Improvements in fatigue following anti-TNF therapy do not reflect reductions in inflammation and are instead a downstream consequence of pain improvements: results from the British Society for Rheumatology Biologics Register for Rheumatoid Arthritis

Background – Rheumatoid Arthritis (RA) related fatigue improves substantially following anti-TNF therapy commencement. It has been shown that, rather than being produced directly by reductions in disease activity, these improvements are largely mediated by changes in pain. Yet, as pain has a complex aetiology, it is not clear whether the observed improvements in fatigue reflect reductions in centrally or peripherally (e.g. inflammation) driven pain. In clarifying whether it is inflammatory pain which should be targeted to reduce fatigue, this study sought to inform optimised future management of this burdensome symptom. Methods –The British Society for Rheumatology Biologics Register for Rheumatoid Arthritis (BSRBR-RA) comprises a large cohort of patients who commenced anti-TNF therapies between October 2000 and November 2008. Data of interest, from 2652 BSRBR-RA participants with high levels of fatigue (SF36 Vitality ≤12.5), comprised change over 6 months in fatigue, pain (SF36 Vitality and Bodily Pain, respectively) and disease activity constituents (DAS28; ESR, global health, swollen and tender joints). Principal Components Factor Analysis (PCFA), a data-reduction method commonly used to collapse or cluster multiple items into fewer theoretical, or unmeasured, constructs (factors), was used to determine common clusters of change in symptoms. The factors which explained a greater proportion of total variance of all items (indicated eigenvalue>1) than any individual item (variance standardized to 1.0) were accepted. To maximize factor-loading strength on one factor and minimize cross-loading (loading>0.3 on ≥2 factors) the correlation between each item and the factors were “rotated” using varimax rotation. Results –Two factors explained more variance in the data than single items alone (eigenvalues; factor 1=2.39, factor 2=1.14). Changes in both swollen and tender joint counts, along with change in ESR loaded most strongly on the first factor, while changes in pain and fatigue, loaded on the second; change in VAS global health showed loadings >0.3 on both factors. With a loading of 0.3, changes in ESR only just met factor loading criteria, suggesting a weak association was present. Conclusions – Our findings suggest that changes in fatigue observed following anti-TNF therapies share an underlying construct with changes in pain, but not with changes in inflammation. Therefore changes in fatigue do not appear to be a direct effect of the drug reducing inflammatory pain. Instead, improvements likely represent the downstream consequence of anti-TNF therapies, reducing pain and associated improvements in general well-being. Accordingly, it is suggested that, in addition to immunomodulation, reductions in fatigue could be promoted by a multidisciplinary management strategy specifically to reduce pain.