higher fasting insulin level (p < .05; mean = 7.0 µIU/ml). Subsequent analyses defined persons with TCDD levels below 15 ppt as “normal.” Comparison of the normal and high (>15 ppt) groups with respect to gender, age, body mass index, or total lipids failed to find any significant differences. No differences were found for fasting glucose or for glucose levels after a 75-g glucose challenge. However, insulin levels were significantly higher in the group with TCDD greater than 15 ppt at 30, 60, and 120 minutes. The ORs for high insulin among individuals with high TCDD relative to those with levels less than 15 ppt were 8.5 (1.5 –49.4) at fasting (high insulin, >4.5 µIU/ml); 12 (2.2–70.1) at 60 minutes postchallenge (high insulin, 228µIU/ml); and 56 (5.7–556) at 120 minutes (high insulin, 97.7 µIU/ml). The high insulin levels were detemained using the ninetieth percentile at each time point. The authors concluded that the study provides evidence that TCDD may cause insulin resistance.

The study provides useful data from a group of nondiabetic healthy individuals sampled from a community with potential TCDD exposure. Insulin resistance was not measured directly, but the presence of hyperinsulinemia provides indirect supporting evidence for TCDD-induced effects on insulin regulation. The comparison groups appeared to be relatively similar on several characteristics that may be potential confounding factors. The study was limited by the sample size, with only 15 individuals in the “high” (>15 ppt) TCDD group.


As anticipated, the methodologic challenges described in the “Issues Related to the Epidemiologic Study of Exposure to Herbicides and Type 2 Diabetes” section earlier in this report hampered clear assessment of the data relevant to a possible association between herbicide or dioxin exposure and the subsequent development of Type 2 diabetes. The committee identified several specific issues, some of which may be addressed through additional research. One concern is that the diabetes rates reported in some studies may be underestimated .15 The committee strongly recommends that a rigorous and consistent case definition of diabetes be applied in all studies, which would allow comparison of findings across studies and comparisons with available population data. It specifically recommends use of the ADA criteria. The committee further recommends replication of the analyses described by Longnecker and Michalek (2000) of serum TCDD level, diabetes incidence, and serum insulin level, examining other populations with background levels of serum dioxin. It is noted that a recommendation made in Update 1998 for a combined analysis of the data generated by the Ranch Hand and NIOSH studies is being pursued. The committee welcomes this effort to further examine the possibility that herbicide or dioxin exposure leads to an increased risk of diabetes.


This is not an issue in the Ranch Hand cohort, in which ascertainment is unusually thorough for an epidemiologic study, and the use of a 100-g load increases the sensitivity of the oral glucose tolerance test.

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement