Science News Magazine
Week of Dec. 21, 2002; Vol. 162, No. 25
In the spring of 2000, one of Jane M. Hightower's patients had been concerned about hair loss, so the internist referred the woman to a specialist in her building. That dermatologist probed the woman's medical history but could find no explanation. That is, until she suddenly recalled a radio broadcast about mercury poisoning in people who had been eating lots of fish from tainted lakes. Their symptoms included hair loss.
So, the dermatologist asked her patient if she ate much fish. Indeed, the woman said, she loved it. The doctor quickly arranged for the woman to get a blood test and then faxed the results back to Hightower. After reviewing the findings, which suggested the patient's mercury concentrations were in fact somewhat elevated, Hightower put the document atop the papers in her in-box.
Which is where it was still sitting, when a patient came in complaining, "My house is poisoning me!"
Hightower listened as the woman described how she sometimes felt so enervated that she could barely summon the will to get out of bed. Other times, especially while traveling abroad for months on end, the woman felt fine.
Oh yes, one other thing: The patient's thinning hair had become such a problem that the woman turned to Rogaine. She told Hightower she had been using this antibalding drug for 2 years.
Glancing at the in-box and her other patient's mercury data, Hightower asked whether her new patient ate much fish. "And she said, 'Yes, as a matter of fact—nine times a week,'" Hightower recalls. This "serendipitous" pairing of cases launched the doctor on a quest to understand whether a taste for fish might be poisoning any of her other patients.
For the next year, Hightower formally surveyed the fish-consumption patterns of every person who came through her practice. Among those 720 people, 123 appeared to be eating fairly high concentrations of fish.
She then convinced 113 of these fish eaters—several of whom also showed symptoms indicative of possible mercury poisoning—to get tested for the metal. All but seven had blood drawn for testing. The remainder, including several children, submitted only their hair for testing. Most of tested individuals exhibited elevated mercury concentrations despite having little or no known exposure to mercury besides eating fish, report Hightower and Dan Moore of the California Pacific Medical Center, also in San Francisco, in an upcoming issue of Environmental Health Perspectives. Among the patients who had blood tests, 89 percent had blood concentrations exceeding 5 micrograms per liter ( m g/L). Indeed, 16 percent had blood concentrations over 20 mg/L of blood—and 4 individuals surpassed 50 mg/L.
Because fetal exposure to mercury can later play out as IQ deficits, the National Academy of Sciences in 2000 recommended that women of childbearing age should try to keep mercury concentrations in their blood to less than 5 mg/L(or hair concentrations to below 1 mg/L). They didn't address other parts of the population.
Hightower advised all her patients with blood or hair values well above those cutoffs to pare fish from their diets over the next few months. And though follow-up blood tests showed that their bodies indeed began shedding mercury, the drop was slow. In some cases, even 21 weeks later, the patients' mercury concentrations remained elevated well above the NAS guideline figures.
Among adults, most symptoms abated as their blood concentrations dropped. Alas, Hightower says, that didn't spare one child, who initially was screened with nearly 15 times the NAS recommended ceiling concentrations for mercury. Hightower notes that this boy had experienced a documented "mental decline" during the 4 years he had regularly been eating not only canned tuna but also fresh tuna and salmon steaks. Though his parents eventually purged fish from his diet, the boy retains a significant neurological impairment, Hightower says.
Since her initial study ended, she has continued to evaluate fish consumption in her patients. Another 60 or so of them turned out to be at risk for subtle mercury poisoning. Perhaps most troubling, Hightower told Science News Online, was that her patients—much like herself—had viewed fish as a healthy food. Study after study had extolled the heart benefits of fish-rich diets. She asked: How could her patients have been so seriously misled? Why weren't they aware that this food can also serve as the vehicle for a potent poison?
In a Nov. 20 letter to President George Bush, she asked for actions to help consumers avoid unnecessary exposure. For instance, she requested that the government continue testing fish for mercury tainting and that the results—and any necessary fish advisories about mercury—"be readily available where fish are sold." But they ate pricey fish. . .
That fish can serve as a dietary vehicle for bringing mercury to the dinner table is hardly new. Mercury is the most commonly cited basis for state warnings that locally caught fish might be dangerous to consumers' health.
However, Hightower says, those advisories generally addressed only freshwater species caught by noncommercial anglers from especially tainted waters. Her patients were eating primarily marine fish. Moreover, these bankers, scientists, physicians, business executives, investment brokers, and Internet entrepreneurs weren't hauling in their own catch of the day. They either ordered it from the counter of a local food retailer or from the menus of white-tablecloth restaurants.
A message that federal health officials have failed to effectively communicate to the public, she says, is that many large, predatory, and long-lived oceanic species also accumulate plenty of heavy metals, including mercury. Many of Hightower's patients noted that they had been selecting precisely these large, predatory marine species because they tasted least fishy and their bones were easy to remove.
Overall, elevated mercury readings among her patients tended to correlate most strongly with any consumption of swordfish. However, many with high mercury scores also ate plenty of tuna—especially steaks—and salmon.
The heart of the matter
In her readings on health effects of mercury, Hightower ran across a 1999 Italian study in the Journal of the American College of Cardiology. It described finding highly elevated concentrations of mercury in heart—but not other muscle—of patients who had died from heart failure related to a condition known as idiopathic dilated cardiomyopathy. Because none of the patients had known elevated exposures to mercury, the data hinted that heart muscle might selectively accumulate the metal, leading to its selective poisoning.
On November 28, the New England Journal of Medicine published two epidemiological studies offering further support for a heart sensitivity to methylmercury—the organic form of the metal found in fish.
In one international study probing cardiovascular risks, Eliseo Guallar
of the Johns Hopkins Medical Institutions and his colleagues correlated
risk of first
heart attack with toenail concentrations of mercury and concentrations of a fish oil (docosahexaenoic acid, or DHA) in body fat. Their data came from 684 men who had had a heart attack and another 724 who hadn't.
In this study, increasing concentrations of mercury in toenails—which serve as a relatively long-term record of exposure—were "directly associated" with increasing risk of heart attack, the study found, whereas DHA concentrations in body fat appeared protective against heart attack. Guallar and his colleagues say that their data suggest that mercury tainting of fish diminishes the cardioprotective effect normally associated with heavy consumption of DHA and oily fish.
The authors noted that they had not collected information on the sources
of mercury or DHA among their participants—nor data on fish intake. However,
they noted, the substantial DHA concentrations measured in some subpopulations
of the participants would suggest their mercury likely derived
from consumption of marine fish.
To date, health advisories against eating mercury-tainted fish have tended to focus on pregnant women and children, with a goal of protecting the neurological development in youngsters, Guallar's group observes. "Our results raise the possibility that this advice should be extended to the general adult population," the researchers say. They recommend that people should not eschew fish, just judiciously choose species that are not likely to be heavily contaminated.
According to a table of data that the U.S. Food and Drug Administration compiled nearly 2 years ago, tilefish, swordfish, shark, and king mackerel lead the list with mean mercury concentrations of between 0.7 and 1.4 parts per million (ppm). Although the agency had fewer samples from a number of other popular marine species, among them red snapper, moonfish, orange roughy, marine bass, and marlin also tended to be fairly heavily tainted, typically averaging 0.4 to 0.6 ppm.
FDA reported somewhat lower—but still far from negligible—mercury tainting in grouper, tuna, halibut, pollock, cod, whitefish, and herring. All were down in the 0.2 to 0.15 ppm range. Canned tuna had less contamination than fresh or frozen. Some shellfish also fall in that category, with lobster containing more mercury than crab.
Seafood with the least mercury contamination includes tilapia, salmon, shrimp, oysters, clams, sole, and flounder.
Bolger, P.M., and B.A. Schwetz. 2002. Mercury and health. New
England Journal of Medicine 347(Nov. 28):1736.
Frustaci, A., et al. 1999. Marked elevation of myocardial trace elements
in idiopathic dilated cardiomyopathy compared with secondary cardiac
function. Journal of the American College of Cardiology
Guallar, E., et al. 2002. Mercury, fish oils, and the risk of myocardial
infarction. New England Journal of Medicine 347(Nov. 28):1747-1754.
Abstract available at
Hightower, J.M., and D. Moore. In press. Mercury levels in high-end
consumers of fish. Environmental Health Perspectives. Abstract
available at http://ehpnet1.niehs.nih.gov/docs/2003/5837/abstract.html.
Yoshizawa, K., et al. 2002. Mercury and the risk of coronary heart
disease in men. New England Journal of Medicine 347(Nov.
28):1755-1760. Abstract available at
2002. Mercury associated with risk of heart attack. Johns Hopkins
University press release. Nov. 27. Available at
Center for Food Safety and Applied Nutrition, Office of Seafood. 2001.
References: Consumer advisory on methylmercury in commercial
seafood. U.S. Food and Drug Administration (May). Available at
Raloff, J. 2001. Landfills make mercury more toxic. Science News
160(July 7)4. Available at
______. 2001. A dietary cost of our appetite for gold. Science News
Online (May 12). Available at
______. 2000. China: A mercury megapolluter. Science News 158(July
______. 2000. Methylmercury's toxic toll. Science News 158(July
______. 1991. Mercurial risks from acid's reign. Science News
Center for Food Safety and Nutrition
Food and Drug Administration
5100 Paint Branch Parkway
College Park, MD 20740-3835
Web site: http://www.cfsan.fda.gov/
Welch Center for Prevention, Epidemiology, and Clinical Research
Johns Hopkins Medical Institutions
2024 E. Monument Street, Suite 2-639
Baltimore, MD 21205-2223
Jane M. Hightower
2100 Webster Street, Suite 418
San Francisco, CA 94115
Harvard School of Public Health
Boston, MA 02115
Acrylamide—From Spuds to
Gingerbread (Dec. 14, 2002)
Home Cooking on the Wane (Dec. 7, 2002)
Diets to Ward Off Diabetes (Nov. 30, 2002)