Schecter, A, HT Quynh, M Pavuk, O Päpke, R Malisch and JD Constable. 2003. Food as a source of dioxin exposure in the residents of Bien Hoa City, Vietnam. Journal of Occupational and Environmental Medicine 45:781–788.

30 years after Agent Orange spraying ended in Vietnam, a team of research scientists reports that some food being eaten by Vietnamese today remains heavily contaminated by dioxin and related chemicals. Their results help understand why high levels of dioxin continue to be found in Vietnamese, decades after the initial exposures. They also warn that Vietnamese children today, even though never directly exposed to spraying, are at risk to the health effects of dioxin because of the food they eat.

Not all types of food are problematic. While duck was exceptionally high, and chicken and some fish far above thresholds for health concerns, beef and pork were relatively clean. In contaminated animal meat, the highest levels were observed within the fat, as is expected because of dioxin's chemical characteristics. This adds to health concerns, however, because some fats are considered a delicacy in Vietnam.

What did they do? Schecter and his co-workers obtained samples of various food types from Vietnamese markets in Bien Hoa City, an area known for high dioxin contamination, and shipped them on dry ice to Germany for state-of-the-art chemical analysis. The 16 samples they analyzed spanned free-ranging and cooped chickens, free-ranging ducks, pork, beef, fish and a toad. Compounds measured included a series of dioxin congeners, including TCDD, dibenzofurans, some PCBs as well as a series of organochlorine pesticides like DDT and HCB. Because of the high levels of dioxin observed by the German lab, three samples were also analyzed by a second, independent laboratory, also in Germany.

What did they find? Dioxin and related contaminants were detected in all 16 samples analyzed. This is summarized in the graph to the right, adapted from Schecter et al. TEQ is an index based on toxicity that reflects the combined effect of dioxin-like contaminants. Different colors within the bars reflect how much different classes of contaminants contribute to total TEQ. A few sampled items stood out as "markedly elevated," notably two sampled free-ranging ducks. High levels were also observed in two free-ranging chickens, a toad and a fish. Caged chicken were lower. Beef and pork were relatively low, meeting European standards (below) , as were four of the fish sampled.

Schecter et al. note that TCDD, the most toxic form of dioxin, made the largest contribution to total TEQ toxicity. This is shown in the graph above (white vs. purple vs. red portions of the bars). For comparison, TEQ levels in US food typically are well below 1 TEQ ppt, as indicated in the table to the left (from the US National Academy of Sciences, Dioxins and dioxin-like compounds in the food supply, 2003, p97). Clearly the levels observed in Vietnamese food are literally "off-the-charts" compared to what occurs in US food. Amounts encountered in duck, toad, some chicken and some fish are also extremely high compared to UK and European standards:1-3 ppt for meat, 4 ppt for fish, 0.75-3 ppt for fat and oils, etc.fish (4 ng WHO-TEQ/kg fresh weight), eggs.

What does it mean? These results provide a clear explanation for why residents of Bien Hoa City still carry high dioxin levels, 30 years after spraying of Agent Orange ended. Sufficient dioxin and dioxin-like contaminants remain distributed in the environment to reach extremely high levels in some food items important to the Vietnamese diet.

Schecter et al. comment on the wide variability of TEQ levels in food items taken from one local area, ranging from 0.11 ppt TEQ in a sample of beef to 343 ppt TEQ in a free-ranging duck. These data suggest it may be possible to develop guidelines for food consumption that could be used to reduce human exposures. More data, including replicate samples from different food types, will be necessary to be confident in the utility of such guidelines, but already the data indicate that free-ranging chickens and ducks should be avoided. Four of five species of fish sampled were relatively uncontaminated; the one fish with high levels burrows in the bottom mud of a heavily contaminated lake. Further research might therefore clarify which species can be consumed with comparative safety.

The wide variability in TEQ levels in local foods also complicates efforts to use relatively crude estimates of dioxin exposure, such as geographic patterns of spraying, as a basis for epidemiological assessments of dioxin risk. The food variability will bias such studies toward false negatives, particularly if dietary differences among people within and among areas lead to large differences in dioxin consumption.