• Open Access

The Effects of Deferoxamine Mesylate on Iron Elimination after Blood Transfusion in Neonatal Foals

Authors


  • Dr Javsicas is presently affiliated with Upstate Equine Medical Center, Schuylerville, NY. This study was completed at the University of Florida College of Veterinary Medicine. A portion of this work was presented at the 2010 ACVIM Forum.

Corresponding author: Steeve Giguère, Department of Large Animal Medicine, College of Veterinary Medicine, University of Georgia, 501 DW Brooks Drive, Athens, GA 32602; email: gigueres@uga.edu.

Abstract

Background: Hepatic failure is one of the more common complications in foals requiring blood transfusion to treat neonatal isoerythrolysis. Iron intoxication is likely the cause of hepatic injury.

Objectives: To determine the effects of deferoxamine on iron elimination in normal foals.

Animals: Thirteen neonatal foals.

Methods: Randomized-controlled trial. At 1–3 days of age, foals received either 3 L of washed packed dam's red blood cells (RBC) or 3 L of saline IV once. Foals were treated with deferoxamine (1 g) or saline (5 mL) SC twice daily for 14 days. Foals were randomly assigned to 1 of 3 groups: RBC/deferoxamine (deferoxamine), RBC/saline (placebo), or saline/saline (control). Blood and urine samples and liver biopsy specimens were collected for measurement of hematological, biochemical, and iron metabolism variables.

Results: There was a significant (P < .05) increase in hematocrit, RBC count, and hemoglobin in the groups transfused with packed RBC as compared with controls at all times. Biochemical variables and liver biopsy scores were not significantly different between groups at any time. Urine iron concentrations and fractional excretion of iron were significantly higher in deferoxamine treated foals. By 14 days after transfusion, liver iron concentrations in foals treated with deferoxamine (79.9 ± 30.9 ppm) were significantly lower than that of foals receiving placebo (145 ± 53.0 ppm) and similar to that of controls (44.8 ± 4.09 ppm).

Conclusions and Clinical Importance: Deferoxamine enhances urinary iron elimination and decreases hepatic iron accumulation after blood transfusion in foals.

Abbreviations:
NI

neonatal isoerythrolysis

RBC

red blood cells

WBC

white blood cells

Neonatal isoerythrolysis (NI) is the leading cause of hemolytic anemia and icterus in neonatal foals. Most foals diagnosed with NI require one or more blood transfusions. There is an approximately 25% case fatality rate for foals treated for NI, with the most common causes of death being hepatic failure, kernicterus, and complications resulting from sepsis.1 There is a strong association between the volume of blood administered and the likelihood of development of liver failure in foals with NI. Foals treated with ≥4 L of blood or blood products were 19.5 times more likely to develop hepatic failure than those treated with a lower volume.1 Hepatic histologic lesions of foals with NI that develop liver failure1 are identical to those after oral administration of ferrous fumarate to newborn foals.2–5 The highly significant association between volume of blood or blood products administered and the development of liver failure as well as the nature of the histologic lesions leads to the hypothesis that iron overload from transfused red blood cells (RBC) might either be the cause or at least contribute to the development of liver disease in foals with NI.

The dose of iron required to induce toxicity is high in most species, with the LD50 of various iron supplements ranging between 300 and 900 mg/kg of body weight in rabbits, guinea pigs, and mice.6 Parenteral administration of iron to pigs at a dose approximately 300 mg/kg of body weight significantly increases hepatic iron content but does not result in liver failure.7 The dose of oral iron that causes intoxication in neonatal foals is strikingly low when compared with dosages reported in other species. Oral administration of iron fumarate to newborn foals (16.5 mg of iron/kg body weight) results in death from liver failure in 5 of 8 foals and severe liver lesions in all 8 foals.5 Assuming normal hemoglobin concentration in the donor, each liter of blood transfused delivers approximately 400–500 mg of iron, and there is no active mechanism for excretion of this excess iron.8,9

Hemosiderosis is an inevitable complication of repeated transfusion therapy in humans with chronic RBC disorders.8 The most important complications of transfusional siderosis in humans are cardiac and hepatic disease. Deferoxamine mesylate is a drug that chelates iron by forming a stable complex that prevents the iron from entering into further chemical reactions.10 The introduction of deferoxamine for the treatment of transfusional iron overload has had a major positive impact on the survival and well-being of patients needing repeated blood transfusions.10 The objectives of this study were to determine if large volume of packed RBC would cause hepatic injury in newborn foals and to determine if deferoxamine can increase urinary excretion of iron and prevent hepatic iron overload secondary to blood transfusion. Our hypotheses were that deferoxamine would increase urinary iron elimination and decrease hepatic iron deposition after blood transfusion.

Materials and Methods

Animals and Experimental Design

Thirteen Thoroughbred or Thoroughbred cross foals (7 fillies, 6 colts) were used beginning at 1–3 days of age. Inclusion criteria were a normal physical examination and an IgG concentrationa >800 mg/dL by 24 hours of age. Ten mares were used as RBC donors for their respective foals. Mares and foals were housed in a paddock with ad libitum access to grass hay and water, and commercial equine sweet feed containing 12% crude protein twice daily. Each mare and foal pair was brought into a stall for sample collection and returned to the paddock upon completion. Foals received either 3 L of packed washed maternal RBC or 3 L of 0.9% saline IV once. Foals were treated with deferoxamineb (1 g, 5 mL) or 0.9% saline (5 mL) SC twice daily for 14 days beginning immediately before transfusion. Foals were randomly assigned to 1 of 3 groups: RBC/deferoxamine (deferoxamine; n = 5), RBC/saline (placebo; n = 5), or saline/saline (control; n = 3). All procedures were approved by the University of Florida Institutional Animal Care and Use Committee.

RBC Collection from Mares and Administration of RBC or Saline to Foals

Hemoglobin and hematocrit were measured from the mares of foals assigned to receive washed packed maternal RBC. The skin overlying 1 jugular vein was clipped and aseptically prepared. A 14 G needle was aseptically inserted into the jugular vein and 6 L of whole blood was collected in a closed system.c Plasma was removed and the RBC were washed twice in a closed system with a commercial RBC washer.d The packed RBC were resuspended to a total volume of 3 L in 0.9% saline. The skin over one jugular vein of each foal was clipped and aseptically prepared. After injection of 0.5–1.0 mL of 2% lidocaine,e a 16 G, 6 cm Teflon catheter was aseptically placed and secured into the jugular vein. Ten milliliters of blood was withdrawn from the catheter for baseline hematological, biochemical, and iron metabolism variables. Urine was collected either by a free catch sample or by aseptically passing a soft Foley urinary catheter under IV sedation with xylazinef (0.5 mg/kg) and butorphanolg (0.07 mg/kg). For foals assigned to the packed maternal RBC group, the transfusion was started at 50 mL/h for 10 minutes followed by infusion at 2 L/h while the foal was closely monitored for signs of transfusion reaction. Control foals received 3 L of 0.9% saline administered at 2 L/h.

Sample Collection and Monitoring

A complete physical examination including recording of temperature, heart rate, and respiratory rate was performed twice a day for 14 days by an investigator unaware of treatment group. Blood samples were collected by jugular venipuncture for hematological evaluation on the day after transfusion (day 1), day 7, and day 14. Blood and urine were collected for biochemical and iron metabolism variables on days 4, 7, and 14 of the study. Body weight was recorded on days 0, 4, 7, and 14. On days 7 and 14, foals were sedated with xylazine (1.0 mg/kg) and butorphanol (0.07 mg/kg) IV, and anesthetized with ketamineh (2.2 mg/kg) IV for collection of liver biopsy specimens. Foals were placed in left lateral recumbency and the appropriate site for biopsy specimens was selected by transabdominal ultrasonography. The skin overlying the site was clipped and aseptically prepared. Liver biopsies were performed under ultrasonographic guidance with a 14 G spring activated biopsy instrument.i Four samples were collected at each time point; 2 samples were fixed in 10% buffered formalin at room temperature for subsequent histopathologic evaluation and 2 samples were frozen in containers without additive, sealed with parafilm, and stored at −70°C until micromineral analysis.

Hematological variables measured included white blood cell (WBC) count, hematocrit, RBC count, mean cell volume, mean corpuscular hemoglobin, mean corpuscular hemoglobin concentration, platelet count, and fibrinogen concentrations. Biochemical variables measured in plasma included sodium, potassium, chloride, total CO2, anion gap, total calcium, phosphorus, total magnesium, blood urea nitrogen, creatinine, glucose, total protein, albumin, globulin, total bilirubin, triglycerides, alkaline phosphatase, γ-glutamyl transferase, aspartate aminotransferase, creatine kinase, sorbitol dehydrogenase, bile acids, and ammonia. Plasma, serum, and urine samples were submitted to a commercial laboratoryj for measurement of total iron binding capacity, ferritin, as well as serum and urine iron. The fractional urinary excretion of iron was calculated using the following equation: [(UFe×PCr)/(UCr×SFe) × 100], where UFe, PCr, UCr, and SFe represent urine iron, plasma creatinine, urine creatinine, and serum iron concentrations, respectively.

Hepatic Micromineral Concentrations and Histopathology

Hepatic iron, aluminum, cobalt, copper, lead, selenium, and zinc concentrations were measured by a commercial laboratoryk by inductively coupled plasma mass spectroscopy, as described previously.11 After fixation, liver samples were processed routinely and embedded in paraffin. Serial sections, 5 μm thick, were obtained and stained with hematoxylin and eosin, Masson's trichrome, and Prussian blue stains. Liver biopsy specimens were examined by a board-certified pathologist (L.L.F.) unaware of treatment groups for the presence or absence of necrosis, portal fibrosis, iron deposition, bile duct hyperplasia, and arteriolar hyperplasia. The number of mitoses per ten 400 × fields was also determined.11 The lesions were scored as reported previously.11

Estimation of Transfused Maternal RBC Half-life

For each foal transfused with packed maternal RBC, estimation of transfused maternal RBC count at each time point was calculated as measured RBC count minus RBC count before transfusion. For each foal, the slope (λ) of the terminal phase of the transfused maternal RBC elimination was determined by linear regression of the logarithmic transfused maternal RBC count versus time curve using 3 data points. Half-life of transfused maternal RBC was calculated as 0.693/λ.

Statistical Analysis

Data were analyzed for normality and equality of variance by the Kolmogorov-Smirnov and Levene's tests, respectively. Variables that did not meet the assumptions of the ANOVA were log- or rank-transformed before analysis. A 2-way repeated measures ANOVA was used to determine the effects of group (deferoxamine, placebo, control), time, and the interaction between treatment and time on each measured variable. When appropriate, multiple pairwise comparisons were done by the Holm-Sidak test. Half-life of transfused maternal RBC in foals treated with deferoxamine was compared with that of foals in the placebo group by the Mann-Whitney U-test. For all analyses, significance was set at P < .05.

Results

The median hematocrit of the mares immediately before blood collection was 35% (32–40%) and the median hemoglobin was 14.0 mg/dL (12.8–17.6 mg/dL). Transfusion of washed maternal RBC did not result in clinically apparent adverse reactions. All foals remained clinically healthy throughout the study. Foals in all 3 groups had mild edema at the site of SC injections. Baseline hematological, biochemical, and iron metabolism variables were not significantly different between groups. Hematocrit, RBC count, and hemoglobin concentrations were significantly higher in the deferoxamine and placebo groups compared with controls at all times after transfusion (Table 1). Platelet count was significantly lower in deferoxamine and placebo groups on day 7 and in the placebo group on day 14 compared with control foals (Table 1). The median calculated half-life of transfused maternal RBC was 14.6 (range = 9.1–29.1) days. The median half-life of transfused maternal RBC in foals treated with deferoxamine (15.0 days) was not significantly different from that of foals in the placebo group (14.0 days).

Table 1.   Hematologic variables in foals before (day 0) and after (days 1, 7, and 14) transfusion with packed maternal RBC or administration of saline (controls; n = 3).
VariableGroupDay 0Day 1Day 7Day 14
  • Transfused foals were treated with deferoxamine (n = 5) or with a placebo (n = 5). Mean ± SD.

  • RBC, red blood cells; WBC, white blood cells.

  • *

    Statistically significant difference (P < .05) compared with controls and compared with pretransfusion values (day 0).

  • #

    Statistically significant difference (P < .05) compared with controls.

Hematocrit (%)Control35.6 ± 3.132.6 ± 6.531.6 ± 2.330.7 ± 2.6
Placebo35.9 ± 4.255.6 ± 5.2*50.8 ± 5.1*44.9 ± 7.0*
Deferoxamine35.3 ± 3.856.8 ± 6.0*50.3 ± 5.9*46.9 ± 4.5*
RBC (× 106/μL)Control9.53 ± 0.598.79 ± 1.528.57 ± 0.478.66 ± 0.61
Placebo9.20 ± 1.0713.20 ± 1.07*12.20 ± 1.08*11.01 ± 1.72*
Deferoxamine9.65 ± 1.1914.16 ± 1.64*12.89 ± 1.48*12.31 ± 1.41*
Hemoglobin (mg/dL)Control13.47 ± 1.0012.50 ± 2.2111.97 ± 0.6711.70 ± 0.75
Placebo13.40 ± 1.6420.92 ± 1.80*19.34 ± 1.76*17.04 ± 2.44*
Deferoxamine13.22 ± 1.3621.26 ± 1.95*19.24 ± 1.95*17.92 ± 1.64*
Total solids (g/dL)Control7.1 ± 0.67.0 ± 0.56.7 ± 0.46.5 ± 0.3
Placebo6.6 ± 0.57.0 ± 0.26.7 ± 0.36.5 ± 0.2
Deferoxamine6.1 ± 0.67.3 ± 0.46.8 ± 0.46.7 ± 0.4
WBC (× 103/μL)Control8.25 ± 4.06.89 ± 1.411.78 ± 1.29.73 ± 1.1
Placebo6.76 ± 2.96.08 ± 2.413.35 ± 4.59.44 ± 1.7
Deferoxamine8.26 ± 3.17.202 ± 2.112.39 ± 4.212.71 ± 4.8
Platelets (× 103/μL)Control231 ± 45143 ± 128297 ± 33343 ± 29
Placebo239 ± 5392 ± 31145 ± 33#246 ± 61#
Deferoxamine267 ± 41117 ± 38146 ± 52#267 ± 39

Mean body weight (±SD) was significantly lower in the placebo group (67.2 ± 3.0 kg) compared with both the deferoxamine (71.8 ± 8.2 kg) and control (80.3 ± 12.1 kg) groups on day 14 only. Physical examination parameters, WBC count, fibrinogen concentrations, total protein, all plasma biochemical variables (Table 2), serum iron, ferritin, and total iron binding capacity (Table 3) were not significantly different between groups.

Table 2.   Plasma biochemical variables in foals before (day 0) and after (days 4, 7, and 14) transfusion with packed maternal RBC or administration of saline (controls; n = 3).
VariableGroupDay 0Day 4Day 7Day 14
  1. Transfused foals were treated with deferoxamine (n = 5) or with a placebo (n = 5). Mean ± SD.

  2. ALP, alkaline phosphatase; AST, aspartate aminotransferase; BUN, Blood urea nitrogen; GGT, γ-glutamyl transferase; SDH, sorbitol dehydrogenase; CK, creatine kinase; RBC, red blood cells.

ALP (U/L)Control1421 ± 270833 ± 44892 ± 196745 ± 60
 Placebo1111 ± 301683 ± 79528 ± 1651041 ± 651
 Deferoxamine1590 ± 599972 ± 411720 ± 188587 ± 52
AST (U/L)Control150 ± 15195 ± 74220 ± 79265 ± 75
 Placebo156 ± 24147 ± 11147 ± 19186 ± 25
 Deferoxamine137 ± 9173 ± 28247 ± 136211 ± 45
Total bilirubin (mg/dL)Control3.9 ± 1.03.1 ± 1.02.8 ± 0.52.2 ± 0.5
 Placebo3.8 ± 0.73.8 ± 0.53.4 ± 0.83.5 ± 1.1
 Deferoxamine3.9 ± 1.05.1 ± 1.64.7 ± 1.53.1 ± 0.6
Total protein (g/dL)Control6.6 ± 0.86.6 ± 0.56.5 ± 0.56.2 ± 0.2
 Placebo5.8 ± 0.86.1 ± 1.05.9 ± 0.86.0 ± 0.7
 Deferoxamine5.6 ± 1.06.3 ± 0.36.1 ± 0.35.8 ± 0.4
Albumin (g/dL)Control3.0 ± 0.33.0 ± 0.33.0 ± 0.13.0 ± 0.1
 Placebo2.9 ± 0.42.9 ± 0.32.8 ± 0.13.1 ± 0.2
 Deferoxamine3.0 ± 0.23.2 ± 0.33.2 ± 0.13.0 ± 0.2
Globulin (g/dL)Control3.6 ± 0.53.6 ± 0.43.5 ± 0.43.2 ± 0.2
 Placebo3.0 ± 0.63.2 ± 0.73.1 ± 0.72.9 ± 0.7
 Deferoxamine2.7 ± 0.93.1 ± 0.33.0 ± 0.32.8 ± 0.3
Calcium (mg/dL)Control10.7 ± 0.411.8 ± 1.012.0 ± 0.512.3 ± 0.7
 Placebo10.9 ± 1.111.5 ± 1.011.4 ± 1.112.3 ± 0.4
 Deferoxamine11.1 ± 0.712.0 ± 0.712.1 ± 0.412.0 ± 1.0
Phosphorus (mg/dL)Control5.4 ± 0.76.0 ± 0.86.8 ± 0.77.4 ± 0.4
 Placebo5.4 ± 0.66.0 ± 0.46.9 ± 0.56.8 ± 1.5
 Deferoxamine5.3 ± 0.66.3 ± 1.36.9 ± 0.86.9 ± 1.2
Creatinine (mg/dL)Control1.1 ± 0.10.8 ± 0.10.9 ± 0.11.2 ± 0.2
 Placebo1.2 ± 0.40.8 ± 0.10.8 ± 0.11.3 ± 0.5
 Deferoxamine1.2 ± 0.30.8 ± 0.10.8 ± 0.11.1 ± 0.2
BUN (mg/dL)Control20 ± 16 ± 06 ± 25 ± 2
 Placebo19 ± 79 ± 29 ± 114 ± 10
 Deferoxamine14 ± 112 ± 210 ± 410 ± 4
Glucose (mg/dL)Control178 ± 19153 ± 13161 ± 10150 ± 15
 Placebo159 ± 16135 ± 26138 ± 8145 ± 34
 Deferoxamine165 ± 24153 ± 36154 ± 9136 ± 16
Triglycerides (mg/dL)Control80 ± 58120 ± 32113 ± 9687 ± 50
 Placebo46 ± 2390 ± 4069 ± 4460 ± 36
 Deferoxamine82 ± 7596 ± 2256 ± 1562 ± 19
Sodium (mEq/L)Control139 ± 2142 ± 1144 ± 4145 ± 4
 Placebo139 ± 6142 ± 4141 ± 2145 ± 8
 Deferoxamine136 ± 8140 ± 2141 ± 3139 ± 12
Potassium (mEq/L)Control3.7 ± 0.24.4 ± 0.44.0 ± 0.34.3 ± 0.3
 Placebo3.7 ± 0.34.1 ± 0.23.9 ± 0.33.9 ± 0.3
 Deferoxamine3.6 ± 0.64.1 ± 0.53.7 ± 0.53.8 ± 0.9
Chloride (mEq/L)Control103 ± 2106 ± 3105 ± 3102 ± 2
 Placebo103 ± 4105 ± 2103 ± 1105 ± 5
 Deferoxamine103 ± 6105 ± 3104 ± 3102 ± 9
Magnesium (mg/dL)Control2.5 ± 0.42.1 ± 0.42.3 ± 0.22.1 ± 0.1
 Placebo2.3 ± 0.42.1 ± 0.32.2 ± 0.22.4 ± 0.3
 Deferoxamine2.4 ± 0.32.1 ± 0.12.2 ± 0.22.1 ± 0.2
GGT (U/L)Control23 ± 745 ± 3955 ± 2471 ± 24
 Placebo22 ± 1123 ± 1225 ± 1126 ± 8
 Deferoxamine18 ± 519 ± 633 ± 534 ± 5
SDH (U/L)Control4.9 ± 1.13.8 ± 3.24.4 ± 2.44.3 ± 2.2
 Placebo4.0 ± 2.33.9 ± 1.44.0 ± 2.43.9 ± 1.8
 Deferoxamine4.1 ± 1.75.0 ± 2.18.1 ± 9.24.1 ± 1.1
Bile acid (μmol/L)Control12.9 ± 4.014.7 ± 2.815.5 ± 4.89.8 ± 1.5
 Placebo14.7 ± 6.110.1 ± 3.27.9 ± 2.213.7 ± 12.4
 Deferoxamine19.6 ± 21.57.8 ± 2.48.1 ± 2.77.6 ± 2.2
CK (U/L)Control333 ± 162182 ± 27197 ± 55254 ± 68
 Placebo238 ± 75146 ± 43152 ± 48318 ± 174
 Deferoxamine529 ± 618268 ± 103247 ± 47283 ± 233
Total CO2 (mg/dL)Control28 ± 527 ± 429 ± 128 ± 1
 Placebo25 ± 126 ± 127 ± 228 ± 2
 Deferoxamine25 ± 325 ± 628 ± 329 ± 4
Table 3.   Serum iron metabolism variables and hepatic micromineral concentrations in foals before (day 0) and after (days 4, 7, and 14) transfusion with packed maternal RBC or administration of saline (controls; n = 3).
VariableGroupDay 0Day 4Day 7Day 14
  1. Transfused foals were treated with deferoxamine (n = 5) or with a placebo (n = 5). Mean ± SD.

  2. TIBC, total iron binding capacity; ND, not determined; RBC, red blood cells.

Iron (μg/dL)Control197.3 ± 98.2171.0 ± 73.1130.0 ± 40.452.7 ± 17.7
 Placebo245.6 ± 57.7214.6 ± 45.5209.6 ± 58.4177.4 ± 76.4
 Deferoxamine175.4 ± 118.5234 ± 90.3165.6 ± 79.7258.2 ± 66.4
Ferritin (ng/mL)Control213 ± 119166 ± 64140 ± 7756 ± 13
 Placebo204 ± 103214 ± 81260 ± 60249 ± 57
 Deferoxamine175 ± 50181 ± 31264 ± 160176 ± 26
TIBC (μg/dL)Control330 ± 35339 ± 42340 ± 14445 ± 50
 Placebo341 ± 42324 ± 55323 ± 36361 ± 67
 Deferoxamine311 ± 84347 ± 24340 ± 49429 ± 78
AluminumControlNDND0.98 ± 1.20.91 ± 1.2
(ppm)PlaceboNDND0.88 ± 1.11.1 ± 1.4
 DeferoxamineNDND0.97 ± 1.20.94 ± 1.2
CobaltControlNDND0.028 ± 0.0170.027 ± 0.017
(ppm)PlaceboNDND0.032 ± 0.0210.033 ± 0.023
 DeferoxamineNDND0.036 ± 0.0260.033 ± 0.025
CopperControlNDND93.9 ± 32.486.3 ± 32.5
(ppm)PlaceboNDND87.1 ± 33.085.5 ± 34.1
 DeferoxamineNDND86.3 ± 30.286.4 ± 30.2
LeadControlNDND0.097 ± 0.0800.085 ± 0.071
(ppm)PlaceboNDND0.095 ± 0.0740.094 ± 0.075
 DeferoxamineNDND0.086 ± 0.0670.086 ± 0.067
ZincControlNDND86.9 ± 34.994.4 ± 39.4
(ppm)PlaceboNDND90.3 ± 36.495.1 ± 39.1
 DeferoxamineNDND94.9 ± 36.1395.5 ± 35.8

Baseline urine iron and fractional urinary excretion of iron were not significantly different between groups. Urine iron concentrations were significantly higher in foals treated with deferoxamine when compared with foals receiving a placebo on day 4 and compared with foals receiving a placebo and control foals on day 14 (Fig 1A). Urine iron concentrations in foals treated with deferoxamine were significantly higher on days 4 and 14 compared with day 0. Urine iron concentrations did not change significantly with time in the placebo and control groups. Fractional urinary excretion of iron was significantly higher in foals treated with deferoxamine at all times after transfusion compared with foals in the placebo and control groups (Fig 1B). Fractional excretion of iron was significantly higher in foals treated with deferoxamine on day 7 than on day 0 and in controls on day 14 compared with day 4.

Figure 1.

 Urine iron concentrations (A) or fractional excretion of iron (B) in foals before (day 0) and after (days 4, 7, and 14) transfusion with packed maternal red blood cells or administration of saline (n = 3). Transfused foals were treated with deferoxamine (n = 5) or with a placebo (n = 5). *Significant difference from all other groups, ^significant difference from placebo, #significant difference from day 0, +significant difference from day 4. Mean ± SD.

Histopathological examination of liver biopsy specimens from 1 foal in the deferoxamine group revealed mild hepatocellular necrosis and neutrophilic inflammation on day 7. Necrosis and inflammation were no longer apparent by day 14, but there was mild fibrosis and biliary hyperplasia. Histopathological examination of all other liver specimens did not reveal important abnormalities and liver biopsy scores were not significantly different between groups. Liver iron concentrations were significantly higher in foals in the placebo group compared with foals treated with deferoxamine and control foals on day 14 (Fig 2). There was a significant decrease in liver iron from days 7 to 14 in foals treated with deferoxamine (Fig 2). There were no significant differences between groups on day 7. There was a significant effect of treatment group on liver selenium concentrations (P= .011). Mean selenium concentration in foals in the placebo group (0.575 ± 0.13 ppm) was significantly higher than that of foals treated with deferoxamine (0.477 ± 0.06 ppm) and control foals (0.435 ± 0.03 ppm). Liver aluminum, cobalt, copper, lead, and zinc concentrations were not significantly different between groups (Table 3).

Figure 2.

 Liver iron concentrations in foals after transfusion with packed maternal red blood cells or administration of saline (n = 3). Transfused foals were treated with deferoxamine (n = 5) or with a placebo (n = 5). *Significant difference from other groups, #significant difference from day 7. Individual data points and median.

Discussion

The present study demonstrates that deferoxamine at a dose of approximately 20 mg/kg SC twice daily increases urinary iron elimination and decreases hepatic iron accumulation after administration of a blood transfusion to normal neonatal foals. Deferoxamine is approved for the treatment of acute iron intoxication and of chronic iron overload because of transfusion-dependent anemias in humans.12 Because of the discomfort resulting from chronic IV, IM, or SC deferoxamine therapy, there has been a need for the development of new PO administered iron chelators. Deferiprone, a PO administered iron chelator, has been used extensively as a substitute for deferoxamine in Europe and Asia. Recently, deferasirox has been approved for oral use in the United States. However, studies have shown that these drugs are less effective than deferoxamine and their potential for hepatotoxicity is an issue of current controversy.10,13,14 In addition, many drugs with adequate oral bioavailability in humans are poorly absorbed in horses. Therefore, deferoxamine was chosen as the optimal iron chelator for use in foals in the present study. The dosage selected was extrapolated from dosages used in children. Because we evaluated a single dosage, it is possible that a lower dosage would be sufficient to enhance urinary iron excretion of iron or that a higher dosage would be more effective. Continuous SC infusion of deferoxamine has long been considered the optimal method of treatment in humans.10 However, recent studies in both adults and children indicate that twice daily SC bolus administration is as effective as continuous infusion.15,16 In contrast, the same total dose administered as a single SC bolus is not as effective.17 Because continuous SC infusion would not be practical for use in foals in most clinical settings, we elected to administer deferoxamine as twice daily SC boluses. Additional studies will be required to determine the optimal dosage and route of administration for deferoxamine in foals.

Deferoxamine readily chelates iron from ferritin and hemosiderin but not readily from transferrin; it does not combine with the iron from cytochromes and hemoglobin.12 In humans with RBC disorders receiving regular blood transfusions, administration of deferoxamine results in urinary excretion of approximately 5–10 mg of iron per day.15,16 Assuming a normal mean urine production of 148 mL/kg/d,18 mean daily urinary iron excretion in the foals of the present study 4 days after initiation of therapy with deferoxamine was approximately 5.9 mg/d. The effect of deferoxamine treatment, as measured by urinary iron excretion, is directly proportional to the severity of iron overload.10 Hence, treatment in foals with more pronounced and clinically significant iron overload would be expected to result in even greater urinary iron excretion than reported in the present study. In the present study, hepatic selenium concentrations in foals in the placebo group were significantly higher than that of foals treated with deferoxamine and control foals. This would suggest a significant effect of deferoxamine on elimination of selenium. A significant decrease in plasma concentrations of selenium has also been reported in human patients treated with deferoxamine.19 The decrease in platelet count observed on day 1 in all 3 groups and on day 7 in foals of the placebo and deferoxamine groups is likely the result of short-term hemodilution from saline and persisting hemodilution resulting from administration of packed RBC.

Although there is a strong association between administration of ≥4 L of blood and development of liver failure in neonatal foals with NI,1 administration of packed RBC (equivalent to 6 L of whole blood) did not induce liver disease in healthy neonatal foals in the present study. This finding in not unexpected and does not rule out a role for iron intoxication as a cause of liver failure in foals with NI. Foals with NI receive blood transfusions because they have already lysed a high proportion of their own RBC, releasing iron that cannot be eliminated by physiologic means. Moreover, foals with NI that develop liver failure typically require multiple blood transfusions (median of 3) over a period of a few days, indicating that the life span of transfused RBC in these foals is remarkably short.1 Rapid lysis of transfused RBC further contributes to iron accumulation. Iron fumarate administered PO (16.5 mg/kg of body weight) is sufficient to induce hepatic failure in normal neonatal foals. In the present study, each foal received approximately 2,400 mg of iron (approximately 48 mg/kg of body weight) within transfused RBC. However, the long half-life of transfused RBC in the present study resulted in slow release of a small proportion of the iron contained within hemoglobin over a prolonged period of time. In addition, differences in formulation of iron (iron fumarate versus iron from hemoglobin) might affect the degree of iron accumulation and toxicity. Hence, hepatic iron accumulation was minimal in the present study. Although hepatic iron concentrations in transfused foals assigned to the placebo group (153.7 ± 48.3 ppm) were significantly higher than those of nontransfused controls (64.4 ± 23.2 ppm), these concentrations were considerably below liver iron concentrations in foals with liver failure from experimental administration of iron fumarate (335–603 ppm).5

The half-life of transfused maternal RBC estimated in the present study (median = 14.6 days) was considerably longer than the median half-life of 5.1 days reported previously in foals of the same age transfused with washed maternal RBC labeled with 50Cr.20 It has been suggested that equine erythrocytes are more susceptible to the toxic effects of chromate treatment than erythrocytes from other species.20 Trivalent chromate inhibits gluthathione reductase and enhances Heinz-body formation in human erythrocytes.21 Equine erythrocytes have a decreased ability to reduce glutathione from its oxidized form compared with other species.22 In one study, the survival of autologously transfused 50Cr-labeled erythrocytes was considerably less than the expected life span of equine erythrocytes as measured with a precursor method, indicating that 50Cr-labeling considerably underestimates the life span of erythrocytes in the horse.20 Maternal RBC were not labeled before administration to foals in the present study. As a result, the half-life was estimated from the progressive decrease in the difference between post- and pretransfusion RBC counts. Hydration status, excitement, and stress may all alter RBC concentrations. Given that the foals were kept in the same environment and had ad libitum access to nursing and water, it is unlikely that dehydration and stress had a profound impact on RBC concentrations in the present study. Additional studies by sensitive and safe methods of RBC labeling will be required to determine the exact half-life of transfused RBC in neonatal foals.

Although transfused foals in this study did not develop hepatic failure, foals in the placebo group had decreased weight gain at 14 days compared with foals in the deferoxamine and control groups. This indicates a possible subclinical effect of transfusion on health status of the foal. As there were no significant differences in liver enzymes or bile acids between groups, it is possible that the effects on body weight were related to the effects of polycythemia. Clinical signs of erythrocytosis in horses include lethargy, weight loss, hyperemic mucous membranes, and exercise intolerance.23 However, the significantly higher weight gain in the deferoxamine group despite the same degree of polycythemia suggests that the decreased weight gain is not the result of polycythemia alone and is likely related to iron accumulation. Patients with iron overload diseases frequently report clinical signs of weakness and fatigue.8 In addition, administration of iron to mice results in exercise intolerance, muscle atrophy, and decreased body weight.24 In the present study, no overt clinical signs or changes in muscle mass were detected in transfused foals but suckling behavior may have been diminished leading to the decrease in weight gain.

The results of this study suggest that deferoxamine increases urinary iron elimination after blood transfusion in healthy neonatal foals. Additional studies are needed to evaluate the effects of deferoxamine on iron elimination in foals with NI requiring multiple transfusions.

Footnotes

aDVM Stat, Corporation for Advanced Applications, Newburg, WI

bDeferoxamine mesylate, Hospira Inc, Lake Forest, IL

cCaridian BCT Inc, Lakewood, CO

d2991 Cell Processor Model 1, COBE BCT Inc, Lakewood, CO

eSparhawk Laboratories Inc, Lenexa, KS

fRompun, Bayer Co, Shawnee Mission, KS

gTorbugesic, Fort Dodge Animal Health, Fort Dodge, IA

hVetaKet, Phoenix Scientific Inc, St Joseph, MO

iCR Bard Inc, Covington, GA

jKansas State University Veterinary Diagnostic Laboratory, Manhattan, KS

kUtah Veterinary Diagnostic Laboratory, Logan, UT

Acknowledgement

The study was supported by funds from the Florida Thoroughbred Breeders' & Owners' Association and the University of Florida 2009 Spring Faculty Competition.

Ancillary