factors included age older than 65 years and daily prednisolone dose greater than 10 mg/d within 1 year before ETN initiation. Concomitant methotrexate, leflunomide, salazopyrin, or hydroxychloroquine administration had a protective effect on ETN discontinuation in patients without PD during ETN treatment, but the protective effect by leflunomide, salazopyrin, and hydroxychloroquine was attenuated in patients with PD during ETN treatment (P for interaction <0.05). Conclusions: A PD history within 5 years before ETN administration was associated with increased ETN discontinuation risk in anti-TNF-naive patients with RA.
- Drug survival
- Rheumatoid arthritis