Article Text

FRI0181 The Course of Fatigue in Rheumatoid Arthritis; A 10 Year Longitudinal Cohort Study
  1. H.W. Van Steenbergen1,
  2. R. Tsonaka2,
  3. T.W. Huizinga1,
  4. A.H. van der Helm-van Mil1
  1. 1Rheumatology
  2. 2Medical Statistics, LUMC, Leiden, Netherlands


Background Fatigue is a commonly experienced symptom in rheumatoid arthritis (RA) and considered by patients as one of the most important consequences of the disease, as it negatively influences many aspects of daily functioning. The cause of fatigue is unknown, but a multidimensional origin with interactions between disease processes, cognitive and personal factors is presumed. Studies thus far have been inconsistent in whether or not inflammation is associated with fatigue. It is also unknown whether other factors that are generally considered as severity factors (for instance presence of auto-antibodies) are associated with fatigue. Furthermore, the long-term course of fatigue in RA has never been studied thoroughly.

Objectives We aimed to study associations between “known RA severity factors” and the severity of fatigue over time.

Methods Patients included in the Leiden Early Arthritic Clinic were studied; of the 2549 patients that were included since 1993, 999 patients fulfilled the 1987-ACR RA criteria and were studied during a maximal follow-up of 10 years. Fatigue severity was evaluated using a visual analog scale (VAS) (scale 0-100 mm). 68-Tender joint counts (TJC), 66-swollen joint counts (SJC), C-reactive protein levels (CRP) and haemoglobin (Hb) levels were yearly measured. Yearly made hands and feet radiographs were scored according to the Sharp-van der Heijde method. To avoid possible bias induced by missing data, multiple imputation was performed by creating five datasets. The repeated measurements were analyzed using univariable and multivariable linear quantile mixed models, as the data were non-normally distributed. Treatment strategies changed over time for the periods 1993-1995, 1996-1998 and ≥1999.

Results At first visit, patients with RA were more fatigue than other arthritis patients. In RA, the median VAS fatigue at baseline was 49 mm (IQR 15-69), it decreased during the first year till a median score of 34 mm and remained rather stable thereafter. Over the 10 years, female RA patients remained more fatigue than male (difference 6.1 mm, p<0.001). Furthermore, patients that over time had higher TJCs, SJCs and CRP levels reported to be more fatigue (all p<0.001). Also RF-negative patients reported to be more fatigue, though the absolute difference was small (2.7 mm, p=0.029); no difference was observed for ACPA. Also no association was observed for Hb. In multivariable analysis female gender, higher TJC and higher CRP were significantly associated with more severe fatigue over time (p=0.004, p<0.001 and p<0.001, beta for TJC 0.8 mm per tender joint, beta for CRP 1.0 mm per 10 mg/L). Although the radiographic progression rate decreased significantly with improved treatment strategies, the course of fatigue was not improved when better treatment strategies were applied (p=0.93). Also the course of fatigue was not associated with radiographic severity over 10 years (p=0.80). Patients experiencing more fatigue, had higher HAQ scores over time (p<0.001).

Conclusions In this first large longitudinal study on fatigue in RA, fatigue was associated with markers of inflammation, though the effect sizes were small. The beneficial effect of improved treatments on radiographic outcome was not accompanied by a beneficial effect on fatigue. Fatigue remains an “unmet need” in RA that is associated with functional loss.

Disclosure of Interest None declared

DOI 10.1136/annrheumdis-2014-eular.3489

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.