Background Patients with rheumatoid arthritis (RA) show higher mortality rates than the general population (1). The increase has been correlated with several features as the long-term activity of the disease, corticosteroid use and autoantibody seropositivity. Regarding autoantibodies, an increase in mortality has been associated with rheumatoid factor (RF) and anti-citrullinated protein antibodies (ACPA), but there is not consensus on the relative role of them. Recently, the anti-carbamylated protein antibodies (anti-CarP) have also been associated with increased mortality in RA patients (2), but in the early RA patients that were studied only RF seemed to be independently associated with all-cause mortality.
Objectives The aim of this study was to determine the association of RF, ACPA and anti-CarP autoantibodies with all-cause mortality in patients with established RA.
Methods Patients with established RA were recruited from 2001 to 2009 and followed until November 2015. Full data were available for 274 patients and only this subset was included in the present analysis. Anti-CarP was assessed by homemade ELISA with in vitro carbamylated fetal bovine serum (3). Cox proportional hazards regression models for all-cause mortality were used adjusting for age, sex and smoking status, and hazard ratios (HR) of ACPA, anti-CarP and FR positivity were obtained in antibody-specific and -combination models.
Results At recruitment, patients showed the following characteristics: age =68.6 years (SD=7.95), disease duration =16.1 years (11.28), RF =61.8%, ACPA =66.4%, anti-CarP =40.2%, ever smokers =19.7%. At the end of the study 171 patients have died (62.4%). Age and smoking at recruitment showed increased hazard ratios for all-cause mortality, whereas gender was not associated.
Each of the three RA autoantibodies was associated with increased all-cause mortality in antibody-specific analysis with very similar HR: RF =1.59 (1.03–2.44, p=0.036), ACPA =1.57 (1.01–2.45, p=0.046), and anti-CarP =1.60 (1.07–2.39, p=0.024). The similarity of effects was also observed in the combined analysis including a variable for each of the antibodies, with the following HR: RF =1.30 (0.79–2.14), ACPA =1.22 (0.72–2.05), and anti-CarP =1.37 (0.88–2.13), none of them significant. Therefore, it seemed appropriate to consider the three autoantibodies as additive factors: the additive showed HR for the number of autoantibodies =1.29 (1.08–1.56, P=0.007) per each additional antibody.
Conclusions The three RA autoantibodies were associated with increased all-cause mortality without clear differences between them, making it appropriate to consider the antibodies as additive factors contributing to risk stratification. This additive contribution is similar to the described for other aspects of RA activity and severity where patients with the largest number of autoantibodies are the most affected.
Sparks JA, et al. Arthritis Care Res (Hoboken). 2015. doi: 10.1002/acr.22752
Ajeganova S, et al. Ann Rheum Dis. 2016. doi: 10.1136/annrheumdis-2015-208579
Shi J, et al. Proc Natl Acad Sci USA 2011; 108:17372–7
Acknowledgement Funding was provided by grants PI12/01209, PI14/01561 and RD12/009/008 of the Instituto de Salud Carlos III (Spain) with participation of the European Regional Development Fund of the EU
Disclosure of Interest None declared