Incidence and prevalence of polyautoimmunity in seropositive compared with seronegative patients with rheumatoid arthritis: a nationwide cohort study

A. Hagelskjær, R. Cordtz, S. Bliddal, A. S. Mortensen, S. Kristensen, C. H. Nielsen, U. Feldt-Rasmussen, C. Torp-Pedersen, L. Dreyer

Research output: Contribution to journalConference abstract in journalResearchpeer-review


Background: Rheumatoid arthritis (RA) is categorized as seropositive or seronegative referring to the presence or absence of IgM rheumatoid factor (IgM-RF) and/ or anti-citrullinated protein antibodies (ACPA).Patients with an autoimmune disease are more likely to develop additional autoimmune conditions than individuals without pre-existing autoimmune disease.Objectives: The aims of this study were to examine patterns of polyautoimmunity in seropositive compared with seronegative patients with recent-onset RA.Methods: The study was register-based and used the nationwide DANBIO register (identification of recent-onset (<1 year) seropositive, ICD-10 M05, and seronegative, M06, RA patients) linked to the Danish National Patient Registry and the Danish National Prescription Registry to obtain information on additional autoimmune diseases (see Table 1 for autoimmune conditions).View this table:Table 1. Characteristics, prevalence, and incidence of polyautoimmunity in incident RA patients.Using age and sex adjusted logistic regression analysis, the odds ratio (OR) of prevalent polyautoimmunity in seropositive compared with seronegative patients at the time of RA diagnosis was calculated.To estimate the hazard ratio (HR) for developing yet another autoimmune disease in the 5 years after RA diagnosis, adjusted cause-specific Cox regression models were performed. Several sensitivity analyses were carried out including alternative exposure and outcome definitions.Results: In total, 12,517 patients with recent-onset RA were included. The groups were similar in terms of disease characteristics and DMARD treatment, but seropositive patients were younger and included more women, see Table 1. Patients with seropositive RA had an OR of 0.79 (95% CI 0.71-0.89) for baseline presence of polyautoimmunity compared with seronegative patients, whereas the 5-year HR was 0.86 (95% CI 0.71-1.05) for incident polyautoimmunity. The results remained similar in all sensitivity analyses.Conclusion: Patients with seropositive RA had a slightly lower prevalence and incidence of polyautoimmunity compared to seropositive patients. The results were somewhat surprising, yet very robust, and thus raises the question if seronegative RA is in fact “more autoimmune” despite the absence of (identified) autoantibodies.Acknowledgements: We acknowledge all patients and all Danish departments of rheumatology contributing to the DANBIO registry.Disclosure of Interests: Amalie Hagelskjær: None declared, René Cordtz: None declared, Sofie Bliddal: None declared, Anders Sandermann Mortensen: None declared, Salome Kristensen: None declared, Claus Henrik Nielsen: None declared, Ulla Feldt-Rasmussen: None declared, Christian Torp-Pedersen: None declared, Lene Dreyer Grant/research support from: Grants from BMS, Galderma, and Eli Lilly
Original languageEnglish
JournalAnnals of the Rheumatic Diseases
Issue numberSuppl. 1
Pages (from-to)382-383
Number of pages2
Publication statusPublished - 1 Jun 2021
EventAnnual European Congress of Rheumatology, EULAR 2021 - Virtuel
Duration: 2 Jun 20215 Jun 2021


ConferenceAnnual European Congress of Rheumatology, EULAR 2021
Internet address

Bibliographical note

Abstract no. POS0313

Cite this