- Email: [email protected]
Sodium ferric gluconate complex in sucrose is safe and effective in hemodialysis patients: North American clinical trial
Sodium Ferric Gluconate Complex in Sucrose Is Safe and Effective in Hemodialysis Patients: North American Clinical Trial Allen R. Nissenson, MD, Robert M. Lindsay, MD, Suzanne Swan, MD, Paul Seligman, MD, and Jur Strobos, MD ● A new intravenous (IV) iron compound, sodium ferric gluconate complex in sucrose (Ferrlecit, R&D Laboratories, Inc, Marina Del Rey, CA), was administered over 8 consecutive dialysis days in equally divided doses to a total of either 0.5 or 1.0 g in a controlled, open, multicenter, randomized clinical study of anemic, iron-deficient hemodialysis patients receiving recombinant human erythropoietin (rHuEPO). Effectiveness was assessed by increase in hemoglobin and hematocrit and changes of iron parameters. Results were compared with historically matched controls on oral iron. High-dose IV treatment with 1.0 g sodium ferric gluconate complex in sucrose resulted in significantly greater improvement in hemoglobin, hematocrit, iron saturation, and serum ferritin at all time points, as compared with low-dose IV (0.5 g) or oral iron treatment. Despite an initial improvement in mean serum ferritin and transferrin saturation, 500 mg IV therapy did not result in a significant improvement in hemoglobin at any time. Eighty-three of 88 patients completed treatment with sodium ferric gluconate complex in sucrose: 44 in the high-dose and 39 in the low-dose group. Two patients discontinued for personal reasons. The other three discontinued because of a rash, nausea and rash, and chest pain with pruritus, respectively. In comparison with 25 matched control patients, adverse events could not be linked to drug therapy, nor was there a dose effect. In conclusion, sodium ferric gluconate complex in sucrose is safe and effective in the management of iron-deficiency anemia in severely iron-deficient and anemic hemodialysis patients receiving rHuEPO. This study confirms the concepts regarding iron therapy expressed in the National Kidney Foundation Dialysis Outcomes Quality Initiative (NKF-DOQI) that hemodialysis patients with serum ferritin below 100 ng/mL or transferrin saturations below 18% need supplementation with parenteral iron in excess of 1.0 g to achieve optimal response in hemoglobin and hematocrit levels. 娀 1999 by the National Kidney Foundation, Inc. INDEX WORDS: Iron; iron-deficiency anemia; functional iron-deficiency anemia; hemodialysis; end-stage renal disease; Ferrlecit Injection; sodium ferric gluconate complex in sucrose.
Editorial, p. 595
A
NEMIA is present in most patients with chronic renal failure (CRF) on hemodialysis1,2 and is due primarily to a deficiency of endogenous erythropoietin.3,4 Although most hemodialysis patients are now receiving recombinant human erythropoietin (rHuEPO), anemia persists, with iron deficiency the most frequent cause of initial poor response to rHuEPO, or acquired resistance to rHuEPO.5-7 Recent clinical practice guidelines developed by the National Kidney Foundation Dialysis Outcomes Quality Initiative (NKF-DOQI) document that most patients receiving hemodialysis require parenteral iron to maintain iron status.8 The basis for this recommendation is the understanding that hemodialysis patients receiving rHuEPO are highly likely to develop iron deficiency, because of regular blood (and iron) loss incident to the dialysis procedure, and the increased demands for iron when rHuEPO is being administered and is driving erythropoiesis. This high iron requirement exceeds the amount of iron that can generally be provided orally.9-13
Both the clinical management of dialysis patients with iron deficiency and clinical research into iron metabolism, monitoring, and administration have been hampered by the current availability in the United States of only a single form of parenteral iron, iron dextran. Although generally safe, severe allergic reactions can occur, and deaths after allergic reactions have been reported.14-16 From the University of California at Los Angeles Medical Center, Los Angeles, CA; the Renal Unit, Victoria Hospital, London, Ontario, Canada; Hennepin County Medical Center, Department of Medicine, University of Minnesota, Minneapolis, MN; Division of Hematology, University of Colorado Health Sciences Center, Denver, CO; and private practice, Washington, DC. Received August 13, 1998; accepted in revised form September 25, 1998. Supported by R&D Laboratories, Inc. Clinical supplies in the form of Ferrlecit Injection (sodium ferric gluconate complex in sucrose) were provided by the manufacturer, Rhoˆne-Poulenc Rorer GmBH, Dagenham, United Kingdom. Address reprint requests to Allen R. Nissenson, MD, UCLA Medical Center, 200 Medical Plaza, Suite 565; Los Angeles, CA 90095. E-mail: [email protected]
娀 1999 by the National Kidney Foundation, Inc. 0272-6386/99/3303-0007$3.00/0
American Journal of Kidney Diseases, Vol 33, No 3 (March), 1999: pp 471-482
471
472
NISSENSON ET AL
Sodium ferric gluconate complex in sucrose injection is a high-molecular-weight complex (350,000 daltons) composed of ferric saccharate centers chelated by gluconate, which has a high affinity for ferric ions and the capacity to form a powerful bridge between adjacent ferric oxide centers.* The absence of the need to degrade a dextran shell or solubilize a ferric oxide core also should provide a more rapidly bioavailable iron after administration. Sodium ferric gluconate complex in sucrose injection has been used since 1959 in more than 20 countries outside the United States for the intravenous administration of iron to patients with iron-deficiency anemia. Currently, the principal intravenous usage is among the 48,000 hemodialysis patients in Germany (R. Schaefer, personal communication, 1997). Several groups have reported excellent results with sodium ferric gluconate complex in sucrose when administered to iron-deficient hemodialysis patients,17-21 but these reports are in small numbers of patients, from single centers, and are largely descriptive. The current study was designed to assess the safety and efficacy of two different doses of parenteral sodium ferric gluconate complex in sucrose in iron-deficient hemodialysis patients. The randomized, controlled study design also permitted the assessment of the minimum iron needs in this patient population, as well as a confirmation of the concept of functional iron deficiency. MATERIALS AND METHODS Sodium Ferric Gluconate Complex in Sucrose The study drug was provided in the form of Ferrlecit by R&D Laboratories, Inc. (Marina Del Rey, CA) through the manufacturer, Rhoˆne-Poulenc Rorer GmBH (Dagenham, United Kingdom). Each ampoule contained 5 mL Ferrlecit for intravenous injection containing 62.5 mg of elemental iron as the sodium salt of a ferric ion carbohydrate complex in an alkaline aqueous solution with approximately 20% sucrose.
Purpose This was an open-label, randomized study of the treatment of iron-deficiency anemia in adult chronic hemodialysis patients receiving rHuEPO that compared the safety and efficacy of iron repletion using two different dosing regimens of intravenous sodium ferric gluconate complex in *Proprietary Data supplied by R&D Laboratories, Inc, Marina Del Rey, CA.
sucrose, totaling either 1.0 or 0.5 g administered over a course of eight sequential dialysis sessions.
Entry Criteria and Patient Population Adult patients receiving chronic hemodialysis with either serum ferritin levels less than 100 ng/mL or iron saturations less than 18% AND either a hemoglobin level less than 10 g/dL or hematocrit of 32% or less were eligible to participate in the study. Patients were excluded if they had received parenteral iron products or any investigational drug that might interfere with iron metabolism in the 2 months preceding enrollment. Other exclusion criteria were significant comorbid conditions and baseline rHuEPO requirements greater than 10,000 units three times weekly. Patients with a history of allergic conditions or a specific non–life-threatening allergic reaction to iron dextran were not excluded. No oral iron or blood transfusions were administered during the study. The study design required that starting rHuEPO doses remain unchanged throughout the study.
Drug Administration Patients were randomly assigned to a course of either low-dose or high-dose sodium ferric gluconate complex in sucrose before administration of a test dose of 25 mg. Drug was administered intravenously as 62.5 mg in 50 mL saline or 125 mg in 100 mL saline over 30 or 60 minutes, respectively, in eight divided doses over eight sequential dialysis sessions. The protocol precluded alternative oral or parenteral iron administration, blood transfusion, or alteration in rHuEPO dose during the course of the 47-day study.
Efficacy Analysis Demographic and medical history variables as well as baseline efficacy values were compared in the two groups. Mean changes within group for hemoglobin and hematocrit from baseline to 2, 14, and 30 days after completion of the full course of drug administration were compared by paired t-test. Mean hemoglobin and hematocrit changes between high-dose and low-dose groups at the same times were compared by an analysis of covariance (ANOVA) using a mixed model for analysis of repeated measures. Transferrin saturation (Tsat) and serum ferritin were analyzed and compared in the two groups at the same points. A matched control group of 25 patients treated only with oral iron (ferrous sulfate 325 mg or ferrous gluconate 650 mg thrice daily) and meeting the same entry criteria as the patients receiving parenteral iron was studied as well. Efficacy data for the control group were available for days 30 and 60 after initiation of oral iron therapy. Analyses for all groups used an intent-to-treat approach.
Safety Analysis Potential adverse events in the patients receiving sodium ferric gluconate complex in sucrose were identified through history, physical examination, and active inquiry of patients by using a form containing Coding System for a Thesaurus of Adverse Reaction Terms (COSTART), which was completed at each drug administration and at 2 days after cessation of drug therapy. Adverse events were rated by the investigator for possible relationship to study drug (none, unlikely, possible, probable, or definite). The incidence of adverse events was compared between dose groups and
SODIUM FERRIC GLUCONATE IN HEMODIALYSIS
against a matched control group by using Fisher’s exact test. Routine blood samples for laboratory assessment of adverse organ-system effects were collected at baseline and on days 2, 14, and 30 after cessation of therapy. Laboratory measurements included alkaline phosphatase, alanine transaminase, aspartate transaminase, bilirubin, blood urea nitrogen, creatinine, glucose, platelets, and white blood cell counts and differentials. Laboratory safety data were assessed by using shift tables, and Fisher’s exact test was used to compare dose groups.
RESULTS
Eighty-three of 88 patients completed the study, including 44 in the high-dose and 39 in the low-dose group. Five patients were withdrawn from the study after entry but before day 2 when first efficacy data were obtained. Two patients discontinued for logistical reasons; one patient was withdrawn after development of pruritus and chest pain immediately after the test dose administration; one patient was withdrawn after development of nausea, abdominal and flank pain, fatigue, and rash after the first full dose; and one patient was withdrawn on development of a red blotchy rash after the first full dose. Baseline demographic and vital signs as well as statistical analysis of comparability are shown in Table 1. Randomization resulted in comparable groups. As shown in Table 2, baseline values for hemoglobin, hematocrit, serum ferritin, serum iron, mean corpuscular hemoglobin (MCH), mean corpuscular volume (MCV), and mean corpuscular hemoglobin content (MCHC) were similar between the low-dose and highdose treatment groups. However, percent iron saturation was significantly (P ⫽ 0.045) lower in the high-dose group compared with the low-dose group. The oral iron group was comparable to the two treatment groups except for baseline serum ferritin, which was significantly elevated in the oral iron group. Anemia—Mean Change in Hemoglobin and Hematocrit
High-Dose Group (1 gram) The mean increase in hemoglobin in the highdose group from baseline (9.6 g/dL) to days 2, 14, and 30 was 1.0 g/dL, 1.1 g/dL, and 1.3 g/dL, respectively. Each of these values was statistically significant by paired t-test (P ⬍ 0.0001, P ⬍ 0.0001, and P ⬍ 0.0001, respectively). The mean increase in hematocrit in the high-dose
473
group from baseline (28.8%) to days 2, 14, and 30 was 3.1%, 3.6%, and 3.5%, respectively. Each of these values was statistically significant by paired test (P ⬍ 0.0001, P ⬍ 0.0001, and P ⬍ 0.0001, respectively). Low-Dose Group (500 mg) The mean increase in hemoglobin in the lowdose group from baseline (9.8 g/dL) to days 2, 14, and 30 was 0.3 g/dL, 0.3 g/dL, and 0.5 g/dL, respectively. Only the value for day 30 was statistically significant by paired t-test (P ⫽ 0.071, P ⫽ 0.072, and P ⫽ 0.01, respectively). The mean increase in hematocrit in the low-dose group from baseline (29.4%) to days 2, 14, and 30 was 1.1%, 1.4%, and 1.4%, respectively. Only the values for days 14 and 31 were statistically significant by paired test (P ⫽ 0.06, P ⫽ 0.018, and P ⫽ 0.024, respectively). Oral Iron (Control) Group The mean increase in hemoglobin in the oral iron group from baseline (9.4 g/dL) to days 14 and 30 was 0.4 g/dL and 0.4 g/dL, respectively. Each of these values was statistically significant by paired t-test (P ⫽ 0.016). The mean increase in hematocrit in the oral dose group from baseline (28.6%) to days 14 and 30 was 0.8% and 0.4%, respectively. None of these values were statistically significant by paired t-test (P ⫽ 0.112 and P ⫽ 0.112, respectively). The comparative increase in mean hemoglobin in the high- versus the low-dose group was statistically significant (P ⬍ 0.001) at all times using an ANOVA that included the effects of center as well as treatment. The comparative increase in mean hematocrit was statistically significant (P ⬍ 0.002) at all times using the ANOVA. The comparative increase in mean hemoglobin in the high-dose versus the oral-dose group was statistically significant (P ⫽ 0.001) at all times by ANOVA. The comparative increase in mean hematocrit in the high-dose versus the oral-dose group was statistically significant (P ⬍ 0.001) at all times by ANOVA. The comparative increase in mean hemoglobin and hematocrit in the low-dose group versus the oral-dose group was not statistically different at any time by ANOVA (P ⫽ 0.337 and 0.065, respectively). The mean hemoglobin levels at baseline and
474
NISSENSON ET AL Table 1. Baseline Patient Demographics and Vital Signs P*
Treatment Group
Variable
Age (yr) n Mean (Std) Minimum Maximum Sex, n (%) Female Male Missing Race, n (%) White Black Latin Asian Hispanic Native-American Persian Unknown Missing Systolic BP (mm Hg) n Mean (Std) Minimum Maximum Diastolic BP (mm Hg) n Mean (Std) Minimum Maximum Pulse (bpm) n Mean (Std) Minimum Maximum Respirations (bpm) n Mean (Std) Minimum Maximum
500 mg (n ⫽ 41)
1,000 mg (n ⫽ 47)
Control (n ⫽ 25)
500 v 1,000 mg
500 mg v Control
1,000 mg v Control
Overall
41 55.0 (17.7) 22.0 80.0
47 57.1 (17.7) 20.0 87.0
25 52.2 (16.6) 25.0 84.0
0.565
0.536
0.259
0.522
23 (56.1) 18 (40.9) 0 (0.0)
25 (50.2) 21 (44.7) 1 (2.1)
17 (68.0) 8 (32.0) 0 (0.0)
1.000
0.408
0.328
0.536
30 (73.2) 8 (19.5) 2 (4.9) 1 (2.4) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0)
36 (76.6) 5 (10.6) 2 (4.3) 1 (2.1) 0 (0.0) 1 (2.1) 1 (2.1) 0 (0.0) 1 (2.1)
10 (40.0) 8 (32.0) 0 (0.0) 1 (4.0) 5 (20.0) 0 (0.0) 0 (0.0) 1 (4.0) 0 (0.0)
0.813
0.003
⬍0.001
0.001
39 144.5 (24.1) 72.0 196.0
43 155.0 (19.7) 120.1 200.0
13 170.0 (19.9) 139.0 199.0
0.030
⬍0.001
0.032
0.001
39 77.2 (13.9) 45.0 100.0
43 82.5 (12.2) 50.0 114.0
13 86.6 (16.2) 61.0 112.0
0.077
0.031
0.337
0.057
40 86.0 (10.2) 62.0 113.0
43 83.4 (12.8) 59.0 112.0
12 78.9 (10.1) 56.0 96.0
0.327
0.072
0.246
0.184
31 18.0 (3.2) 12.0 22.0
33 18.3 (3.0) 12.0 22.0
—† —† —† —†
0.980
N/A
N/A
0.980
*For a continuous variable, an ANOVA model with effects for treatment group was used to compare the group means, and the P value was associated with the P test. For a categorical variable, the P value was associated with the Fisher’s exact test. †No data were available. Abbreviation: N/A, not applicable.
days 2, 14, and 30 for the low- and high-dose groups are shown in Fig 1. The mean hematocrit level at baseline and days 2, 14, and 30 for the low-dose and high-dose groups are presented in Fig 2. Figure 3 provides the mean hemoglobin levels at baseline and days 14 and 30 for low-, high-, and oral-dose groups. Figure 4 provides mean hematocrit levels at baseline and days 14 and 30 for low-, high-, and oral-dose groups.
Iron ‘‘stores’’—serum ferritin and iron saturation. The mean increase in serum ferritin in the high-dose group from baseline (88 ng/dL) was 320 ng/dL, 199 nm/dL, and 134 nm/dL on days 2, 14, and 30, respectively. Each of these values was statistically significant by paired t-test (P ⬍ 0.0001, P ⬍ 0.0001, and P ⬍ 0.0001, respectively). The mean increase in Tsat in the highdose group from baseline (15.6%) to days 2, 14,
SODIUM FERRIC GLUCONATE IN HEMODIALYSIS
475
Table 2. Baseline Hematological Data P*
Treatment Group
Variable
500 mg (n ⫽ 41)
Baseline hemoglobin (g/dL) n 39 Mean (Std) 9.8 (0.8) Minimum 7.1 Maximum 11.1 Baseline hematocrit (%) n 39 Mean (Std) 29.4 (2.3) Minimum 23.0 Maximum 33.0 Baseline percent iron saturation (%) n 39 Mean (Std) 20.1 (11.9) Minimum 4.0 Maximum 58.0 Baseline serum ferritin (mg/mL) n 37 Mean (Std) 105.6 (116.3) Minimum 19.0 Maximum 660.0 Baseline serum iron (ug/dL) n 39 Mean (Std) 49.7 (27.7) Minimum 11.0 Maximum 112.0 Baseline MCH (pg) n 40 Mean (Std) 30.0 (3.2) Minimum 20.0 Maximum 35.0 Baseline MCV (fl) n 40 Mean (Std) 89.5 (7.7) Minimum 67.0 Maximum 103.6 Baseline MCHC (g/dL) n 40 Mean (Std) 33.4 (1.3) Minimum 30.0 Maximum 35.3
1,000 mg (n ⫽ 47)
Control (n ⫽ 25)
500 v 1,000 mg
44 9.6 (0.9) 6.5 10.9
25 9.4 (0.8) 7.8 10.4
0.298
0.038
0.230
0.124
44 28.8 (2.6) 20.0 32.0
25 28.6 (2.1) 24.0 31.4
0.305
0.212
0.705
0.401
43 15.6 (7.5) 2.0 35.0
25 14.2 (4.4) 6.7 27.0
0.026
0.012
0.530
0.020
0.678
⬍0.001
⬍0.001
⬍0.001
0.112
0.007
0.162
0.024
0.691
0.055
0.022
0.060
0.705
0.001
⬍0.001
⬍0.001
0.566
0.036
0.009
0.028
43 12 88.4 (143.2) 605.6 (390.9) 8.0 41.0 941.0 1000 44 41.6 (22.4) 6.0 101.0
25 33.5 (13.6) 16.0 81.0
44 29.7 (2.2) 25.0 34.0
25 31.4 (3.3) 26.6 39.1
43 88.9 (5.3) 73.6 100.1
25 96.0 (9.0) 84.6 116.0
44 33.5 (1.2) 31.0 36.0
25 32.7 (0.6) 31.4 34.2
500 mg v Control
1,000 mg v Control
Overall
*For a continuous variable, an ANOVA model with effects for treatment group was used to compare the group means, and the P value was associated with the P test. For a categorical variable, the P value was associated with the Fisher’s exact test.
and 30 was 9.0%, 8.5%, and 5.5%, respectively. Each of these values was statistically significant by paired test (P ⬍ 0.0001, P ⬍ 0.0001, and P ⫽ 0.002, respectively). The mean increases in serum ferritin in the low-dose group from baseline (106 ng/dL) were 130 ng/dL, 132 ng/dL, and 65 ng/dL on days 2, 14, and 30, respectively. The values for days 2, 14, and 30 were statistically significant by paired t-test (P ⬍ 0.0001, P ⫽ 0.003, and P ⫽ 0.01, respectively). The mean
change in iron saturation in the low-dose group from baseline (20.1%) was 1.2%, 2.8%, and ⫺0.6% on days 2, 14, and 30, respectively. None of the values were statistically significantly different from baseline by paired test (P ⫽ 0.47, P ⫽ 0.156, and P ⫽ 0.72, respectively). Iron saturation and serum ferritin levels fell rapidly after cessation of iron therapy. There was not a statistically significant difference in the increase in mean ferritin in the high-
476
NISSENSON ET AL
Fig 1. Mean values for hemoglobin in renal dialysis patients receiving low- (䊏) or high-dose (䊐) sodium ferric gluconate complex in sucrose.
versus the low-dose group at any time. Conversely, there was a significant increase in mean iron saturation in the high- versus the low-dose group (P ⬍ 0.017) at all points under ANOVA. The mean serum ferritin levels from baseline to days 2, 14, and 30 for the low-dose and high-dose groups are presented in Fig 5. The mean Tsat values from baseline to days 2, 14, and 30 for the low-dose and high-dose groups are presented in Fig 6. Safety analysis. A total of 88 patients enrolled in the study and received a test dose of 25 mg sodium ferric gluconate complex in sucrose before the beginning of the study. All 88 patients
Fig 2. Mean values for hematocrit in renal dialysis patients receiving low- (䊏) or high-dose (䊐) sodium ferric gluconate complex in sucrose.
were included in the safety analysis. No adverse event was considered by the investigators to be definitely related to sodium ferric gluconate complex in sucrose treatment. No patient experienced a type I immediate hypersensitivity reaction. There were no hospitalizations or deaths related to drug administration. Most adverse events were considered to be of mild or moderate severity and not definitely related to sodium ferric gluconate complex in sucrose administration. Adverse events that were possibly or probably related to treatment by investigator report were experienced by nine patients and included nausea in four patients,
SODIUM FERRIC GLUCONATE IN HEMODIALYSIS
477
Fig 3. Mean values for hemoglobin in renal dialysis patients receiving low- (䊏) or high-dose (䊐) sodium ferric gluconate complex in sucrose or oral-dose iron (°).
vomiting in three, rash in two, and abdominal pain, back pain, fatigue, syncope, cramps, agitation, menorrhagia, pruritus, chest pain, paresthesia, and abnormal erythrocytes in one patient each. An analysis of the incidence of adverse events in the two treatment groups showed that the number of patients experiencing each type of event was similar between the groups. Adverse events were further analyzed by comparison with published data on rates of adverse events in patients on routine hemodialysis.22 The adverse events reported in the patients receiving sodium ferric gluconate complex in sucrose occurred at no greater frequency statistically than these same
events in patients undergoing routine dialysis but not receiving parenteral iron as assessed by Fisher’s exact test. No clinically significant changes in laboratory values or vital signs were observed during the study. DISCUSSION Safety and Efficacy of Sodium Ferric Gluconate Complex in Sucrose
This study shows that sodium ferric gluconate complex in sucrose is safe and effective in correcting anemia and improving iron status in adult patients on chronic hemodialysis and rHuEPO
Fig 4. Mean values for hematocrit in renal dialysis patients receiving low- (䊏) or high-dose (䊐) sodium ferric gluconate complex in sucrose or oral-dose iron (°).
478
NISSENSON ET AL
Fig 5. Mean values for serum ferritin in renal dialysis patients receiving low- (䊏) or high-dose (䊐) sodium ferric gluconate complex in sucrose.
therapy. Although hospitalizations are common in this patient population (17 study patients were hospitalized at some time during the study), there were no serious adverse events, hospitalizations, deaths, or laboratory abnormalities related to drug administration. No patient experienced a type I immediate hypersensitivity reaction. We previously reported on nine hemodialysis patients who had experienced serious reactions to iron dextran treatment and who were successfully treated with sodium ferric gluconate complex in sucrose.23 These data are consistent with
Fig 6. Mean values for transferrin saturation in renal dialysis patients receiving low- (䊏) or high-dose (䊐) sodium ferric gluconate complex in sucrose.
a recent review of worldwide adverse events reporting from 1976 to 1996 on usage of sodium ferric gluconate complex in sucrose in Germany, where no known hypersensitivity deaths have occurred.24 The current study, however, did not include patients randomized to iron dextran, limiting any final conclusions about the relative safety of these two compounds. Iron-Deficiency Anemia in Hemodialysis Patients
Iron-deficiency anemia results from the absence of sufficient available body iron stores to
SODIUM FERRIC GLUCONATE IN HEMODIALYSIS
sustain hematopoiesis. Sixty-five percent of body iron is present in circulating red blood cells and myoglobin, with the rest in the erythroid marrow, or other longer-term storage sites, such as the liver and macrophages within the reticuloendothelial system (RES).25,26 Iron-deficiency anemia most commonly results from blood loss.25,26 Correction of iron deficiency with oral iron is generally simple and effective unless patients have chronic continuing or uncontrolled blood loss, such as occurs with regular hemodialysis treatment.25-27 rHuEPO therapy, which produces an increase in serum hemoglobin levels, results in further increases in iron loss during hemodialysis even though the total blood loss may remain constant.7,28,29 Iron-deficiency anemia, common in hemodialysis patients because of ongoing blood loss, is thus exacerbated by rHuEPO therapy.28-31 There are other factors that also may contribute to iron deficiency in the hemodialysis patient, including a diet that is low in bioavailable iron, less than optimal iron absorption, and ongoing gastrointestinal tract losses of iron.32,33 A number of studies have shown that the theoretical inability of hemodialysis patients receiving rHuEPO to keep up with continuing iron demands by oral iron administration is actually seen clinically.9-13 Only a few studies support the effectiveness of oral iron usage in this population, and then only when oral iron is administered under a carefully controlled regimen. Effectiveness can only be shown for a minority of patients.34-36 As a result, most hemodialysis patients require frequent courses of parenteral iron. Unfortunately, the most commonly used clinical tests to define iron deficiency in dialysis patients have major shortcomings. In non-CRF patients, for example, serum ferritin levels have been associated with the amount of iron bound to intracellular ferritin within the RES and are known to vary with intracellular or hepatic iron stores. Uncontrolled clinical studies have shown that in anemic renal failure patients, however, a hematopoietic response to iron administration can be obtained despite serum ferritin levels that are significantly higher than 10 ng/mL, thought to be diagnostic of iron deficiency in nonrenal patients, and even 100 ng/mL, usually cited as diagnostic of iron deficiency in dialysis patients.6,7 A hematopoietic response to exogenous iron in this setting indicates the presence of ‘‘functional’’ iron deficiency.9
479
It has been suggested that functional iron deficiency may reflect a state in which the serum ferritin value is consistent with the presence of sufficient iron bound to intracellular ferritin to meet hematopoietic needs, but because of an inability of the RES to release iron (RES blockade), this stored iron is unavailable. The administration of supplemental intravenous iron in this situation would overcome the RES blockade either by directly providing bioavailable iron or by increasing iron stores to nonphysiological levels to overcome the blockade.9 Alternatively, patients with functional iron deficiency may be frankly iron deficient, but serum ferritin levels simply may not accurately reflect iron stores. For some time, serum ferritin was thought to represent leakage of intracellular RES ferritin. Three lines of evidence now refute this thesis. First, serum ferritin is largely present as apoferritin, without any bound iron, even when intracellular ferritin is sufficiently iron overloaded to result in deposition of intracellular hemosiderin. Second, the molecular structure of serum ferritin is different from that of intracellular ferritin.37 Third, the control over the synthesis of serum ferritin and intracellular ferritin is different. The synthesis of serum ferritin is stimulated by interleukin-1 and interleukin-6, which are elevated in CRF patients.38 Thus, although there is generally a positive correlation between intrahepatic iron stores and serum ferritin, the relationship is not consistent and the correlation coefficient is low with a broad confidence interval.26,27 Many chronic diseases, including CRF, are associated with high serum ferritin levels despite low tissue iron levels. Tsat is a measure of serum iron bound to circulating transferrin. Low saturations are generally indicative of either (1) an increase in total transferrin concentration and synthesis that is increased under autoregulation by iron deficiency; or (2) a low concentration of total circulating iron occupying available transferrin-binding sites. In the non-CRF patient, Tsat below 10% to 15% is considered indicative of iron deficiency. Tsat is also not an ideal indicator of iron status in the dialysis patient. Tsat varies diurnally and only assesses the small quantity of iron currently available for erythropoiesis in the circulation. Exogenous rHuEPO therapy may depress Tsat levels directly by increasing
480
erythroid marrow uptake or transferrin-binding receptor sites. rHuEPO also increases red blood cell turnover and causes the release of iron from iron storage sites.26,27 Thus, Tsat has become an imperfect measure of iron deficiency in the CRF patient. In patients on dialysis, the hematopoietic responses to iron therapy in patients with higher serum ferritin or Tsat levels has led to the supposition that these patients have functional iron deficiency. In this study, iron-deficient hemodialysis patients randomized to treatment with 500 mg sodium ferric gluconate complex in sucrose did show normalization of average patient laboratory parameters of iron deficiency. However, these patients had no improvement in hemoglobin and only minimal improvement in hematocrit. By contrast, equivalent randomized patients receiving an additional 500 mg sodium ferric gluconate complex in sucrose had a brisk and therapeutic response in hemoglobin. Thus, the ‘‘undertreated’’ 500 mg of sodium ferric gluconate complex in sucrose patients continued to be iron deficient despite normalization of laboratory parameters of iron deficiency. The low-dose treatment group was, therefore, ‘‘functionally’’ iron deficient in the sense that these patients had normal parameters of iron stores—serum ferritins greater than 200 ng/mL—but would respond to additional iron therapy with an increase in hemoglobin, as was shown by the high-dose group. Thus, a total dosage of 500 mg sodium ferric gluconate complex in sucrose is inadequate to significantly correct anemia in these patients despite improvement in parameters of iron stores, and, based on our study, this lower dosage group proves the existence of functional iron deficiency. The average iron-deficient hemodialysis patient given 1 g of iron becomes replete but not optimally replenished. Threshold for Initiation and Continuation of Intravenous Iron Therapy
This study confirms, in a controlled clinical trial, that hemodialysis patients with either a Tsat less than 18% or a serum ferritin less than 100 ng/mL are iron deficient and respond to adequate iron replacement with an increase in hemoglobin and hematocrit. Furthermore, this study confirms that the iron deficit in these patients is greater than 500 mg, because the low-dose group did not show an increase in hemoglobin or hematocrit.
NISSENSON ET AL
Furthermore, 14 days after completion of the low-dose therapy, serum ferritin levels were greater than 240 ng/mL and Tsat was greater than 23% despite no significant change in hemoglobin or hematocrit. These findings confirm NKFDOQI guideline recommendations that iron repletion should continue empirically in hemodialysis patients, until the target hematocrit or hemoglobin is reached notwithstanding rises in serum ferritin to greater than 100 ng/mL or Tsat to greater than 20%. Controlled Clinical Confirmation of Ineffectiveness of Oral Iron in Iron Repletion
Previous studies have shown that maintenance of hemoglobin and hematocrit in hemodialysis patients with oral iron is difficult or impossible and that maintenance intravenous iron dextran is superior to maintenance oral iron in achieving and sustaining a given target hemoglobin and hematocrit at a lower dose of rHuEPO.9 The current study controlled for the effects of rHuEPO dose and showed that oral iron is ineffective in iron repletion in hemodialysis patients with frank iron-deficiency anemia. The failure of oral iron in this study was similar to the failure of the low-dose parenteral iron group. This suggests that these approaches to iron replacement are unable to sustain the necessary high rates of bioavailable iron delivery in the face of chronic continuing blood loss and the demands of rHuEPO-driven erythropoiesis. Controlled Clinical Confirmation of NKF-DOQI Iron Dosing Recommendations
Treatment with high-dose sodium ferric gluconate complex in sucrose (1.0 g) caused a significantly greater increase in hemoglobin, hematocrit, and iron saturation as compared with low-dose (0.5 g) treatment. As compared with baseline, high-dose therapy resulted in statistically significant increases in all efficacy parameters at all points after therapy, including hemoglobin, hematocrit, serum ferritin, and iron saturation. Low-dose therapy resulted in much smaller improvements in the foregoing parameters over the efficacy times. Clinical improvement in anemia in irondeficient hemodialysis patients, therefore, can be accomplished with a minimum of an eight-dose
SODIUM FERRIC GLUCONATE IN HEMODIALYSIS
repletion course of sodium ferric gluconate complex in sucrose administered as 125 mg diluted in 100 mL of saline over 1 hour during eight sequential dialysis sessions. Although the total repletion dosage of 1.0 g produces a significant but suboptimal improvement in hemoglobin and hematocrit 30 days after completion of therapy, patients have begun a return to baseline measures of iron stores by that time. NKF-DOQI guidelines include recommendations to continue therapy with intravenous iron after an initial 1.0-g course of iron in patients with documented iron-deficiency anemia. This study, therefore, provides scientific evidence that confirms this opinion-based DOQI recommendation of continued empiric iron therapy in this patient population to maintain a satisfactory target hemoglobin or hematocrit unless empiric iron therapy has been shown to be ineffective. Judicious use of intravenous iron is appropriate, because the level of ferritin or Tsat above which adverse effects, such as infection, are seen remains unknown. Treatment that achieves the values of these measures of iron status within the ranges recommended by the NKF-DOQI would seem prudent. This demonstration of the safety and efficacy of sodium ferric gluconate complex in sucrose will, it is hoped, lead to the availability of this compound for use by clinicians and researchers in the United States. The more widespread use and study of this agent will enable additional insights into the optimal methods of diagnosing and treating iron deficiency in this vulnerable patient population. REFERENCES 1. Loge JP, Lange RD, Moore CV: Characterization of the anemia associated with chronic renal insufficiency. Am J Med 24:4-18, 1958 2. Medicare Intermediary Manual. Health Care Financing Administration. Section 3907, 1996 3. Eschbach J, Egrie J, Downing M: Correction of anemia of end stage renal disease with recombinant human erythropoietin. N Engl J Med 316:73-78, 1987 4. Berliner N, Duffy TP, Abelson HT: Approach to the adult and child with anemia, in Hoffman R, Benz EJ Jr, Shattil SJ, Furie B, Cohen HJ, Silberstein LE (eds): Hematology: Basic Principles and Practice. Philadelphia, PA, Churchill Livingstone, 1995, pp 287-325 5. Health Care Financing Administration: 1995 Annual Report, End Stage Renal Disease Core Indicators Project. Baltimore, MD, Department of Health and Human Services, Health Care Financing Administration, Office of Clinical Standards and Quality, December 1995
481
6. Nissenson AR: National cooperative rHuEPO study in patients with chronic renal failure: A phase IV multicenter study. Am J Kidney Dis 18:24-33, 1991 (suppl 1) 7. Kleiner MJ, Van Wyck DB, Kaupke CJ, Kirlin LF: The role of iron and other factors in patients unresponsive to EPO therapy. Semin Dial 7:288-292, 1994 8. National Kidney Foundation Study Group: NKFDOQI Clinical Practice Guidelines for the Treatment of Anemia of Chronic Renal Failure. Am J Kidney Dis 30:S192S237, 1997 (suppl 3) 9. Fishbane S, Frei GL, Maesaka J: Reduction in recombinant human erythropoietin doses by the use of chronic intravenous iron supplementation. Am J Kidney Dis 26:4146, 1995 10. Ho¨rl WH, Dreyling K, Steinhauer HB, Engelhardt R, Schollmeyer P: Iron status of dialysis patients under rHuEPO therapy. Contrib Nephrol 87:78-86, 1990 11. Macdougall IC, Tucker B, Thompson J, Tomson CRV, Baker LRI, Raine AEG: A randomized controlled study of iron supplementation in patients treated with erythropoietin. Kidney Int 50:1694-1699, 1996 12. Van Wyck DB, Stivelman JC, Ruiz J, Kirlin LF, Katz MA, Ogden DA: Iron status in patients receiving erythropoietin for dialysis-associated anemia. Kidney Int 35:712716,1989 13. Wingard RL, Parker RA, Ismail N, Hakim RH: Efficacy of oral iron therapy in patients receiving recombinant human erythropoietin. Am J Kidney Dis 25:433-439, 1995 14. Fishbane S, Ungureanu V, Maesaka J, Kaupke CJ, Lim V, Wish J: The safety of intravenous iron dextran in hemodialysis patients. Am J Kidney Dis 28:529-534, 1996 15. Paull J: A prospective study of dextran-induced anaphylactoid reactions in 5745 patients. Anaesth Intensive Care 15:163-167, 1987 16. Hamstra RD, Block MH, Schocket AL: Intravenous iron dextran in clinical practice. JAMA 243:1726-1731, 1980 17. Navarro JF, Teruel JL, Liano F, Marcep R, Ortuno J: Effectiveness of intravenous administration of Fe-Gluconate-Na complex to maintain adequate body iron stores in hemodialysis patients. Am J Nephrol 16:268-272, 1996 18. Taylor JE, Peat N, Porter C, Morgan AG: Regular low-dose intravenous iron therapy improves response to erythropoietin in hemodialysis patients. Nephrol Dial Transplant 11:1079-1083, 1996 19. Pascual J, Teruel JL, Liano F, Sureda A, Ortuno J: Intravenous Fe-Gluconate-Na for iron-deficient patients on hemodialysis. Nephron 60:121, 1992 (letter) 20. Pascual J, Teruel JL, Liano E, Sureda A, Ortuno J: Sodium ferric gluconate complex given intravenously for iron deficiency in hemodialysis. Clin Nephrol 35:87, 1991 (letter) 21. Allegra V, Mengozzi G, Vasile A: Iron deficiency in maintenance hemodialysis patients: assessment of diagnosis criteria and of three different iron treatments. Nephron 57:175-182, 1991 22. Amerling R, Cu GA, Dubrow A, Levin NW, Osheroff RJ: Complications during hemodialysis, in Nissenson AR, Fine RN, Gentile DE (eds): Clinical Dialysis. Norwalk, CT, Appleton & Lange, pp 235-267 23. Nissenson AR, Swan S, Lambrecht LL, Anderson P, Schweitzer S: Ferric gluconate (FerrlecitÆ) is safe in hemo-
482
dialysis patients who react to iron dextran. J Am Soc Nephrol 7:1460A, 1996 (abstr) 24. Faich G, Strobos J: Sodium ferric gluconate complex in sucrose: Safer than iron dextran. J Am Soc Nephrol 8:192A, 1997 (abstr) 25. Bridges KR, Seligman PA: Disorders of iron metabolism, in Handin RI, Lux SE, Stossel TP (eds): Blood: Principles and Practice of Hematology. Philadelphia, PA, Lippincott, 1995 26. Brittenham GM: Disorders of iron metabolism, in Hoffman R, Benz EJ, Shattil SJ, Furie B, Cohen HJ, Silberstein LE (eds): Hematology: Basic Principles and Practice. New York, NY, Churchill Livingstone, 1995 27. Nissenson AR, Strobos J: Iron deficiency in patients with renal failure. Kidney Int 55:518-521, 1999 (Suppl 69) 28. Stivelman JV: Optimization of iron therapy in hemodialysis patients treated with rHuEPO. Semin Dial 6:59-69, 1993 29. Lindsay RM, Burton JA, Edward N, Dargle JH, Prentice CRM, Kennedy AC: Dialyzer blood loss. Clin Nephrol 1:29-34, 1973 30. Longnecker RE, Goffinet JA, Hendler ED: Blood loss form maintenance hemodialysis. Trans Am Soc Artif Internal Organs 20:135-141, 1974 31. Muller-Wiefel DE, Sinn H, Gilli G, Scharer K: Hemo-
NISSENSON ET AL
lysis and blood loss in children with chronic renal failure. Clin Nephrol 8:481-486, 1977 32. Eschbach J, Cook JD, Scribner BH, Finch CA: Iron balance in hemodialysis patients. Ann Intern Med 87:710713, 1977 33. Kooistra MP, van Es A, Struyvenberg A, Marx JJM: Iron metabolism in patients with anaemia of end-stage renal disease during treatment with recombinant human erythropoietin. Br J Haematol 79:634-639, 1991 34. Dunea G, Swagel MA, Bodiwala U, Arruda JAL: Intradialytic oral iron therapy. Int J Artif Organs 17:261-264, 1994 35. Anastassiades EG, Howarth D, Howarth J, Shanks D, Waters HM, Hyde K, Geary CG, Liu Yin JA, Gokal R: Monitoring iron requirements in renal patients on erythropoietin. Nephrol Dial Transplant 8:846-853, 1993 36. Kotaki M, Uday K, Henriquez M, Blum S, Dave M: Maintenance therapy with intravenous iron in hemodialysis patients receiving erythropoietin. Clin Nephrol 48:63-64, 1997 37. Linder MC, Schaffer KJ, Hazegh-Azam M, Zhou CY, Tran TN, Nagel GM: Serum ferritin: Does it differ from tissue ferritin? J Gastroenterol Hepatol 11:1033-1036, 1996 38. Rogers JT: Ferritin translation by interleukin-6: The role of sequences upstream of the start codons of the heavy and light subunit genes. Blood 87:2525-2537, 1996
Editorial, p. 595
A
NEMIA is present in most patients with chronic renal failure (CRF) on hemodialysis1,2 and is due primarily to a deficiency of endogenous erythropoietin.3,4 Although most hemodialysis patients are now receiving recombinant human erythropoietin (rHuEPO), anemia persists, with iron deficiency the most frequent cause of initial poor response to rHuEPO, or acquired resistance to rHuEPO.5-7 Recent clinical practice guidelines developed by the National Kidney Foundation Dialysis Outcomes Quality Initiative (NKF-DOQI) document that most patients receiving hemodialysis require parenteral iron to maintain iron status.8 The basis for this recommendation is the understanding that hemodialysis patients receiving rHuEPO are highly likely to develop iron deficiency, because of regular blood (and iron) loss incident to the dialysis procedure, and the increased demands for iron when rHuEPO is being administered and is driving erythropoiesis. This high iron requirement exceeds the amount of iron that can generally be provided orally.9-13
Both the clinical management of dialysis patients with iron deficiency and clinical research into iron metabolism, monitoring, and administration have been hampered by the current availability in the United States of only a single form of parenteral iron, iron dextran. Although generally safe, severe allergic reactions can occur, and deaths after allergic reactions have been reported.14-16 From the University of California at Los Angeles Medical Center, Los Angeles, CA; the Renal Unit, Victoria Hospital, London, Ontario, Canada; Hennepin County Medical Center, Department of Medicine, University of Minnesota, Minneapolis, MN; Division of Hematology, University of Colorado Health Sciences Center, Denver, CO; and private practice, Washington, DC. Received August 13, 1998; accepted in revised form September 25, 1998. Supported by R&D Laboratories, Inc. Clinical supplies in the form of Ferrlecit Injection (sodium ferric gluconate complex in sucrose) were provided by the manufacturer, Rhoˆne-Poulenc Rorer GmBH, Dagenham, United Kingdom. Address reprint requests to Allen R. Nissenson, MD, UCLA Medical Center, 200 Medical Plaza, Suite 565; Los Angeles, CA 90095. E-mail: [email protected]
娀 1999 by the National Kidney Foundation, Inc. 0272-6386/99/3303-0007$3.00/0
American Journal of Kidney Diseases, Vol 33, No 3 (March), 1999: pp 471-482
471
472
NISSENSON ET AL
Sodium ferric gluconate complex in sucrose injection is a high-molecular-weight complex (350,000 daltons) composed of ferric saccharate centers chelated by gluconate, which has a high affinity for ferric ions and the capacity to form a powerful bridge between adjacent ferric oxide centers.* The absence of the need to degrade a dextran shell or solubilize a ferric oxide core also should provide a more rapidly bioavailable iron after administration. Sodium ferric gluconate complex in sucrose injection has been used since 1959 in more than 20 countries outside the United States for the intravenous administration of iron to patients with iron-deficiency anemia. Currently, the principal intravenous usage is among the 48,000 hemodialysis patients in Germany (R. Schaefer, personal communication, 1997). Several groups have reported excellent results with sodium ferric gluconate complex in sucrose when administered to iron-deficient hemodialysis patients,17-21 but these reports are in small numbers of patients, from single centers, and are largely descriptive. The current study was designed to assess the safety and efficacy of two different doses of parenteral sodium ferric gluconate complex in sucrose in iron-deficient hemodialysis patients. The randomized, controlled study design also permitted the assessment of the minimum iron needs in this patient population, as well as a confirmation of the concept of functional iron deficiency. MATERIALS AND METHODS Sodium Ferric Gluconate Complex in Sucrose The study drug was provided in the form of Ferrlecit by R&D Laboratories, Inc. (Marina Del Rey, CA) through the manufacturer, Rhoˆne-Poulenc Rorer GmBH (Dagenham, United Kingdom). Each ampoule contained 5 mL Ferrlecit for intravenous injection containing 62.5 mg of elemental iron as the sodium salt of a ferric ion carbohydrate complex in an alkaline aqueous solution with approximately 20% sucrose.
Purpose This was an open-label, randomized study of the treatment of iron-deficiency anemia in adult chronic hemodialysis patients receiving rHuEPO that compared the safety and efficacy of iron repletion using two different dosing regimens of intravenous sodium ferric gluconate complex in *Proprietary Data supplied by R&D Laboratories, Inc, Marina Del Rey, CA.
sucrose, totaling either 1.0 or 0.5 g administered over a course of eight sequential dialysis sessions.
Entry Criteria and Patient Population Adult patients receiving chronic hemodialysis with either serum ferritin levels less than 100 ng/mL or iron saturations less than 18% AND either a hemoglobin level less than 10 g/dL or hematocrit of 32% or less were eligible to participate in the study. Patients were excluded if they had received parenteral iron products or any investigational drug that might interfere with iron metabolism in the 2 months preceding enrollment. Other exclusion criteria were significant comorbid conditions and baseline rHuEPO requirements greater than 10,000 units three times weekly. Patients with a history of allergic conditions or a specific non–life-threatening allergic reaction to iron dextran were not excluded. No oral iron or blood transfusions were administered during the study. The study design required that starting rHuEPO doses remain unchanged throughout the study.
Drug Administration Patients were randomly assigned to a course of either low-dose or high-dose sodium ferric gluconate complex in sucrose before administration of a test dose of 25 mg. Drug was administered intravenously as 62.5 mg in 50 mL saline or 125 mg in 100 mL saline over 30 or 60 minutes, respectively, in eight divided doses over eight sequential dialysis sessions. The protocol precluded alternative oral or parenteral iron administration, blood transfusion, or alteration in rHuEPO dose during the course of the 47-day study.
Efficacy Analysis Demographic and medical history variables as well as baseline efficacy values were compared in the two groups. Mean changes within group for hemoglobin and hematocrit from baseline to 2, 14, and 30 days after completion of the full course of drug administration were compared by paired t-test. Mean hemoglobin and hematocrit changes between high-dose and low-dose groups at the same times were compared by an analysis of covariance (ANOVA) using a mixed model for analysis of repeated measures. Transferrin saturation (Tsat) and serum ferritin were analyzed and compared in the two groups at the same points. A matched control group of 25 patients treated only with oral iron (ferrous sulfate 325 mg or ferrous gluconate 650 mg thrice daily) and meeting the same entry criteria as the patients receiving parenteral iron was studied as well. Efficacy data for the control group were available for days 30 and 60 after initiation of oral iron therapy. Analyses for all groups used an intent-to-treat approach.
Safety Analysis Potential adverse events in the patients receiving sodium ferric gluconate complex in sucrose were identified through history, physical examination, and active inquiry of patients by using a form containing Coding System for a Thesaurus of Adverse Reaction Terms (COSTART), which was completed at each drug administration and at 2 days after cessation of drug therapy. Adverse events were rated by the investigator for possible relationship to study drug (none, unlikely, possible, probable, or definite). The incidence of adverse events was compared between dose groups and
SODIUM FERRIC GLUCONATE IN HEMODIALYSIS
against a matched control group by using Fisher’s exact test. Routine blood samples for laboratory assessment of adverse organ-system effects were collected at baseline and on days 2, 14, and 30 after cessation of therapy. Laboratory measurements included alkaline phosphatase, alanine transaminase, aspartate transaminase, bilirubin, blood urea nitrogen, creatinine, glucose, platelets, and white blood cell counts and differentials. Laboratory safety data were assessed by using shift tables, and Fisher’s exact test was used to compare dose groups.
RESULTS
Eighty-three of 88 patients completed the study, including 44 in the high-dose and 39 in the low-dose group. Five patients were withdrawn from the study after entry but before day 2 when first efficacy data were obtained. Two patients discontinued for logistical reasons; one patient was withdrawn after development of pruritus and chest pain immediately after the test dose administration; one patient was withdrawn after development of nausea, abdominal and flank pain, fatigue, and rash after the first full dose; and one patient was withdrawn on development of a red blotchy rash after the first full dose. Baseline demographic and vital signs as well as statistical analysis of comparability are shown in Table 1. Randomization resulted in comparable groups. As shown in Table 2, baseline values for hemoglobin, hematocrit, serum ferritin, serum iron, mean corpuscular hemoglobin (MCH), mean corpuscular volume (MCV), and mean corpuscular hemoglobin content (MCHC) were similar between the low-dose and highdose treatment groups. However, percent iron saturation was significantly (P ⫽ 0.045) lower in the high-dose group compared with the low-dose group. The oral iron group was comparable to the two treatment groups except for baseline serum ferritin, which was significantly elevated in the oral iron group. Anemia—Mean Change in Hemoglobin and Hematocrit
High-Dose Group (1 gram) The mean increase in hemoglobin in the highdose group from baseline (9.6 g/dL) to days 2, 14, and 30 was 1.0 g/dL, 1.1 g/dL, and 1.3 g/dL, respectively. Each of these values was statistically significant by paired t-test (P ⬍ 0.0001, P ⬍ 0.0001, and P ⬍ 0.0001, respectively). The mean increase in hematocrit in the high-dose
473
group from baseline (28.8%) to days 2, 14, and 30 was 3.1%, 3.6%, and 3.5%, respectively. Each of these values was statistically significant by paired test (P ⬍ 0.0001, P ⬍ 0.0001, and P ⬍ 0.0001, respectively). Low-Dose Group (500 mg) The mean increase in hemoglobin in the lowdose group from baseline (9.8 g/dL) to days 2, 14, and 30 was 0.3 g/dL, 0.3 g/dL, and 0.5 g/dL, respectively. Only the value for day 30 was statistically significant by paired t-test (P ⫽ 0.071, P ⫽ 0.072, and P ⫽ 0.01, respectively). The mean increase in hematocrit in the low-dose group from baseline (29.4%) to days 2, 14, and 30 was 1.1%, 1.4%, and 1.4%, respectively. Only the values for days 14 and 31 were statistically significant by paired test (P ⫽ 0.06, P ⫽ 0.018, and P ⫽ 0.024, respectively). Oral Iron (Control) Group The mean increase in hemoglobin in the oral iron group from baseline (9.4 g/dL) to days 14 and 30 was 0.4 g/dL and 0.4 g/dL, respectively. Each of these values was statistically significant by paired t-test (P ⫽ 0.016). The mean increase in hematocrit in the oral dose group from baseline (28.6%) to days 14 and 30 was 0.8% and 0.4%, respectively. None of these values were statistically significant by paired t-test (P ⫽ 0.112 and P ⫽ 0.112, respectively). The comparative increase in mean hemoglobin in the high- versus the low-dose group was statistically significant (P ⬍ 0.001) at all times using an ANOVA that included the effects of center as well as treatment. The comparative increase in mean hematocrit was statistically significant (P ⬍ 0.002) at all times using the ANOVA. The comparative increase in mean hemoglobin in the high-dose versus the oral-dose group was statistically significant (P ⫽ 0.001) at all times by ANOVA. The comparative increase in mean hematocrit in the high-dose versus the oral-dose group was statistically significant (P ⬍ 0.001) at all times by ANOVA. The comparative increase in mean hemoglobin and hematocrit in the low-dose group versus the oral-dose group was not statistically different at any time by ANOVA (P ⫽ 0.337 and 0.065, respectively). The mean hemoglobin levels at baseline and
474
NISSENSON ET AL Table 1. Baseline Patient Demographics and Vital Signs P*
Treatment Group
Variable
Age (yr) n Mean (Std) Minimum Maximum Sex, n (%) Female Male Missing Race, n (%) White Black Latin Asian Hispanic Native-American Persian Unknown Missing Systolic BP (mm Hg) n Mean (Std) Minimum Maximum Diastolic BP (mm Hg) n Mean (Std) Minimum Maximum Pulse (bpm) n Mean (Std) Minimum Maximum Respirations (bpm) n Mean (Std) Minimum Maximum
500 mg (n ⫽ 41)
1,000 mg (n ⫽ 47)
Control (n ⫽ 25)
500 v 1,000 mg
500 mg v Control
1,000 mg v Control
Overall
41 55.0 (17.7) 22.0 80.0
47 57.1 (17.7) 20.0 87.0
25 52.2 (16.6) 25.0 84.0
0.565
0.536
0.259
0.522
23 (56.1) 18 (40.9) 0 (0.0)
25 (50.2) 21 (44.7) 1 (2.1)
17 (68.0) 8 (32.0) 0 (0.0)
1.000
0.408
0.328
0.536
30 (73.2) 8 (19.5) 2 (4.9) 1 (2.4) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0)
36 (76.6) 5 (10.6) 2 (4.3) 1 (2.1) 0 (0.0) 1 (2.1) 1 (2.1) 0 (0.0) 1 (2.1)
10 (40.0) 8 (32.0) 0 (0.0) 1 (4.0) 5 (20.0) 0 (0.0) 0 (0.0) 1 (4.0) 0 (0.0)
0.813
0.003
⬍0.001
0.001
39 144.5 (24.1) 72.0 196.0
43 155.0 (19.7) 120.1 200.0
13 170.0 (19.9) 139.0 199.0
0.030
⬍0.001
0.032
0.001
39 77.2 (13.9) 45.0 100.0
43 82.5 (12.2) 50.0 114.0
13 86.6 (16.2) 61.0 112.0
0.077
0.031
0.337
0.057
40 86.0 (10.2) 62.0 113.0
43 83.4 (12.8) 59.0 112.0
12 78.9 (10.1) 56.0 96.0
0.327
0.072
0.246
0.184
31 18.0 (3.2) 12.0 22.0
33 18.3 (3.0) 12.0 22.0
—† —† —† —†
0.980
N/A
N/A
0.980
*For a continuous variable, an ANOVA model with effects for treatment group was used to compare the group means, and the P value was associated with the P test. For a categorical variable, the P value was associated with the Fisher’s exact test. †No data were available. Abbreviation: N/A, not applicable.
days 2, 14, and 30 for the low- and high-dose groups are shown in Fig 1. The mean hematocrit level at baseline and days 2, 14, and 30 for the low-dose and high-dose groups are presented in Fig 2. Figure 3 provides the mean hemoglobin levels at baseline and days 14 and 30 for low-, high-, and oral-dose groups. Figure 4 provides mean hematocrit levels at baseline and days 14 and 30 for low-, high-, and oral-dose groups.
Iron ‘‘stores’’—serum ferritin and iron saturation. The mean increase in serum ferritin in the high-dose group from baseline (88 ng/dL) was 320 ng/dL, 199 nm/dL, and 134 nm/dL on days 2, 14, and 30, respectively. Each of these values was statistically significant by paired t-test (P ⬍ 0.0001, P ⬍ 0.0001, and P ⬍ 0.0001, respectively). The mean increase in Tsat in the highdose group from baseline (15.6%) to days 2, 14,
SODIUM FERRIC GLUCONATE IN HEMODIALYSIS
475
Table 2. Baseline Hematological Data P*
Treatment Group
Variable
500 mg (n ⫽ 41)
Baseline hemoglobin (g/dL) n 39 Mean (Std) 9.8 (0.8) Minimum 7.1 Maximum 11.1 Baseline hematocrit (%) n 39 Mean (Std) 29.4 (2.3) Minimum 23.0 Maximum 33.0 Baseline percent iron saturation (%) n 39 Mean (Std) 20.1 (11.9) Minimum 4.0 Maximum 58.0 Baseline serum ferritin (mg/mL) n 37 Mean (Std) 105.6 (116.3) Minimum 19.0 Maximum 660.0 Baseline serum iron (ug/dL) n 39 Mean (Std) 49.7 (27.7) Minimum 11.0 Maximum 112.0 Baseline MCH (pg) n 40 Mean (Std) 30.0 (3.2) Minimum 20.0 Maximum 35.0 Baseline MCV (fl) n 40 Mean (Std) 89.5 (7.7) Minimum 67.0 Maximum 103.6 Baseline MCHC (g/dL) n 40 Mean (Std) 33.4 (1.3) Minimum 30.0 Maximum 35.3
1,000 mg (n ⫽ 47)
Control (n ⫽ 25)
500 v 1,000 mg
44 9.6 (0.9) 6.5 10.9
25 9.4 (0.8) 7.8 10.4
0.298
0.038
0.230
0.124
44 28.8 (2.6) 20.0 32.0
25 28.6 (2.1) 24.0 31.4
0.305
0.212
0.705
0.401
43 15.6 (7.5) 2.0 35.0
25 14.2 (4.4) 6.7 27.0
0.026
0.012
0.530
0.020
0.678
⬍0.001
⬍0.001
⬍0.001
0.112
0.007
0.162
0.024
0.691
0.055
0.022
0.060
0.705
0.001
⬍0.001
⬍0.001
0.566
0.036
0.009
0.028
43 12 88.4 (143.2) 605.6 (390.9) 8.0 41.0 941.0 1000 44 41.6 (22.4) 6.0 101.0
25 33.5 (13.6) 16.0 81.0
44 29.7 (2.2) 25.0 34.0
25 31.4 (3.3) 26.6 39.1
43 88.9 (5.3) 73.6 100.1
25 96.0 (9.0) 84.6 116.0
44 33.5 (1.2) 31.0 36.0
25 32.7 (0.6) 31.4 34.2
500 mg v Control
1,000 mg v Control
Overall
*For a continuous variable, an ANOVA model with effects for treatment group was used to compare the group means, and the P value was associated with the P test. For a categorical variable, the P value was associated with the Fisher’s exact test.
and 30 was 9.0%, 8.5%, and 5.5%, respectively. Each of these values was statistically significant by paired test (P ⬍ 0.0001, P ⬍ 0.0001, and P ⫽ 0.002, respectively). The mean increases in serum ferritin in the low-dose group from baseline (106 ng/dL) were 130 ng/dL, 132 ng/dL, and 65 ng/dL on days 2, 14, and 30, respectively. The values for days 2, 14, and 30 were statistically significant by paired t-test (P ⬍ 0.0001, P ⫽ 0.003, and P ⫽ 0.01, respectively). The mean
change in iron saturation in the low-dose group from baseline (20.1%) was 1.2%, 2.8%, and ⫺0.6% on days 2, 14, and 30, respectively. None of the values were statistically significantly different from baseline by paired test (P ⫽ 0.47, P ⫽ 0.156, and P ⫽ 0.72, respectively). Iron saturation and serum ferritin levels fell rapidly after cessation of iron therapy. There was not a statistically significant difference in the increase in mean ferritin in the high-
476
NISSENSON ET AL
Fig 1. Mean values for hemoglobin in renal dialysis patients receiving low- (䊏) or high-dose (䊐) sodium ferric gluconate complex in sucrose.
versus the low-dose group at any time. Conversely, there was a significant increase in mean iron saturation in the high- versus the low-dose group (P ⬍ 0.017) at all points under ANOVA. The mean serum ferritin levels from baseline to days 2, 14, and 30 for the low-dose and high-dose groups are presented in Fig 5. The mean Tsat values from baseline to days 2, 14, and 30 for the low-dose and high-dose groups are presented in Fig 6. Safety analysis. A total of 88 patients enrolled in the study and received a test dose of 25 mg sodium ferric gluconate complex in sucrose before the beginning of the study. All 88 patients
Fig 2. Mean values for hematocrit in renal dialysis patients receiving low- (䊏) or high-dose (䊐) sodium ferric gluconate complex in sucrose.
were included in the safety analysis. No adverse event was considered by the investigators to be definitely related to sodium ferric gluconate complex in sucrose treatment. No patient experienced a type I immediate hypersensitivity reaction. There were no hospitalizations or deaths related to drug administration. Most adverse events were considered to be of mild or moderate severity and not definitely related to sodium ferric gluconate complex in sucrose administration. Adverse events that were possibly or probably related to treatment by investigator report were experienced by nine patients and included nausea in four patients,
SODIUM FERRIC GLUCONATE IN HEMODIALYSIS
477
Fig 3. Mean values for hemoglobin in renal dialysis patients receiving low- (䊏) or high-dose (䊐) sodium ferric gluconate complex in sucrose or oral-dose iron (°).
vomiting in three, rash in two, and abdominal pain, back pain, fatigue, syncope, cramps, agitation, menorrhagia, pruritus, chest pain, paresthesia, and abnormal erythrocytes in one patient each. An analysis of the incidence of adverse events in the two treatment groups showed that the number of patients experiencing each type of event was similar between the groups. Adverse events were further analyzed by comparison with published data on rates of adverse events in patients on routine hemodialysis.22 The adverse events reported in the patients receiving sodium ferric gluconate complex in sucrose occurred at no greater frequency statistically than these same
events in patients undergoing routine dialysis but not receiving parenteral iron as assessed by Fisher’s exact test. No clinically significant changes in laboratory values or vital signs were observed during the study. DISCUSSION Safety and Efficacy of Sodium Ferric Gluconate Complex in Sucrose
This study shows that sodium ferric gluconate complex in sucrose is safe and effective in correcting anemia and improving iron status in adult patients on chronic hemodialysis and rHuEPO
Fig 4. Mean values for hematocrit in renal dialysis patients receiving low- (䊏) or high-dose (䊐) sodium ferric gluconate complex in sucrose or oral-dose iron (°).
478
NISSENSON ET AL
Fig 5. Mean values for serum ferritin in renal dialysis patients receiving low- (䊏) or high-dose (䊐) sodium ferric gluconate complex in sucrose.
therapy. Although hospitalizations are common in this patient population (17 study patients were hospitalized at some time during the study), there were no serious adverse events, hospitalizations, deaths, or laboratory abnormalities related to drug administration. No patient experienced a type I immediate hypersensitivity reaction. We previously reported on nine hemodialysis patients who had experienced serious reactions to iron dextran treatment and who were successfully treated with sodium ferric gluconate complex in sucrose.23 These data are consistent with
Fig 6. Mean values for transferrin saturation in renal dialysis patients receiving low- (䊏) or high-dose (䊐) sodium ferric gluconate complex in sucrose.
a recent review of worldwide adverse events reporting from 1976 to 1996 on usage of sodium ferric gluconate complex in sucrose in Germany, where no known hypersensitivity deaths have occurred.24 The current study, however, did not include patients randomized to iron dextran, limiting any final conclusions about the relative safety of these two compounds. Iron-Deficiency Anemia in Hemodialysis Patients
Iron-deficiency anemia results from the absence of sufficient available body iron stores to
SODIUM FERRIC GLUCONATE IN HEMODIALYSIS
sustain hematopoiesis. Sixty-five percent of body iron is present in circulating red blood cells and myoglobin, with the rest in the erythroid marrow, or other longer-term storage sites, such as the liver and macrophages within the reticuloendothelial system (RES).25,26 Iron-deficiency anemia most commonly results from blood loss.25,26 Correction of iron deficiency with oral iron is generally simple and effective unless patients have chronic continuing or uncontrolled blood loss, such as occurs with regular hemodialysis treatment.25-27 rHuEPO therapy, which produces an increase in serum hemoglobin levels, results in further increases in iron loss during hemodialysis even though the total blood loss may remain constant.7,28,29 Iron-deficiency anemia, common in hemodialysis patients because of ongoing blood loss, is thus exacerbated by rHuEPO therapy.28-31 There are other factors that also may contribute to iron deficiency in the hemodialysis patient, including a diet that is low in bioavailable iron, less than optimal iron absorption, and ongoing gastrointestinal tract losses of iron.32,33 A number of studies have shown that the theoretical inability of hemodialysis patients receiving rHuEPO to keep up with continuing iron demands by oral iron administration is actually seen clinically.9-13 Only a few studies support the effectiveness of oral iron usage in this population, and then only when oral iron is administered under a carefully controlled regimen. Effectiveness can only be shown for a minority of patients.34-36 As a result, most hemodialysis patients require frequent courses of parenteral iron. Unfortunately, the most commonly used clinical tests to define iron deficiency in dialysis patients have major shortcomings. In non-CRF patients, for example, serum ferritin levels have been associated with the amount of iron bound to intracellular ferritin within the RES and are known to vary with intracellular or hepatic iron stores. Uncontrolled clinical studies have shown that in anemic renal failure patients, however, a hematopoietic response to iron administration can be obtained despite serum ferritin levels that are significantly higher than 10 ng/mL, thought to be diagnostic of iron deficiency in nonrenal patients, and even 100 ng/mL, usually cited as diagnostic of iron deficiency in dialysis patients.6,7 A hematopoietic response to exogenous iron in this setting indicates the presence of ‘‘functional’’ iron deficiency.9
479
It has been suggested that functional iron deficiency may reflect a state in which the serum ferritin value is consistent with the presence of sufficient iron bound to intracellular ferritin to meet hematopoietic needs, but because of an inability of the RES to release iron (RES blockade), this stored iron is unavailable. The administration of supplemental intravenous iron in this situation would overcome the RES blockade either by directly providing bioavailable iron or by increasing iron stores to nonphysiological levels to overcome the blockade.9 Alternatively, patients with functional iron deficiency may be frankly iron deficient, but serum ferritin levels simply may not accurately reflect iron stores. For some time, serum ferritin was thought to represent leakage of intracellular RES ferritin. Three lines of evidence now refute this thesis. First, serum ferritin is largely present as apoferritin, without any bound iron, even when intracellular ferritin is sufficiently iron overloaded to result in deposition of intracellular hemosiderin. Second, the molecular structure of serum ferritin is different from that of intracellular ferritin.37 Third, the control over the synthesis of serum ferritin and intracellular ferritin is different. The synthesis of serum ferritin is stimulated by interleukin-1 and interleukin-6, which are elevated in CRF patients.38 Thus, although there is generally a positive correlation between intrahepatic iron stores and serum ferritin, the relationship is not consistent and the correlation coefficient is low with a broad confidence interval.26,27 Many chronic diseases, including CRF, are associated with high serum ferritin levels despite low tissue iron levels. Tsat is a measure of serum iron bound to circulating transferrin. Low saturations are generally indicative of either (1) an increase in total transferrin concentration and synthesis that is increased under autoregulation by iron deficiency; or (2) a low concentration of total circulating iron occupying available transferrin-binding sites. In the non-CRF patient, Tsat below 10% to 15% is considered indicative of iron deficiency. Tsat is also not an ideal indicator of iron status in the dialysis patient. Tsat varies diurnally and only assesses the small quantity of iron currently available for erythropoiesis in the circulation. Exogenous rHuEPO therapy may depress Tsat levels directly by increasing
480
erythroid marrow uptake or transferrin-binding receptor sites. rHuEPO also increases red blood cell turnover and causes the release of iron from iron storage sites.26,27 Thus, Tsat has become an imperfect measure of iron deficiency in the CRF patient. In patients on dialysis, the hematopoietic responses to iron therapy in patients with higher serum ferritin or Tsat levels has led to the supposition that these patients have functional iron deficiency. In this study, iron-deficient hemodialysis patients randomized to treatment with 500 mg sodium ferric gluconate complex in sucrose did show normalization of average patient laboratory parameters of iron deficiency. However, these patients had no improvement in hemoglobin and only minimal improvement in hematocrit. By contrast, equivalent randomized patients receiving an additional 500 mg sodium ferric gluconate complex in sucrose had a brisk and therapeutic response in hemoglobin. Thus, the ‘‘undertreated’’ 500 mg of sodium ferric gluconate complex in sucrose patients continued to be iron deficient despite normalization of laboratory parameters of iron deficiency. The low-dose treatment group was, therefore, ‘‘functionally’’ iron deficient in the sense that these patients had normal parameters of iron stores—serum ferritins greater than 200 ng/mL—but would respond to additional iron therapy with an increase in hemoglobin, as was shown by the high-dose group. Thus, a total dosage of 500 mg sodium ferric gluconate complex in sucrose is inadequate to significantly correct anemia in these patients despite improvement in parameters of iron stores, and, based on our study, this lower dosage group proves the existence of functional iron deficiency. The average iron-deficient hemodialysis patient given 1 g of iron becomes replete but not optimally replenished. Threshold for Initiation and Continuation of Intravenous Iron Therapy
This study confirms, in a controlled clinical trial, that hemodialysis patients with either a Tsat less than 18% or a serum ferritin less than 100 ng/mL are iron deficient and respond to adequate iron replacement with an increase in hemoglobin and hematocrit. Furthermore, this study confirms that the iron deficit in these patients is greater than 500 mg, because the low-dose group did not show an increase in hemoglobin or hematocrit.
NISSENSON ET AL
Furthermore, 14 days after completion of the low-dose therapy, serum ferritin levels were greater than 240 ng/mL and Tsat was greater than 23% despite no significant change in hemoglobin or hematocrit. These findings confirm NKFDOQI guideline recommendations that iron repletion should continue empirically in hemodialysis patients, until the target hematocrit or hemoglobin is reached notwithstanding rises in serum ferritin to greater than 100 ng/mL or Tsat to greater than 20%. Controlled Clinical Confirmation of Ineffectiveness of Oral Iron in Iron Repletion
Previous studies have shown that maintenance of hemoglobin and hematocrit in hemodialysis patients with oral iron is difficult or impossible and that maintenance intravenous iron dextran is superior to maintenance oral iron in achieving and sustaining a given target hemoglobin and hematocrit at a lower dose of rHuEPO.9 The current study controlled for the effects of rHuEPO dose and showed that oral iron is ineffective in iron repletion in hemodialysis patients with frank iron-deficiency anemia. The failure of oral iron in this study was similar to the failure of the low-dose parenteral iron group. This suggests that these approaches to iron replacement are unable to sustain the necessary high rates of bioavailable iron delivery in the face of chronic continuing blood loss and the demands of rHuEPO-driven erythropoiesis. Controlled Clinical Confirmation of NKF-DOQI Iron Dosing Recommendations
Treatment with high-dose sodium ferric gluconate complex in sucrose (1.0 g) caused a significantly greater increase in hemoglobin, hematocrit, and iron saturation as compared with low-dose (0.5 g) treatment. As compared with baseline, high-dose therapy resulted in statistically significant increases in all efficacy parameters at all points after therapy, including hemoglobin, hematocrit, serum ferritin, and iron saturation. Low-dose therapy resulted in much smaller improvements in the foregoing parameters over the efficacy times. Clinical improvement in anemia in irondeficient hemodialysis patients, therefore, can be accomplished with a minimum of an eight-dose
SODIUM FERRIC GLUCONATE IN HEMODIALYSIS
repletion course of sodium ferric gluconate complex in sucrose administered as 125 mg diluted in 100 mL of saline over 1 hour during eight sequential dialysis sessions. Although the total repletion dosage of 1.0 g produces a significant but suboptimal improvement in hemoglobin and hematocrit 30 days after completion of therapy, patients have begun a return to baseline measures of iron stores by that time. NKF-DOQI guidelines include recommendations to continue therapy with intravenous iron after an initial 1.0-g course of iron in patients with documented iron-deficiency anemia. This study, therefore, provides scientific evidence that confirms this opinion-based DOQI recommendation of continued empiric iron therapy in this patient population to maintain a satisfactory target hemoglobin or hematocrit unless empiric iron therapy has been shown to be ineffective. Judicious use of intravenous iron is appropriate, because the level of ferritin or Tsat above which adverse effects, such as infection, are seen remains unknown. Treatment that achieves the values of these measures of iron status within the ranges recommended by the NKF-DOQI would seem prudent. This demonstration of the safety and efficacy of sodium ferric gluconate complex in sucrose will, it is hoped, lead to the availability of this compound for use by clinicians and researchers in the United States. The more widespread use and study of this agent will enable additional insights into the optimal methods of diagnosing and treating iron deficiency in this vulnerable patient population. REFERENCES 1. Loge JP, Lange RD, Moore CV: Characterization of the anemia associated with chronic renal insufficiency. Am J Med 24:4-18, 1958 2. Medicare Intermediary Manual. Health Care Financing Administration. Section 3907, 1996 3. Eschbach J, Egrie J, Downing M: Correction of anemia of end stage renal disease with recombinant human erythropoietin. N Engl J Med 316:73-78, 1987 4. Berliner N, Duffy TP, Abelson HT: Approach to the adult and child with anemia, in Hoffman R, Benz EJ Jr, Shattil SJ, Furie B, Cohen HJ, Silberstein LE (eds): Hematology: Basic Principles and Practice. Philadelphia, PA, Churchill Livingstone, 1995, pp 287-325 5. Health Care Financing Administration: 1995 Annual Report, End Stage Renal Disease Core Indicators Project. Baltimore, MD, Department of Health and Human Services, Health Care Financing Administration, Office of Clinical Standards and Quality, December 1995
481
6. Nissenson AR: National cooperative rHuEPO study in patients with chronic renal failure: A phase IV multicenter study. Am J Kidney Dis 18:24-33, 1991 (suppl 1) 7. Kleiner MJ, Van Wyck DB, Kaupke CJ, Kirlin LF: The role of iron and other factors in patients unresponsive to EPO therapy. Semin Dial 7:288-292, 1994 8. National Kidney Foundation Study Group: NKFDOQI Clinical Practice Guidelines for the Treatment of Anemia of Chronic Renal Failure. Am J Kidney Dis 30:S192S237, 1997 (suppl 3) 9. Fishbane S, Frei GL, Maesaka J: Reduction in recombinant human erythropoietin doses by the use of chronic intravenous iron supplementation. Am J Kidney Dis 26:4146, 1995 10. Ho¨rl WH, Dreyling K, Steinhauer HB, Engelhardt R, Schollmeyer P: Iron status of dialysis patients under rHuEPO therapy. Contrib Nephrol 87:78-86, 1990 11. Macdougall IC, Tucker B, Thompson J, Tomson CRV, Baker LRI, Raine AEG: A randomized controlled study of iron supplementation in patients treated with erythropoietin. Kidney Int 50:1694-1699, 1996 12. Van Wyck DB, Stivelman JC, Ruiz J, Kirlin LF, Katz MA, Ogden DA: Iron status in patients receiving erythropoietin for dialysis-associated anemia. Kidney Int 35:712716,1989 13. Wingard RL, Parker RA, Ismail N, Hakim RH: Efficacy of oral iron therapy in patients receiving recombinant human erythropoietin. Am J Kidney Dis 25:433-439, 1995 14. Fishbane S, Ungureanu V, Maesaka J, Kaupke CJ, Lim V, Wish J: The safety of intravenous iron dextran in hemodialysis patients. Am J Kidney Dis 28:529-534, 1996 15. Paull J: A prospective study of dextran-induced anaphylactoid reactions in 5745 patients. Anaesth Intensive Care 15:163-167, 1987 16. Hamstra RD, Block MH, Schocket AL: Intravenous iron dextran in clinical practice. JAMA 243:1726-1731, 1980 17. Navarro JF, Teruel JL, Liano F, Marcep R, Ortuno J: Effectiveness of intravenous administration of Fe-Gluconate-Na complex to maintain adequate body iron stores in hemodialysis patients. Am J Nephrol 16:268-272, 1996 18. Taylor JE, Peat N, Porter C, Morgan AG: Regular low-dose intravenous iron therapy improves response to erythropoietin in hemodialysis patients. Nephrol Dial Transplant 11:1079-1083, 1996 19. Pascual J, Teruel JL, Liano F, Sureda A, Ortuno J: Intravenous Fe-Gluconate-Na for iron-deficient patients on hemodialysis. Nephron 60:121, 1992 (letter) 20. Pascual J, Teruel JL, Liano E, Sureda A, Ortuno J: Sodium ferric gluconate complex given intravenously for iron deficiency in hemodialysis. Clin Nephrol 35:87, 1991 (letter) 21. Allegra V, Mengozzi G, Vasile A: Iron deficiency in maintenance hemodialysis patients: assessment of diagnosis criteria and of three different iron treatments. Nephron 57:175-182, 1991 22. Amerling R, Cu GA, Dubrow A, Levin NW, Osheroff RJ: Complications during hemodialysis, in Nissenson AR, Fine RN, Gentile DE (eds): Clinical Dialysis. Norwalk, CT, Appleton & Lange, pp 235-267 23. Nissenson AR, Swan S, Lambrecht LL, Anderson P, Schweitzer S: Ferric gluconate (FerrlecitÆ) is safe in hemo-
482
dialysis patients who react to iron dextran. J Am Soc Nephrol 7:1460A, 1996 (abstr) 24. Faich G, Strobos J: Sodium ferric gluconate complex in sucrose: Safer than iron dextran. J Am Soc Nephrol 8:192A, 1997 (abstr) 25. Bridges KR, Seligman PA: Disorders of iron metabolism, in Handin RI, Lux SE, Stossel TP (eds): Blood: Principles and Practice of Hematology. Philadelphia, PA, Lippincott, 1995 26. Brittenham GM: Disorders of iron metabolism, in Hoffman R, Benz EJ, Shattil SJ, Furie B, Cohen HJ, Silberstein LE (eds): Hematology: Basic Principles and Practice. New York, NY, Churchill Livingstone, 1995 27. Nissenson AR, Strobos J: Iron deficiency in patients with renal failure. Kidney Int 55:518-521, 1999 (Suppl 69) 28. Stivelman JV: Optimization of iron therapy in hemodialysis patients treated with rHuEPO. Semin Dial 6:59-69, 1993 29. Lindsay RM, Burton JA, Edward N, Dargle JH, Prentice CRM, Kennedy AC: Dialyzer blood loss. Clin Nephrol 1:29-34, 1973 30. Longnecker RE, Goffinet JA, Hendler ED: Blood loss form maintenance hemodialysis. Trans Am Soc Artif Internal Organs 20:135-141, 1974 31. Muller-Wiefel DE, Sinn H, Gilli G, Scharer K: Hemo-
NISSENSON ET AL
lysis and blood loss in children with chronic renal failure. Clin Nephrol 8:481-486, 1977 32. Eschbach J, Cook JD, Scribner BH, Finch CA: Iron balance in hemodialysis patients. Ann Intern Med 87:710713, 1977 33. Kooistra MP, van Es A, Struyvenberg A, Marx JJM: Iron metabolism in patients with anaemia of end-stage renal disease during treatment with recombinant human erythropoietin. Br J Haematol 79:634-639, 1991 34. Dunea G, Swagel MA, Bodiwala U, Arruda JAL: Intradialytic oral iron therapy. Int J Artif Organs 17:261-264, 1994 35. Anastassiades EG, Howarth D, Howarth J, Shanks D, Waters HM, Hyde K, Geary CG, Liu Yin JA, Gokal R: Monitoring iron requirements in renal patients on erythropoietin. Nephrol Dial Transplant 8:846-853, 1993 36. Kotaki M, Uday K, Henriquez M, Blum S, Dave M: Maintenance therapy with intravenous iron in hemodialysis patients receiving erythropoietin. Clin Nephrol 48:63-64, 1997 37. Linder MC, Schaffer KJ, Hazegh-Azam M, Zhou CY, Tran TN, Nagel GM: Serum ferritin: Does it differ from tissue ferritin? J Gastroenterol Hepatol 11:1033-1036, 1996 38. Rogers JT: Ferritin translation by interleukin-6: The role of sequences upstream of the start codons of the heavy and light subunit genes. Blood 87:2525-2537, 1996