EDITORIAL: A Call To Arms For Medical Toxicologists: The Dose, Not The Detection, Makes The Poison

Charles A. McKay Jr. MD FACMT, FACEP, ABIM
Associate Professor of Emergency Medicine
Associate Medical Director, Connecticut Poison Control Center
University of Connecticut School of Medicine
Medical Director, Occupational Health Services
Department of Traumatology and Emergency Medicine
Hartford Hospital
80 Seymour St.
Hartford, CT 06102-5037
Phone: (860) 545-5411
Fax: (860) 545-2137
Email: cmckay@harthosp.org

Michael G. Holland, MD, FACMT, FACOEM, FACEP
Assistant Professor of Emergency Medicine
Penn State University College of Medicine
The Milton S. Hershey Medical Center
Occupational and Environmental Toxicologist
Penn State Poison Center
PO Box 850, MC H043
500 University Drive
Hershey, PA 17033-0850
Phone: (717) 531-7057
Fax: (717) 531-4441
Email: mholland@psu.edu

Lewis S. Nelson MD, FACMT, FACEP
Assistant Professor of Surgery (Emergency Medicine)
New York University School of Medicine
455 First Ave., Room 123, Box 81
New York City, NY 10016
Phone: (212) 447-8151
Fax: (212) 447-8223
Email: lnelson@pol.net

Int J Med Toxicol 2003; 6(1):1


Over the last several decades, the analytic capability to measure very small concentrations of an increasingly vast array of chemical structures has increased dramatically. Analytic chemists can now measure certain purported toxicants at a fraction of a part per trillion.[1] To give some idea of this level of detection, the proverbial "drop in a bucket" would be measuring things at the parts per million range; parts per trillion is equivalent to a "drop in a lake"!

Unfortunately, our ability to determine what to do with this data has not progressed as fast as the analytic technology. Although a tenet of toxicology is that "the dose makes the poison", many people inappropriately fear that the very detection of a substance must equate with toxicity. As medical toxicologists, we focus on the patient's symptoms and signs and their association with exposure and delivered dose. However, many of us are faced with patients coming from other practitioners with laboratory data from a multi-element panel indicating toxicity by mercury, arsenic, or other heavy metals or excesses or deficiencies of a wide array of trace elements or hydrocarbons (so-called environmental pollutants). These laboratory tests are often presented as de facto evidence of toxicity or "systemic imbalance or insufficiency" without any evidence of excessive dose or exposure. Furthermore these test results are then considered the cause of a variety of poorly characterized or general symptoms. Unfortunately, "environmental ecologists" and other practitioners[2] often use these test results, which we consider clinically irrelevant, as support for a variety of scientifically unproven or clinically non-indicated treatments.

We define esoteric testing to be uncommonly performed laboratory analyses for trace elements, environmental contaminants, or endogenous enzymes obtained from samples of blood, urine, hair or other body tissue. These tests or matrices generally lack a published reporting of validated reference ranges or suffer from significant procedural difficulties. While a large number of potentially valid analytes or methods may fall into this broad definition, the widespread use of certain testing panels and laboratories by certain groups of practitioners present obvious examples of aberrant practices with which we are all familiar. (the so-called "know it when you see it" definition of quackery).

We present the following composite case and a rationale for a proposed set of criteria to assist physicians in the decision to perform esoteric testing and in the interpretation and application of results already obtained.

Case Example

A 52 year old woman presents to the toxicology clinic complaining of generalized fatigue, difficulty with memory, and anxiety. There is a history of some weight loss over the last few months and difficulty sleeping. The patient is an ex-smoker and consumes occasional ethanol. A general physical exam is unremarkable, as is a neurological and mini-mental status exam. During further questioning, as the toxicologist formulates a wide differential (including a number of non-toxicologic diagnoses), the patient declares, "My other doctors found I was out of balance and have too much mercury in my system. I want to know if I should have my dental fillings removed because I don't feel much better after chelation." With further discussion, it becomes clear that the patient has been to a number of practitioners, some of whom have used "alternative practices" such as kinesiology to determine she has an excess of heavy metal contamination, while others have given courses of dimercaptopropane sulphonate (DMPS) followed by urinary mercury collection and hair mercury analysis.


While poisoning by a wide variety of naturally-occurring heavy metals or industrial contaminants is well-described, the "low-level" toxicity of mercury, arsenic, and other heavy metals is more problematic. Even for elements, such as mercury, where it is generally accepted that hair analysis is a valid analytic technique[3], proper collection, analysis and interpretation is still necessary. Furthermore, the distinction between public health concerns and individual toxicity is very important. For example, it is generally accepted that mercury contamination of the environment has contributed to an increase in the mercury concentration in marine animals. All states have health advisories regarding the consumption of fresh-water fish because of concerns about mercury (and PCB) contamination. Yet these advisories are focused on the possible risk for neurotoxicity for the unborn child of a pregnant woman. While various studies have raised questions about subtle population neurodevelopmental effects from amounts of mercury 10-100 times that of the average American diet (resulting in maternal hair mercury measurements far above what is commonly reported as abnormal by hair analysis laboratories), even these authors state that none of their subjects demonstrated clinical mercury poisoning.[4] Can we reassure the vast majority of patients with vague symptoms and abnormal heavy metal screens without glossing over the patient who is truly poisoned? We believe such a balance is possible and should be one component of the medical toxicologists' practice. On an individual basis, we can educate practitioners and the general populace in our area regarding some of the cautions to take with available laboratory testing. Each of the following points deserves careful consideration:

1) The decision to perform laboratory testing should be based on a differential diagnosis, rather than indiscriminate testing.

It is often tempting to run a large battery of tests on patients with poorly characterized or complex presentations. Patients who carry diagnoses such as chronic fatigue syndrome, multiple chemical sensitivity, fibromyalgia are especially prone to this type of testing, since these "conditions" are essentially symptom complexes and have no known organic or toxic etiology. Also, patients with chronic, progressive or incurable disorders such as multiple sclerosis and autism may be tested for toxicants. Some physicians will order trace mineral analyses searching for a cause of these syndromes, but many unscrupulous practitioners order these tests to "prove" to patients the need for chelation or other unnecessary, and potentially dangerous, "treatments". Unfortunately, this reliance on analytic testing is often misplaced. By pure chance, the statistical likelihood of finding a test result outside a population norm will increase as the number of tests increases. In the absence of good clinical correlation, these results are usually meaningless, but can cause a good deal of confusion and concern in both patient and physician.[5] As mentioned above, the dose determines toxicity. In addition, most toxicants produce a characteristic pattern of effects; this specificity of effect should be carefully sought in the history and physical exam, which then should guide testing patterns.

2) Critical methodological steps regarding specimen collection and laboratory analysis must be heeded.

All of these tests measure very small amounts of chemical compounds. As such, even low-level contamination of collection materials or procedures can result in false positive reports. This problem is well described with lead biomonitoring, where elevated capillary blood measurements from fingerstick testing must be confirmed with a venous sample because skin contamination with lead may result in falsely elevated blood levels. This can also occur with heavy metal testing of hair, due to external contamination by metals found in hair treatments, public water supplies or air pollution.[6] Similar problems arise with blood or urine collections.[7] In addition, dietary restrictions are necessary when analyzing body burden of heavy metals or trace elements to prevent false elevations from such agents as dietary supplements or seafood. As an example, the presence of largely non-toxic arsenobetaine and arsenocholine - "fish arsenic" - from seafood interferes with the assessment of arsenic exposure.[8] Although a further testing refinement (i.e. speciation of arsenic type) can be used for this element if there are concerns about the patient's dietary contribution, few laboratories provide this expensive service. Furthermore, this would not distinguish the contribution of arsenosugars that are present in marine algal products (often present in supplements).[9] Finally, many labs will analyze a urine specimen collected for six hours after a chelation challenge, and then compare this result with a norm based on a non-challenged collection. This result will almost always be higher than the non-challenged test but does not reflect an abnormal body burden of the presumed toxicant.[10,11,12] As an example, normal subjects may excrete several fold more mercury post-chelation than in their own pre-chelation test.[12] The results then are "flagged" as abnormal when in fact the testing has done little more than document a normal response to the chelator.

3) Laboratory tests should have well-validated reference ranges. These are lacking for many esoteric tests.

Population norms are often not standardized or are based on small numbers. In fact, some of these laboratories have developed their own reference ranges that are much lower than widely accepted ranges such as that published for hair mercury by the National Centers for Environmental Health of the CDC. This represents their belief that these toxins are more poisonous than mainstream medical science believes. The end result is many patients' results will be flagged as abnormal. In addition, accuracy is very poor for some analyses, such as hair testing by popular laboratories.[13] Many of these laboratories claim Clinical Laboratory Improvement Amendments (CLIA) certification, a federal standard for certain analytic tests, yet no such certification specifically exists for hair mineral analyses. Proficiency testing standards for hair testing do not exist, and individual labs devise their own verification methods and criteria for accuracy. Analytic laboratories should demonstrate some validity of testing, both internal (precision) and compared to standards (accuracy). Even when this is done,[14] information regarding measurements in a target population, such as those with known clinical effects from excesses or deficiencies of the given analyte, should be included.

4) Exceedance of a reference value does not necessarily imply that a patient is poisoned.

Interpretation of laboratory tests is best done in the clinical setting. Often additional clinical, epidemiological and laboratory data are necessary to establish a scientific basis for linking an elevated lab value with the presence or future risk of an adverse health outcome. In fact, for some elements and enzymes, the biologic or physiologic human health effects are not well characterized. As with the heavy metals, the effects of gross deficiencies (e.g. selenium)[15] or excesses (e.g. manganese)[16] are well described, while the effects of smaller variations from a population norm are less clear. Indeed, the experiences of certain unusual populations, such as two-three fold increases in serum manganese in patients receiving total parenteral nutrition, suggest no clinical adverse effects from these excesses.[17,18] Again, laboratories will often report determinations, usually in hair or red blood cells, compared to an unvalidated population norm, rather than as correlated with health or disease. Laboratories should provide normal ranges based on validated control populations. It is inappropriate for a laboratory to provide treatment recommendations. This is particularly true when the laboratory is associated with industries that distribute or otherwise promote treatments for the purported intoxications or deficiencies they claim to document.


In general, testing for heavy metals, nutritional elements present at extremely low concentrations, or so-called environmental contaminants, should only be obtained in the following situations and with the indicated precautions:

  • A properly performed clinical history and physical exam suggests the lack or excess of these chemicals or minerals/metals.
  • Proper patient preparation may include dietary avoidance of food and supplements that contain the substance of interest for several days prior to the sample collection.
  • The use of collections after chelation is usually unwarranted.
  • If post-chelation collections are used, the range of normals must be adjusted accordingly, and the results must be interpreted with extreme caution.
  • Collection should be done through a certified laboratory that is experienced in the collection and handling of these specimens to avoid contamination.
  • Analysis should be at a reputable laboratory that provides data on their normative population, including the selection and number of controls, and validation of their analytic procedures.
  • The laboratory should not provide treatment recommendations or sell therapy to the patient.

There are many factors to consider before ordering a large array of esoteric laboratory tests and a number of important considerations in the interpretation of these tests. The current popularity of broad trace element or pollutant screening with subsequent "detoxification" treatment, is often inappropriate. At this time, many of these tests are best utilized as research tools, such as the current population evaluations by the National Center for Environmental Health of the Centers for Disease Control and Prevention.[19] Application of these test results to individual patients is fraught with problems. Current concerns about environmental-related illness have been misappropriated by a number of practitioners to vindicate non-indicated treatments. A large portion of our toxicology clinic population is convinced their symptoms are due to poisoning, when neither their symptom complex nor laboratory testing justify such a conclusion. It is our contention that medical toxicologists should be at the forefront in the discussion regarding the appropriateness of toxicologic testing and its interpretation. In addition, we should be active in protecting patients from the misapplication of these tests.


The proceedings of an ATSDR panel on hair analysis have been published recently. The reference is: Harkins DK, Susten AS. Hair Analysis: Exploring the State of the Science. Environ Health Persp 2003;111:576-578.

  1. Olson ML, DeWild JF. Low-level collection techniques and species-specific analytical methods for mercury in water, sediment and biota. http://wi.water.usgs.gov/pubs/WRIR-99-4018-B/wrir-99-4018b.pdf (last accessed 9/12/01).
  2. http://www.acam.org/ (last accessed 1/10/03).
  3. ATSDR. Public health statement for mercury. http://www.atsdr.cdc.gov/toxprofiles/phs46.html (last accessed 3/26/02).
  4. Grandjean P, Weihe P, White RF, et al. Cognitive deficit in 7-year-old children with prenatal exposure to methylmercury. Neurotoxicol Teratol. 1997;19:417-428.
  5. Schneiderman LJ, DeSalvo L, Baylor S et al. The "abnormal" screening laboratory results. Its effect on physician and patient. Arch Intern Med 1972;129:88-90.
  6. Flanagan R. Hair today… Syva Drug Monitor 1999;3:29-31. http://ramindy.sghms.ac.uk/~ltg/dbehring.htm (last accessed 9/18/01).
  7. Wax PM, Goldfarb A, Cernichiari E. Mercury contamination of heavy metal collection containers. Vet Hum Toxicol 2000;42:22-25.
  8. Arbouine MW, Wilson HK. The effect of seafood consumption on the assessment of occupational exposure to arsenic by urinary arsenic speciation measurements. J Trace Elem Electrolytes Health Dis 1992;6:153-160.
  9. National Academy of Sciences 2001 Update. Arsenic in drinking water. National Academy Press; 2001:152. http://stills.nap.edu/books/0309076293/html (last accessed March 14, 2002).
  10. Gonzalez-Ramirez D, Maiorino RM, Zuniga-Charles M, et al. Sodium 2,3-Dimercaptopropane-1-Sulfonate challenge test for mercury in humans: II. Urinary mercury, porphyrins and neurobehavioral changes of dental workers in Monterrey, Mexico. J Pharm Exp Ther 1995; 272:264-274.
  11. Maiorino RM, Gonzalez-Ramirez D, Zuniga-Charles M, et al. Sodium 2,3-Dimercaptopropane-1-Sulfonate challenge test for mercury in humans: III. Urinary mercury after exposure to mercurous chloride. J Pharm Exp Ther 1996; 277:938-944.
  12. Dargan PI, Jones AL, Bogle RG, House IM. The effects of DMSA on urinary lead and mercury excretion in healthy volunteers. J Toxicol Clin Toxicol 2001;39(5):521 (abstract 122).
  13. Seidel S, Kreutzer R, Smith D, et al. Assessment of commercial laboratories performing hair mineral analysis. JAMA 2001;285:67-72.
  14. Bass DA, Hickok D, Quig D, et al. Trace element analysis in hair: factors determining accuracy, precision, and reliability. Altern Med Rev 2001;6(5):472-481. [Portions presented as abstract: Validation of trace metal analysis of hair for measurement by ICP-MS. Clin Chem 2001;47(6S):A64 (abstract 207)].
  15. Fryer MJ. Selenium and human health. Lancet 2000;356:943.
  16. Sato K, Ueyama H, Arakawa R, et al. A case of welder presenting with parkinsonism after chronic manganese exposure. Rinsho Shinkeigaku - Clinical Neurology 2000;40:1110-1115.
  17. Reimund JM, Dietemann JL, Warter JM et al. Factors associated to hypermanganesemia in patients receiving home parenteral nutrition. Clin Nutr 2000;19:343-348.
  18. Fitzgerald K, Mikalunas V, Rubin H, et al. Hypermanganesemia in patients receiving total parenteral nutrition. JPEN J Parenter Enteral Nutr. 1999;23:333-336.
  19. http://www.cdc.gov/nceh/dls/report/ (last accessed 9/18/01)

Journals Home  | Past Issues | Search | Send Comments to ACMTNet

Copyright 1999-2003, American College of Medical Toxicology.