Article Text

THU0088 Iron Deficiency and Low Erythropoietin Concentrations Are Frequent Anemia Causing Factors in Rheumatoid Arthritis
  1. B. Möller1,
  2. A.B. Leichtle2,
  3. P. Meyer3,
  4. M. Fiedler2,
  5. A. Finckh4,
  6. D. Kyburz5,
  7. C. Gabay4,
  8. P.M. Villiger1
  1. 1Rheumatology, Immunology & Allergology
  2. 2Clinical Chemistry
  3. 3Rheumatology & Clinical Immunology, Inselspital - Bern University Hospital, Bern
  4. 4Rheumatology, University Hospitals of Geneva, Geneva
  5. 5Rheumatology, Basel University Hospital, Basel, Switzerland


Background Anemia was recently re-evaluated as a predicting parameter for radiographic joint damage progression in addition to standard disease activity measures in rheumatoid arthritis (RA) [1].

Objectives To characterize the origin of anemia in RA.

Methods All patients in the Swiss SCQM-RA cohort with available hemoglobin concentrations and a biobank sample were included. Anemia was defined according to WHO definitions (hemoglobin <12g/dl in women and <13 g/dl in men). Mean corpuscular volume (MCV) and hemoglobin (MCH) were locally acquired. Serum samples were collected and stored according to stringent pre-analytical quality requirements. Serum iron, transferrin, ferritin, soluble transferrin receptor (sTfR) and erythropoietin (EPO) concentrations were measured using standard procedures in an accredited central laboratory. Non-RA patients with normal CRP served as control for logEPO/Hb quotient interpretation. Iron deficiency anemia (IDA) was defined according to [2] in case of 1. t-sat ≤20% & ferritin ≤30 μg/l or 2. sTfR index, the quotient of sTfR divided by log-transformed ferritin concentrations ≥1mg/μg. Anemia of chronic disorders (ACD) was diagnosed by signs of inflammation plus 1. T-sat ≤20% & ferritin ≥100 μg/ml or 2. sTfR index <1mg/μg & ferritin ≥30 μg/l. Serum hepcidin-25 was quantified using commercially available ELISA kits from Peninsula Laboratory, LLC, San Carlos, CA.

Results 106 (14%) of the 779 patients included were anemic at the time of sample collection. Anemic patients were significantly older, had more comorbidities, higher DAS-28 and used more often corticosteroids, but did not differ in NSAID, conventional (75%) or biological (70%) DMARD use. They had lower MCV, lower MCH and lower transferrin levels (p<0.04) than non-anemic patients, but other serum parameters were not significantly different. sTfR was more sensitive than transferrin concentrations to detect possible iron deficiency (14% versus 1%).

IDA was detected in 98 (92%) of the 106 anemic patients. ACD was diagnosed in 61 (58%) of the anemic cases, when active disease was defined by CRP >10 mg/l or ESR >30 mm/h, or in 82 (77%) of subjects after including DAS-28 >3.2 as additional marker of inflammation. Only 3 respectively 6 cases could be attributed to ACD alone, while 58 (55%) respectively 76 (72%) of anemia cases had ACD and IDA in combination, dependent of the definition of inflammation. In a multivariate linear regression analysis (r2=0.38, n=1075, p<0.0001), logEPO concentrations depended on hemoglobin concentrations (-0.1, 95%CI -0.11 to -0.09, p<0.0001), which increased significantly less in RA (-0.29, 95%CI -0.34 to -0.24, p<0.001) with decreasing hemoglobin concentrations than in controls.

Conclusions Iron deficiency was the most common cause of anemia in this large RA cohort with representative prevalence of anemia for an intensively treated RA population. We further found inadequately low EPO levels in relation to observed hemoglobin concentrations among RA patients, which may further contribute to anemia in these patients. Their relation to radiographic damage progression needs to be determined.

  1. Moller B, et al. Ann Rheum Dis 2014;73:691–6.

  2. van Santen S, et al. Arthritis Rheum 2011;63:3672–80.

Disclosure of Interest B. Möller Grant/research support from: VIFOR Pharma, Villars-sur-Glâne, Switzerland, A. Leichtle: None declared, P. Meyer: None declared, M. Fiedler: None declared, A. Finckh: None declared, D. Kyburz: None declared, C. Gabay: None declared, P. Villiger: 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.