Background
Anemia is a frequent complication of chronic kidney disease (CKD) [
1]. In CKD, the main causes of anemia are deficiency of erythropoietin, iron-restricted erythropoiesis and anemia of the chronic disease (ACD) [
2‐
4]. The latter originates from the chronic inflammation that is a hallmark of CKD patients and has been shown to be associated with adverse outcomes such as cardiovascular events, end-stage renal disease, increased mortality, and decreased quality of life [
5]. In ACD, pro-inflammatory cytokines upregulate hepcidin production in the liver which subsequently hampers iron uptake from the gut and iron release from the reticulo-endothelial system [
6,
7] which leads to functional iron deficiency that negatively affects erythropoiesis. Furthermore, increased iron utilization due to the use of erythropoiesis stimulating agents (ESA), and iron loss as a result of dialysis-related blood loss contribute to the high prevalence of anemia in patients with CKD [
8].
Oral administration of iron has limited efficaciousness and is associated with gastrointestinal side effects. By means of intravenous iron, gastrointestinal absorption is bypassed and incorporated more rapidly [
5]. Indeed, it has been established that intravenous iron supplementation as a treatment for iron deficiency anemia is superior to oral iron supplementation in non-dialysis dependent CKD, hemodialysis (HD) and peritoneal dialysis patients [
9‐
11]. Furthermore, ESA requirements have been shown to be decreased in patients receiving intravenous iron [
12].
Nowadays, several intravenous iron supplementations are available, of which iron sucrose (IS) and a more recently introduced intravenous (IV) iron compound, ferric carboxymaltose (FCM) (brand names Venofer and Ferinject, respectively) are frequently used. Although the efficacy of IS and that of FCM for treatment of anemia in CKD patients have been studied individually by comparing the formulations to oral iron supplementation [
10,
13], a head to head comparison of these two intravenous formulations has never been performed in HD patients. Therefore, the goal of this audit is to analyze the effects of a switch from IS to equally dosed FCM. Hence, we performed an audit on records of HD patients in three dialysis centers that switched from IS to FCM.
Discussion
In this analysis, we have shown that the switch from IS to FCM in HD patients was associated with a significant improvement of iron status, unrelated to iron dose. After the switch to FCM hemoglobin, hematocrit, serum ferritin, TSAT, and MCV increased, while transferrin levels decreased, reflecting erythropoiesis which is less restricted by iron deficiency [
17]. Furthermore, ESA dose decreased after the switch to FCM. This is, to our knowledge, the first investigation to compare the effects on iron status of a switch to FCM from IS in dialysis-dependent CKD patients.
In the subgroup of anemic patients at baseline especially, the effect of the switch on iron status was even more pronounced, despite the fact that these patients were prescribed significantly less iron and a smaller ESA dose after the switch. Switching from IS to FCM resulted in a marked increase in TSAT in both groups, a large increase of MCV in the anemic group, and an increase in serum ferritin in both groups. Although the baseline hemoglobin and iron status in the anemic and iron deficient groups were worse, the large improvement of iron status in these groups seems attributable to the effect of switching from IS to FCM since we did not identify a statistically significant general trend based solitarily on the factor time.
These results seem to indicate that switching from IS to similarly dosed FCM causes an increase in ferritin and TSAT in all groups, which is more pronounced in anemic patients. When speculating on possible mechanisms why FCM seems to be more effective than IS in replenishing iron stores, one could put forth the argument that FCM has increased bioavailability of elemental iron compared to IS. FCM and IS are both composed of an iron(III)-hydroxide core, surrounded by a carbohydrate shell (carboxymaltose and sucrose, respectively) [
18]. FCM has a higher molecular weight than IS (150,000 v 43,300 Da) and a longer half-life (7–12 v 5–6 h) which increase the area under the curve, indicating that the bioavailability of FCM is greater than that of IS [
18]. A second explanation might be that FCM is much more stable than IS, which prevents release of labile iron into the blood, where it can saturate transferrin and lead to significant amounts of non-transferrin bound iron (NTBI) [
19]. This causes not only a less efficient uptake of iron by reticuloendothelial macrophages [
18], but the NTBI may also lead to oxidative stress [
20]. At this point it should be noted that there is no long-term data available on the pharmacokinetics of both compounds; therefore a direct comparison between the two compounds cannot be performed. A head to head pharmacokinetic study would be required before conclusions can be drawn.
These results are in accordance with the results of the REPAIR-IDA trial, a randomized controlled trial comparing FCM to IS in a group of 2584 non-renal IDA patients. Onken et al. describe a significant increase in serum ferritin and TSAT in favor of FCM, 56 days after drug administration [
21]. We have demonstrated that over a period of 9 months, a significant difference in serum ferritin and TSAT continues to exist due to administration of FCM.
One of the adverse effects of FCM described is hypophosphatemia. We observed a nonsignificant 0.03 mmol/L decrease in serum phosphate levels; hence, hypophosphatemia did not become more prevalent as a result of the switch from IS to FCM. This is in accordance with a clinical trial in inflammatory bowel disease patients, which described a transient decrease of serum phosphate levels in the FCM group which resolved between week 4 and week 12 after the switch [
22]. Our results contradict the findings of a study by Hardy et al. specifically assessing the effect of FCM on serum phosphate levels compared to IS in iron deficiency anemia patients. The authors found that FCM caused significantly more hypophosphatemia than IS, which resolved after a mean duration of 6 months [
23]. It should be noted, however, that the latter study did not comprise solitarily chronic kidney disease patients. Hence, the diminished renal function and as such the lower risk of developing hypophosphatemia in our study might have resulted in a substantive difference between our identified prevalence and the results as described by Hardy and colleagues.
It is known, at least theoretically, that administration of intravenous iron in CKD patients on the long term might lead to an iron overload, which may produce endothelial dysfunction, cardiovascular disease, and immune dysfunction [
24]. It seems plausible, based on our results, that we can correct iron status parameters more efficiently with FCM, at a lower dose, and as such a putative iron overload can be prevented. Furthermore, we can also speculate that FCM administration could be more cost-efficient than IS due to the lower iron and ESA dose, which may lead to financial savings in the long term [
25]. A well performed cost-benefit analysis is needed to substantiate this possible advantage of using FCM as compared to IS.
A strength of our investigation is that it comprises a comparison of multiple data points within one patient, meaning that our results correspond better to a real-life situation where a patient is switched from IS to FCM. Moreover, it should be kept in mind that we censored patients who stopped using FCM as data was no longer available, likely contributing to the underestimation of the effect of FCM as these patients will likely have iron status parameters longer in target.
A limitation of our study is the longitudinal study design without randomization, making it difficult to draw firm conclusions. A head-to-head comparison between IS and FCM is needed to confirm our results. Furthermore, we acknowledge as limitation that we did not assess the tolerability or safety of either IV iron preparation even though these factors play an important role when prescribing. Although the safety of FCM has never been compared to IS in the HD population specifically, it has been assessed in other patient groups such as non-dialysis dependent CKD in the REPAIR-IDA trial [
21]. In this trial, Onken et al. found no significant difference in the number of patients that reached a primary composite safety endpoint. Hypertensive events immediately following drug administration occurred significantly more in the FCM group than in the IS group, with 7.45% of patients compared to 4.36% experiencing an event, however, hypertensive events occurred on non-dosing days nearly twice as often in the IS group.