Association of circulating levels of MMP-8 with mortality from respiratory disease in patients with rheumatoid arthritis

Table 5 Multivariate baseline predictors of all-cause and respiratory disease-related mortality in rheumatoid arthritis patients

Step and variable	Hazard ratio (95% CI)	P
All-cause mortality
1. Age (years)	1.08 (1.06, 1.09)	< 0.0001
2. Nodular disease	2.05 (1.27, 3.31)	< 0.0001
3. Taking CVD drugs	1.96 (1.25, 3.05)	0.002
4. Smoking at baseline	1.97 (1.25, 3.10)	0.007
5. MMP-8 (ng/ml)	1.01 (1.002, 1.02)	0.01
Respiratory disease mortality
1. Age (years)	1.12 (1.08, 1.16)	< 0.0001
2. MMP-8 (ng/ml)	1.013 (1.006, 1.02)	< 0.0001
3. Nodular disease	2.76 (1.38, 5.50)	0.0002
4. Rheumatoid factor	2.57 (1.18, 5.58)	0.02

Stepwise Cox proportional hazards regression. Baseline variables included in the analyses were age, sex, disease duration, MMP-8 (ng/ml), rheumatoid factor (+/-), nodular disease (+/-), C-reactive protein (mg/l), erythrocyte sedimentation rate (mm/h), Health Assessment Questionnaire (score 0 to 3), Larsen score, taking drugs for cardiovascular disease (CVD) (+/-) and smoking at entry (+/-). Only significant variables retained in the stepwise analysis are shown.

ISSN: 1478-6362