Cox regression analysis for TNFi discontinuation stratified by diagnosis. Values are HR (95% CI).
| Variables | AS | PsA | ||
|---|---|---|---|---|
| Univariate | Final Model after Backward Selection | Univariate | Final Model after Backward Selection | |
| Sex, male vs female | 0.66 (0.48–0.93)a | 0.73 (0.51–1.04)* | 0.64 (0.46–0.88)b | 0.62 (0.45–0.86)b |
| Age, per 10 yrs | 0.94 (0.74–1.32) | 0.91 (0.64–1.29) | ||
| Symptom duration, < vs ≥ 5 yrs | 1.40 (1.00–1.97)* | 0.90 (0.65–1.26) | ||
| TNFi agent used, ETN vs other | 2.17 (1.50–3.15)c | 1.77 (1.19–2.63)b | ||
| TNFi agent used | ||||
| IFX, ref | 1.00 (ref) | |||
| ETN | 1.22 (0.81–1.83) | |||
| ADA | 1.25 (0.75–2.06) | |||
| GOL | 2.35 (1.12–4.91)a | |||
| Year of TNFi therapy start, per 2 yrs | 1.24 (1.11–1.38)c | 1.17 (1.04–1.31)a | 1.04 (0.92–1.18) | |
| Previous csDMARD, yes vs no | 0.72 (0.53–0.99)a | 0.65 (0.47–0.90)b | 1.06 (0.65–1.74) | |
| MTX co-therapy, yes vs no | 0.89 (0.62–1.30) | 0.65 (0.46–0.92)a | 0.61 (0.43–0.87)b | |
| Axial disease, yes vs no | 0.91 (0.60–1.37) | |||
| Peripheral disease, yes vs no | 1.41 (1.02–1.94)a | 1.53 (1.11–2.10)b | 1.98 (0.82–4.76)** | 2.20 (0.90–5.34)* |
| CRP, > vs ≤ 1.2 mg/dl | 0.83 (0.58–1.19) | 0.74 (0.49–1.11)** | ||
| VAS global, > vs ≤ 60 | 1.15 (0.82–1.63) | 1.09 (0.76–1.56) | ||
| PGA, > vs ≤ 2 | 1.07 (0.73–1.58) | 1.06 (0.71–1.58) | ||
| BASDAI, > vs ≤ 5 | 1.07 (0.74–1.53) | |||
| BASFI, > vs ≤ 5 | 1.27 (0.84–1.94) | |||
| SJC-28, > vs ≤ 2 | 1.05 (0.72–1.52) | |||
| TJC-28, > vs ≤ 3 | 1.24 (0.84–1.84) | |||
| DAS28-ESR baseline, > vs ≤ 4.5 | 1.06 (0.70–1.62) | |||
↵* p < 0.1.
↵**: p < 0.2.
↵a p < 0.05.
↵b p < 0.01.
↵c p < 0.001. TNFi: tumor necrosis factor inhibitor; AS: ankylosing spondylitis; PsA: psoriatic arthritis; IFX: infliximab; ETN: etanercept; ADA: adalimumab; GOL: golimumab; csDMARD: conventional synthetic disease-modifying antirheumatic drugs; MTX: methotrexate; CRP: C-reactive protein; VAS: visual analog scale; PGA: physician’s global assessment; BASDAI: Bath Ankylosing Spondylitis Disease Activity Score; BASFI: Bath Ankylosing Spondylitis Functional Index; SJC-28: swollen joint count in 28 joints; TJC-28: tender joint count in 28 joints; DAS28-ESR: 28-joint Disease Activity Score in 28 joints using erythrocyte sedimentation rate.