Blood levels of polychlorinated biphenyls (PCBs) and diabetes are associated. Several studies consistently support this,1–4 including the one by Vasiliu et al5 reported in this issue of Epidemiology. The new study is not only much larger than its predecessors—it is also the only one that is prospective. But does confirmation in a prospective study suggest causality?
First, however, let us dispense with a detail: although Vasiliu et al5 emphasize the findings in women, their data also show some support for a relationship in men. For the sake of argument, having noted the possible sex difference, let us set it aside.
In studies of workers occupationally exposed to PCBs, relevant findings exist. In one such study, serum glucose was unrelated to serum PCBs.6 In 2 other cohorts, death from diabetes was essentially unrelated to PCB exposure.7,8 Although the mixture of PCBs present in industrial settings differed from that among background-exposed subjects, the congeners found in the general population were present in the workers. At any event, because among highly exposed workers no findings resonate with those among the background-exposed, the evidence of a causal association is weakened.
The data on PCBs and diabetes resemble those on dioxin and diabetes in several respects—most importantly, that on the whole they present a conundrum.9 As with exposure to PCBs, workers exposed to dioxin were not at increased risk of diabetes,10 and yet an association was present among those with only background-level exposure.11
Several types of persistent organic pollutants tend to be increased in patients with diabetes with background-level exposures. Although the specific compounds that are increased depends on place, time, and choice of analytes, Rylander et al,4 for example, found increases in DDT metabolites among women, whereas Fierens et al3 reported increases in dioxins and PCBs. This lack of specificity also weighs somewhat against causality, though dioxin and some PCBs act via the same mechanism.12
If there is no causal relation why would PCB blood levels be related to subsequent risk of type 2 diabetes mellitus in a prospective study? Recall that because PCBs are lipophilic they are drawn from the adipose reservoir into plasma by higher plasma lipid levels. In the Vasliu et al study,5 PCB levels were not adjusted for serum lipid level, but a sensitivity analysis showed that even with exclusion of cases diagnosed in the first 17 years of follow-up, the association persisted. If subjects at increased risk of diabetes have elevated lipid levels more than 17 years before diagnosis, this fact could account for the association. Some evidence supports this possibility.13
Arguably, a more interesting possible explanation is that the level of PCBs in a background-exposed person reflects, along with intake and other factors, their clearance of PCBs, and clearance is determined, in turn, by factors related to risk of diabetes, either innate or dietary. Because the PCBs found in greatest concentration in human blood have half-lives on the order of years,14 small differences in clearance could, with sufficient time, account for substantial differences in blood levels. Body mass index may be inversely related to clearance of PCBs,15 and is a risk factor for type 2 diabetes mellitus. Vasiliu et al5 included baseline body mass index in their models. But some other factor, such as waist circumference, that independently predicts risk16 could also affect clearance. Such a bias might not apply in the occupational studies, because either blood levels were not measured7,8 or exposure was so high and recent that it was the main determinant of blood levels.6
The PCB levels among the subjects in the Vasiliu et al5 cohort are difficult to compare with those in earlier studies because of the older laboratory methods. However, the levels were probably similar to those in other U.S. studies in the general population from the same period,17 and several-fold higher than today.18 Some convincing evidence of biologic effects within this range of exposures exists, although causality and clinical significance remain to be determined.19
The impressive capabilities of modern analytical chemistry are being applied with increasing frequency in epidemiologic studies. With lower exposures being evaluated, and lower concentrations being measured, it seems opportune to bear in mind that a great proportion of variation in measured levels among subjects may be accounted for by differences in metabolism and excretion.20 The measurements we obtain may afford only a glimpse of a byproduct of the underlying pharmacokinetics, systems biology, and pathogenesis. While these are not entirely new concepts,21 their importance may be clearer today than ever before.15
ABOUT THE AUTHOR
MATTHEW LONGNECKER is a Senior Investigator in the Epidemiology Branch, National Institute of Environmental Health Sciences, National Institutes of Health. He has studied health effects of persistent organic pollutants for over 10 years, and was the first to report increased levels of PCBs in diabetics. His research interests include environmental risk factors for type 2 and gestational diabetes mellitus.
1. Longnecker MP, Klebanoff MA, Brock JW, et al. Polychlorinated biphenyl serum levels in pregnant subjects with diabetes. Diabetes Care
2. Glynn AW, Granath F, Aune M, et al. Organochlorines in Swedish women: determinants of serum concentrations. Environ Health Perspect
3. Fierens S, Mairesse H, Heilier JF, et al. Dioxin/polychlorinated biphenyl body burden, diabetes and endometriosis: findings in a population-based study in Belgium. Biomarkers
4. Rylander L, Rignell-Hydbom A, Hagmar L. A cross-sectional study of the association between persistent organochlorine pollutants and diabetes. Environ Health
5. Vasiliu O, Cameron L, Gardiner J, et al. Polybrominated biphenyls, polychlorinated biphenyls, body weight and incidence of adult-onset diabetes mellitus. Epidemiology
6. Lawton RW, Ross MR, Feingold J, et al. Effects of PCB exposure on biochemical and hematological findings in capacitor workers. Environ Health Perspect
7. Loomis D, Browning SR, Schenck AP, et al. Cancer mortality among electric utility workers exposed to polychlorinated biphenyls. Occup Environ Med
8. Kimbrough RD, Doemland ML, Mandel JS. A mortality update of male and female capacitor workers exposed to polychlorinated biphenyls. J Occup Environ Med
9. Longnecker MP, Daniels JL. Environmental contaminants as etiologic factors for diabetes. Environ Health Perspect
. 2001;109(Suppl 6):871–876.
10. Steenland K, Calvert G, Ketchum N, et al. Dioxin and diabetes mellitus: an analysis of the combined NIOSH and Ranch Hand data. Occup Environ Med
11. Longnecker MP, Michalek JE. Serum dioxin level in relation to diabetes mellitus among Air Force veterans with background levels of exposure. Epidemiology
12. van den Berg M, Peterson RE, Schrenk D. Human risk assessment and TEFs. Food Addit Contam
13. Bjornholt JV, Erikssen G, Liestol K, et al. Prediction of Type 2 diabetes in healthy middle-aged men with special emphasis on glucose homeostasis. Results from 22.5 years’ follow-up. Diabet Med
14. Wolff MS, Fischbein A, Selikoff IJ. Changes in PCB serum concentrations among capacitor manufacturing workers. Environ Res
15. Wolff MS, Britton JA, Teitelbaum SL, et al. Improving organochlorine biomarker models for cancer research. Cancer Epidemiol Biomarkers Prev
16. Wang Y, Rimm EB, Stampfer MJ, et al. Comparison of abdominal adiposity and overall obesity in predicting risk of type 2 diabetes among men. Am J Clin Nutr
17. Longnecker MP, Wolff MS, Gladen BC, et al. Comparison of polychlorinated biphenyl levels across studies of human neurodevelopment. Environ Health Perspect
18. Centers for Disease Control and Prevention. Third National Report on Human Exposure to Environmental Chemicals. Atlanta, GA, 2005. Available at: http://www.cdc.gov/exposurereport/3rd/
. Accessed April 11, 2006.
19. Fitzgerald EF, Hwang SA, Lambert G, et al. PCB exposure and in vivo CYP1A2 activity among Native Americans. Environ Health Perspect
20. Dorne JL, Renwick AG. The refinement of uncertainty/safety factors in risk assessment by the incorporation of data on toxicokinetic variability in humans. Toxicol Sci
21. Wilcosky TC. Criteria for selecting and evaluating markers. In: Hulka BS, Wilcosky TC, Griffith JC, eds. Biological Markers in Epidemiology
. New York: Oxford University Press; 1990:28–55.