1873: Relationship Between Baseline Characteristics of Rheumatoid Arthritis and Types of Mortality: Study of Cohort Followed Prospectively

Cristina Corrales¹, Fabricio Benavides Villanueva¹, Ivan Ferraz Amaro², Nuria Vegas Revenga³, Virgi Portilla⁴, Ricardo Blanco⁵, Miguel Ángel González-Gay⁶ and Alfonso Corrales⁴, ¹Hospital Universitario Marqués de Valdecilla, Santander, Spain, ²Division of Rheumatology. Hospital Universitario de Canarias. Spain., Santa Cruz de Tenerife, Spain, ³Hospital Galdakao- Usansolo, Galdakao, Spain, ⁴Research Group on Genetic Epidemiology and Atherosclerosis in Systemic Diseases and in Metabolic Bone Diseases of the Musculoskeletal System, IDIVAL; and Department of Rheumatology, Hospital Universitario Marqués de Valdecilla, Santander, Spain, ⁵Hospital Universitario Marqués de Valdecilla, IDIVAL, Santander, Spain, ⁶Department of Medicine and Psychiatry, Universidad de Cantabria; Rheumatology Division, Hospital Universitario Marqués de Valdecilla; Research group on genetic epidemiology and atherosclerosis in systemic diseases and in metabolic diseases of the musculoskeletal system, IDIVAL, Santander, Spain. Cardiovascular Pathophysiology and Genomics Research Unit, School of Physiology, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa

Background/Purpose: Patients with rheumatoid arthritis (RA) have an increased risk of mortality mainly due to cardiovascular (CV) events, infections and malignancies. The relationship between the characteristics of the disease at the baseline and the different types of mortality is not well established in RA. The main objective of this abstract is to establish the characteristics related to the disease that could be predictors of the most frequent types of mortality in patients with RA.

Methods: We conducted a prospective longitudinal study of 673 patients with RA in a single tertiary center. The different types of mortality and their association with patient- and disease-dependent factors were analyzed using multivariate Cox regression.

Results: We included 673 patients with RA, women (75%), mean age 61±13 years. The main baseline characteristics, CV risk factors, activity data, and treatments are detailed in Table 1.

After a follow-up of 4367 person-year (mean time 6.4±1.4 years), 67 deaths were observed. The causes of death in order of frequency were: infections (N=23) (34%), malignancies (N=18) (27%), CV events (N=12) (18%), lung disease (N=2) (3%), and others (N=12) (18%). The main results of the study are summarized in Table 2. In relation to mortality due to infections, arterial hypertension (HR 4.43 (1.64-11.94), p=0.003) and Diabetes Mellitus (DM) (HR 3.87 (1.65-9.17), p=0.002) showed a statistically significant association in the univariate analysis. In the multivariate regression analysis adjusted for CV risk factors, the characteristics of the disease that presented statistical significance were the presence of RF (HR 3.18 (1.24-8.14), p=0.016) and ACPA (HR 2.95 (1.19-7.33). ), p=0.020), high values of CRP (HR 1.03 (1.02-1.05), p < 0.001), DAS28-ESR (HR 1.53 (1.10-2.13), p=0.012) and DAS28-CRP (HR 1.66 (1.16 -2.37), p= 0.006), as well as Prednisone treatment (HR 2.73 (1.01-7.42), p=0.048) and Rituximab (HR 8.76 (2.39-32.08), p=0.001). In mortality due to malignancies, only the male gender (HR 3.12 (1.24-7.87), p = 0.016), abdominal circumference (HR 1.05 (1.03-1.09), p < 0.001) and DM (HR 3.56 (1.33-9.48), p=0.011) showed statistical significance. However, none of the disease-related factors were statistically associated with cancer mortality. In relation with CV mortality, only arterial hypertension (HR 3.90 (1.05-14.43), p=0.04) presented a statistically significant association in the univariate analysis. In multivariate analysis adjusted for CV risk factors, disease-related factors associated with CV mortality were elevated ESR (HR 1.04 (1.02-1.06), < 0.001) and DAS28-ESR (HR 1.79 (1.13- 2.82), p=0.012) and NSAIDs treatment (HR 5.48 (1.46-20.53), p=0.012).

Conclusion: Factors related to disease activity and different types of therapy at baseline were associated with mortality from CV events and infections in RA patients.
Table 1. Baseline characteristics of 673 RA patients.

Table 2. Multivariate analysis adjusted for age, sex and those CV risk factors that showed a statistically significant association in the univariate analysis. In bold, statistically significant results (HR, p < 0.05)
Disclosures: C. Corrales, None; F. Benavides Villanueva, None; I. Ferraz Amaro, AbbVie/Abbott, Merck/MSD, Janssen, Roche, AbbVie/Abbott, Pfizer, Roche, Amgen, Celgene, Merck/MSD; N. Vegas Revenga, None; V. Portilla, None; R. Blanco, Eli Lilly, Pfizer, Roche, Janssen, MSD, AbbVie, Amgen, AstraZeneca, Bristol Myers Squibb, Galapagos, Novartis, Sanofi; M. González-Gay, AbbVie/Abbott, Merck/MSD, Janssen, Roche, AbbVie/Abbott, Roche, Sanofi, Eli Lilly, Celgene, Sobi, Merck/MSD; A. Corrales, None.