Antibiotic Residues

Although most countries have banned the antibiotics chloramphenicol and nitrofurans from animal food production due to their toxicity to humans, traces of the drugs have been detected in shrimp and other aquaculture products. Chloramphenicol can cause potentially fatal aplastic anemia and leukemia, and nitrofurans are carcinogenic. Therefore, the seafood industry fully supports regulations that control the drugs to assure wholesome foods for consumers.
 
At issue in the current antibiotic residue debate is the distinction between detection and toxicity. Advancing analytical technology is allowing detection of substances at ever-diminishing levels. For example, heavy metals, pesticides and carcinogens can be highly toxic, yet are found in virtually all wholesome foods at trace levels. Studies have shown that such low levels are innocuous. With increasing analytical capability, it is unrealistic to expect foods to be free of any detectable level of hazardous substances.
 
In order to determine the point at which a hazardous substance presents a health risk, food safety experts have developed the concept of the maximum residue limit (MRL). This is the amount of residue considered to have no significant toxicological risk for human health. MRLs are based on "acceptable daily intakes," which in turn are typically based on "no observable adverse effects" levels derived from animal and in vitro trials.
 
Environmental Contamination
Increasing human population is contributing greater contamination to global water bodies. A recent study by the United States Geological Survey ("Pharmaceuticals, Hormones, and Other Organic Waste Water Contaminants in U.S. Streams, 1999-2000: A National Reconnaisance," published in the May 15, 2002 issue of Environmental Science & Technology) analyzed water samples from 139 streams across 30 states. Results indicated the presence of antibiotics in 48% of the samples at combined residual levels of 3.6 ppb.
 
The existence of such pervasive environmental contamination implies that it will not be possible to completely eliminate antibiotic residues from aquaculture products, even if antibiotic usage by aquaculture is completely stopped. Indeed, preliminary analyses of a variety of sea-caught products, as well as terrestrial animal foods revealed traces of chloramphenicol and nitrofurans.
 
A regulatory precedent exists for this type of situation with dioxins, which are also ubiquitous. Dioxin regulations reflect this reality by setting achievable limits -- not zero tolerance.
 
Regulatory Discrepancies
In the case of chloramphenicol and nitrofurans, insufficient toxicity data are available to establish MRLs. Consequently, regulations promulgated during the 1980s and 1990s banned the use of these antibiotics in food production and established a zero-tolerance policy. In other words, no detectable residues are permissible in animal foodstuffs. However, interpretation of zero tolerance varies widely among countries.
 
In Japan, the zero-tolerance threshold for chloramphenicol is defined as 50 ppb. In the United States, the Food and Drug Administration defined its zero-tolerance threshold as 5 ppb, which was the limit of detection technology at the time the regulation was developed. FDA has detected no chloramphenicol residues at this limit in approximately 40 samples of imported shrimp analyzed during the last year.
 
In the U.S., the Consumer Advocacy Group, Inc., based in Inglewood, California, has filed intention to sue 18 companies for selling imported shrimp in California that contained chloramphenicol and failing to label their products under the Safe Drinking Water and Toxic Enforcement Act of 1986, better known as Proposition 65. Before 2002, when the terms of this law changed, it was the responsibility of the alleged violator to prove his or her innocence. In the case facing shrimp companies, this means proving that no one was exposed to chloramphenicol residue on shrimp products. A business found in violation of Proposition 65 is subject to civil penalties up to $2,500 per day for each violation dating back at least a year.
 
To this point, no chloramphenicol residues have been detected in shrimp imported into the U.S., but all measurements have been based on FDA threshold levels. In the case of nitrofurans, the U.S. has not yet developed a standard detection method nor a threshold level for zero tolerance.
 
In the European Union, the zero-tolerance threshold is based on the minimum detectable limit. This is a moving target that continues to decline with advancing analytical technology. In recent months, shrimp have been rejected at chloramphenicol levels as low as 0.01 ppb. At detection levels of less than 1 ppb, hundreds of positive samples have been detected. This led to a ban on imports of Chinese products of animal origin. Shipments from other Asian countries are on a mandatory detention list. Not only is the European Commission rejecting shipments that exceed its tolerable limits, it recently stipulated that such shipments must be destroyed, rather than returned to the country of origin.
 
It is important to clarify that the E.C.'s chloramphenicol threshold levels are not based on new toxicity information. They are purely a reflection of advancing detection technology and application of the precautionary principle. The E.C. policy ignores the impact of these actions on importers and exporters in terms of disruption in trade and loss of income.
 
In the case of nitrofurans, zero tolerance in the E.U. is based on minimum detectable levels, which are typically 0.3 ppb at this point.
 
Dutch Clinical Studies
Little quantitative risk data is available on chloramphenicol and furizolidone toxicity, but studies conducted in the Netherlands help to give an order-of-magnitude evaluation of the risk of consuming contaminated shrimp.
 
Research with mice has shown that chloramphenicol exposure increases the risk of cancer in one out of 1 million individuals when the lifelong exposure level is 1-5 ug/kg/day. Assuming an average weekly shrimp intake of 8.4 g, chloramphenicol contamination of 10 ug/kg and average body weight of 70 kg, the chloramphenicol exposure would be 0.17 ng/kg/day. This is 5,000-fold less than the dose projected to increase cancer risk in one out of 1 million individuals.
 
In a similar Dutch study, furazolidone increased the risk of cancer in one out of 1 million individuals when the lifelong exposure level was 100 ng/kg/day. Assuming an average weekly shrimp intake of 8.4 g, furazolidone contamination level of 22 ug/kg, and average female body weight of 65 kg, the furadolidone exposure would be 0.4 ng/kg/day. This is 250 times less than the dose projected to increase cancer risk in one out of 1 million individuals.