Hematologic parameters predicting a response to oral iron therapy in chronic inflammation.

In anemic patients with chronic inflammatory conditions, the assessment of iron status and the selection of those who may benefit from iron therapy frequently presents diagnostic challenges. Traditional biochemical iron parameters, such as ferritin, serum iron and transferrin, are affected by inflammation making them less suitable indicators. Recently, iron parameters have become available that provide direct information on iron availability for erythropoiesis: the central iron regulatory hormone hepcidin and the hemoglobin content of reticulocytes and erythrocytes (Ret-He and RBC-He). We hypothesized that changes in these functional iron parameters shortly after start of iron supplements may predict an increase in hemoglobin in a longer course of iron supplements in patients with chronic inflammation. Forty adult patients diagnosed with a chronic rheumatic disease and anemia were selected during routine visits in 2012 at the rheumatology outpatient clinic of the Radboud University Medical Center, Nijmegen, the Netherlands. Anemia was defined as a hemoglobin concentration below 12.1 g/dL for males and below 11.3 g/dL for female, respectively. Exclusion criteria were signs of hemolysis (increased levels of lactate dehydrogenase), end stage renal disease, liver disease (two times the upper limit of the reference value of alanine aminotransaminase (ALT)), known malignancies, vitamin B12 or folic acid deficiency, use of iron supplements, erythropoietin or medications with potential drug interactions with the iron supplements and a blood transfusion in the previous two months. Patients’ characteristics, including age, sex, medication and the type of inflammatory condition, were recorded at baseline (Table 1). Rheumatoid arthritis was the most common diagnosis (58%) and disease-modifying drugs (60%) and nonsteroidal anti-inflammatory drugs (53%) were frequently used. Ethical clearance was granted by the Radboud University Medical Center, and written informed consent was obtained from all participants. Ferrofumarate 200 mg BD (130 mg of elemental iron per day) was prescribed during six weeks. Blood samples were collected at baseline, after one, two and six weeks. Responders to iron therapy were defined as patients with an increase in hemoglobin of at least 0.81 g/dL (0.5 mmol/L) after six weeks of oral iron supplementation. The hemoglobin content parameters were measured on a Sysmex XE5000 analyer (Sysmex Corporation). Serum hepcidin-25 measurements were performed using a combination of weak cation exchange chromatography and time-of-flight mass spectrometry as described previously. The other iron parameters were determined as previously described. At week 6, 2 patients had withdrawn from the study: one at personal request and one because of pneumonia. Sixteen of 38 participants who completed the study were responders. Responders had significantly lower base-line values of ferritin, hepcidin, Ret-He and RBC-He and transferrin saturation (TS), and higher total iron binding capacity (TIBC), soluble transferrin receptor (sTfR) and its ratio to log ferritin (sTfR-index) levels in comparison to nonresponders (Figure 1). Levels of CRP were similar between the groups. At one week of iron therapy, responders had a significantly larger increase in Ret-He level (0.28 ± 0.26 fmol) compared to the non-responders (0.18 ±0.04 fmol) (P<0.000) and a larger change in reticulocyte numbers (Figure 1). There was no increase in Ret-He between weeks 1 and 2. In contrast, there were no significant differences in the change in hepcidin levels at either time point. Moreover, responders had significantly larger increases in serum iron and TS values at one week with a concurrent significant decline in sTfR-index levels after two weeks of iron therapy. Regression analyses were applied to predict the change in hemoglobin and the outcome of being a responder after six weeks of iron therapy. The strongest association with the hemoglobin response at six weeks was found for baseline RBC-He, ferritin, Ret-He and hepcidin levels, and the increase after one week in values of TS, serum iron, RetHe, reticulocytes and ferritin, both in descending order (Table 2). The diagnostic performance of parameters was subsequently assessed using Receiver Operating Characteristic (ROC) analyses. The change in values of serum iron, TS, Ret-He and reticulocytes at one week had the highest areas under the curve (AUC) (Table 2). At one week, an increase in serum iron and Ret-He of more than 4 mmol/L and 0.11 fmol, respectively, had a sensitivity of 80% and 81%, and specificity of 91% and 86%, respectively. At baseline, values of ferritin, hepcidin and the sTfR-index had the highest AUC values, but the change in these parameters at week 1 was not a useful predictor of the response to iron therapy. The combination of the change in values of Ret-He and TS at one week provided the highest discriminatory power (AUC 0.906, SE 0.054; P<0.000). Overall, the increase after one week of oral iron therapy in Ret-He, TS, serum iron level and the reticulocyte count best predicted an increase in hemoglobin level of at least