Coordination Chemistry Reviews 284 (2015) 278285

Contents lists available at ScienceDirect

Coordination Chemistry Reviews

journal homepage: www.elsevier.com/locate/ccr

Review

Kill or cure: Misuse of chelation therapy for human diseases

Guido Crisponi a, , Valeria M. Nurchi a , Joanna I. Lachowicz a ,
Miriam Crespo-Alonso a , Maria A. Zoroddu b , Massimiliano Peana b
a
b

Dipartimento di Scienze Chimiche e Geologiche, University of Cagliari, Cittadella Universitaria, 09042 Monserrato, Italy
Dipartimento di Chimica e Farmacia, University of Sassari, Via Vienna 2, 07100 Sassari, Italy

Contents
1.
2.

3.
4.
5.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Coronary and peripheral arteriopathies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.1.
Theories on which CT is based . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.2.
Reports of reliable trials conducted on CT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.3.
Main effects of CT on homeostasis of essential metal ions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.4.
Ascertained deaths among subjects receiving CT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5.
Concerns related to protocols used for ascertainment of patient poisoning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Autism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Further applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

a r t i c l e

i n f o

Article history:
Received 20 March 2014
Received in revised form 14 April 2014
Accepted 24 April 2014
Available online 5 May 2014
Keywords:
Chelation therapy
Cardiovascular disease
Autism
EDTA
Speciation

a b s t r a c t
Chelation therapy is a consolidated medical procedure used primarily to reduce the toxic effects of metal
ions on human tissues. Its application spans a broad spectrum of disorders, ranging from acute metal
intoxication to genetic metal-overload. The use of chelating agents is compromised by a number of
serious side effects, mainly attributable to perturbed equilibrium of essential metal ion homeostasis
and dislocation of complexed metal ions to dangerous body sites. For this reason, chelation therapy has
been limited to specic critical and otherwise untreatable conditions and needs to be monitored within
an appropriate clinical context. The aim of this review is to discuss how this false chelation therapy
developed and in which diseases it is currently applied.
2014 Elsevier B.V. All rights reserved.

1. Introduction
Chelation therapy (CT) is a consolidated medical procedure used
primarily to reduce the toxic effects of metal ions in humans.

Abbreviations: ACC, American College of Cardiology; AHA, American Heart Association; AMA, American Medical Association; CAD, coronary artery disease; CT,
chelation therapy; DFO, desferal; DMPS, 2,3-dimercapto-1-propanesulfonic acid;
DMSA, meso-2,3-dimercaptosuccinic acid; EDTA, ethylene diaminetetraacetic acid;
FDA, Food and Drug Administration; MI, myocardial infarction; NIH, National Institute of Health; TACT, Trial to Assess Chelation Therapy.
Corresponding author at: Dipartimento di Scienze Chimiche e Geologiche, Universit di Cagliari, Cittadella Universitaria, 09042 Monserrato-Cagliari, Italy.
Tel.: +39 070 675 4476; fax: +39 070 675 4478.
E-mail address: crisponi@unica.it (G. Crisponi).
http://dx.doi.org/10.1016/j.ccr.2014.04.023
0010-8545/ 2014 Elsevier B.V. All rights reserved.

278
279
280
280
281
282
282
283
284
284
284
284

Current treatment approaches have different mechanisms of action

depending upon what occurs after complexing of the targeted
metal ions: the complexed metal can be removed from the organism or it can be dislocated to tissues where it does not exert
any toxic action or, alternatively, to where its toxic effects can
be attenuated through the formation of complexes. Chelating
agents embrace a broad spectrum of applications, ranging from
acute intoxication and chronic toxicity deriving from occupational,
environmental and even iatrogenic causes to the metal toxicity
observed in certain genetic diseases. However, their use is constellated by a number of potentially serious side effects, mainly
attributable to an imbalance in essential metal ion homeostasis.
This disequilibrium depends upon a variety of interrelated factors
that are extremely hard to control. Nevertheless, when used in a

G. Crisponi et al. / Coordination Chemistry Reviews 284 (2015) 278285

279

Na
OH
OH

OH

OH

Ca2+

N
O

Na

O
O

Na2EDTA

EDTA

Na2CaEDTA

SH

SH

HO
OH
O

O
Na

N
O

OH

OH

Na

HS

OH
S
O

SH

DMSA

DMPS

Scheme 1. Structure of EDTA and of its two most used salts, and of DMSA and DMPS.

controlled clinical setting, CT has proven to be an invaluable tool

in solving a wide range of clinical problems [111]. Desferal (DFO)
for the treatment of blood transfused -thalassemia patients represents the most unequivocal success. Indeed, this drug, has led
to dramatic improvements in the quality of life and overall life
expectancy of thalassemia patients. According to Bernhardt [12]
-thalassemia patients now in their 50s who have undergone DFO
chelation therapy since childhood are living proof of the value of
this drug. Those aficted with this disease prior to the emergence
of DFO, or who have been unable to cope with the demands or cost
of DFO therapy, have typically died in their teens.
Chelation therapy has been approved for systemic iron overload, lead poisoning and other metal toxicities. Several chelating
agents have been approved for use in humans with each one having specic afnity for a given metal ion or set of metal ions.
Despite the limited approved indications for the use of chelators,
advertisements for the treatment of numerous other conditions can
easily be found on the web. Although the advertised treatments
have no scientic basis, they are proposed to the public in such an
appealing way that it may be difcult to effectively counteract the
phenomenon.
Many of these expensive, lengthy treatments are administered directly into the bloodstream in an ofce setting. Others are
advertised for use at home. Purported uses include:
1. treating arteriopathies (in particular atherosclerosis) by removing calcium from arterial plaques;
2. treating intermittent claudication (painful leg cramps sometimes due to atherosclerosis);
3. curing or improving symptoms of autism;
4. preventing or curing neurodegenerative diseases and multiple
sclerosis.
In all these cases, improvement up to full recovery is largely
advertised on the basis of anecdotal patient reports. Unapproved
use of chelation therapy has resulted in harm and even death.
This is an unacceptable risk in the absence of a proven benet.
Several unapproved chelators are marketed for home use. The
U.S. Food and Drug Administration (FDA) recently warned several
companies that they are breaking the law by marketing such
therapies, reminding them that all approved chelating drugs
require a prescription.
Based on the National Health Statistics Report published in 2007,
the use of chelation has increased by 68% since 2002, passing from

66,000 to 111,000 adults using CT [13], most of whom with no clinical or laboratory indications for this kind of therapy. Overall, there
appears to be a prevalence for the inappropriate use of chelation
therapy in cardiovascular disease [14].
In our opinion, scientists with years of experience in these topics
have both the duty and responsibility to warn the members of the
scientic community of these facts and to provide them with precise and detailed information on the possible toxic effects of CT. The
purpose of this review is to describe how false chelation therapy
historically developed, and in which disorders CT is presumed to
be effective, including coronary and peripheral arteriopathies and
autism.
2. Coronary and peripheral arteriopathies
The rst report of chelation therapy dates back to the early 1950s
and describes the use of ethylenediaminetetraacetic acid (EDTA)
(see Scheme 1) in patients affected by lead toxicity [15]. These preliminary ndings prompted further investigation into the possible
applications of EDTA. The apparent success of EDTA in reducing
calcium deposits [16] led Clarke et al. [17] to use this chelator in
patients affected by angina, and other authors to use it in various forms of atherosclerosis [1820]. Subsequently, CT evolved
to constitute infusions of disodium EDTA (a molecule that binds
divalent and trivalent cations, including calcium, magnesium, lead,
cadmium, zinc, iron, aluminum and copper and facilitates their
excretion via the kidneys) and vitamins [20,21].1
Throughout the decades, despite the clear lack of scientic evidence in support of CT for vascular diseases and the mounting
opposition of Health and Medical Associations, chelation practitioners have increasingly used EDTA for the treatment of coronary
and peripheral arteriopathies on the sole basis of favorable anecdotal case reports. They claim that EDTA chelation therapy is helpful
against atherosclerosis, cardiac disease and peripheral vascular diseases, such as intermittent claudication. This has resulted in an
extensive and inappropriate use of CT, aggravated by the fact that
many patients have been led to consider this treatment approach a
valid substitute for other well-recognized medical treatments. The

1
This typical Chelation therapy includes a series of intravenous infusions of a
solution containing 0.5 L sterile water, besides Na2 EDTA (3 g), MgCl2 (2 g), KCl
(2 mEquiv.) procaine HCl (0.100 g), ascorbate (7 g), NaHCO3 (0.840 g), pantothenic
acid (0.250 g), heparin (2500 units), thiamine (0.100 g) [22].

280

G. Crisponi et al. / Coordination Chemistry Reviews 284 (2015) 278285

promoters of CT report their theories in four books published in

the 80s of the last century [2125] in which they assert to have
treated 300,000500,000 patients who have safely beneted from
treatment with EDTA.
The following section addresses some important points concerning the application of EDTA chelation to cardiovascular
diseases:
1.
2.
3.
4.
5.

theories on which CT is based;

reports of reliable trials conducted on CT;
main effects of CT on homeostasis of essential metal ions;
ascertained deaths among subjects receiving CT;
concerns related to protocols used for ascertainment of patient
poisoning.

2.1. Theories on which CT is based

CT proponents claim that the ability of EDTA to remove calcium from atherosclerotic plaques induces plaque regression and
frees clogged atherosclerotic arteries [26]. Different theories have
been proposed to explain the mechanisms of action underlying CT,
though none have been scientically conrmed:
One theory asserts that EDTA operates by scavenging calcium
from fatty plaques, provoking their breakdown. The promoters
of this theory retain that once the calcium is removed by regular
administration of EDTA, the remaining elements in the plaque
break up and clear away, restoring the lumen of the narrowed
arteries [17].
A second theory suggests that CT stimulates the release of a hormone which favors calcium removal from plaques and/or lowers
cholesterol level.
According to a third theory, CT possibly limits the harmful effects
of oxidative stress on endothelial cells by reducing the production of reactive oxygen species (ROS). This reduction in oxidative
stress is thought to reduce inammation in the arteries and
improve blood vessel function [26].
According to Clarke et al. [17], EDTA treatment (5 g/day, 5 days
a week for a mean of 35 treatments) had positive effects on 20
patients with angina pectoris. Nineteen of these patients reported
subjective improvement of their symptoms. One patient in this
series died suddenly. The authors speculated that the death could
have been caused by a calcium embolus to the brain, originating
from a decalcied atherosclerotic plaque.
In 1960, Meltzer et al. [27], reported that there was no objective evidence of improvement in any of ten patients with angina
pectoris that could be ascribed to EDTA chelation treatment.
Nonetheless, most patients reported unusual improvement in their
symptoms.
During the years, numerous reports have emerged from both
patients and their physicians, claiming improvement in coronary
artery disease (CAD) following CT. However, the large majority of
reports substantiating the efcacy of EDTA chelation are in the

format of case reports, case series or uncontrolled clinical trials. Although these studies describe a reduction in angina, they
are mostly based on uncontrolled clinical observations or retrospective data, generally with a limited number of participants. A
systematic Cochrane Data Review analyzing 5 different CT studies
showed that the acquired evidence was not sufcient to support
or negate the advantages of CT [28]. Although pioneering studies
in the sixties indicated a possible efcacy of EDTA in the treatment
of atherosclerosis, the results of well designed studies (discussed
in the following section) demonstrated that the use of EDTA is not
effective in the treatment of patients with heart disease. On these
bases, several medical organizations in the United States, including
the American Medical Association (AMA), the National Institutes
of Health (NIH), the American College of Cardiology (ACC) and the
American Heart Association (AHA), disapproved and condemned
the use of disodium EDTA in the cure of cardiovascular diseases
[29].
As a general observation, the literature both in support and in
opposition to CT is rather supercial: the afrmations may look
scientically sound, but even the citations are often unreliable and
need to be accurately revised and conrmed.
2.2. Reports of reliable trials conducted on CT
The details of reliable clinical trials on the application of CT
in various cardiovascular diseases are schematically reported in
Table 1. The results of these 7 randomized clinical trials are summarized below.
Olszewer et al. [30] demonstrated improvement in walking distance in a trial of 10 patients with peripheral arterial disease; the
patients were divided into 2 groups of 5 patients each and therapy
was only partially blinded.
Sloth-Nielsen et al. [32], in a study of 153 patients with peripheral vascular disease did not nd signicant differences between
the EDTA group versus the placebo group.
Guldager et al. [33], in a double blind, placebo-controlled trial
enrolling 153 individuals with peripheral arterial disease (75 EDTA
and 78 placebo) did not observe any signicant differences for EDTA
on walking time or ankle-brachial blood pressure.
A randomized, double-blind, placebo-controlled trial of walking time and ankle-brachial blood pressure indices, conducted
by van Rij et al. [34] on 32 patients (15 EDTA and 17 placebo)
with intermittent claudication did not observe any benecial
effects in patients with peripheral vascular or ischemic heart
disease.
The study by Knudtson et al. [35], the best-designed trial of
chelation therapy in CAD, involved 41 participants in the EDTA
group and 43 in the placebo group. The authors concluded that
CT was not associated with benecial effects on exercise time,
functional reserve for exercise, and quality-of-life in patients with
proven ischemic heart disease. A sub-study of this trial reported by
Anderson et al. [36] did not show any improvement when CT was
added to the well established treatment schemes of patients with
atherosclerotic risk factors.

Table 1
Details of the 7 randomized, double blind, controlled clinical trials.
Reference

Diseasea

Inclusion number

Design

Benet

[31]
[30]
[32]
[33]
[34]
[35]
[36]

CAD
PAD
PVD
PAD
PVD/IH
CAD
CAD

28
10
153
153
32
84
47

Double-blind not clearly explained

Initially double-blind, after 10 sessions single blind
Double-blind
Double-blind
Double-blind
Double-blind; randomized; placebo controlled
Double blind; randomized; placebo controlled

No
Improvement in walking distance
No
No
No
No
No

CAD, Coronary artery disease; PAD, Peripheral arterial disease; PVD, Peripheral vascular disease; IH, Ischemic heart.

G. Crisponi et al. / Coordination Chemistry Reviews 284 (2015) 278285

281

Table 2
Metal ion amounts in human plasma. Data from the International Commission on radiological Protection, Report of the task group on Reference Man ICRP publication 23,
Pergamon Press, Oxford 1994 [38]. Fe (non haeme) from [39]. These data refer to a 70 kg man with 5200 mL plasma with a density of 1.06 g/mL. In the fourth column the Total
content in mg has been converted in Total mmol, and in the fth the mM concentration in 5.2 L of plasma has been reported. In the last row the total mmol corresponding
to 3 g of EDTA are reported together to its concentration in 5.2 L of plasma.
Element

Total mg

Atomic weight

Total mmol

mM

Al
As
Bi
Ca
Co
Cu
Fe (non haeme)
Pb
Mg
Mn
Hg
Ag
Zn
Sb
Cd
Ni

1.9
2.5
6.2 102
3.1 102
1.7 103
5.6
4.0
1.4
2.1 102
1.4 101
2.6 102
2.9 101
34
2.4 102
3.6 102
1.6 101

26.98
74.92
208.98
40.08
58.93
63.55
55.84
207.2
24.30
54.93
200.59
107.87
65.39
121.76
112.41
58.69

7.0 102
3.3 102
3.0 104
7.8
2.9 105
8.8 102
7.2 102
6.8 103
8.6
2.5 103
1.3 104
2.7 103
5.2 101
2.0 104
3.2 104
2.7 103

1.3 102
6.4 103
5.7 105
1.5
5.5 106
1.7 102
1.4 102
1.3 103
1.7
4.9 104
2.5 105
5.2 104
1.0 101
3.8 105
6.2 105
5.2 104

336.2

8.92

1.72

Na2 EDTA

3.0 103

Contrary to the previously mentioned uncontrolled investigations of CT reporting symptomatic improvements, these few
controlled trials provide evidence that the possible benets of CT
can simply be ascribed to placebo effects. The potential risks of CT
and the non-existent or weak scientic evidence of its efcacy in
CAD are sufcient cause for its rejection in favor of proven therapies.
The Trial to Assess Chelation Therapy (TACT) performed by the
National Institute of Health (NIH) began in 2003 and was scheduled to be completed in 2009. TACT was designed to investigate the
public health problems posed by EDTA chelation therapy, i.e. large
numbers of patients being exposed to undened risks for unproven
benets. After clinical start up, TACT had to face a lot of criticism.
In particular, Atwood et al. [37] argued that the TACT was unethical, dangerous, pointless, and wasteful and should be abandoned.
Indeed, the nal results of TACT may serve as a guide for further
research but are not sufcient to support the routine use of chelation therapy for treatment of patients who have had myocardial
infarction (MI) [14].

2.3. Main effects of CT on homeostasis of essential metal ions

This section evaluates the effects of CT with EDTA on the homeostasis of essential metal ions. It has to be remembered that the
acronym EDTA stands for ethylenediaminetetraacetic acid. The
same acronym EDTA is often ambiguously used to refer to both
the two most commonly used salts, Na2 EDTA and Na2 CaEDTA.2
The Food and Drug Administration provides detailed information on the clinical applications of these two chelating agents.
Na2 EDTA, marketed as Endrate and also known with the chemical name of disodium edetate, has been approved for selected
patients with high blood calcium levels (hypercalcemia) as well
as for patients with heart-rhythm problems due to intoxication
with digitalis. Na2 CaEDTA, marketed as Calcium Disodium Versenate and known as calcium disodium edetate, has been approved
to lower blood lead levels in patients with lead poisoning.

2
These two common ways of writing the names of EDTA salts can present in turn
ambiguity. Actually, in Na2 EDTA two sodium ions substitute two of the four acidic
proptons of EDTA, in Na2 CaEDTA all the four acidic protons are substituted by two
sodium ions and one calcium ion.

To quantitatively evaluate the effects of Na2 EDTA on essential

metal ion homeostasis a speciation study has to be performed on
the basis of:
metal ion amounts in human plasma,
protonation constants of EDTA and its complex formation
constants with the involved metal ions.
Accurate values of mean amount in plasma of a number of elements can be found in the Report of the Task Group on Reference
Man [38], referring to a man weighing 70 kg with 5.2 L circulating plasma (density = 1.06 mg/mL) (Table 2). Complex formation
constants of EDTA with selected metal ions can be found in the
IUPAC Database of Stability Constants [40] (Table 3).
First of all, a large excess of three metal ions in plasma was
assumed, i.e. the two alkaline earth cations Ca2+ (1.49 mM) and
Mg2+ (1.66 mM), and Zn2+ (0.10 mM). EDTA was 1.72 mM, corresponding to 3 g sodium edetate. Using a speciation program it is
possible to calculate the concentration of EDTA complexes and the
percentage of complexed metal ions. Table 4 shows the complex
concentrations and the percentage of complexed metal ions calculated with the speciation program HySS [47].
The total concentrations of Ca2+ and Mg2+ in plasma were very
similar (1.49 mM and 1.66 mM, respectively), and close to that of
EDTA (1.72 mM). At these concentrations, the calcium and magnesium ions share the available EDTA according to their conditional
complex formation constant at pH 7.4, resulting in 92.6% complexed calcium and 13.9% complexed magnesium at 20 C, with
slightly different values (89.3% and 17.2%, respectively) at 37 C.
The lower amount of zinc was totally complexed at both temperatures because of its much higher complex stability. Even more
so, all the other metal ions considered in Table 4 were fully coordinated by EDTA. It must be mentioned that the possible competition
between EDTA and endogenous ligands for the same metal ions,
such as transferrin for iron, was not considered in these speciation
calculations in virtue of the extremely high stability of EDTA metal
complexes.
It is apparent that from a thermodynamic point of view the
delivery of Na2 EDTA in a molar amount comparable to the total
amount of calcium circulating in plasma can completely perturb the
homeostatic equilibrium of this metal ion. Obviously, the kinetic
effects connected to the slow delivery of EDTA attenuate the

282

G. Crisponi et al. / Coordination Chemistry Reviews 284 (2015) 278285

Table 3
EDTA complex formation constants with some selected metal ions, measured at 20 C and at 0.1 M ionic strength and/or at 37 C and 0.15 M ionic strength.
Metal ion
2+

Ca
Mg2+
Zn2+
Cu2+
Pb2+
Hg2+
Mn2+
Fe3+

log 11 at 20 C

References

log 11 at 37 C

References

10.59
8.69
16.26
18.79
18.3
21.80
14.04
25.10

[41]
[41]
[41]
[41]
[41]
[41]
[41]
[45,46]

9.36
7.75
14.61
18.30
16.62

[42]
[42]
[42]
[43]
[44]

12.42

[42]

Representing EDTA as the tetraprotic acid H4 Y, the log 11 values refer to the complex formation reaction
Mn+ + Y4 = MY(n4)+
EDTA cumulative protonation constants, relative to the four protonation equilibria Y4 + H+ = HY3 , Y4 + 2H+ = H2 Y2 , Y4 + 3H+ = H3 Y , Y4 + 4H+ = H4 Y at 20 C and at 0.1 M
ionic strength, are log 1 = 10.26, log 2 = 16.42, log 3 = 19.09, log 4 = 21.08 [41], and at 37 C and 0.15 M ionic strength are log 1 = 9.12 and log 2 = 15.03 [42], or log 1 = 9.74
and log 2 = 15.74 [43].

otherwise devastating effects of the introduction of such an amount

of EDTA. At any rate, the depletion of other essential metal ions,
such as zinc and copper is highly detrimental in the long term.
In this respect, the data presented by Allain et al. [48] are
extremely interesting: 1 g of CaNa2 EDTA dissolved in 250 ml
(1000/374.24 mmol) was infused intravenously over 1 h in 10
healthy subjects. Urines were collected over 24 h, the day before
and on the day of the CaNa2 EDTA infusion test. The elements Al,
B, Ba, Cu, Fe, Mn, Si, Sr, Zn, Na, K, Ca, Mg, S and P were measured
by ICP-AES, and Pb was measured by ICP-MS. The CaNa2 EDTA infusion increased the 24 h elimination of Al from 9.8 g to 58 g, Fe
from 66 to 121 g, Mn from 2.9 to 16.5 g, Pb from 9.8 to 56 g
and Zn from 623 to 8847 g, with an induced ratio of the increase
of urinary elimination of about 2 for Fe, 5 for Al, Pb and Mn, and
15 for Zn. These data show that 9 mg of the 34 mg of zinc (about
) circulating in plasma, and 16.5 mg of the 140 mg Mn in plasma
(about 1/9) were completely scavenged by the chelator. This information is extremely valuable in the evaluation of the safety levels
of CT. Similar results for zinc were previously reported by Thomas
and Chisolm [49] in their study on CT treatment of lead-poisoned
children.
2.4. Ascertained deaths among subjects receiving CT
Important information published on the Food and Drug Administration website [50] refer to the dramatic consequences generated
by the improper and confusing term EDTA used for both the
two different drugs Na2 EDTA and CaNa2 EDTA. Because of this
confusion, some patients are given the wrong drug with fatal consequences in some cases and serious adverse reactions in others.
This medication error is particularly dangerous when Na2 EDTA
is given to a patient who should be treated with CaNa2 EDTA.
During the time period 1971 through 2007, the FDA received 11
reports of deaths associated with the use of disodium edetate. Nine
of these deaths were reported after administration of disodium

edetate. A specic EDTA drug was not identied in two cases. These
two reports simply referred to the use of EDTA. Seven of the 11
deaths resulted from confusion of disodium edetate with another
drug. In ve cases, disodium edetate was administered instead
of calcium disodium edetate. In two cases, disodium edetate was
administered instead of the drug Etomidate. Here is it important
to mention that Etomidate is not a form of EDTA. Since the time
of FDA approval of disodium edetate for use in the treatment of
hypercalcemia or digitalis toxicity, multiple new drugs have been
developed and marketed for treatment of these conditions. Based
on the availability of other drugs that can be used for the indications for which disodium edetate was approved and the safety
issues related to disodium edetate, FDA is currently reconsidering the overall risks and benets of disodium edetate to determine
the most appropriate action. FDA is also issuing a Public Health
Advisory regarding disodium edetate: The FDA has not approved
the EDTA drugs for treatment of autism or hardening of the arteries. FDA has not received clinical data sufcient to support either
the safety or efcacy of EDTA drugs when they are used for these
specic chelation purposes.
Even though brief treatments with CaNa2 EDTA are believed safe
[49,51], prolonged treatments at high doses have been reported to
cause acute proximal tubular necrosis and nephrotoxic reactions in
children and adults that are not always reversible and can be fatal
[5258].
2.5. Concerns related to protocols used for ascertainment of
patient poisoning
Since health assurance policies do not cover the use of CT in
cardiovascular diseases, there are many circumstances in which
patients invoke at least some metal poisoning to justify CT.
Some considerations on the protocols used for the evaluation
of patient poisoning are contained in the Laboratory Medicine
Practice Guidelines sponsored by the National Academy of Clinical

Table 4
EDTA complex concentration of some selected metal ions, calculated at pH 7.4, 20 C and 0.1 M ionic strength and at the same pH, 37 C and 0.15 M ionic strength.
20 C,  = 0.10 M
Metal ion
2+

Ca
Mg2+
Zn2+
Fe3+
Cu2+
Mn2+
Pb2+

[Mn+ EDTA]

[Mn+ ]Tot
3

1.49 10
1.66 103
1.00 104
1.38 102
1.70 105
0.49 106
1.30 106

1.38 10
0.23 103
1.00 104
1.38 102
1.70 105
0.49 106
1.30 106

37 C,  = 0.15 M
% Mn+ chelated
92.6
13.9
100
100
100
100
100

[Mn+ EDTA]
3

1.33 10
0.28 103
1.00 104
1.38 102
1.70 105
0.49 106
1.30 106

% Mn+ chelated
89.3
17.2
100
100
100
100
100

G. Crisponi et al. / Coordination Chemistry Reviews 284 (2015) 278285

Biochemistry (NACB) under the subheading Broad-spectrum

screening for trace elements and environmental pollutants [59]. In
this Section, the Committee denes the tests used for screening as
uncommon laboratory analyses used to search for trace elements in
biological samples such as blood, urine and hair. The authors point
out that these tests generally lack a published report of certied reference intervals and present a number of problems associated with
patient preparation, sample collection, analysis and reliability. The
authors retain that screening for trace elements is inappropriate
when apparent occupational and environmental causes are lacking. Of major concern to the Committee was that the screening
tests had not been validated in a broad patient population and
lacked data such as detection and quantication limits, applicability to different media (e.g. urine, saliva, plasma, serum, red blood
cell, hair), instructions for specimen collection and stability during transport, as well as specicity for given diseases. Coupled to
these concerns were the nancial connections between some of
the practitioners and the testing laboratories which aroused considerable alarm among the Committee members and led them to
adopt a precautionary approach. In conclusion, the Committee recommended biomonitoring exclusively for occupational exposure
until research and clinical studies could satisfactorily address all
the concerns raised.
A debated method, promoted by many laboratories as a simple and reliable tool for the detection of metal ion poisoning, is
hair mineral analysis. One of the rst studies to disqualify this test
was conducted in 1985 [60]. A follow-up study was performed in
2001 by the ofcials of various Californian Health Services to assess
whether the reliability of data from commercial laboratories performing multimineral hair analyses [61] had improved since 1985.
Hair samples collected from a single healthy volunteer were submitted for analysis to the six commercial laboratories responsible
for 90% of the samples submitted for mineral analysis in the United
States. It is noteworthy that these laboratories make an annual
prot on hair mineral tests of about 9.6 million US dollars. The study
concluded that mineral hair analysis was unreliable for several reasons:
a number of factors ranging from external contamination to hair
treatment strongly affect mineral concentration;
even if the above variables are strictly controlled, the analytical
variability among laboratories performing hair mineral analysis
makes it difcult to interpret the results;
the dose-response data between hair concentrations and effects
on target organs are largely unavailable.
For all the above mentioned reasons the authors invite both Public Health Agencies and Consumer Protection Agencies to warn the
public of the unreliability of these tests.

3. Autism
Autism is a syndrome characterized by social impairments, communication difculties, repetitive behavior, abnormal movements
and sensory dysfunction. Recent epidemiological studies estimate
that autism affects 1 in 150 children of the United States.
CT was introduced for the treatment of autism based on
the belief that mercury contained in the vaccine preservative
thimerosal could be the cause of autism. This belief started to take
shape in 2001 when Bernard et al. published two reviews of medical
literature and US government data suggesting that: (i) many cases
of idiopathic autism are induced by early mercury exposure from
thimerosal; (ii) this type of autism represents an unrecognized mercurial syndrome; and (iii) genetic and non-genetic factors establish

283

a predisposition whereby thimerosal adverse effects occur only in

some children [62,63].
In the same year, 2 papers were published by Kidd in support
of the mercury-autism causative hypothesis. In these reports, Kidd
afrms that prenatal toxic exposures are consistent with autism
spectrum symptomatology and that repeated vaccinations containing live virus and toxic mercurial content (thimerosal) are
a plausible etiologic factor [64]. In conclusion, he suggests that
detoxication of mercury and other heavy metals by DMSA/DMPS
chelation may result in signicant benet for patients [65].
A more recent report by Nelson and Bauman [66] questioned
the results obtained in these reports. After thorough examination
of the hypotheses forwarded by Bernard et al. [62,63] asserting that
I. the clinical signs of mercury toxicity are similar to those of
autism,
II. the increase in the diagnosis of autism parallels exposure to
thimerosal,
III. there are higher mercury levels in persons with than without
autism
the authors reached the following conclusions:
at a sufcient dose mercury is indeed a neurotoxin, but the typical
clinical signs of mercurism are not similar to the typical clinical
signs of autism;
changes in environmental exposure and medical care within the
same period can all be considered plausible biological contributors to the occurrence and severity of autism;
no peer-reviewed papers in the literature report an abnormal
body burden of mercury, or mercury excess in the blood, urine
or hair of subjects affected by autism [67]. The one paper that
sought a relationship between mercury levels in hair and autism
did not nd such a relationship [68].
Therefore, on the basis of current evidence, these authors tend
to exclude a causal relationship between thimerosal-containing
vaccines and autism.
In 2004, the Immunization Safety Review Committee Board
on Health Promotion and Disease Prevention of the Institute of
Medicine of the National Academies published a review on Vaccines and Autism [69].
Also in this case, the bulk of epidemiological evidence excluded
the possibility of a causal relationship between thimerosal and
autism. In fact, three controlled observational studies [6971]
and two uncontrolled observational studies [72,73] consistently
provided evidence of no association, despite the fact that these
studies used different methods and examined different populations
(Sweden, Denmark, The United States and the United Kingdom).
Davis presented a systematic review of the literature on the
effects of CT on symptoms of autism spectrum disorders (ASD) [74].
Five studies met the criteria for inclusion in this review [7579].
Eighty-two children aged 3 to 14 years, 58 of whom were males,
were treated with CT. Only in the two-phase study by Adams [76],
high levels of toxic metal ions were an inclusion criteria. Forty-one
participants received CT in the rst phase, and only those with a
high excretion of toxic metals in the rst phase continued to the
second phase: one group of participants received additional CT,
while a second group received placebo. An inclusion criteria in
the study by Geier [75] was a medical history of mercury exposure. Eppright et al. [77] instead reported that all participants in
their study had proven high blood levels of lead. In the conclusions
of their review, Davis et al. [74] remarked that (i) the literature
is extremely limited by the small number of studies and their
weak methodological approaches, (ii) the results do not support
CT as a treatment for ASD. According to these authors CT should be

284

G. Crisponi et al. / Coordination Chemistry Reviews 284 (2015) 278285

considered a cart before the horse because its use was not validated prior to its application.
Despite the lack of supporting evidence, the hypothesized link
between toxic levels of metals in the body and ASD is one reason
why chelation treatment continues to be administered to many
individuals with ASD.
To provide some insight into this issue, the National Institute of
Mental Health proposed a Clinical Trial on Mercury Chelation to
Treat Autism. However, this study was suspended in September
2006, prior to enrollment, since there was no clear evidence of
a direct benet to the children and the proposed chelation trial
presented more than minimal risk.
4. Further applications
In recent years a growing number of new applications for CT
have been launched and passed off as being the panacea for a variety
of disorders, including mercury overload caused by mercurycontaining pharmaceuticals [80], rheumatoid arthritis [81], and
multiple sclerosis [82]. Considering the paucity of data available
in the scientic literature, the magnitude reached by chelation
therapy on the web is astonishing. The effects of the inappropriate
use of CT on public health in the near future may be devastating.
For this reason we carefully need to consider the best way to reduce
and eventually eradicate this harmful practice.
5. Conclusions
R.J.P Williams, in his amazing Bakerian lecture 1981 on Natural
selection of the chemical elements [83], proposed a very general
overview of the role of these elements in biological samples. In
particular he pointed out that selection has rened the functions of
individual elements so that each element can play a distinct role, proteins providing the evolutionary media for the development of function.
It was the recognition and separation of each element in their specic
sites (proteins) that allowed elements to be positioned in space. In
turn the spatial organization generates, through feedback, the ow of
other elements. These extraordinary interrelationships constitute
the grounds for the fragile homeostatic equilibrium of elements
whose safeguard hampers any attempts to disrupt it and thus promotes the existence of living organisms.
Therefore, the voluntary introduction into the human body of a
perturbing factor, such as a chelating agent, in the absence of a well
ascertained vital health problem that requires it, is detrimental and
dangerous.
Acknowledgments
We thank Prof. Gavino Faa for helpful discussions and review
of medical terms and statements. We are grateful to the Sardinian
Regional Government authorities for the nancial support received
by GC and JIL for the project Integrated approach in the design of
metal chelators for human diseases, MCA for the Master and Back
Program, and VMN, MP and MAZ for the project CRP-26712.
References
[1] O. Andersen, Chem. Rev. 99 (1999) 26832710.
[2] M. Blanusa, V.M. Varnai, M. Piasek, K. Kostial, Curr. Med. Chem. 12 (2005)
27712794.
[3] E.J. Baran, Curr. Med. Chem. 17 (2010) 36583672.
[4] M.S. Sinicropi, D. Amantea, A. Caruso, C. Saturnino, Arch. Toxicol. 84 (2010)
501520.
[5] G. Crisponi, V.M. Nurchi, M. Crespo-Alonso, L. Toso, Curr. Med. Chem. 19 (2012)
27942815.
[6] G. Crisponi, M. Remelli, Coord. Chem. Rev. 252 (2008) 12251240.
[7] Y. Ma, T. Zhou, X. Kong, R.C. Hider, Curr. Med. Chem. 19 (2012) 28162827.
[8] G. Crisponi, V.M. Nurchi, D. Fanni, C. Gerosa, S. Nemolato, G. Faa, Coord. Chem.
Rev. 254 (2010) 876889.

[9] G. Crisponi, V.M. Nurchi, V. Bertolasi, M. Remelli, G. Faa, Coord. Chem. Rev. 256
(2012) 89104.
[10] M.A. Santos, Coord. Chem. Rev. 228 (2002) 187203.
[11] G. Berthon, Coord. Chem. Rev. 228 (2002) 319341.
[12] P.V. Bernhardt, Dalton Trans. (2007) 32143220.
[13] P.M. Barnes, B. Bloom, R.L. Nahin, Natl. Health Stat. Rep. (2008) 123.
[14] G.A. Lamas, C. Goertz, R. Boineau, D.B. Mark, T. Rozema, R.L. Nahin, L. Lindblad,
E.F. Lewis, J. Drisko, K.L. Lee, JAMA 309 (2013) 12411250.
[15] S.P. Bessman, H. Ried, M. Rubin, Med. Ann. Dist. Columbia 21 (1952) 312315.
[16] N.E. Clarke, C.N. Clarke, R.E. Mosher, Am. J. Med. Sci. 229 (1955) 142149.
[17] N.E. Clarke, C.N. Clarke, R.E. Mosher, Am. J. Med. Sci. 232 (1956) 654666.
[18] H.R. Casdorph, C.H. Farr, J. Holistic Med. 5 (1983) 315.
[19] M.T. Grier, D.G. Meyers, C. Dalmady-Israel, M. Perreault, Ann. Pharmacother.
27 (1993) 15041509.
[20] C.J. Rudolph, E.W. McDonagh, R.K. Barber, J. Adv. Med. 4 (1991) 3945.
[21] E.M. Cranton, A Textbook on EDTA Chelation Therapy, Hampton Roads Publishing, Newburyport, MA, 2001, pp. 503539.
[22] G.A. Lamas, C. Goertz, R. Boineau, D.B. Mark, T. Rozema, R.L. Nahin, J.A. Drisko,
K.L. Lee, Am. Heart J. 163 (2012) 712.
[23] M. Walker, G. Gordon, The Chelation Answer: How to Prevent Hardening of the
Arteries and Rejuvenate Your Cardiovascular System, M. Evans & Company,
Inc., New York, 1982.
[24] J.P. Trowbridge, M. Walker, Chelation Therapy The Key to Unclogging Your
Arteries Improving Oxygenation, Treating Vision Problems, 3rd ed., Better
Health Booklet Service, Marble Falls, TX, 1990.
[25] E. Cranton, A. Brecher, Bypassing Bypass The New Technique of Chelation Therapy, 2nd ed., 1990.
[26] E. Ernst, Circulation 96 (1997) 10311033.
[27] L.E. Meltzer, M.E. Oral, J.R. Kitchell, in: M.J. Seven (Ed.), Metal-Binding in
Medicine, J.B. Lippincott Co, Philadelphia, 1960, pp. 132136.
[28] M.V. Villarruz, A. Dans, F. Tan, Cochrane Database Syst. Rev. (2002), CD002785.
[29] University of Maryland Medical Center http://umm.edu/health/medical/
altmed/supplement/ethylenediaminetetraacetic-acid
[30] E. Olszewer, J.P. Carter, Med. Hypoth. 27 (1988) 4149.
[31] J.R. Kitchell, F. Palmon, N. Aytan, L.E. Meltzer, Am. J. Cardiol. 11 (1963) 501506.
[32] J. Sloth-Nielsen, B. Guldager, C. Mouritzen, E.B. Lund, M. Egeblad, O. Norregaard,
S.J. Jorgensen, R. Jelnes, Am. J. Surg. 162 (1991) 122125.
[33] B. Guldager, R. Jelnes, S.J. Jorgensen, J.S. Nielsen, A. Klaerke, K. Mogensen,
K.E. Larsen, E. Reimer, J. Holm, S. Ottesen, J. Intern. Med. 231 (1992)
261267.
[34] A.M. Van Rij, C. Solomon, S.G.K. Packer, W.G. Hopkins, Circulation 90 (1994)
11941199.
[35] M.L. Knudtson, D.G. Wyse, P.D. Galbraith, R. Brant, K. Hildebrand, D. Paterson,
D. Richardson, C. Burkart, E. Burgess, JAMA 287 (2002) 481486.
[36] T.J. Anderson, J. Hubacek, D. George Wyse, M.L. Knudtson, J. Am. Coll. Cardiol.
41 (2003) 420425.
[37] K.C. Atwood Iv, E. Woockner, R.S. Baratz, W.I. Sampson, Medscape J. Med. 10
(2008) 115124.
[38] Report of the Task Group on Reference Man, The International Commission on
Radiological Protection, ICRP Publication 23, Pergamon Press, 1994.
[39] R.R. Crichton, M. Charloteaux-Wauters, Eur. J. Biochem. 164 (1987) 485506.
[40] L.D. Pettit, K.J. Powell, The IUPAC Stability Constants Database, ver. 5.7, Academic Software and IUPAC, Otley, UK, 2001.
[41] G. Schwarzenbach, R. Gut, G. Anderegg, Helv. Chim. Acta 37 (1954) 937957.
[42] J.R. Dufeld, P.M. May, D.R. Williams, J. Inorg. Biochem. 20 (1984) 199214.
[43] G. Arena, S. Musumeci, E. Rizzarelli, S. Sammartano, D.R. Williams, Ann. Chim.
(Rome) 68 (1979) 535542.
[44] M.J. Willes, D.R. Williams, Inorg. Chim. Acta 80 (1983) L35L36.
[45] G. Schwarzenbach, J. Heller, Helv. Chim. Acta 34 (1951) 576591.
[46] E. Bottari, G. Anderegg, Helv. Chim. Acta 50 (1967) 23492356.
[47] L. Alderighi, P. Gans, A. Ienco, D. Peters, A. Sabatini, A. Vacca, Coord. Chem. Rev.
184 (1999) 311318.
[48] P. Allain, Y. Mauras, A. Premel-Cabic, S. Islam, J.P. Herve, J. Cledes, Br. J. Clin.
Pharmacol. 31 (1991) 347349.
[49] D.J. Thomas, J.J. Chisolm Jr., J. Pharmacol. Exp. Ther. 239 (1986) 829835.
[50] http://www.fda.gov/Drugs/DrugSafety/PostmarketDrugSafetyInformationfor
(updated
08/23/13;
accessed
PatientsandProviders/ucm113738.htm
01/03/14).
[51] R.P. Wedeen, V. Batuman, E. Landy, Environ. Res. 30 (1983) 5862.
[52] H. Foreman, C. Finnegan, C.C. Lushbaugh, JAMA 160 (1956) 10421046.
[53] M.D. Reuber, J. Bradley, JAMA 174 (1960) 263269.
[54] A.J. Khan, U. Patel, M. Rafeeq, A. Myerson, K. Kumar, H.E. Evans, J. Pediatr. 102
(1983) 147149.
[55] D.I. Moel, K. Kumar, Pediatrics 70 (1982) 259262.
[56] E. Weining, W. Schwerd, Muench. Med. Woch. 100 (1955) 17881789.
[57] S. Moeschlin, Schweiz. Med. Woch. 87 (1957) 10911096.
[58] W. Vogt, H. Cottier, Schweiz. Med. Woch. 87 (1957) 665667.
[59] A.H.B. Wu, C. McKay, L.A. Broussard, R.S. Hoffman, T.C. Kwong, T.P. Moyer, E.M.
Otten, S.L. Welch, P. Wax, Clin. Chem. 49 (2003) 357379.
[60] S. Barrett, JAMA 254 (1985) 10411045.
[61] S. Seidel, R. Kreutzer, D. Smith, S. McNeel, D. Gilliss, JAMA 285 (2001) 6772.
[62] S. Bernard, A. Enayati, L. Redwood, H. Roger, T. Binstock, Med. Hypoth. 56 (2001)
462471.
[63] S. Bernard, A. Enayati, H. Roger, T. Binstock, L. Redwood, Mol. Psychiatry 7
(2002) S42S43.
[64] P.M. Kidd, Altern. Med. Rev. 7 (2002) 292316.

G. Crisponi et al. / Coordination Chemistry Reviews 284 (2015) 278285

[65]
[66]
[67]
[68]
[69]

[70]
[71]
[72]
[73]

P.M. Kidd, Altern. Med. Rev. 7 (2002) 472499.

K.B. Nelson, M.L. Bauman, Pediatrics 111 (2003) 674679.
M. Aschner, S.J. Walker, Mol. Psychiatry 7 (2002) S40S41.
L. Wecker, S.B. Miller, S.R. Cochran, D.L. Dugger, W.D. Johnson, J. Ment. Dec.
Res. 29 (1985) 1522.
Institute of Medicine (US) Immunization Safety Review Committee, Immunization Safety Review: Vaccines and Autism, National Academies Press
(US), Washington, DC, 2004, Available from: http://www.ncbi.nlm.nih.gov/
books/NBK25344/
A. Hviid, M. Stellfeld, J. Wohlfahrt, M. Melbye, JAMA 290 (2003) 17631766.
T. Verstraeten, R.L. Davis, F. DeStefano, T.A. Lieu, P.H. Rhodes, S.B. Black, H.
Shineeld, R.T. Chen, Pediatrics 112 (2003) 10391048.
K.M. Madsen, M.B. Lauritsen, C.B. Pedersen, P. Thorsen, A.M. Plesner, P.H. Andersen, P.B. Mortensen, Pediatrics 112 (2003) 604606.
P. Stehr-Green, P. Tull, M. Stellfeld, P.B. Mortenson, D. Simpson, Am. J. Prev.
Med. 25 (2003) 101106.

285

[74] T.N. Davis, M. OReilly, S. Kang, R. Lang, M. Rispoli, J. Sigafoos, G. Lancioni,

D. Copeland, S. Attai, A. Mulloy, Res. Autism Spect. Disord. 7 (2013)
4955.
[75] D.A. Geier, M.R. Geier, Neuroendocrinol. Lett. 27 (2006) 833838.
[76] J.B. Adams, M. Baral, E. Geis, J. Mitchell, J. Ingram, A. Hensley, I. Zappia, S.
Newmark, E. Gehn, R.A. Rubin, K. Mitchell, J. Bradstreet, J. El-Dahr, BMC Clin.
Pharmacol. 9 (2009) 16.
[77] T.D. Eppright, J.A. Sanfacon, E.A. Horwitz, Mo. Med. 93 (1996) 136138.
[78] K. Patel, L.T. Curtis, J. Altern. Complement. Med. 13 (2007) 10911097.
[79] H.G. Senel, J. Autism Dev. Disord. 40 (2010) 494503.
[80] S. Corsello, A. Fulgenzi, D. Vietti, M.E. Ferrero, Cases J. 2 (2009) 199203.
[81] F. Bamonti, A. Fulgenzi, C. Novembrino, M.E. Ferrero, Biometals 24 (2011)
10931098.
[82] A. Fulgenzi, S.G. Zanella, M.M. Mariani, D. Vietti, M.E. Ferrero, Biometals 25
(2012) 569576.
[83] R.J.P. Williams, Proc. R. Soc. Lond. B 213 (1981) 361397.