Does non-erosive rheumatoid arthritis exist? A cross-sectional analysis and a systematic literature review Jenny Amaya-Amaya a,b , Omar-Javier Calixto a,b , Sandra Saade-Lemus a , Enrique Calvo-Paramo a,c , Ruben-Dario Mantilla a,b , Adriana Rojas-Villarraga a,b , Juan-Manuel Anaya a,b,n a Center for Autoimmune Diseases Research (CREA), School of Medicine and Health Sciences, Universidad del Rosario, Carrera 24 No. 63C-69,11001000 Bogotá, Colombia b Mederi, Hospital Universitario Mayor, Bogotá, Colombia c School of Medicine, Universidad Nacional de Colombia, Bogotá, Colombia article info Keywords: Rheumatoid arthritis Erosions X-ray Ultrasonography Computed tomography abstract Objective: To evaluate the prevalence and factors associated with non-erosive rheumatoid arthritis (RA). Methods: First, a cross-sectional analytical study was performed. Non-erosive disease, defined as the absence of any erosion on X-rays after 5 years of RA, was evaluated in 500 patients. Further and additional evaluations including ultrasonography (US) and computed tomography (CT) were performed in those patients meeting the eligibility criteria. The Spearman correlation coefficient, kappa analysis, and Kendall's W test were used to analyze the data. Second, a systematic literature review (SLR) was performed following the PRISMA guidelines. Results: Of a total of 40 patients meeting the eligibility criteria for non-erosive RA, eight patients were confirmed to have non-erosive RA by the three methods. A positive correlation between non-erosive RA and shorter disease duration, antinuclear antibodies positivity, lower rheumatoid factor (RF) and C-reactive protein titers, lower global visual analog scale values, toxic exposures, and lower disease activity-(RAPID3) was found. In addition, an inverse correlation with anticyclic citrullinated peptide antibodies (ACPA) positivity and medication use was observed. From the SLR, it was corroborated that factors associated with this subphenotype were shorter disease duration, younger disease onset, negative ACPA and RF titers, low cytokine levels, and some genetic markers. Conclusion: Non-erosive RA is rare, occurring in less than 2% of cases. These findings improve on the understanding of RA patients who present without erosions and are likely to have less severe disease. & 2014 Elsevier Inc. All rights reserved. Introduction Rheumatoid arthritis (RA) is the most common inflammatory arthropathy worldwide. It is a chronic, multifactorial, and hetero- geneous autoimmune disease (AD) characterized by the presence of long-standing inflammation of the diarthrodial joints, resulting in symmetric polyarthritis and systemic manifestations [1]. As with most ADs, it predominantly affects women, and its preva- lence has been reported to be between 0.3% and 1.6% [2]. Chronic inflammation ultimately affects the joints, causing synovial membrane hypertrophy and bone and cartilage destruction. Erosion is the hallmark of the disease and is generally progressive and considered irreversible [1]. The presence of erosion at diagnosis is one of the factors most suggestive of a poor prognosis. Erosion reflects the cumulative history of the disease and correlates with deformity, disability (i.e., mainly in the hands [3] and the feet [4]), decreased functional work capacity, premature mortality, and reduced quality of life [5]. Radiographic assessment of joint damage is the most widely accepted standard tool for diagnosis, determination of disease extent, and RA follow-up. Despite demonstrating a high specif- icity in erosion detection, radiographs lack sensitivity in early disease [6]. Consequently, newer imaging techniques have been developed to improve diagnosis and follow-up of bone erosions [7]. Contents lists available at ScienceDirect journal homepage: www.elsevier.com/locate/semarthrit Seminars in Arthritis and Rheumatism http://dx.doi.org/10.1016/j.semarthrit.2014.09.006 0049-0172/& 2014 Elsevier Inc. All rights reserved. This work was supported by “Fondo de Investigación de la Universidad del Rosario (FIUR),” Bogota, Colombia. n Corresponding author at: Center for Autoimmune Diseases Research (CREA), School of Medicine and Health Sciences, Universidad del Rosario, Carrera 24 No. 63C-69, Bogotá 11001000, Colombia. E-mail address: juan.anaya@urosario.edu.co (J.-M. Anaya). Seminars in Arthritis and Rheumatism 44 (2015) 489–498