Article Text

SAT0386 Extreme patient reported outcome (PRO) in early spondyloarthritis: a surrogate for fibromyalgia? its impact on TNF-ALPHA blockers treatment effect?


Background In case of a concomitant fibromyalgia (FM) with axial spondyloarthritis (axSpA), there is a risk of misclassifying a patient as active (e.g. BASDAI >4), and to falsely consider him as refractory to NSAIDs/ biologics, since FM patients often report higher level of pain and fatigue1. We hypothesized that not only are extreme patient reported outcome (PRO) potentially surrogate marker of FM in axSpA, but also that this extreme PRO may have an impact on the TNF-α blockers (TNFb) treatment effect

Objectives A) To describe the prevalence of extreme PRO in an early axSpA cohort. B) To compare the phenotype of axSpA patients with and without extreme PRO. C) To assess the impact of extreme PRO on the TNFb efficacy.

Methods This analysis was performed on the DESIR cohort which included 708 adult patients (>18 and <50 years) with inflammatory back pain suggestive of axSpA (according to the rheumatologist's conviction of ≥5/10) for >3 months but <3 years duration. All patients were biologic naïve at inclusion and were followed up every 6 months for the first 2 years. At baseline, data pertaining to demographics, BASDAI, history of depression, type of medications (antidepressants or muscle relaxants) used and concomitant diseases were collected. It is worth noting that no systematic assessment of the fulfilment of the ACR criteria for FM was performed in this study; thus, we created a “FM gold standard” according to the available data (i.e., presence of either Muscle relaxants/ Depression/ Anti-depressant drug treatment/ Fibromyalgia comorbidity reported in the CRF). BASDAI was the selected PRO for this study due to its widespread use in clinical practice. BASDAI was tested against this “FM gold standard”: we plotted ROC curves to define the best cut-off to define an “extreme PRO” for BASDAI. Phenotype of patient's with/without extreme PRO scores was compared. Impact of extreme PRO score on TNFb efficacy was assessed by comparing the retention rate of the first TNFb by Cox analysis.

Results ROC curves to define an “extreme PRO” determined a different cut-off for each BASDAI question (i.e. >6, >5, >1, >4, >5 and >3 for question 1 to 6, respectively), and the need of at least 4 out of these 6 cut-offs to fulfil the “extreme PRO” condition; giving us a prevalence of 42.9% (304 patients) of extreme PRO in DESIR. Phenotypically, this group with extreme PRO, consisted of older patients (34.6 (8.3) vs 33.1 (8.8)), had more females (184 (60.5%) vs 195 (48.8%)), reported less sacroiliitis [radiographic and MRI, (36 (12%) vs 76 (19.4%)) and (82 (27.6%) vs 148 (37.8%)), respectively], showed less HLA B27 positivity (160 (52.6%) vs 248 (62.2%)), had higher CRP values (102 (34.3%) vs 102 (26.5%)), and more arthritis/ enthesitis history (212 (69.7%) vs 190 (47.5%)/ 182 (59.9%) vs 166 (41.5%)). A lower retention rate was observed in the group of patients with “extreme PRO” (Figure 1)

Conclusions Coexistence of extreme PRO might be considered as a surrogate marker for FM in axSpA patients. Moreover, it appears to have a negative impact on TNF-α blockers retention rate.


  1. Wendling D, Prati C. Spondyloarthritis and fibromyalgia: interfering association or differential diagnosis? Clin Rheumatol 2016;35:2141–43.


Acknowledgements The DESIR cohort is financially supported by unrestricted grants from the French Society of Rheumatology and Pfizer France.

Disclosure of Interest B. Meghnathi Grant/research support from: ASAS Society (ASAS FELLOW), A. Etcheto: None declared, P. Claudepierre: None declared, M. Dougados: None declared, A. Moltό: None declared

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.