Iron-mobilizing properties of the gadolinium-DTPA complex: clinical and experimental observations.
ABSTRACT Gadolinium (Gd) magnetic resonance imaging (MRI) contrast agents are considered to be safe in patients with impaired renal function. Our study investigates a mechanism of severe iron intoxication with life-threatening serum iron levels in a haemodialysis patient following MRI with Gd-diethylenetriaminepentaacetic acid (Gd-DTPA) administration. His previous history was remarkable for multiple blood transfusions and biochemical evidence of iron overload. We hypothesized that Gd-DTPA may have an iron-mobilizing effect in specific conditions of iron overload combined with prolonged exposure to the agent.
For the in vitro study, Gd-DTPA was added to mice liver homogenate and iron metabolism parameters were measured after incubation in comparison with the same samples incubated with saline only. For the in vivo study, an experimental model of acute renal failure in iron-overloaded rats was designed. Previously iron-overloaded and normally fed rats underwent bilateral nephrectomy by renal pedicle ligation, followed by Gd-DTPA or saline injection. Iron and iron saturation levels were checked before and 24 h after Gd-DTPA or vehicle administration.
Significant mobilization of iron from mice liver tissue homogenate in mixtures with Gd in vitro was seen in the control (saline) and in the experimental (Gd) groups (513+/-99.1 vs 1117.8+/-360.8 microg/dl, respectively; P<0.05). Administration of Gd-DTPA to iron-overloaded rats after renal pedicle ligation caused marked elevation of serum iron from baseline 143+/-3.4 to 570+/-8 microg/dl (P<0.0001). There were no changes of the named parameter, either in iron-overloaded anuric rats after saline injection or in normal diet uraemic animals, following Gd-DTPA administration.
The combination of iron overload and lack of adequate clearance of Gd chelates may cause massive liberation of iron with dangerous elevation of free serum iron. It is highly recommended that after Gd contrast study, end-stage renal disease patients with probable iron overload should undergo prompt and intensive haemodialysis for prevention of this serious complication.
- SourceAvailable from: Anil Chandraker[Show abstract] [Hide abstract]
ABSTRACT: 11-13 which were previously considered to be safe in patients with kidney disease. 14 In an upcoming report, Todd et al demonstrate that 24-month mortality is increased in outpatient hemodialysis patients with the cutaneous changes of NSF (48% versus 21%), with an adjusted hazard ratio of 2.9 (95% CI, 1. 4-5.9. 15
- Der Nephrologe 01/2009; 4(1):33-41.
Iron-mobilizing properties of the gadolinium–DTPA complex:
clinical and experimental observations
Marina Vorobiov1, Anna Basok1, David Tovbin1, Alla Shnaider1, Leonid Katchko2and
1Department of Nephrology and2Department of Pathology, Soroka Medical Center, Ben Gurion University of the
Negev, Faculty of Health Sciences, Beer Sheva, Israel
ging (MRI) contrast agents are considered to be safe in
patients with impaired renal function. Our study investi-
gates a mechanism of severe iron intoxication with life-
threatening serum iron levels in a haemodialysis patient
following MRI with Gd–diethylenetriaminepentaacetic
acid (Gd–DTPA) administration. His previous history
was remarkable for multiple blood transfusions and
biochemical evidence of iron overload. We hypothesized
that Gd–DTPA may have an iron-mobilizing effect
in specific conditions of iron overload combined with
prolonged exposure to the agent.
Methods. For the in vitro study, Gd–DTPA was added
to mice liver homogenate and iron metabolism para-
meters were measured after incubation in comparison
with the same samples incubated with saline only. For
the in vivo study, an experimental model of acute
renal failure in iron-overloaded rats was designed.
Previously iron-overloaded and normally fed rats
underwent bilateral nephrectomy by renal pedicle
ligation, followed by Gd–DTPA or saline injection.
Iron and iron saturation levels were checked before
and 24 h after Gd–DTPA or vehicle administration.
Results. Significant mobilization of iron from mice
liver tissue homogenate in mixtures with Gd in vitro
was seen in the control (saline) and in the experimental
(Gd) groups (513"99.1 vs 1117.8"360.8 mgudl, respec-
tively; P-0.05). Administration of Gd–DTPA to iron-
overloaded rats after renal pedicle ligation caused
marked elevation of serum iron from baseline 143"3.4
to 570"8 mgudl (P-0.0001). There were no changes of
the named parameter, either in iron-overloaded anuric
rats after saline injection or in normal diet uraemic
animals, following Gd–DTPA administration.
Conclusions. The combination of iron overload and
lack of adequate clearance of Gd chelates may cause
massive liberation of iron with dangerous elevation of
free serum iron. It is highly recommended that after
Gd contrast study, end-stage renal disease patients
with probable iron overload should undergo prompt
and intensive haemodialysis for prevention of this
Keywords: chelate; dialysis; gadolinium; iron overload;
renal failure; Gd–DTPA
Gadolinium (Gd)-containing agents are widely used as
contrast media for magnetic resonance imaging (MRI)
studies and are thought to be safe  even in patients
with impaired renal function. Although they are rapidly
cleared with a half-life of -2 h in patients with normal
renal function, in chronic renal failure half-life is pro-
longed and may exceed 30 h . While haemodialysis
enables more effective clearance of Gd-containing
contrast material than peritoneal dialysis, lack of
adequate dialysis may significantly prolong Gd clear-
ance . Possible side effects may occur due to in vivo
dissociation of the Gd–ligand complex into metal ion
and ligand . This phenomenon is thought to launch a
ligand competition reaction and metal ion exchange
(transmetallation) [5,6]. Very little has appeared in the
literature relating to the clinical significance of these
transmetallation processes in conditions of prolonged
exposure, such as occurs in end-stage renal disease
(ESRD) patients. We present here our clinical observa-
tions and the results of our laboratory investigation in
relation to this phenomenon.
Subjects and methods
A 19-year-old male presented for his routine haemodialysis
Correspondence and offprint requests to: Dr Marina Vorobiov,
Department of Nephrology, Soroka Medical Center, Ben Gurion
University of the Negev, Center of Health Sciences, PO Box 151,
Beer Sheva 84101, Israel. Email: firstname.lastname@example.org
Nephrol Dial Transplant (2003) 18: 884–887
#2003 European Renal Association–European Dialysis and Transplant Association
by guest on May 30, 2013
undergone an MRI with Gd–diethylenetriaminepentaacetic
acid (Gd–DTPA) because of recurrent episodes of abdo-
minal pain. Haemodialysis was discontinued soon after it
was begun because of high fever.
The patient has a history of reflux nephropathy and
kidney transplant from a living related donor in 1998. Six
months after kidney transplantation, non-Hodgkin’s lym-
phoma and Kaposi’s sarcoma with lymph node involvement
were diagnosed. Immunosuppressive therapy was stopped
and interferon (IFN) started. IFN therapy was followed by
severe autoimmune haemolytic anaemia, requiring multiple
blood transfusions and eventual graft rejection. Iron over-
load was diagnosed with values of blood ferritin )5000 nguml.
After treatment with erythropoietin and chelation therapy
by deferoxamine, ferritin levels descended to ;1000 nguml.
Haemodialysis treatment was reinstated after rejection of
On admission his clinical examination was unremarkable.
Laboratory values were consistent with those expected in
haemodialysis patients. Chest X-ray was normal. Blood
cultures were negative. Expanded chemistry analysis showed
an extremely elevated serum iron level of 741 mgudl, com-
pared with 146 mgudl several days prior to admission, which
continued to rise up to 1377 mgudl. It should be noted that
transferrin values were normal. Elevated levels of liver
enzymes of serum glutamic oxaloacetic transaminase (SGOT)
and lactate dehydrogenase (LDH) were also found. LDH
level increased to 1752 Uul 1 day after his admission,
dropping to normal values (482 Uul) after 4 days. Similarly,
SGOT elevated to a maximal point of 73 Uul, returning to
normal levels (39 Uul) on the third day of hospitalization.
Two days after admission there was a marked rise in ferritin
levels,reaching)30 000 nguml.Afterthreeconsecutivehaemo-
dialysis sessions, iron levels dropped and returned to baseline
within 1 week. The patient also improved clinically with com-
The course of this patient’s clinical status led us to
hypothesize that Gd may have an iron-mobilizing effect in
patients with kidney failure and severe iron overload. This
may be due to the prolonged exposure to Gd as a result of
impaired renal clearance in uraemic patients. In order to
prove this supposition we designed two experimental models:
an in vitro model of Gd–DTPA exposure with iron-rich tissue
and an in vivo model of acute renal failure in iron-overloaded
animals combined with Gd–DTPA administration.
In vitro study. Mouse liver was homogenized in tissue culture
medium with the addition of 20% autologous serum. The
samples were divided into equal aliquots in the control and
the experimental tissue culture tubes. The basal level of iron-
metabolism parameters was taken and 0.05 ml of 469 mguml
solution Gd–DTPA to 10 ml of homogenate (experimental)
or equal amount of saline (control) tubes were added. The
tubes were then incubated in a humidified atmosphere of
5% CO2 in air at 378C. Consecutive samples of medium
after short and gentle centrifugation were taken at the
beginning of the experiment and after 2, 4, 24 and 72 h of
incubation. Iron and transferrin levels were measured.
Serum iron levels were measured by in vitro colorimetric
assay for the quantitative determination of iron in human
serum on an automated clinical chemistry analyser (Roche
Diagnostics, GmbH, Mannheim, Germany). Transferrin
determination was performed by in vitro immunoturbidi-
metric assay for the quantitative determination of trans-
ferrin in human serum on automated clinical chemistry
analyser (Roche Diagnostics).
All thetestswere carried
In vivo study. Three groups were included in the experi-
ment. In group 1 we examined Gd–DTPA action on serum
iron parameters in iron-overloaded anuric rats. Group 2
animals were similarly prepared (iron overload q anuria)
and treated by vehicle instead of Gd–DTPA. In group 3,
Gd–DTPA was administered to normal feeding anephric rats
in order to investigate the influence of Gd–DTPA on iron
metabolism in renal failure without previous manipulation of
Iron-overload rat model. Male Sprague–Dawley rats weighing
200–220 g were used. They were housed in cages at room
temperature of 21–238C, kept alight from 08.00 to 20.00 h,
and were given commercial rat pellet chow with free access to
water during the observation period. All animals were
divided into three groups: groups 1 and 2 received iron
supplementation in drinking water (as a carbonyl iron (II),
50 mguml solution), the total dose being 300–350 mgurat over
a period of 3 months . Group 3 was given only regular diet
Acute-renal-failure rat model. After intraperitoneal anaes-
thesia by ketamineuxylasine mixture, both kidneys of all
animals were exposed from midline laparotomy and renal
pedicles were ligated. Absence of urine and macroscopic
appearance of ischaemic kidney evidenced adequacy of
ligation. Immediately after closure of the abdominal cavity,
blood samples were taken from the internal jugular vein and
0.15 mmolukg of Gd–DTPA (groups 1 and 3) or 0.5 ml 0.9%
NaCl (group 2) was injected. Twenty-four hours later the rats
were again anaesthetized and blood samples were drawn by
heart puncture. The animals were then sacrificed by anaes-
thetic overdose. Liver samples were taken for Prussian blue
staining. Serum biochemistry parameters were measured:
urea, creatinine, iron and transferrin levels as described
Statistical analysis of iron and transferrin levels in super-
natant and in blood plasma was performed using ANOVA
test followed by Tuckey’s multiple comparison test. Results
are expressed as means"SEM. A P-value of -0.05 was
As shown in Figure 1, after 72 h of exposure of liver
homogenate to Gd–DTPA, iron concentrations in
the supernatant compared with control tubes were
In the in vivo studies, following ligation of the
renal pedicle, all animals became anuric and after 24 h,
serum creatinine rose to a concentration of 3.442"
0.126 mgudl from a baseline of 0.699"0.02, thus
guaranteeing the absence of renal clearance of Gd.
885 Effect of Gd–DTPA in iron overload
by guest on May 30, 2013
The histology of the liver from the iron-fed animals
showed extensive haemosiderosis.
570"8 mgudl before and 24 h after administration
of Gd to the oral iron-loaded renal failure animals
(group 1), respectively (P-0.0001). No changes were
observed in the other two groups (Figure 2). Liver
enzyme levels (data not shown) were identical in all
three groups of animals.
There were no changes in transferrin levels in any of
We describe a haemodialysis patient who developed
extremely high iron concentrations in the plasma. The
patient had undergone an MRI investigation a few
days prior to admission in which Gd–DTPA was
administered as a contrast agent. We hypothesized
that the iron intoxication was a result of chelation of
stored iron accumulated from previous numerous
blood transfusions. This hypothesis was confirmed by
in vitro experiments in which a Gd-containing agent
was incubated with liver homogenates and in an in vivo
model of chronic iron-overloaded renal failure rats to
whom Gd–DTPA had been administered.
Free Gd ion [Gd(III)] solubility is poor and can
form in vivo precipitates of salts with anions phos-
phate, carbonate or hydroxyl, which are deposited in
liver, bone and muscle. Incorporation of Gd ion within
organic ligand forms an ionically stable compound
(chelate) with improved solubility, tissue distribution
and renal clearance, making it safe for clinical use.
Many agents presently in use as contrast material have
a linear polyaminocarboxylate molecule derived from
ethylenediaminetetraacetic acid (EDTA). DTPA is the
first derivative that contains five negatively charged
carboxylate moieties. This agent is known as ‘gado-
pentetate dimeglumine’ (Magnevist2) . In vivo dis-
sociation of Gd complexes into metal ion and ligand
are possibly responsible for certain adverse effects of
the agent. This process can be facilitated both by endo-
genous metals (zinc [Zn(II)], copper [Cu(II)], calcium
[Ca(II)] and iron [Fe(III)]) and endogenous acids,
destabilizing the complex and leading to its dissocia-
tion. Displacement of Gd from its ligand by other
metals through competitive ionic binding is known as
Several chelating agents are used to remove iron
from its body stores, namely deferoxamine, Ca–DTPA,
Ca3Na–DTPA and EDTA. These agents differ in their
sites of activity (intra- or extracellular) and potency,
which is in inverse correlation with complex stability
. Gd–DTPA, in spite of its generic relationship with
DTPA,isthoughttobe averyweakchelator without any
known effects on iron metabolism [1,4].
Iron cations tend to be tightly bound to haemo-
siderin and ferritin in vivo and are consequently
thought to be poor candidates for transmetallation
because of their low free plasma concentration.
Acidaemia, resulting from inflammation or tissue
hypoxia, may promote conversion of haemosiderin to
iron donor, especially in iron-overload conditions
[8,9]. In renal failure, the combination of metabolic
acidosis and the absence of adequate clearance of
Gd-containing agent may favour clinically significant
transmetallation with formation of iron–ligand com-
plexes. Abundance of iron ions may also promote
concurrent cation exchange in ligand site .
Simple addition of Gd–DTPA to the whole blood in
our preliminary experiments did not change the iron or
iron-saturation levels either immediately or after 24 h
of incubation. Furthermore, we did not find changes
of iron metabolism after the MRI procedure in other
dialysis patients even 48 h after Gd administration
(data not shown).
In vitro experiments demonstrated a significant
increase of iron concentrations in the supernatant of
liver homogenates incubated with Gd–DTPA as
compared with experiments in which the liver homo-
genate was incubated with saline. This increase may
indicate a process of transmetallation.
Our in vivo study showed that a combination of iron
overload and prolonged Gd–DTPA exposure causes
significant elevation of serum iron and iron-saturation
levels in uraemic rats. This rise in plasma iron
concentrations was not detected in the iron-loaded
uraemic animals treated with saline only or in the
uraemic non-iron-loaded animals after Gd–DTPA
Fig. 1. Iron level in mice liver tissue homogenate after 24 h
incubation with Gd or saline. *P-0.05.
Fig. 2. Serum iron level in anuric rats after Gd or saline adminis-
tration: group 1, iron overload q Gd; group 2, iron overload q
saline; group 3, normal fed q Gd. *P-0.0001.
886M. Vorobiov et al.
by guest on May 30, 2013
This observation supports our hypothesis that
Gd–DTPA displays its chelating properties only in
iron-overload conditions. Rapid Gd–DTPA clearance
in normal renal function probably hinders the trans-
became more pronounced after prolonged exposure
Haemodialysis enables more effective clearance of
Gd-containing agents in ESRD patients than perito-
neal dialysis [3,10]. Lack of effective dialysis in our
patient (due to fever and general discomfort) probably
was the first participating factor of the subsequent
Abnormal liver function tests can be explained by
endogenous iron liberation and its toxic effect on the
liver . The transient character of this abnormality
renders other aetiologies for the liver impairment most
Since elevated ferritin levels in our patient appeared
only after multiple blood transfusions, a diagnosis of
primary haemochromatosis is very unlikely [12,13].
Reaching the initial ferritin level after three haemo-
dialysis sessions could indicate chelated iron (by Gd–
ligand) complex elimination, since iron can be removed
by haemodialysis only in chelated form [14–16]. The
sharp elevation of iron levels in our patient 1 day after
the MRI study with Gd–DTPA and the rapid drop in
iron levels after haemodialysis sessions support our
theory that chelation of iron indeed occurred and that
the complex iron–chelate was removed by dialysis.
We conclude from our study that prolonged
exposure to Gd chelates in conditions of iron overload
may cause a transmetallation phenomenon with release
of iron from stores. Patients with advanced renal
failure on haemodialysis who are suffering from iron
overload combined with metabolic acidosis may
develop significant elevation of serum iron levels with
clinical signs of iron toxicity. Prompt and intensive
haemodialysis is essential for the prevention of this
Chaimovitz for his encouragement and invaluable input, to
Mrs Batia Vesler for her important contribution in the clinical
pathology laboratory, and Mrs Rita Soullam for her excellent
We aretruly gratefulto Prof. Cidio
1. Runge VM. Safety of approved MR contrast media for
intravenous injection. J Magn Reson Imaging 2000; 12: 205–213
2. Townsend RR, Cohen DL, Katholy R, Swan SK et al. Safety of
intravenous gadolinium (Gd–BOPTA) infusion in patients with
renal insufficiency. Am J Kidney Dis 2000; 36: 1207–1212
3. Joffe P, Thomsen HS, Meusel M. Pharmacokinetics of
gadodiamide injection in patients with severe renal insufficiency
and patients undergoing hemodialysis or continuous ambulatory
peritoneal dialysis. Acad Radiol 1998; 5: 491–502
4. Mann JS. Stability of gadolinium complexes in vitro and in vivo.
J Comput Assist Tomogr 1993; 17: S19–S23
5. Idee JM, Berthommier C, Goulas C et al. Haemodynamic effects
of macrocyclic and linear gadolinium chelates in rats: role of
calcium and transmetallation. Biometals 1998; 11: 113–123
6. Cacheris WP, Quay SC, Rocklage SM. The relationship between
thermodynamics and the toxicity of gadolinium complexes.
J Magn Reson Imaging 1990; 8: 467–481
7. Papanastasiou DA, Vayenas DV, Vassilopoulos A, Repanti M.
Concentration of iron and distribution of iron and transferrin
after experimental iron overload in rat tissues in vivo: study of
the liver, the spleen, the central nervous system and other organs.
Pathol Res Pract 2000; 196: 47–54
8. Pippard MJ, Jackson MJ, Hoffman K, Petrou M, Modell CB.
Iron chelation using subcutaneous infusions of diethylene
triamine penta-acetic acid (DTPA). Scand J Haematol 1986;
9. Ozaki M, Kawabata T, Awai M. Iron release from haemosiderin
and production of iron-catalysed hydroxyl radicals in vitro.
Biochem J 1988; 250: 589–595
10. Rakba N, Aouad F, Henry C et al. Iron mobilisation and
cellular protection by a new synthetic chelator O-Trensox.
Biochem Pharmacol 1998; 55: 1797–1806
11. Damade R, Rosenthal E, Cacoub P. [Hyperferritinaemia]. Ann
Med Interne (Paris) 2000; 151: 169–177
12. Dorsam J,Knopp MV, Schad L,Piesche S,Carl S, Oesingmann N.
Nephrol Dial Transplant 1995; 10: 1228–1230
13. Crowley JP, Nealey TA, Metzger J, Ponto L, Chazan JA.
Transfusion and long-term hemodialysis. Arch Intern Med 1987;
14. Tapia G, Troncoso G, Galleano M, Fernandez V, Puntarulo S,
Videl LA. Time course study of the influence of acute iron
overload on Kupffer cell functioning and hepatotoxicity assessed
in the isolated perfused rat liver. Hepatology 1998; 27: 1311–1316
15. Hatton RC, Portakles IT, Finlay A, Ross EA. Removal of iron
dextran by hemodialysis: an in vitro study. Am J Kidney Dis
1995; 26: 327–330
16. Locsey L, Papp L. Study of trace elements in patients on
haemodialysis. Int Urol Nephrol 1983; 15: 289–295
by peritoneal dialysis.
Received for publication: 26.2.02
Accepted in revised form: 20.12.02
887 Effect of Gd–DTPA in iron overload
by guest on May 30, 2013