| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
a author for correspondence: fax 303-315-8477, e-mail Paul.Seligman{at}UCHSC.edu
Serum iron concentrations >42.3 µmol/L (>250 µg%) seen in childhood iron poisoning may be high enough to exceed the total number of transferrin binding sites in serum and are associated with marked toxicity (1)(2). Concerns over iron toxicity related to this "oversaturation" of transferrin have recently been applied to circumstances in which iron is given intravenously as either iron dextran (used in the US) or iron gluconate to patients on hemodialysis receiving erythropoietin therapy(3)(4). A recent study describing oversaturation of transferrin, as calculated by the measurement of high serum iron concentrations shortly after intravenous iron gluconate treatment, suggested that with this product, transferrin saturation is commonly >100% (serum iron/total iron binding capacity for transferrin binding sites x 100 = % saturation)(3). When transferrin is fully saturated, serum may contain "free" iron or iron loosely bound to other plasma proteins, a situation that can cause acute toxicity, including tissue damage, cardiac arrhythmias, and/or hypotension (1)(2). It has been suggested that this oversaturation soon after intravenous iron infusion in hemodialysis patients may account for these types of complications (3). However, parenteral iron is a necessary treatment in this patient population, and similar complications, particularly hypotension, are common during hemodialysis sessions even when iron is not given (4).
Because iron bound to iron dextran or iron gluconate does not cause acute iron toxicity (5), we hypothesized that serum assays for iron measure the iron present in these complexes after intravenous infusion, which produces misleadingly high results for the oversaturation of transferrin.
Pharmacokinetic studies have indicated that after infusion, serum concentrations of dextran-bound iron are distributed almost entirely in the plasma, with removal limited to the reticulo-endothelial system at a constant rate (3)(5)(6)(7). Previous studies in iron-deficient patients have suggested that dextran has a variable, but in some patients a relatively long, half-life of >20 h in the circulation (5)(6)(7). In a recent study, a mean half-life of 58.9 h was measured in patients with iron deficiency and chronic disease, including renal failure patients on hemodialysis receiving Dexferrum (8). Various modifications of methodology have given conflicting results as to whether iron as iron dextran is measured as "serum iron" (5)(7). When one methodology was used (5), <3% of the iron present as iron dextran was measured (5), and it was suggested this small percentage was bioavailable iron already dissociated from the iron dextran complex (5). However, another study, in which a different methodology was used, indicated that serum iron measurements up to 2 weeks after iron dextran injection could overestimate the amount of serum iron bound to transferrin (bioavailable iron); this overestimation was attributed to dissociation of the iron from iron dextran in the assay used (6)(7).
The potential for incorrect diagnosis of oversaturation of transferrin may increase with the introduction of an iron gluconate preparation that recently was approved for use in the US (9). To address this oversaturation question and to assess routine clinical monitoring guidelines (4), we measured serum iron in vitro in the presence of added iron gluconate as well as iron dextran.
Two methods for serum iron determination were used. One method uses an acetate buffer (pH 4.5) with 15 g/L hydroxylamine hydrochloride (Sigma Diagnostics) to release the iron from transferrin (10). The initial absorbance is read at 560 nm on a Hewlett Packard 8452A diode array spectrophotometer. The released iron reacts with 8.5 g/L Ferrozine reagent during a 15-min incubation at 37 °C to produce a magenta-colored complex, which is read a second time. Because the assay is linear, a single calibrator with an iron concentration of 84.7 µmol/L (500 µg/dL) is used. The second method, often used in European countries (11) and now used in the US in certain automated systems, uses an acetate buffer (0.1 mol/L, pH 4.8) containing ascorbic acid (56.8 mmol/L) and guanidine (6 mol/L) to release transferrin-bound iron. The absorbance is measured at 595 nm(3)(11) before and after a 5-min incubation at 37 °C with 36 mmol/L Ferrene S. Again the assay is linear, and a single calibrator is used [30–84.7 µmol/L (177–500 µg/dL)]. Neither of these methods measures the iron present in heme. We used the form of iron gluconate currently used in Europe and approved for use in the US, sodium ferric gluconate complex in sucrose (Ferrlecit), from R & D Laboratories, and iron dextran (InFed) from Schein Pharmaceuticals.
The results obtained for each of the two methods after various
incubation times are shown in Fig. 1
. Serum iron was measured in samples obtained after blood was
drawn into red-stoppered tubes, allowed to clot, and centrifuged to
separate the serum. Iron as iron gluconate or iron dextran was added at
a concentration of 204 µmol/L (1206 µg/dL). This is the amount of
added iron, based on what is known about iron dextran distribution and
kinetics (see above) (3)(5)(6)(7), that would be
expected to be present in a patient's serum after approximately one
half-life of disappearance of the iron preparations if 100 mg was given
intravenously; this same amount would be present at least several
half-lives (i.e., 1 week or more) after "total dose" iron dextran
infusion (1000 mg) (12). The total iron present was verified
by the addition of sodium dithionite, an agent known to release iron
from gluconate or dextran complexes (5). The method that
uses the acetate/hydroxylamine hydrochloride buffer system measured
2.2% (± 1.2%) of the iron in iron dextran after the recommended
15-min incubation, whereas the ascorbic acid/guanidine method measured
7.3% of the iron after the recommended 5-min incubation period. Both
assays measured significantly more iron when iron gluconate was added:
7.0% (± 0.4%) in the acetate/hydroxylamine hydrochloride method vs
33% (± 11.0%) in the ascorbic acid/guanidine method. When both
assays and five different concentrations of iron gluconate from 51 to
408 µmol/L (301 to 2412 µg/dL) were used, the percentage of iron
measured from the added iron gluconate remained similar (± 3%) over
the entire range of concentrations. In other experiments, both iron
compounds were added to the serum within 1 h prior to the assay
(as shown in Fig. 1
) or 6 and 24 h prior to the assay and kept
at room temperature with no significant change in the amount of iron
released (difference <10%).
|
As shown in Fig. 1
, the change in iron released with increasing
incubation times changed significantly when the ascorbic acid/guanidine
buffer was used. Many clinical laboratories in the US now use automated
systems with a similar ascorbic acid/guanidine buffer (Boehringer
Mannheim) for serum iron determinations. This methodology might, based
on minor differences in incubation time or other details of the assay,
give results different from those shown in Fig. 1
. Three different
serum samples from three healthy subjects with different
starting iron concentrations and the same samples containing 204
µmol/L iron as iron gluconate or iron dextran were sent to three
clinical laboratories, one (Veterans Administration Hospital, Denver,
CO) that uses a hydroxylamine method (Beckman) and two (University
Hospital, Denver, CO, and Presbyterian-St. Luke's Medical
Center, Denver, CO) that use variations of an ascorbic acid method with
different automated equipment (Hitachi 917 and Johnson & Johnson Vitros
system, respectively). The results obtained with the hydroxylamine
method (Table 1
) for both iron dextran and iron gluconate were similar to our
manual results shown in Fig. 1
. The results obtained from the two
variations of the ascorbic acid method are somewhat different from each
other and the results in Fig. 1
, but still show significantly increased
release of iron from both complexes compared with the hydroxylamine
method. When pooled serum samples (from three other subjects)
containing two different concentrations of iron dextran or iron
gluconate (204 and 33.9 µmol/L) were sent in duplicate to each
laboratory, almost identical results were obtained for from each
laboratory, as shown in Table 1
.
|
Limited studies were performed using both iron gluconate and iron dextran diluted in pH 7.4 buffer, with the assay performed in the same neutral buffer (UIBC; Sigma Chemical Co.). In this assay, the iron present in iron dextran was either not measurable or barely measurable (<1%); the iron measurable in added iron gluconate was significantly higher (mean, 2.8%) than in duplicate samples assayed with pH 4.1 buffer (mean, 9.1%; P <0.05, paired t-test). Iron measured in pH 7.4 buffer clearly represents free iron present in the compound, although this may underestimate other loosely bound iron that may become bioavailable.
Taken together, these data suggest that both acidic buffers not only release iron from the transferrin-bound iron complex and measure free iron (see above), but also release some iron still bound to the gluconate and dextran complexes. The amount of iron released by the hydroxylamine method is relatively low and consistent for various assay conditions. The higher and more variable amounts of iron released in the ascorbic acid/guanidine method could lead to gross overestimations of serum iron if either of these iron complexes is present in the serum. Thus, serum obtained after infusion of 100 mg of iron as iron dextran or iron gluconate could, for example, add an additional 169 µmol/L (1206 µg%) of iron (the equivalent of one half-life), and iron measurements could falsely indicate oversaturation of transferrin (i.e., if 10–25% of this iron were released, an additional 16.9–42.4 µmol/L (16.9–302.5 µg%) would be measured in the serum iron assay).
Because of concerns about iron deficiency as well as iron toxicity, measurement of serum iron and calculation of the percentage of saturation are performed frequently on all hemodialysis patients receiving intravenous iron preparations (4). Our results indicate that the measurement of serum iron within 24 h after infusion, even if a patient shows signs indicative of iron toxicity, would yield misleading results for transferrin saturation with any of the present methodologies. In addition, measurement of serum iron to assess the response to intravenous iron as reflected by bioavailable iron (increased percentage of saturation) may still reflect the in vitro release of iron from these intravenous iron preparations, which may lead to a change in therapy based on guidelines that recommend using serum iron determinations to monitor iron dosing. For future pharmacokinetic studies of iron gluconate, extensive in vitro studies that include measurement of total iron and free iron (5) are necessary to determine any overestimation of "bioavailable iron" that may be related to the methodology used.
In summary, methodologies that measure serum iron, particularly those using an ascorbic acid/guanidine buffer method, will cause in vitro dissociation of iron bound to gluconate or dextran complexes. To ensure that only bioavailable iron is measured in patients treated with intravenous iron, as suggested by one guideline (4), it is best to wait at least 2 weeks after the last dose of intravenous iron. In addition, careful attention must be paid to any changes in the methodologies used by the laboratories that receive samples from dialysis centers.
Acknowledgments
This study was supported in part by grants from the Dautch Foundation and from Schein Pharmaceuticals. We thank Valarie Allen and Meg Langfur for technical assistance.
Footnotes
School of Medicine, Department of Medicine, Division of Hematology, University of Colorado Health Sciences Center, 4200 East Ninth Ave., Campus Box B-170, Denver, CO 80220
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  B. A. Warady, P. A. Seligman, and N. V. Dahl Single-Dosage Pharmacokinetics of Sodium Ferric Gluconate Complex in Iron-Deficient Pediatric Hemodialysis Patients Clin. J. Am. Soc. Nephrol., November 1, 2007; 2(6): 1140 - 1146. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. F. de Vecchi, C. Novembrino, S. Lonati, S. Ippolito, and F. Bamonti Two different modalities of iron gluconate i.v. administration: effects on iron, oxidative and inflammatory status in peritoneal dialysis patients Nephrol. Dial. Transplant., June 1, 2007; 22(6): 1709 - 1713. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. B. Van Wyck Labile Iron: Manifestations and Clinical Implications J. Am. Soc. Nephrol., December 1, 2004; 15(suppl_2): S107 - S111. [Full Text] [PDF]
|
|
|
|
|
|
B. Scheiber-Mojdehkar, B. Lutzky, R. Schaufler, B. Sturm, and H. Goldenberg Non-Transferrin-Bound Iron in the Serum of Hemodialysis Patients Who Receive Ferric Saccharate: No Correlation to Peroxide Generation J. Am. Soc. Nephrol., June 1, 2004; 15(6): 1648 - 1655. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Van Wyck, J. Anderson, and K. Johnson Labile iron in parenteral iron formulations: a quantitative and comparative study Nephrol. Dial. Transplant., March 1, 2004; 19(3): 561 - 565. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Beutler, A. V. Hoffbrand, and J. D. Cook Iron Deficiency and Overload Hematology, January 1, 2003; 2003(1): 40 - 61. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. K. Salahudeen and R. M. Schaefer High dose of bolus iron vs low dose of weekly infusion: bolusing high dose, a recipe for iron toxicity? Nephrol. Dial. Transplant., May 1, 2002; 17(5): 939 - 940. [Full Text]
|
|
|
|
 |
|
 L. T. Goodnough, B. Skikne, and C. Brugnara Erythropoietin, iron, and erythropoiesis Blood, August 1, 2000; 96(3): 823 - 833. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
