It’s one of the most common quackeries out there, used by a wide variety of practitioners for a wide variety of ailments blamed on “heavy metal toxicity.” Chelation therapy, which involves using chemicals that can bind to the metal ions and allow them to be excreted by the kidneys, is actually standard therapy for certain types of acute heavy metal poisoning, such as iron overload due to transfusion, aluminum overload due to hemodialysis, copper toxicity due to Wilson’s disease, acute heavy metal toxicity, and a handful of other indications.
My personal interest in chelation therapy developed out of its use by unscrupulous practitioners who blamed autism on the mercury-containing thimerosal preservative that used to be in many childhood vaccines until 2001 but has since all but disappeared from such vaccines except for one vaccine (the flu vaccine, for which a thimerosal-free alternative is available) and in trace amounts in some other vaccines. Mercury became a convenient bogeyman to add to the list of “toxins” antivaccinationists hype in vaccines. In fact, my very first post after I introduced myself on this very blog discussed the idea that mercury in vaccines was a significant cause of autism and autism spectrum disorders, and I’ve periodically written about such things ever since, in particular the bad science of Mark and David Geier, whose idea that chemical castration of children with Lupron “works” against “mercury-induced” autism is based on a chemically ridiculous idea that somehow testosterone binds mercury and makes it harder to chelate. Unfortunately, this particular autism quackery has real consequences and has been responsible for the death of a child.
Chelation isn’t just for autism, however. Despite many practitioners advertising it for autism, cancer (often with dubious studies that I might have to take a look at), Alzheimer’s disease (which Hugh Fudenberg has blamed on the flu vaccine, a claim parroted by Bill Maher, of course!), and just about every ailment under the sun, it’s easy to forget that the original use for chelation therapy promoted by “alternative medicine” practitioners was for cardiovascular disease. When it is used for coronary artery disease or autism, on a strictly stoichiometric and pharmacological basis, it is extremely implausible. Moreover, it is not without potential complications, including renal damage and cardiac arrhythmias due to sudden drops in calcium levels. Such arrhythmias can and have led to death in children, and in adults complications such as renal failure and death.
Despite this extreme implausibility, randomized controlled studies showing that chelation is no better than placebo for cardiovascular disease, a veritable cottage industry of chelation therapy for cardiovascular disease has sprung up, fueled by extravagant claims likening chelation to a “Roto-Rooter for your arteries“, an alternative to angioplasty and coronary artery bypass surgery, and portraying the hostility of SBM to it as not based on medicine but rather on the “need” to protect the “billion dollar industry of angioplasty and CABG.” With most regimens costing $100 to $150 per treatment and “requiring” 30 to 40 doses, it’s a tidy little profit center for “alternative” physicians.
The belief that chelating toxic metals out of people can treat cardiovascular disease has no basis in physiology, biology, or pharmacology, but it’s a major treatment modality favored by naturopaths and many other “alternative” practitioners. Given the infiltration of quackademic medicine into medical academia, it should not be surprising that its advocates promoted clinical trials of this disproven modality. In the early 2000s, they succeeded in the form of the National Institutes of Health (NIH) Trial to Assess Chelation Therapy (TACT), a five year phase 3 trial begun in 2003 to test office-based, intravenous disodium ethylene-diamine-tetra-acetic acid (Na2EDTA) as a treatment for coronary artery disease (CAD). A while back, I learned that the results of this trial would finally be revealed on Sunday (i.e., yesterday) at the American Heart Association’s annual meeting in a session on late-breaking clinical trials in the form of two abstracts:
- Results of the Trial to Assess Chelation Therapy
- Quality of Life Outcomes in the Trial to Assess Chelation Therapy (TACT)
The story was embargoed until last night, but now the embargo is lifted and we can write about it freely, thanks to press releases and reported results of the trial. I expect that the results are—shall we say?—disappointing to chelationists. Certainly if this were a conventional medical therapy it would not be viewed as a particularly favorable trial. However, there are enough equivocal findings that the alt-med websites will soon be touting this study as ironclad evidence that chelation therapy works as well as bypass surgery. I guarantee it. In fact, I wouldn’t be surprised if such articles appear at the same time as this post. Chelationists have access to press information too and tend not to be as fastidious as us in honoring press embargoes.
The story of this $30 million trial is long and depressing and was documented ably and in extreme detail in 2008 by our very own Kimball Atwood and Wally Sampson, as well as Elizabeth Woeckner and Robert Baratz, in an article for the Medscape Journal of Medicine entitled Why the NIH Trial to Assess Chelation Therapy (TACT) Should Be Abandoned. In it, Atwood et al. documented the long and dubious history of TACT, how it came about through the political influence far more than scientific merit (of which it has virtually none), and how the investigators are utterly unqualified to carry out such a large multicenter trial, concluding that the TACT is “pointless, dangerous, unethical, and wasteful.” It’s worth reading the article in full detail, as well as other posts right here on SBM that have examined the trial. In addition, you should check out R. W. Donnell’s Magical Mystery Tour of NCCAM Chelation Study Sites, Part I, Part II, Part III, Part IV, Part V, Part VI, and Part VII. Seriously. As Dr. Donnell points out, only 12 of the 110 TACT study sites were academic medical centers. Many of the study sites were highly dubious clinics touting highly dubious therapies, including heavy metal analysis for chronic fatigue, intravenous infusions of vitamins and minerals (I could never figure out how infusing minerals could be reconciled with chelation therapy to remove minerals, but that’s just me), anti-aging therapies, assessment of hormone status by saliva testing, and much more. Dr. Donnell also points out that the blinding of the study groups to local investigators was likely to have been faulty. So right off the bat, this study was dubious for so many reasons, not the least of which was that some of its site investigators were felons, a problem blithely dismissed by the NIH as being in essence irrelevant to whether the study could be done safely.
Efficacy? What efficacy?
Let’s take a look at the results of TACT, starting with the main outcome measures first, and then we’ll move on to the presentation describing quality of life (QOL) measures. The first presentation is by Gervasio Lamas, MD, a professor of clinical medicine at the Columbia University Division of Cardiology and now Chairman of Medicine at Mount Sinai Medical Center. and was entitled The Trial To Assess Chelation Therapy (TACT): Chelation-Placebo Comparison. One notes right away that the study was funded by the National Center for Complementary and Alternative Medicine (NCCAM, grant U01AT001156) and the National Heart, Lung and Blood Institute (NHLBI, U01HL092607). What’s interesting here is that the study was originally funded by NCCAM and then taken over by NHLBI later. Indeed, when Steve Novella, Kimball Atwood, and I met with Dr. Josephine Briggs, the director of NCCAM, and brought up TACT with her as an example of a scientifically dubious, expensive, and unethical study, she was very quick to dismiss it as no longer within NCCAM’s purview. She was clearly embarrassed by our question and also dismissed it as having come into existence before she took over as director at NCCAM. The second thing I noticed was that this was funded under the NIH U01 mechanism. This mechanism is designed to fund multi-institution collaborations to “discrete, specified, circumscribed projects to be performed by investigator(s) in an area representing their specific interests and competencies.” In practice, what I gather from more senior investigators is that U01 grants are less like R01 grants and more like contracts (more specifically, they are cooperative agreements) to carry out specific projects. As such, they appear to be a bit more amenable to political pressures to be granted—or at least were.
So what about the study itself? The clinical trial at the heart of TACT was a 2 x 2 factorial design that looked at:
- Chelation plus high oral high dose vitamin and mineral supplement
- Chelation placebo plus oral high dose vitamin and mineral supplement
- Chelation plus oral high dose vitamin and mineral supplement placebo
- Chelation placebo plus oral high dose vitamin and mineral supplement placebo
The regimen was also quite rigorous and is described in detail in a recent publication. One notes that the vitamin supplements included doses ranging from 25% to 6,667% of the RDA for these vitamins. For instance, the dose of vitamin C was 2,000% of the RDA; thiamin, 6,667%; and vitamin A, 500%. There were a total of 40 infusions that took three hours each. Thirty of these were administered weekly, followed by ten “maintenance” infusions administered two to eight weeks apart. Now let’s look at the nitty-gritty, followed by the results. First up, the inclusion criteria were:
- Age 50 or older
- MI > 6 months prior
- Creatinine <2.0 mg/dL
- No coronary or carotid revascularization within 6 months
- No active heart failure or heart failure hospitalization within 6 months
- Able to tolerate 500cc infusions weekly
- No cigarette smoking within 3 months
- Informed consent
The exclusion criteria were:
- Chelation therapy within 5 years
- Allergy to any study drug
- Coronary or carotid revascularization within 6 months
- Planned revascularization
- Symptomatic or clinically evident heart failure
- Heart failure hospitalization within 6 months
- Blood pressure > 160/100
- No venous access
- Serum creatinine > 2.0 mg/dL
- Platelet count <100000/mm3
- Cigarette smoking within the last 3 months
- Liver disease or ALT or AST >2.0 times the upper limit of normal
- Diseases of copper, iron, or calcium metabolism
- Inability to tolerate 500 mL of fluids weekly
- Inability to keep to study schedules
- Medical condition likely to affect patient survival within 4 years
- Women of child-bearing potential
The placebo infusion consisted of 500 mL normal saline and 1.2% dextrose, while the chelation infusion consisted of:
- disodium EDTA, 3 grams, adjusted downward based on eGFR,
- ascorbic acid, 7 grams
- magnesium chloride, 2 grams
- potassium chloride, 2 mEq
- sodium bicarbonate, 840 mg
- pantothenic acid, thiamine, pyridoxine,
- procaine, 100 mg
- unfractionated heparin, 2500 U
- sterile water to 500 mL
I find it very interesting that the investigators included procaine (a product of evil big pharma) in the chelation mixture. Yes, I know that disodium EDTA is also a product of the evil big pharma, but one can’t very well do chelation therapy without the chelating agent, can one? In any case, procaine is actually Novocaine and is a topical anesthetic. Its inclusion makes me wonder how much the chelation concoction being tested in TACT hurts as it is injected, as we usually don’t add local anesthetics to infusions unless the infusion causes significant pain at the injection site. Could this also contribute to the ability of patients to know if they’re getting the “real” drug (and, yes, disodium EDTA is a drug)? One wonders, one does.
One last piece needs to be put into place, and that’s to let you know the outcomes that were tested in TACT. The primary endpoint is a composite of death, myocardial infarction, stroke, coronary revascularization, and hospitalization for angina. In other words, investigators took all of these endpoints and added them together. The significance of this design will become clear in a moment, after I reveal the results. Another important point was that the original plan was to randomize 2,372 patients with a one year followup, a design that was estimated to have 85% power to detect a 25% difference. However, in 2009, due to low enrollment (more on that later), it was pointed out that “blinded investigators asked for a reduction of total sample size to 1700, with a compensatory increase in follow-up to maintain same unconditional power.” I find it rather interesting that the word “blinded” is used, instead of just saying that the investigators asked for a decrease in number and a longer followup time in order to try to compensate for low accrual. The implication to me is that there were some investigators who were not blinded. Whether I’m reading too much into this or not, I don’t know, but it sounds odd. In any case, the request was approved.
So, on to the results. The result being touted by the investigators is described in the press release:
Heart attack patients given weekly infusions of chemicals used for chelation therapy had fewer cardiovascular events than those who received identical appearing placebo infusions, according to late-breaking clinical trial results presented at the American Heart Association’s Scientific Sessions 2012.
In the multicenter, double-blind efficacy trial, Trial to Assess Chelation Therapy (TACT),1,708 heart attack patients were randomized to receive 40 infusions of a 500 mL chelation solution or a placebo infusion, with a second randomization to an oral vitamin and mineral regimen or an oral placebo. The chelation solution contained three grams of the synthetic amino acid ethylene diamine tetra-acetic (EDTA), seven grams of vitamin C, B-vitamins, electrolytes, a local anesthetic and heparin, an anti-clotting drug. The placebo infusion was salt water and a small amount of sugar.
Researchers found that patients receiving the chelation solution had fewer serious cardiovascular events than the control group (26 percent vs. 30 percent). Cardiovascular events were defined as death, heart attack, stroke, coronary revascularization and hospitalization for angina.
Although participants with diabetes appeared to have a particular benefit from the infusions, the study team cautioned that subgroup analyses can be unreliable and need to be reproduced.
There’s the spin. Let’s look at the results. The primary endpoint (i.e., the aggregated serious cardiovascular events) did indeed show a modest difference, namely 30% of placebo subjects versus 26.5% of the EDTA chelation subjects (hazard ratio 0.82 for chelation). However, the result is just barely statistically significant, p = 0.035, with the 99% confidence interval for the hazard ratio ranging from 0.69 to 0.99. Note that the predetermined level for statistical significance for purposes of this study was 0.036. More importantly, if you look at the individual endpoints that make up that aggregate, there was no statistically significant difference in death, myocardial infarction, stroke, coronary revascularization, and hospitalization for angina. Subgroup analysis (always a questionable analysis that requires replication, even when preplanned, as in TACT) purported to show a much greater benefit for diabetics, with a hazard ratio of 0.61 (p=0.002), while patients without diabetes showed no statistically significant difference in any of the outcome measures, including the aggregated total bad outcomes (I like that term better).
There were some problems, as you might imagine. First off, only 65% of subjects finished all infusions, with only 76% finishing at least 30. That’s a high drop-out rate, and is likely largely due to just how grueling it is to have to undergo weekly three hour infusions for well over six months, followed by several more months of less frequent infusions. Moreover, 17% withdrew consent, resulting in missing data. I’m not sure how the investigators tried to correct for this (there are standard ways to do it), but these issues are serious. They might not be so serious if there had been a much more convincing treatment effect, but when you get equivocal results such as this such issues loom much larger. Indeed, critics have correctly pointed this out, as in this AP report:
Other experts questioned the results, especially because 60 more people in the group getting dummy infusions withdrew from the study than in the group getting chelation. Usually, more people in a treatment group drop out because of side effects, said Dr. Christie Ballantyne, a Baylor College of Medicine heart specialist. To find the opposite is “a red flag” that suggests those who got dummy treatments found that out and decided to drop out.
“There’s something funky going on here,” Ballantyne said. “It raises questions about study conduct,” especially since a difference of one or two people or complications could have nullified the small overall benefit researchers reported.
Dr. Clyde Yancy, a Northwestern University cardiologist and a former Heart Association president, agreed.
“It’s funny business,” he said. “I’ve never seen a study in which one in five people withdrew consent.”
As Kimball has pointed out, there was a lot more than “funny business going on here.” Moreover, this concern is quite consistent with worries expressed over six years ago by Dr. R. W. Donnell about the adequacy of the blinding of the trial. In light of such concerns, the differential drop-out rate between the two groups makes a lot more sense. Too bad that Dr. Lamias apparently didn’t see fit to include specifics about the drop-out rate for each group in his press release or in his slide set.
Finally, no study would be complete without a consideration of adverse events. After all, in determining whether a therapy is worth pursuing, it is important to weigh its efficaciousness versus its safety. Overall, 79 adverse events were observed forcing discontinuation of infusions. Reasons included: reaching an endpoint; heart failure; other cardiac issues; GI problems; hematological problems; and a variety of other problems. Unfortunately, the presentation slides did not break down how many of these 79 adverse events occurred for patients in the treatment group versus patients from the placebo groups. In fact, given that there were four groups, these adverse events needed to be broken down into four groups but were not. There were a total of four unexpected severe adverse events possibly or definitely related to study therapy, two in the placebo group with one death and two in the treatment group, with one death. Kimball Atwood discussed some of the serious adverse events before in detail. They also don’t appear to line up exactly with what I saw in the presentation:
Reports of Serious Adverse Events (SAE) include two deaths. One occurred in a 74 yo man with critical aortic stenosis and moderately poor left ventricular function who was admitted to a hospital in florid congestive heart failure 6 days after his 32nd chelation infusion. In the other, a 59 yo woman with a history of congestive heart failure suffered an acute episode of left ventricular failure and pulmonary edema 14 hours after her 33rd chelation infusion, followed shortly by cardiac arrest. In each case “The investigator assessed the causal relationship between the study drug and the serious adverse event as not associated,” citing the subjects’ underlying diseases.
In the first case, however, during the subject’s most recent chelation infusion “the investigator [had] noticed evidence of gross lower extremity edema and administered an extra dose of Bumex 2 milligrams followed by as needed treatment.” There is no report that “as needed treatment” occurred, but the onset of signs of congestive heart failure in a subject with critical aortic stenosis constitutes a medical emergency. It is not grounds for an unmonitored extra dose of Bumex (a diuretic that can paradoxically worsen the circulatory problems associated with critical aortic stenosis).
I will be awaiting the full publication, in which (hopefully) more will be revealed about these two deaths. In the meantime, it is clear to me that, even if these results are valid and there is a small benefit to chelation therapy, it’s a long run for a risky short slide.
Quality of life? What quality of life?
Still, even though the results are so unconvincing that the study investigators concluded that TACT “does not constitute evidence to recommend the clinical application of chelation therapy” and that (of course!) “additional research will be needed to confirm or refute our results and explore possible mechanisms of therapy” while also saying TACT showed “some evidence of a potentially important treatment signal in post-MI patients already on evidence-based therapy,” maybe the therapy does something for quality of life. It’s possible, albeit highly implausible. So in parallel, QOL outcomes were measured and presented as a second abstract at AHA presented by Daniel B. Mark, MD, MPH.
Several QOL tools were administered to participants in TACT. These included:
- DASI: A cardiac-related functional status tool that ranges from 0-58 and reflects the ability of patients to do physical activities without difficulty or assistance in 12 domains
- MHI-5: A tool that measures psychological well-being, including both depression and anxiety. Investigators normalized scores to 50±10, with clinically significant difference being >2.5 points.
- Other measures: SAQ (frequency, stability, QOL), SF-36, EQ-5D
In brief, 911 (53%) of the 1,708 main TACT subjects were randomly selected for the QOL substudy, with structured interviews at baseline, 6 months, 12 months, and 24 months. Let me tell you, though, that this study will be much easier to discuss than the main TACT trial for one very simple reason: there were no statistically significant differences in QOL outcomes measured for any of the assessment tools used at any of the time points examined.
So what do you do when your study is completely negative? Easy! You do subgroup analysis, and that’s what Mark et al. did. Well, actually, these subgroup analyses were prespecified; so it’s basically legitimate. Even so, they couldn’t find much. All they could find is that patients with angina symptoms at baseline showed a modest treatment effect in favor of chelation therapy at one year, but at none of the other time points. A good rule of thumb is that for repeated measures, seeing an effect at only one time point is strongly suggestive that the difference for that time point is spurious and not real; so there isn’t much to say about this other than that this was about as close to a completely negative study as one might imagine, and even the press release had to acknowledge that:
“We didn’t see any effect on the quality of life of chelation therapy patients,” said Daniel B. Mark, M.D., M.P.H., lead author of the sub-study and professor of medicine, director of outcomes research at Duke University Medical Center and Duke Clinical Research Institute in Durham, N.C. “Patients weren’t any worse, but they weren’t any better.
One of the tools used to measure quality of life was the Duke Activity Status Index, DASI, to measure patients’ ability to complete daily tasks. The lowest score of 0 means the patient couldn’t do any chores associated with their own care such as feeding, toileting and dressing themselves. The highest score of 48 would be achieved by a professional athlete, Mark said.
At the beginning of the study, patients taking chelation therapy had a score of 24.6 and after two years it went up to 27.1. Those on placebo, dummy infusions that contained no medicine, had a baseline score of 23.5 that went up to 25.1. The small difference between chelation and placebo wasn’t significant enough to show a notable impact on how patients functioned in their daily lives.
The results were similar when researchers used the SF-36, the Short Form Health Survey, which assesses mental wellbeing or stress. After two years of chelation or placebo, patients reported similar scores.
“We thought it might make people feel better, but we didn’t see that consistently enough,” Mark said.
No, Dr. Mark didn’t see it at all, just as the trial investigators apparently didn’t see Edzard Ernst’s criticisms of their trial design when they cited his review article as a source. As Dr. Ernst put it:
The TACT has been criticized as being “unethical, dangerous, pointless and wasteful.” Yet, Lamas et al inform us that it went ahead with “reduced sample size” and that “TACT has finished enrolment.” From my perspective, the most puzzling part of the article of Lamas et al is the following sentence: “EDTA chelation of divalent and trivalent ions has been postulated to produce a favorable effect on atherosclerotic plaque, questionably leading to improvement in endothelial function, reductions in symptoms, and major vascular events.” To support this statement, Lamas et al cite my review that shows an “almost total lack of convincing evidence” and concludes that “given the potential of chelation therapy to cause severe adverse effects, this treatment should now be considered obsolete.”
Ernst was correct, and the now-revealed results of TACT only serve to confirm that.
The bottom line
When we criticize NCCAM and the infiltration of quackademic medicine into medical academia, we often point to the many pernicious effects that “integrating” pseudoscience with science- and evidence-based medicine has. One of these is the drive to test highly implausible therapies without adequate preclinical evidence, a practice at odds with the Helsinki Declaration’s requirement that clinical trials be based on firm basic science in preclinical models. This is problematic enough from an ethical standpoint when the treatment being tested is water (i.e., homeopathy), but when it’s an active treatment with real risks, it is completely unethical. That’s why I have said on multiple occasions that TACT is unethical. Worse, TACT was not funded based on a clinical need, scientific or clinical promise, or scientific merit due to its potential to reveal an important previously unsuspected mechanism of disease or target for treatment. Rather, it came into existence because a pro-quackery legislator, Rep. Dan Burton (R-IN) strong-armed the then-director of NCCAM to green light it. Later, it became such an albatross about NCCAM’s neck that NCCAM ceded control to NHBLI and downgraded its involvement to an advisory capacity. Indeed, Dr. Lamas himself seems to indicate that even he didn’t expect any positive results from TACT, although I suppose it’s possible that he means it was unexpected that this trial was, in essence, a negative trial:
“We have to look carefully at these unexpected results,” said Gervasio A. (Tony) Lamas, M.D., lead author of the study and chief of Columbia University Division of Cardiology at Mount Sinai Medical Center in Miami Beach, Fla. “Although not approved by the Food and Drug Administration for treating heart disease, chelation therapy has been used for over 50 years and has generally been believed by conventional medical practitioners and cardiologists to be without value. A definitive answer on chelation therapy will take much additional research. The most exciting part of this study is that there may be an unexpected signal of benefit. We need to understand whether the signal is true, or whether it occurred by chance.”
No, the “signal” only comes from having aggregated a bunch of outcomes into one large outcome, and even then this signal, in a study of over 1,700 patients, strained to reach statistical significance. On each and every individual outcome, the “signal” doesn’t exist!
Let’s step back a moment an look at this. In the case of TACT, the result of the “integration” of quackery with scientific medicine has been to spend $30 million on a trial conducted at highly dubious CAM and “integrative” medicine clinics whose practitioners were completely unqualified to carry it out. This expenditure of scarce research dollars has resulted in a primary finding that is at best equivocal and at worst completely negative, and a secondary finding that is completely negative. Even if this study’s results are taken at face value, we can say that chelation therapy for coronary artery disease does not increase survival, obviate the need for angioplasty or CABG surgery (a prime claim frequently made by chelationists), or even decrease the severity of patients’ angina symptoms or increase their tolerance for physical activity. It is worthless. Actually, it’s worse than worthless, because it’s expensive, arduous, and, even if we took TACT’s reported results at face value, promises minimal benefit.
I am not condemning this trial because it was negative. Sometimes—often, in fact—clinical trials fail to find a benefit from the experimental treatment. There’s nothing wrong with that. However, such clinical trials are based on a sound preclinical evidence base of basic science and animal experimentation that indicates scientific plausibility and a reasonable likelihood that the treatment would be efficacious in humans. TACT had none of these things. Indeed, there was an existing preclinical and clinical evidence base that gave every indication that chelation therapy shouldn’t work.
The acceptance of chelation among CAM practitioners is also as good an example of the CAM double standard as I’ve ever seen. Imagine, if you will, if big pharma produced a treatment like chelation therapy that had no good preclinical evidence suggesting its efficacy and several existing clinical trials suggesting that it does no better than placebo for cardiovascular disease. Imagine that big pharma tried to get FDA approval to market its chelation therapy for cardiovascular disease. Imagine how CAM practitioners would react. Now look for how they react to this trial. I can predict it. (Not that it’s hard or anything.) They’ll make excuses. They’ll cherry pick the one seemingly promising result. They’ll claim that there was something wrong with the protocol or that the wrong chelating agent was used. They’ll demand more studies. In other words, they won’t simply admit that their therapy doesn’t work and move on.
Same as it ever was. Forever and ever and ever, amen.