The Risks for Na
The established method for investigating the immediate and short-term risks for a drug, after sufficient animal studies have been completed, is the phase 1 trial, which measures responses to incremental doses in order to determine the threshold for toxicity while minimizing the risk for serious injury. There were several early studies of toxicity of Na2EDTA in animals, but not all questions were answered, and there have been no formal phase 1 studies.
Early human studies, numbering a few hundred subjects, suggested that the most serious reported toxicities, hypocalcemia and renal failure, could be avoided by limits on dosing and rates of infusion.[250,291] Mucocutaneous lesions around the mouth, face, and scrotum were described as similar to lesions of "B complex deficiency," but were thought by some investigators to be due to zinc depletion caused by chelation.[245,291,292] These lesions were rarely reported after investigators lowered Na2EDTA doses; nevertheless, some began giving zinc, pyridoxine, and riboflavin to subjects undergoing chelation. Burning at infusion sites was common in early studies, but chelation proponents argue that this has been minimized with reduced infusion rates, buffering, and the addition of procaine.
Hypoglycemia in diabetics receiving exogenous insulin, especially insulin to which zinc had been added to reduce solubility, was reported in early studies. This was possibly due to zinc depletion leading to increased solubility, and hence more rapid action, of the exogenous insulin.
Meltzer (1961), which the 2003 TACT protocol erroneously cites as an efficacy series ( Table 2 ), reported the side effects and toxicities of 2000 Na2EDTA infusions in an uncontrolled series of 81 adults. Dosing consisted of 3.0 g Na2EDTA in 500-mL normal saline or glucose solution, infused over 2.5-3 hours. This is similar to the TACT Na2EDTA regimen, although Meltzer's group typically gave 3 infusions per week, whereas the TACT calls for 1 infusion per week. Unlike the TACT, Meltzer's group did not give zinc, B vitamins, or other additives. The findings are summarized in Table 6 , reproduced from the original article. The most frequent side effects were burning at injection sites, hypotension (2 cases were "severe," defined as "a drop of systolic pressure of 30 mm or more with distinct hypotensive symptoms"; 23 cases were "moderate," defined as "a drop of systolic pressure of 30 mm with or without symptoms"), nausea, vomiting, abdominal cramping, and symptoms suggestive of hypocalcemia.
Meltzer and colleagues also reported "a period of asthenia and lethargy described by several patients for 24 hours following an infusion." In 14 subjects whose hands and wrists were studied by "serial roentgenographic examinations using graded densities," the authors found no evidence of bone loss.
The only reliable, if incomplete, risk data from more recent trials come from RCTs. Guldager and colleagues gave 20 infusions of 3.0 g Na2EDTA or placebo, 2-3 times per week, to 159 subjects. Of 6 subjects who dropped out during the treatment period, 5 were in the EDTA group; one of these dropped out due to "impaired renal function," but no more information was given. Proteinuria occurred in 10 chelation subjects and 4 placebo subjects, but there were no overall differences in serum creatinine values between the 2 groups. There were some differences in other side effects: "Faintness" occurred in 11 chelation subjects and 1 placebo subject, but the authors did not report corresponding blood pressures or ECG findings. "Hypocalcemic symptoms" occurred in 6 subjects in the chelation group and 2 subjects in the placebo group, but only 1 subject -- in the chelation group -- "showed subnormal calcium levels." One subject in the chelation group had "Raynaud's phenomenon of two fingers" that persisted for 4 days. In a substudy of 54 subjects, there was biochemical evidence of bone loss in the chelation group.
The van Rij trial, in which 15 subjects received Na2EDTA, reported no differences between active and placebo groups. That was also the case for the Hopf trial, in which 8 subjects received Na2EDTA.
In the PATCH study, with twice-weekly dosing for 15 weeks, 1 of 41 subjects in the Na2EDTA group was withdrawn after 10 treatments because of a rise in serum creatinine from 1.5 to 2.1 mg/dL, for which "no other cause...was found." That subject's creatinine reverted to 1.6 mg/dL after 10 weeks. The PATCH reported no other toxicities suggestive of Na2EDTA per se, but did not mention minor complications or side effects.
Dr. Lamas conducted a "pilot" trial, PACT, with a treatment regimen similar to the TACT, involving 40 subjects, 30 of whom received Na2EDTA. He reported some of the safety results in the 2003 TACT protocol. One subject developed "atrial fibrillation with a slow ventricular rate, and, secondarily, heart failure"; another had "a significant increase in AST and ALT"; another had an episode of "transient hypotension, which resolved within 15 minutes." There was "a benign course of creatinine levels." Lesser "adverse events" may have occurred, but the table reporting them was "withheld in entirety proprietary information" [sic] from the copy that we received through a FOIA request.
Thus safety data from credible studies are limited to about 255 chelation subjects (175 from the RCTs, 80 from Meltzer), most of whom received 20-30 infusions of 3 g Na2EDTA, given 1-3 times per week over 2.5-3 hours. Potentially serious complications occurred in at least 7 or 8 subjects: renal insufficiency in 1 or 2, elevated liver enzymes, hypocalcemia, and congestive heart failure in 1 subject each, and symptomatic hypotension in at least 3 subjects and possibly as many as 37. In the Meltzer study, 20 more subjects had symptomatic evidence of hypocalcemia, a potentially life-threatening complication, but this was not objectively documented. In the Guldager study, 11 subjects had symptomatic evidence of hypotension (faintness), but this was not objectively documented.
Episodes of discomfort in addition to faintness, including burning at the infusion site, gastrointestinal symptoms, and the "period of asthenia and lethargy" described by Meltzer, appear to be more common, but in the Guldager study -- the only RCT to report these in detail -- there was no real difference between the active and the placebo groups. A Guldager substudy suggested that repeated infusions of Na2EDTA cause bone loss. That could be of particular importance in postmenopausal women, who are certainly among the subjects in the TACT.
Overall, these data suggest that in well-monitored circumstances, the risk for a serious, but probably manageable, complication occurring at some time during a course of 20-30 infusions is in the realm of 1% to 5%; the risk for minor but annoying side effects is probably 5% to 20%; there is almost certainly decalcification of bone, because excreted calcium must be replaced from the body's stores, but it may be quickly replenished from dietary sources and hence inconsequential -- or perhaps not, especially in postmenopausal women. We simply don't know. The risks for other long-term or subtle complications, some of which are discussed below, are also unknown.
We are aware that these estimates are based on small samples from disparate sources and are therefore subject to error, possibly considerable error. The point, however, is that there is adequate reason to judge the risks for chelation, even when administered carefully in a monitored research environment, to be substantially more than trivial. The onus is not on us to justify a precise number or range; the onus is on investigators to justify exposing human subjects to such risks.
In addition to trials, other sources provide ample evidence of risks for chelation. These include case reports in the medical literature; case reports not in the medical literature, such as expert witness reports in criminal and disciplinary cases; theoretical risks for Na2EDTA that may be apparent only in certain circumstances; and risks, both theoretical and supported by previous studies, of the supplements used in the TACT.
We have previously cited deaths associated with the chelation practice of H. Ray Evers. These were judged due to renal failure, congestive heart failure, and hypoglycemia. There is evidence that Dr. Evers could have prevented several deaths if he had identified and managed complications in a timely and competent fashion. The initial complications would nevertheless have remained. The only significant difference between the dosing of Na2EDTA in Dr. Evers' cases and that of the TACT is that Evers gave infusions 5 times/week for 3-4 weeks, followed by an indefinite rest period, whereas the TACT gives 1 infusion/week for 30 weeks. That difference may explain Evers' complications, but perhaps not: Dr. J. David Spence, one of the government's expert witnesses, reckoned that over a 2-year period at Meadowbrook Hospital, Evers had treated about 600-700 patients with chelation. The incidence of serious complications on the basis of only deaths -- as few as 13 or as many as 21 -- was thus 1.8% to 3.5%, similar to the incidence of serious, albeit appropriately treated, complications reported by the trials cited above.
At least 9 more chelation-related deaths have been reported in the past 15 years.[13,14,15,16,17,19,20] Some occurred during or shortly after an infusion, or were found to be due to hypocalcemia, thus indicating toxicity of the drug. Others occurred the day after an infusion or at a time unspecified in the report. Those may have been due to drug toxicity or to "heart attack," as stated by a couple of the reports, suggesting another sort of complication: the "opportunity cost" of not having received an effective intervention. These reports are necessarily anecdotal because they occurred outside of human trials and were investigated after the fact by regulators or expert witnesses. It is thus unsurprising that advocates have excused such deaths as not having been due to chelation, or due to chelation administered not "in accordance with established guidelines.[30,92]"
In most cases, however, the practitioners have been members of the ACAM, which promulgates those guidelines. One, Neil Ahner, is a "past director" of the ACAM and "board certified" in chelation therapy.[17,294,295,296] Another, the late Dan Christian Roehm, was listed, in the 1986 membership roster, as a Director of the AAMP (the former name of the ACAM).[17,297] Another, Elmer Cranton, is a former ACAM president and one of the "prominent experts" named to the TACT Liaison Committee.[4,19]
The practitioner impugned in the 2005 hypocalcemic death of an autistic 5-year-old in Pennsylvania is an ACAM member and a self-styled expert in "mercury poisoning.[13,108,298,299]" There is now evidence that he infused 1 g of Na2EDTA into the 20-kg boy over 5 minutes -- an extremely dangerous rate. The naturopath involved in the 2003 hypocalcemic death of a woman in Oregon, who chelated her to "remove heavy metals," is also a member of the ACAM.[13,109,300]
Serious but nonfatal complications of chelation are likely to be at least as common as deaths, but for obvious reasons are less likely to be reported. According to a 2003 Board Action in Connecticut, 1 of 3 charts reviewed in an investigation of a chelationist's practice revealed that in 2002, "during an IV infusion of EDTA, [the patient] became lethargic and incoherent and developed rales in her chest." This suggests hypocalcemia and/or congestive heart failure, 2 well-described, life-threatening complications of Na2EDTA infusions. The chelationist was Robban Sica, an ACAM member and a "Director" of the ABCMT, the ACAM's "certification" organization. Between 1999 and 2001, Dr. Sica had been a defendant in 3 lawsuits alleging insurance fraud, racketeering, and sham ownership of medical professional corporations, each time settled by confidential agreement or consent agreement. In 2005 she signed a Consent Order that placed her Connecticut license on probation for 1 year, pending correction of substandard and dangerous practices involving chelation, including erroneous diagnoses of "heavy metal toxicity." Sica is now a TACT co-investigator.
A 2002 article reported 5 patients evaluated at the Emory University, Atlanta, Georgia, emergency department after they had become "ill 30 min to 2 h into iv chelation therapy at an outpatient
The same article reported that "the Georgia Medical Board has investigated several [other] cases related to complications of chelation therapy treatments." On the basis of the substantial experience of one of us (RSB) in reviewing chelation cases, we believe that complications from chelation "as practiced in the community" are fairly common but underreported.
All observers agree that the more rapidly Na2EDTA is infused, the more likely it is to cause dangerous hypocalcemia.[5,55,291,304] Soffer and colleagues[55,304] reported that the only reliable, early sign of clinically important hypocalcemia in their series had been prolongation of the QT interval:
Hypocalcemic tetany may appear with terrifying suddenness, even moments after Chvostek's and Trousseau's signs are noted to be negative. A rapid increase in heart rate and the appearance of an advanced hypocalcemic electrocardiographic pattern are the best signs of impending crisis.
The Na2EDTA (Endrate) prescribing information, mimicked by the ACAM and TACT protocols, stipulates a dose in adults of no more than 3 g to be infused over at least 3 hours.[5,12,33] Nevertheless, unintended variations in IV flow rates are inevitable, prompting one of the expert witnesses in the Evers case to argue that if Na2EDTA were to be infused at all, it should be done with continuous nursing care, continuous ECG monitoring, and infusion pumps.[65,66] That was 30 years ago. Today, neither the ACAM protocol nor the TACT requires these safeguards.[5,33] Thus occasional episodes of hypocalcemia will continue to be unsurprising.
ACAM members acknowledge that
...there can be a febrile systemic reaction that may occur 4 to 8 hours after infusion of EDTA...characterized by a rapid onset of malaise, fatigue, and excessive thirst followed by the sudden appearance of chills and fever. This, in turn, is followed by severe myalgia, frontal headaches, anorexia, occasional nausea and vomiting, and rarely, increased urinary frequency and urgency.
The prescribing information for Endrate warns, in addition to the complications already mentioned, of postural hypotension, generalized dermatitis, hyperuricemia, and more.
At least 1 author has warned of the enhanced potential for chelation to injure the kidneys of occasional patients who genuinely have elevated levels of certain heavy metals, especially cadmium. Although rare, such people are not excluded from the TACT, because there is no preliminary screening for heavy metals.
We have previously cited evidence for pro-oxidant effects of iron chelated by EDTA.[147,148,149] More recent work has replicated this finding. There is also evidence that the pro-oxidant effects of iron are enhanced by ascorbate,[150,237,238] although the physiologic significance of this is
In the absence of testing for markers of oxidative stress, pro-oxidant effects cannot be measured directly in the TACT. They can only be suggested by an excess of long-term, unwelcome clinical consequences in the active treatment groups, that is, only after some subjects have been harmed. In order to be eligible for the TACT, moreover, all subjects must have CAD, the result of a process that is thought to be accelerated by oxidative stress. Such an entry criterion is useful for detecting reductions in events that would typically be common for the entire study cohort. For the same reason, however, this entry criterion -- in effect, a "proxy" for oxidative stress -- hinders the study's capacity to detect increases in such events. Thus if the chelation regimen has harmful, pro-oxidant effects, the TACT cannot detect them before injuries have occurred, and perhaps not at all.
Antioxidant trials have not only failed to demonstrate efficacy, but have suggested that some of the "high dose vitamin and mineral supplements" advocated by the ACAM and included in the TACT are dangerous. Vitamins E and A and beta-carotene have been associated with a small excess of congestive heart failure and all-cause mortality.[306,307] Vitamin A has been associated with reduced bone density and hip fractures,[308,309] which are of particular importance in postmenopausal women, and even more so in light of evidence that Na2EDTA itself causes bone loss. Vitamin C given orally in doses of 60-2000 mg/day appears to be harmless, but as suggested above, the TACT regimen of 7000 mg given intravenously with Na2EDTA may be a different matter.[4,5]
Vitamin C as administered in the TACT confers another risk that would not have been apparent in other antioxidant trials. Seven grams of ascorbate contain, according to the ACAM protocol, more than 600 mg Na+. Together with 600 mg of Na+ in 3 g of Na2EDTA, this means that each chelation infusion has the same amount of sodium as 1 L of 3% NaCl or 3 L of normal saline. Thus the risk for volume overload is substantially increased by the addition of high doses of vitamin C.
An indirect risk for chelation, suggested by NCCAM Director Straus but not specifically acknowledged in the TACT protocols, is that its advocates frequently denigrate proven interventions.[7,199,228,229,230] The risk that subjects might not receive effective therapies must be judged greater in the TACT, in which the bulk of co-investigators are chelationists, than in other NIH-sponsored trials.
The TACT protocols ignore or minimize several of the risks described above. In regard to responsible trials, the 2003 protocol concludes: "...the data on adverse events are reassuring...; we expect a low rate of adverse events and an overall safe intervention." The first paragraph of the 2003 protocol, citing a 2001 "textbook" edited by Elmer Cranton, states: "It has been estimated that in the last few years, over one million patients received over 20 million infusions 'with no serious adverse effects,' but this has not been well-documented." What have been well documented, but unstated in the protocol, are the serious adverse effects reported above. Cranton himself had a chelation-related death only a year before the publication of his textbook.
The 2001 protocol, which was the application that the NIH approved in awarding the grant to Dr. Lamas, relies mainly on the ACAM's assessment of risk:
...safety considerations may lead to our suspending infusions on some patients; however, based on the clinical experience of the ACAM, we expect this number to be very small.... The chelation protocol used at present has been published by ACAM...is in use in hundreds of thousands of infusions yearly, and is thought to be safe.... The chelation protocol has been in use for years and has been found to be quite safe by chelation practitioners.... The DSMB [Data Safety and Monitoring Board] will remain independent, as it will be selected by the NCCAM.
We wonder whether the NIH Special Emphasis Committee and the relevant IRBs were aware, when they reviewed that document, that the ACAM had consistently misrepresented Na2EDTA and its dangers. It continues to do so, as typified by this press release:
...IV EDTA is an FDA approved treatment for lead toxicity in children and adults, with an excellent track record for safety. Millions of infusions have been administered over the last 30 + years, without any deaths being noted, when used in accordance with established guidelines.
Na2EDTA is not an FDA-approved treatment for lead toxicity. Several deaths have been "noted," as documented above. The same press release claims that the very existence of the TACT shows that the NIH has deemed chelation safe:
In fact, the TACT study (Trial to Assess Chelation Therapy), has been in progress, with a goal to assess the efficacy of Chelation Therapy in close to 2,400 patients. In launching this NIH sponsored study, the safety of IV EDTA was accepted to have been firmly established [sic].
The 2001 protocol, however, states that "...assessment of safety is an important secondary endpoint of the trial." In pitching the TACT to prospective subjects, the NCCAM Web site does not assert that the safety of IV EDTA has been "firmly established." On the contrary, it states that the TACT "will assess whether EDTA chelation therapy and/or high-dose vitamin/mineral supplements are safe and effective in treating individuals with CAD."
The TACT is expected to have an independent Data Safety and Monitoring Board (DSMB). According to the TACT RFA: "At the time of award, the Awardee will be requested to nominate prospective DSMB members to the Director of NCCAM, who will select the DSMB members." We wonder whether Awardee Lamas told Director Straus that Robert Nash, who has identified himself as a member of the TACT DSMB, is an ACAM board member and recent chairman of the ACAM's "credentialing" organization, the ABCMT. The ABCMT, like the ACAM itself, claims that "established detoxification techniques [e.g., chelation] have been proven safe and effective over time...." Nash is the author of a 2004 letter admonishing all state medical boards to accept the "ABCMT Standard of Care for Increased Total Body Burden of Toxic Metals."
TACT Co-Investigator Robban Sica offered Nash as an expert witness in her failed 2004 federal lawsuit attempting to stop the Connecticut Bureau of Health Care Systems from disciplining her for substandard practices involving chelating agents. Nash, Sica, Martin Dayton, Ted Rozema, James Carter, Terry Chappell, and other TACT co-investigators are fellow members or advisors of the ABCMT Board or the board of another close affiliate of the ACAM, the ICIM.[218,311] If Nash is a member of the TACT DSMB, that committee would appear not to be independent of pro-chelation activism.
Neither of the TACT protocols acknowledges reports of chelation-related deaths "in the last 30+ years," although such reports are easy to find. The protocols should have mentioned Evers' deaths in particular, because his "3000 cases" constitute a large fraction of Lamas' "thousands of patients with successful outcomes" ( Table 2 ). Evers' deaths also argue that nephrotoxicity is a more substantial risk than the TACT protocols admit. In minimizing that risk, Dr. Lamas cites an article co-authored by Edward McDonagh,[4,174] the chelationist who admitted under oath to having falsified his data.
The TACT protocols concede that acute hypocalcemia is likely to occur when Na2EDTA is infused too rapidly. Nevertheless, the only systematic safeguard is a recommendation to use "the smallest gauge catheter or a 25 gauge butterfly needle as this will limit the maximum infusion rate." This cannot be relied on to prevent rapid infusions. It may, however, result in inadequate IV access when emergency treatments become necessary.
The 2001 TACT protocol does not raise the possibility of bone loss; the 2003 protocol does, but dismisses it without citing the study that most convincingly suggests that it occurs. Instead it cites another article, again co-authored by admitted defrauder McDonagh, that reported no evidence of bone loss by bone densitometry.
The 2003 TACT protocol does not address evidence that chelation may have prooxidant effects. The 2001 protocol mentions that "iron...may remain in a redox active state when bound to EDTA," but does not elaborate on the clinical implications, and asserts that this doesn't happen "when there is a molar excess of EDTA, as is the case in plasma following high-dose EDTA infusion." In support of that assertion, the protocol cites an article describing in vitro Fe-dependent superoxide modification of low-density lipoprotein (LDL) in preparations containing smooth muscle cells and Fe at a concentration of 10 micromoles (mcM). According to the article, however, "EDTA stimulated Fe-promoted modification [of LDL] in the 1-100 mcM range." Thus EDTA, even at a concentration 10-fold greater than that of the iron in the preparation, increased free radical damage -- contrary to the assertion in the 2001 protocol, but exactly as predicted by other, uncited
The protocols do not mention risks for the high doses of vitamins that are included in 2 of the 4 arms of the trial, although the authors should have been aware of reports suggesting excess mortality and bone loss.[308,309,313,314]
Dr. Lamas has repeatedly asserted that chelation for CAD is in a "state of equipoise" or a "state of clinical equipoise.[4,5,29]" "Equipoise," in regard to clinical trials, refers to the ethical requirement for "a state of genuine uncertainty...regarding the comparative therapeutic merits of each arm in a
According to the 2003 TACT protocol:
In general, the timing of a trial is a delicate matter. On the one hand, there must be sufficient belief in a favorable benefit to risk ratio of the intervention to justify exposing half the subjects. On the other hand, there must be sufficient doubt to justify withholding the intervention from the other half. Thus a state of equipoise exists. From a clinical and public health perspective, chelation is in equipoise....
But is it? Revisiting the "2 x 2" design of the TACT, a state of equipoise requires that current knowledge predict similar risk-benefit ratios for each of the 4 arms. However, the available evidence predicts that the 2 Na2EDTA groups are at substantially greater risk, relative to potential benefits, than the 2 placebo groups. The 2 high-dose supplement groups appear to be at slightly greater risk, relative to potential benefits, than the 2 low-dose supplement groups. Thus there is no state of equipoise.
Paradoxically, the majority of co-investigators in the TACT also disagree with Dr. Lamas that chelation is in a state of equipoise: They are chelationists who profess a certitude opposite to what the available evidence warrants. In 1987 Freedman argued that requiring individual investigators to make judgments of equipoise is unreasonable, because they frequently prefer one treatment over the other(s). He proposed, instead, the "concept of 'clinical equipoise,' the requirement [for which] is satisfied if there is genuine uncertainty within the expert medical community -- not necessarily on the part of the individual investigator -- about the preferred treatment."
Thus if chelation for CAD is to be judged in a state of clinical equipoise, experts in cardiovascular disease and clinical trials should make that judgment.
Such experts, however, have consistently argued otherwise. In 2000 the NHLBI Advisory Council, chaired by NHLBI Director Claude Lenfant, rejected a proposed chelation trial in part because of "the paucity of scientific evidence" and "the potential long-term toxicity for bone disease, especially in post-menopausal women." That was prior to the publication of the Canadian PATCH trial, which would only bolster the scientific evidence against the method.[27,28] The AHA has written: "According to qualified scientists who are familiar with research in heart disease, there's only a very small chance that chelation therapy will work." The AHA, furthermore, would support a human trial only if, in a preliminary study, "EDTA had been proven successful in reducing arterial plaque without dangerous side effects." Because no such study has occurred, it follows that the AHA and "qualified scientists," ie, experts, do not favor the TACT.
Cardiologists in general seem to lack enthusiasm for the TACT. Chelationist Co-Investigator John Gannage reported that the recent expansion of the trial to "chelation sites" in Canada was necessitated by the failure of American cardiologists to recruit subjects. Finally, it is reasonable to suppose that the many learned chelation skeptics writing in the 1970s and 1980s, prior to any controlled trials, would be all the more skeptical now that subsequent RCTs have uniformly failed to support the treatment.[9,10,11,63,65,78,124,125,126,127,128,253]
We wonder whether Dr. Lamas' description of ACAM members as "prominent experts" was born less of naiveté or patronization, and more of cynicism: that to counter real experts refuting his claim of clinical equipoise, he would need to conjure experts of another sort.
Medscape J Med. 2008;10(5):115 © 2008
Cite this: Why the NIH Trial to Assess Chelation Therapy (TACT) Should Be Abandoned - Medscape - May 13, 2008.