The full content of Annals is available to subscribers

Subscribe/Learn More  >
Current Clinical Issues |

Balancing the Risks and Benefits of Fish Consumption

Jennifer Fisher Wilson
[+] Article and Author Information

Potential Financial Conflicts of Interest: None disclosed.

Ann Intern Med. 2004;141(12):977-980. doi:10.7326/0003-4819-141-12-200412210-00024
Text Size: A A A

Most Americans know that eating fish is good for their health. Studies from the past 2 decades have repeatedly linked the consumption of fish—especially fish that is high in ω-3 fatty acids—with healthier hearts in the aging population (1). Scientists have also found associations between fish consumption and a reduced risk for stroke, dementia, asthma, kidney disease, and diabetes. Americans have responded to the news: U.S. fish consumption per capita increased 50% since 1980, according to the U.S. Department of Agriculture. Consumption of salmon alone, the country's third most popular fish, increased 23% between 1987 and 1999.

First Page Preview

View Large
First page PDF preview





Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).


Submit a Comment
Incorrect Statements Regarding Mercury and PCBs in Fish
Posted on December 22, 2004
Dr. Luanne K. Williams
North Carolina Department of Health and Human Services
Conflict of Interest: None Declared

First, the article incorrectly states on page 977 that methylmercury reaches its highest levels in bottom-feeders such as crab. Actually, high levels generally occur in long-lived predatory fish and not in bottom feeders. FDA and North Carolina recommend that women of childbearing age and children consume crab because of the low methylmercury levels reported in crab (see North Carolina's website http://www.epi.state.nc.us/epi/fish/whatfisharesafe.pdf and FDA's websites at http://www.epa.gov/waterscience/fishadvice/advice.html and http://www.cfsan.fda.gov/~frf/sea-mehg.html). FDA collected 59 blue, king and snow crab samples between 1990 and 2002. The average methylmercury level for these species is 0.06 mg/kg which is the same average level reported for pollock, another species recommended by FDA and North Carolina for consumption for women of childbearing age and children. These levels are well below North Carolina's action level for issuing fish consumption advisories of 0.4 mg/kg.

Second, physicians should be made aware of the long list of predatory species that have been reported to have high methylmercury levels (average methylmercury level of 0.4 mg/kg and higher). These predatory marine species include shark, swordfish, king mackerel, Spanish mackerel, tilefish, canned white tuna (albacore), fresh/frozen tuna, grouper (gag, scamp, red, and snowy), marlin, orange roughy, ladyfish, cobia, crevalle jack, greater amberjack, almaco jack, banded rudderfish, and little tunny (see Florida data at http://www.floridamarine.org/education/view_article.asp?id=20125 and FDA data at http://www.cfsan.fda.gov/~frf/sea-mehg.html). Some freshwater species have also been reported to have high methylmercury levels in some states such as largemouth bass, bowfin (blackfish), and jack fish (chain pickerel).

Third, because fish low in methylmercury are good for the developing brain in children and heart for adults, physicians should encourage women of childbearing age (15-44 years) and children (under age 15) to eat at least two meals a week of fish low in methylmercury and the general public to eat at least four meals a week of fish low in methylmercury. Fish that have been reported to have low methylmercury levels or average levels below 0.4 mg/kg are croaker, cod, whitefish, pollock, mahi-mahi, ocean perch , halibut, haddock, flounder, spot, speckled trout (spotted seatrout), herring, kingfish (sea mullet), canned light tuna, shrimp crab, lobster, clams, oysters, scallops, farm-raised fish, trout, crappie, sunfish, white perch, yellow perch, bream, salmon, butterfish, jacksmelt, tilapia, sheepshead, skate, pompano, tripletail, white grunt, pigfish, and black drum (see Florida data at http://www.floridamarine.org/education/view_article.asp?id=20125 and FDA data at http://www.cfsan.fda.gov/~frf/sea-mehg.html).

Fourth, the article incorrectly states on page 977 that methylmercury is released mainly through industrial practices including the burning of fossil fuels and solid wastes. It is the inorganic forms of mercury (e.g., elemental mercury and mercuric chloride) that are released from coal fired power plants not the organic form of mercury or methylmercury. The inorganic forms of mercury are released from industrial pollution and then eventually are deposited into surface water, accumulating in streams and oceans. Bacteria in the water cause chemical changes that transform the inorganic forms of mercury into methylmercury. It is this organic form of mercury that is accumulated in fish and can be toxic. Fish absorb methylmercury from water and their food as they feed on aquatic organisms The larger, predatory fish accumulate higher concentrations of methylmercury (see FDA website http://vm.cfsan.fda.gov/~dms/admehg.html and http://www.epa.gov/ttn/oarpg/t3/reports/volume3.pdf).

Fifth, the PCB levels reported for farm raised salmon of 30 parts per billion or 0.030 parts per million is a very low level. At this level, the Great Lakes, Connecticut, North Carolina and other states would not issue a fish consumption advisory. Using US EPA standard equations and assuming a 50% loss from cooking, a person could safely eat 3 meals a week which is calculated as follows (http://www.epa.gov/ost/fishadvice/volume2/v2ch2.pdf; http://www.epa.gov/iris):

0.00002mg/kg-day(US EPA reference dose)x70 kg/0.03 ppm = 0.046 kg fish/day 0.046 kg fish/day x 30.44 day/mos = 1.4 kg fish/mos 1.4 kg fish/mos x 1 meal/0.227 kg fish (8 oz meal) = 6 meals/mos

6 meals/mos x 2 (assuming 50% loss in cooking) = 12 meals/mos 12 meals/mos x 1 mos/4 weeks = 3 meals/week

Eating 3 meals a week of just farm raised salmon would be safe and the exposure dose would be equal to EPA's recommended dose of 0.00002 mg/kg-day. The cancer risk at this recommended meal limit of 3 meals a week of salmon with average levels of 0.03 ppm would correlate to an excess cancer risk of approximately 1 in 15,000 which is within US EPA target risk range of 1 in 10,000 to 1 in a million. The cancer risk can be calculated as shown (http://www.epa.gov/iris):

2.0 risk/mg/kg-day x 0.00002 mg/kg-day = 0.6/15,000 or approximately 1 out of 15,000

Thank you for the opportunity to provide comments.

Sincerely Dr. Luanne K. Williams, Toxicologist Medical Evaluation and Risk Assessment Unit Occupational and Environmental Epidemiology Branch North Carolina

Conflict of Interest:

None declared

Transcriptional Profiling in Methylmercury Toxicity
Posted on December 30, 2004
Russell A. Wilke
Center for Human Genetics, Marshfield Clinic, Marshfield, WI
Conflict of Interest: None Declared

Dear Editor:

JF Wilson and the editorial staff at Annals are to be applauded for their recent discussion of the risks and benefits of fish consumption [1]. Dr Carpenter and Dr Goldman are to be congratulated on their ongoing efforts to lend clarity to this controversial field. Part of the uniqueness of organic mercurials as neurotoxicants can be assigned to their lipophilicity (hence their potency) [2, 3]. While it is true that "symptoms have usually been described in adults with [mercury] blood levels of 200 mg/L or greater, and sometimes in those with blood levels as low as 50 mg/L," we observed wide variability in the concentrations of methylmercury that were necessary for the induction of neuronal apoptosis in vitro [4]. Through transcriptional profiling studies conducted in animal models, we verified that methylmercury-induced neurotoxicity is dose-dependent [4]. In light of the variability in patient response discussed by Dr Goldman and Dr Kales, this raises several interesting questions about population-based potency issues for this potentially lethal neurotoxicant.

Dr Goldman insightfully comments that a variety of factors "might influence individual response to mercury exposure"¦" and she lists genetics first among these factors. Our transcriptional profiling data suggest that variability in the oxidative stress response is a pivotal contributor to a neuron's decision whether or not to undergo apoptosis following methylmercury exposure [4], and recent studies conducted by others have suggested that a number of polymorphisms exist within the human oxidative stress response genes [5].

We agree with Annals that "the public needs robust prospective studies in populations in which fish is a major part of the diet," and we propose that large-scale, population-based toxicogenetic studies are also needed [6] to define potential markers for risk stratification.

Russell A. Wilke, MD, PhD Center for Human Genetics Marshfield, Wisconsin USA References:

1. Wilson JF: Balancing the risks and benefits of fish consumption. Annals of Internal Medicine 141: 977-980, 2004 (December 21).

2. Windebank AJ: Toxic Neuropathy. In: Adams JH, Duchen LW, editors. Greenfield's neuropathology. 5th Edition. New York: Oxford Press; 2002.

3. Wilke RA and Moyer TP: Neurotoxic Metals. In: Noseworthy JH, editor. Neurological Therapeutics "“ Principles and Practice. London: Martin Dunitz; 2003.

4. Wilke RA, Rahimi RA, Kolbert CP, Windebank AJ: Methylmercury induces apoptosis in cultured rat dorsal root ganglion neurons. Neurotoxicology 24: 369-378, 2003.

5. Landi S, Gemignani F, Gioia-Patricola L, Chabrier A, Canzian F. Evaluation of a microarray for genotyping polymorphisms related to xenobiotic metabolism and DNA repair. Biotechniques 35: 816-20, 2003.

6. McCarty C, Wilke RA, Giampietro P, Wesbrook S, Caldwell MD: The Marshfield Clinic Personalized Medicine Research Project (PMRP) "“ design, methods and initial recruitment results for a population-based DNA biobank. (in press)

Conflict of Interest:

None declared

Non fish sources of long chain omega 3-fatty acids
Posted on January 12, 2005
Edward B. Nelson
Martek Bioscience
Conflict of Interest: None Declared

Dear Sir:

The discussion 'Balancing the Risks and Benefits of Fish Consumption' (1) highlighted the issue of the oceanic contaminants, mercury and PCB's, associated with consuming fish to obtain adequate intake of 'fish oil'. Adequate intake of these oils containing omega-3-long chain fatty acids (LCPUFA) such as docosahexanoic acid,(DHA),is a recommendation of the American Heart Assocation(2). Unfortunately the discussion was incomplete in describing a solution noting "physicians are not certain how to guide their patients". No options were given when in fact a viable alternative is available.

What the article did not detail is that those benefical LCPUFA such as DHA found in 'fish oil' are in fact originally from microalgae that are ingested by fish and ultimately harvested as oil from the fish remains. The microalgae are the first step in the food chain(3). An alternative source of LCPUFA such as DHA is to produce it directly by fermentation processes from pure cultures of these algae. This process is currently used to supply DHA for all major brands of infant and preterm infant formula that are fortified with DHA. As such it has met the FDA standards for purity and safety in those products targeted to the most vulnerable of human populations. This LCPUFA is free of oceanic contaminants and is not from a genetically modified organism. These products are available in capsule form in retail outlets in the U.S. Physicians can recommend these products as an alternative source of LCPUFA knowning oceanic contamination is not an issue.

1.Wilson,JF; Balancing the Risks and Benefits of Fish Consumption. Ann Intern Med.2004;141:977-980

2. Kris-Etherton PM,Harris WS,Appel LJ; American Heart Association. Nutrition Committee. Fish Consumption,fish oil,omega-3-fatty acids, and cardiovascular disease.Circulation.2002;106(21):2747-57. Erratum in : Circulation.2003;107(3):512

3. Sergent J,Bell G, McEvoy L,Tocher D, Estevez,A; Recent developments in the essential fatty acid nutition of fish. Aquaculture.1999;177:191-199

Conflict of Interest:

authors are employees of Martek Biosciences Corporation

Comments on Ms. Wilson's piece
Posted on January 26, 2005
Michael Mogadam
No Affiliation
Conflict of Interest: None Declared

Letter to editor:

Ms. Fisher Wilson's article "Balancing the risk and benefits of fish consumption", in the Current Clinical Issues (1) contains numerous inaccuracies and anecdotal (mis)information that by virtue of appearing in our esteemed Journal, may be sited by news media and self-appointed nutritional gurus to further undermine public's trust in our food supply. For example: "you don't have to eat carcinogens and neurobehavioral toxicants in order to get your omega-3 fatty acids"; "Many Americans are believed to have dangerous levels of methylmercury in their bodies"¦"; "Any more than a single 8-ounce portion of farmed salmon a month poses an unacceptable cancer risk to consumers"; eat only "up to 12 ounces (2 average meals) per week of a variety of fish that are lower in mercury"¦"; and that flaxseed, and canola or soybean oils are good "alternatives" to seafood, "¦ First, what is a carcinogen in a test tube, a cell line or a rodent is not necessarily a carcinogen in humans. Based on data from the US Environmental Protection Agency most "contaminated" lake or farm-raised fish contain 1-5 parts per billion (PPB) of PCB. However, a few analyzed samples contained as much as 40 PPB. Even if we assume that every contaminated fish has 40 PPB, a most implausible assumption, I calculated that one would have to eat approximately 25,000 Kg of such heavily contaminated fish or 0.7 Kg (about 1½ lb) every day for 100 years to receive one gram of PCB! I have not found any study to show that even this amount of PCB can cause any cancer in humans. This explains why there is no record of a single case of any cancer in the US, directly or indirectly attributed to consumption of PCB-contaminated seafood.

Second, the methylmercury phobia is a perpetual myth that has no relevance to non-pregnant, non-lactating adults, millions of whom are at high risk for cardiovascular diseases. A recent meta analysis of 11 studies including 222,364 cases with an average follow-up of 11.8 years showed that consumption of seafood once per week, 2-4 times per week, or >5 times per week was associated with a 15%, 23% and 38% lower CHD mortality, respectively (2). Each 1 oz /day increase in fish intake reduced cardiovascular mortality by 10%. Frequent seafood consumption can also reduce the risk of sudden cardiac death by more than 70% (3); no currently available drug(s) can provide such a dramatic cardioprotection. In addition, among 4775 men and women 65 years or older who were followed- up for12 years, the risk of ischemic stroke was 30% lower in those who ate tuna or other broiled or baked fish 5 or more times a week compared with an intake of less than once per month (4). Over the past 30 years more than 20 million Americans have died of cardiovascular diseases, many in their 30s, 40s or 50s; approximately 5-7 million of whom could have been saved by eating 4-5 seafood meals a week. Yet, during the same period, not a single person has ever died of methylmercury poisoning from seafood. In fact, the recent American Heart Association Scientific Statement recommends that patients without coronary heart disease should eat a variety of fish at least twice a week and those with CHD "should consume > 1g of EPA and DHA per day, preferably from oily fish" (5). Ms Fisher Wilson's dire warnings that consumption of seafood by pregnant women causes "brain damage, mental retardation, blindness and seizures" are also inaccurate and dated. Several recent studies have challenged this anecdotal boomerang. For example, in a 9-year follow-up of 779 mother- infant pairs from Seychelles, where mothers eat, on average, 12 seafood meals per week (>10 times higher than US women), and the methylmercury concentration of seafood is similar to that in the US, there was no evidence of any neurodevelopmental abnormalities that Ms. Fisher Wilson is scaring us with (6). In fact, among 7421 British children, frequent fish intake by mothers during pregnancy and by children was associated with higher developmental scores (7). Also, contrary to her alarmist statement, in the 2001-2002 survey of US women of childbearing age (16-49 years), only 3.9% (vs.7.8% in the 1999-2000) had blood mercury levels that were slightly above the very rigid Environmental Protection Agency's "safe" level of 5.8 µg/l, "below which exposures are considered to be without adverse effects" (8). Even in frequent fish eaters the average level was < 2µg /l, and in children the levels were 3-4 times lower than in women; hardly anywhere near "dangerous" levels.

Third, the vast benefits of omega-3 fatty acids (cardiovascular and others) are due to seafood's highly unsaturated and longer chain omega-3 fats with 20 and 22 carbons in their chain (Eicosapentaeonic acid- EPA=20:5n-3, and Docoasahexaeonic acid-DHA=22:6n-3). Plant omega-3 fatty acid (á-linolenic acid =18:3n-3) has 18 carbons and requires both elongation and desaturation to convert it to EPA and DHA. Although some animals have the necessary elongases and desaturases to accomplish this, humans do not! Thus no more than 1-2 % of á-linolenic acid is converted to EPA and only a negligible amount to DHA. During pregnancy (in a very high estrogen environment), women can convert as much as 5-8% of á-linolenic acid to EPA (a fraction of which is then converted to DHA), thus providing the fetus with these essential fatty acids (9). Although EPA and especially DHA have anticarcinogenic effects against breast, colon and prostate cancers, several studies have shown that high consumption of á- linolenic acid (such as flaxseed oil) increases the risk of advanced prostate cancer (RR 2.02, CI: 1.35-3.03), whereas DHA decreases the risk (RR 0.74, CI: 0.49-1.08) (10). One plausible mechanism is that á- linolenic acid competitively blocks the entry of EPA and DHA into prostatic cell membrane, and deprives the prostate from the anti- inflammatory and anti-carcinogenic action of dietary EPA and DHA. Thus, plant omega-3 fatty acid is not and cannot be considered a good "alternative" to seafood consumption. As physicians (and scientists) we must be able to act as knowledgeable resources for the public to turn to for valid medical and nutritional information. We should not use anecdotal or misleading and sensational sound bites as our own resources.

Michael Mogadan, MD, FACP, FACG Alexandria, VA Michmoga@aol.com


1. Fisher Wilson J. Balancing the risks and benefits of fish consumption. Ann Intern Med 2004; 141: 977-80 2. He K, Song Y, Daviglus ML et al. Accumulated evidence on fish consumption and coronary heart disease mortality. Circulation 2004; 109: 2705-11 3. Albert CM, Campos H, Stampfer MJ et al. Blood levels of long chain n-3 fatty acids and the risk of sudden death. N Eng J Med 2002; 346: 1113-8 4. Mozaffarian D, Longstreth WT, Lemaitre RN, et al. Fish consumption and stroke in elderly individuals. The Cardiovascular Health Study. Arch Intern Med 2005; 165: 200-06 5. Kris-Etherton PM, Harris WL, Appel LJ for the AHA Nutrition Committee. AHA Scientific Statement. Circulation 2002; 106: 2747-57 6. Myers GJ, Davidson PW, Cox C et al. Prenatal methylmercury exposure from ocean fish consumption in Seychelles Child Development study. Lancer 2003; 361: 1686-92 7. Daniels JL, Longnecker MP, Rowland AJ, Golding J. Fish intake during pregnancy and early cognitive development of offspring. Epidemiology 2004; 15: 394-402 8. Schober SC, Sinks TH, Jones RL et al. Blood mercury levels in US children and women of childbearing age. JAMA 2003; 289: 1667-74 9. de Groot RHM, Honstra G, van Houweelingen AC, Roumen F. Effect of á- linolenic acid supplementation during pregnancy on maternal and neonatal polyunsaturated fatty acids status and pregnancy outcome. Am J Clin Nutr 2004; 79: 251-60 10. Leitzman MF, Stampfer MJ, Michaud DJ et al. Dietary intake of n-3 and n-6 fatty acids and the risk of prostate cancer. A

Conflict of Interest:

None declared

Comments on Fish COnsumption
Posted on January 28, 2005
Alan H. Stern
University of Medicine & Dentistry of New Jersey - School of Public
Conflict of Interest: None Declared

To the Editor;

In general, the recent Current Clinical Issues article "Balancing the Risks and Benefits of Fish Consumption" was informative and useful to physicians in understanding the issues facing their fish consuming patients. However, as a member of the former National Academy of Sciences/National Research Council committee responsible for the recommendations for exposure to methylmercury referenced in the article, I believe that it is necessary to clarify information in the article regarding the risk to patients with elevated levels of methylmercury. The article states that "Many Americans are believed to have dangerous levels of methylmercury in their bodies...Levels higher than 5 ug/L in blood or higher than 1 ug/g in hair are potentially hazardous to the developing fetus...This level corresponds to a reference dose of approximately 0.1 ug/kg of body weight per day of methylmercury exposure." This reference dose is, in fact specifically defined by the U.S.EPA as "An estimate (with uncertainty spanning perhaps an order of magnitude) of a daily oral exposure to the human population (including sensitive subgroups) that is likely to be without an appreciable risk of deleterious effects during a lifetime." Thus, exposure at the reference dose, and the related blood and hair concentrations of mercury, corresponds to a virtual safe level of exposure. Exposure exceeding this dose, exceeds this safety benchmark, but does not necessarily imply "danger." In general, patients with moderate exceedences of this benchmark, particularly pregnant women, and women planning to become pregnant, should be encouraged to reduce their exposure by substituting high mercury fish in their diets with fish with characteristically lower levels of mercury. The reference dose is a useful warning sign of elevated exposure, and a prompt for prudent modification of the patient's diet. However, the reference dose should not be interpreted as a bright-line indication of clinical "danger" leading to drastic action on either the physician's or patient's part.

Alan H. Stern, Dr.P.H., D.A.B.T.


Adjunct Associate Professor

Department of Environmental and Occupational Medicine University of Medicine & Dentistry of New Jersey - School of Public Health

Conflict of Interest:

None declared

Submit a Comment

Summary for Patients

Clinical Slide Sets

Terms of Use

The In the Clinic® slide sets are owned and copyrighted by the American College of Physicians (ACP). All text, graphics, trademarks, and other intellectual property incorporated into the slide sets remain the sole and exclusive property of the ACP. The slide sets may be used only by the person who downloads or purchases them and only for the purpose of presenting them during not-for-profit educational activities. Users may incorporate the entire slide set or selected individual slides into their own teaching presentations but may not alter the content of the slides in any way or remove the ACP copyright notice. Users may make print copies for use as hand-outs for the audience the user is personally addressing but may not otherwise reproduce or distribute the slides by any means or media, including but not limited to sending them as e-mail attachments, posting them on Internet or Intranet sites, publishing them in meeting proceedings, or making them available for sale or distribution in any unauthorized form, without the express written permission of the ACP. Unauthorized use of the In the Clinic slide sets will constitute copyright infringement.


Buy Now

to gain full access to the content and tools.

Want to Subscribe?

Learn more about subscription options

Related Articles
Topic Collections
Forgot your password?
Enter your username and email address. We'll send you a reminder to the email address on record.