• Home
• About ATSDR
• Press Room
• A-Z Index
• Employment
• Contact Us
• CDC

Evaluation of Serum Lipid Concentrations among U.S. Workers
Exposed to 2,3,7,8-Tetrachlorodibenzo-p-Dioxin

ABSTRACT. 2,3,7,8-Tetrachlorodibenzo-p-dioxin alters lipid metabolism in animals; however, evidence for such an effect in humans is conflicting. This conflict was addressed using data from a cross-sectional medical study conducted between 1987 and 1988. The exposed participants had been employed at least 15 y earlier in the manufacture of 2,4, 5-trichlorophenol or one of its derivatives at two chemical plants in the United States. A total of 281 workers and 260 unexposed referents participated. Workers had substantial exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin, evidenced by a median serum 2,3,7,8-tetrachlorodibenzo-p-dioxin concentration of 406.6 femtograms/gram of serum (fg/g serum), compared with 36.9 fg/g serum among the referents. A slight association between triglyceride concentration and serum 2,3,7,8-tetrachlorodibenzo-p-dioxin concentration was found (p = .05). Over the range of observed 2,3,7,8-tetrachlorodibenzo-p-dioxin values (i.e., 37–19 000 fg/g serum), triglyceride concentration increased only about 0.4 mmol/l. No association was found between an abnormally elevated triglyceride (i.e., > 2.82 mmol/l) concentration and serum 2,3,7,8-tetrachlorodibenzo-p-dioxin concentration. An association was also found between serum 2,3,7,8-tetrachlorodibenzo-p-dioxin concentration and an abnormal high-density lipoprotein concentration (p = .09). In summary, there was evidence of an effect on lipid metabolism in a group of workers with high serum 2,3,7,8-tetrachlorodibenzo-p-dioxin concentrations. The influence of serum 2,3,7,8-tetrachlorodibenzo-p-dioxin on lipid concentrations, however, was small, compared with the influence of other factors.

DISORDERS of lipid metabolism, characterized by the presence of hyperlipidemia (i.e., hypercholesterolemia and hypertriglyceridemia), are major risk factors for the development of coronary heart disease. Lipid disorders result from primary inborn errors of metabolism or from a variety of secondary causes The secondary causes of lipid metabolism disorders include diabetes mellitus, nephrotic syndrome, liver disease, obesity, a high-fat diet, sedentary lifestyle, and certain drugs Chemical agents are also known to be secondary causes of lipid metabolism disorders (e.g., carbon disulfide, ethanol).

Animal studies have shown that sublethal exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) is associated with effects on lipid metabolism (as reflected by increased serum cholesterol, as well as generally increased serum triglyceride concentrations) However, the evidence in humans for an association between TCDD exposure and disorders of lipid metabolism is conflicting. All human evidence has originated from cross-sectional medical studies. Statistically significant elevations in total cholesterol and/or in triglyceride concentrations among populations exposed to TCDD-contaminated substances have been reported in some studies, whereas declines have been described in others In addition, reduced high-density lipoprotein (HDL) concentrations were observed in one group of workers involved in the manufacture of TCDD-contaminated phenoxy herbicides and among U.S. Air Force Ranch Hand personnel. The HDL concentrations, however, among chemical workers who were exposed to TCDD after a factory explosion were not reduced. The total cholesterol-HDL cholesterol ratio was associated with TCDD exposure in the ranchhand study, but its measurement has not been reported for any other TCDD-exposed cohorts. The statistical analyses in the Ranch Hand study involved the use of lipid-adjusted TCDD concentrations measured in serum collected in 1987. The use of lipid-adjusted TCDD concentrations in the analysis of outcomes correlated to serum lipids complicates interpretation of the results.

Copyright is not claimed; this work is in the public domain.

We undertook a cross-sectional medical study to examine the long-term health effects of occupational exposure to chemicals and materials contaminated with TCDD. We collected data between 1987 and 1988 and analyzed the information to determine if TCDD exposure was associated with (a) an increase in total cholesterol concentration, (b) triglyceride concentration, and (c) total cholesterol/HDL cholesterol ratio. In addition, we analyzed the data to determine if TCDD exposure was associated with a decrease in the HDL cholesterol concentration. To avoid difficulties in interpretation, we assessed TCDD exposure, using serum TCDD concentrations that were not adjusted for total lipids (whole weight TCDD). However, because we used whole-weight TCDD concentrations, the results reported in this article cannot be compared directly with those from the ranch-hand study.

Materials and Method

The details of the study design have been described previously.In summary, living individuals (workers) who had been employed more than 15 y earlier in the production of sodium trichlorophenate (NaTCP) or 2,4,5-trichlorophenoxyacetic ester (2,4,5-T ester), both of which were contaminated with TCDD, were compared with an unexposed comparison group. The workers were employed in one of two plants located in Newark, New Jersey, and in Verona, Missouri. A total of 490 workers who were employed during the time period 1951–1969 at the New Jersey facility engaged in the production of NaTCP, 2,4,5-T ester, and other chemicals. At the facility in Verona, Missouri, 96 individuals were involved in the production of NaTCP or of 2,4,5-T ester. Production of NaTCP and 2,4,5-T ester occurred for approximately 4 mo in 1968, and production of NaTCP also occurred from April 1970 to January 1972. A variety of other chemicals were produced at both plants, but no other TCDD-contaminated phenoxy herbicides were produced. A referent (i.e., comparison) group was established; 1 individual with no self-reported occupational exposure to TCDD-contaminated substances was sought from within the neighborhood of each worker, matched by age (within 5 y), race, and gender. The study protocol was approved by the National Institute of Occupational Safety and Health (NIOSH) Human Subjects Review Board, and informed consent was obtained from each of the participants.

Information about the health of workers and referents was collected via standardized interviews and medical examinations. All individuals who conducted medical histories, examinations, and tests were blind to the exposure status (worker or referent) of the participant. An interviewer-administered lifetime occupational history was elicited from each participant, independent of the medical history. Duration of each job and duration of occupational exposure to many specific substances were recorded, beginning at age 16 y.

Blood was obtained from the participants after they had fasted for at least 12 h, and it was analyzed for TCDD and total cholesterol, triglyceride, and HDL cholesterol. The total cholesterol/HDL cholesterol ratio was calculated for each participant. Total serum lipid concentration was calculated with the values for total cholesterol and triglyceride, as described by Phillips et al. Blood was not drawn from 2 participants (1 worker and 1 referent).

To categorize the participant's levels of total cholesterol, triglyceride, HDL cholesterol, and total cholesterol/HDL cholesterol ratio as normal or abnormal, we obtained reference values from the Report of the National Cholesterol Education Program. A participant's total cholesterol/HDL ratio, total cholesterol concentration, and triglyceride concentration were considered abnormal if they equaled or exceeded their respective reference values. The HDL cholesterol concentration was considered abnormal if it was equal to or less than the reference value. Reference values follow: cholesterol, 6.21 mmol/l (240 mg/dl); triglyceride, 2.82 mmol/l (250 mg/dl); HDL cholesterol, 0.91 mmol/l (35 mg/dl); and total cholesterol/HDL ratio, 5. These reference values were chosen because they represent the threshold above which the risk for coronary artery disease is particularly increased.

Data analysis. In the unadjusted analysis of continuous variables, means of workers and referents were compared, using Student's t test. To evaluate the association between TCDD exposure and each of the four lipid measures of a priori interest (total cholesterol, triglycerides, HDL cholesterol, total cholesterol/HDL cholesterol ratio), we performed linear and logistic regression analyses. The variables examined in the regression analyses are summarized in Table 1. Potential confounders were retained in the final model if they were statistically significant for the outcome or if they created a meaningful difference in the coefficient of the TCDD exposure variable (i.e., more than a 15% difference). Once confounders were identified, pairwise

Table 1.—Variables Evaluated for Each of the Lipids Analyzed

Table 1.—Variables Evaluated for Each of the Lipids Analyzed
	Triglycerides	HDL	Total cholesterol	Total cholesterol/ HDL ratio
Exposure to TCDD	X	X	X	X
Years since last exposure	X	X	X	X
Age at examination (y)	X	X	X	X
Race (white = 1, nonwhite = 0)	X	X	X	X
Gender (male = 1, female = 0)	X	X	X	X
Drank alcohol in past y (yes = 1, no = 0)	X	X	X	X
Former drinker—no alcohol in past y     (yes = 1, no = 0)	X	X	X	X
Alcohol-years*	X	X	X	X
Employed in New Jersey plant     (yes = 1, no = 0)	X	X	X	X
Employed in Missouri plant     (yes = 1, no = 0)	X	X	X	X
Body-weight index (kg/m2)	X	X	X	X
Carbon disulfide exposure	X	X	X	X
Medications affecting triglyceride     (yes = 1, no = 0)	X
Pack-years†	X	X	X	X
Smoked in past y (yes = 1, no = 0)	X	X	X	X
Former smoker—denied smoking in past y     (yes = 1, no = 0)	X	X	X	X
Hypothyroidism (TSH > 8; yes = 1, no = 0)		X	X	X
Used medications that affect cholesterol     (yes = 1, no = 0)		X	X	X
Used thiazide diuretics (yes = 1, no = 0)	X	X	X	X
Use of beta-blockers (yes = 1, no = 0)	X	X		X
Proteinuria (yes = 1, no = 0)	X	X	X	X
Current diabetic (yes = 1, no = 0)	X	X	X	X
Obese (yes = 1, no = 0)	X	X	X	X
*Average number of alcoholic drinks consumed per day multiplied by the number of years alcohol was consumed. †Average number of cigarettes smoked per day multiplied by the number of years cigarettes were smoked. ‡Used within the past 2 wk. §Self-report of physician-diagnosed protein in the urine. //Self-report of physician-diagnosed diabetes mellitus or two fasting serum glucose measurements exceeding 7.7 mmol/l (140 mg/dl). #Defined as actual body weight being 20% higher than the expected body weight, as based on the Metropolitan Life Insurance Reference Weights for height. 23

interactions with each confounder and the exposure variable were examined. All significant interactions are reported.

The effect of exposure to TCDD-contaminated materials was assessed, using serum TCDD concentrations measured at the time of the examination (expressed as whole-weight serum TCDD in femtograms/gram of serum [fg/g serum]). Although serum concentrations of TCDD are typically adjusted for total lipids, this adjustment was not made in these analyses to avoid the problems of interpretation that would arise when adjusting a covariate by the dependent variable. To aid in the interpretation of the dose-response data, we stratified workers into quartiles of approximately equal size, based on serum TCDD concentrations. Workers in the lowest quartile had serum TCDD concentrations less than 158 fg/g serum, thus making them almost equivalent to the referents because all of the referents had serum TCDD concentrations of less than 158 fg/g serum. The remaining 3 quartiles comprised workers with serum dioxin concentrations of 159–520 fg/g serum, 521–1 515 fg/g serum, and 1 516–19 717 fg/g serum. In these quartile analyses, the trend statistic consisted of the p value attached to the regression coefficient for the continuous serum TCDD concentration. Adjusted means for each of the worker quartiles were compared with the referent group, using Student's t test. Eight workers and 1 referent were excluded from the dose-response analyses because serum TCDD concentrations were not available. All analyses were conducted using SAS procedures (SAS Institute [Cary, North Carolina]).

TCDD is lipophilic and partitions into serum lipids; therefore, all things being equal, individuals with higher serum lipid concentrations will have higher serum TCDD concentrations. There was concern that any association we observed between TCDD and a specific lipid component would have resulted from this partitioning phenomenon—and not from an effect on lipid metabolism. In an attempt to address this concern, the regression models were rerun by including a term for

Table 2.—Means of Serum Lipid Tests for Workers and Referents
	Workers				Referents
Laboratory test	n		SD		n		SD
Total cholesterol (mmol/l)	280	5.73	.07		259	5.63	.07
Triglycerides (mmol/l)	280	1.50	.08		259	1.39	.07
HDL cholesterol (mmol/l)	280	1.22	.02		259	1.21	.02
Total cholesterol/HDL      cholesterol ratio	280	5.06	.10		259	4.90	1.47





Table 3.—Mean Total Cholesterol Concentration and Adjusted Odds Ratios (ORs) for an Abnormal Total Cholesterol Concentration, by Serum 2,3,7,8-TCDD Category
Serum 2,3,7,8-TCDD category (fg/g serum)	Total cholesterol
	n	*	SE†	Abnormal (%)‡	OR	95% CI
Referents
< 158	259	5.6	0.1	32.4	1.0
Workers	273	5.7	0.1	34.9	1.1	0.8–1.6
< 158	87	5.8	0.1	37.9	1.3	0.8–2.3
158–520	62	5.8	0.1	35.5	1.0	0.7–2.2
521–1 515	62	5.7	0.1	33.9	1.2	0.7–2.1
1 516–19 717	62	5.7	0.1	29.0	1.0	0.5–1.7
Test for trend		p = .44			p = .71
Notes: fg/g = femtograms/gram, OR = odds ratio, and Cl = confidence interval. *Geometric means (mmol/l), adjusted for pack-years and current diabetes. †Geometric standard error term. ‡Abnormal defined as a total cholesterol concentration ≥ 6.21 mmol/l. §Adjusted for pack-years, age, medication known to affect cholesterol, and current diabetes.





Table 4.—Mean HDL Cholesterol Concentration and Adjusted Odds Ratios (ORs) for an Abnormal HDL Cholesterol Concentration, by Serum 2,3,7,8-TCDD Category
Serum 2,3,7,8-TCDD category (fg/g serum)	HDL cholesterol
	n	*	SE†	Abnormal (%)‡	OR§	95% CI
Referents
< 158	259	1.2	1.01	12.7	1.0
Workers	273	1.2	1.01	16.8	1.2	0.7–2.1
< 158	87	1.3	1.03	9.2	0.6	0.3–1.4
158–520	62	1.2	1.03	21.0	1.6	0.8–3.5
521–1 515	62	1.1	1.03	12.9	1.0	0.4–2.4
1 516–19 717	62	1.1	1.03	25.8	2.2	1.1–4.7
Test for trend		p = .15			p = .09
Notes: fg/g = femtograms/gram, OR = odds ratio, and Cl = confidence interval. *Geometric means (mmol/l), adjusted for gender, body-weight index, pack-years, use of beta-blocker medication, and current diabetes. †Geometric standard error term. ‡Abnormal defined as a HDL cholesterol concentration ≤ 0.91 mmol/l. §Adjusted for body-weight index, use of beta-blocker medication, and current diabetes.

total serum lipids. Inclusion of this term may control for any partitioning effect. Because all of the models were essentially unchanged by inclusion of the term for total serum lipids, we reported only the models without this term.

Results

Descriptive information on the study participants has been reported previously.The median whole-weight serum TCDD concentration was elevated significantly

Table 5.—Mean Total Cholesterol/HDL Cholesterol Ratio and Adjusted Odds Ratios (ORs) for an Abnormal Total Cholesterol/HDL Cholesterol Ratio, by Serum 2,3,7,8-TCDD Category
Serum 2,3,7,8-TCDD category (fg/g serum)	Total cholesterol/HDL cholesterol ratio
	n	*	SE†	Abnormal (%)‡	OR	95% Cl
Referents
< 158	259	4.7	1.02	44.4	1.0
Workers	273	4.8	1.02	47.9	1.1	0.8–1.6
< 158	87	4.5	1.03	41.4	0.9	0.5–1.4
158–520	62	4.9	1.04	46.8	1.1	0.6–1.9
521–1 515	62	4.9	1.04	50.0	1.3	0.8–2.4
1 516–19 717	62	5.0	1.04	58.1	1.5	0.8–2.7
Test for trend		p = .17			p = .17
Notes: fg/g = femtograms/gram, OR = odds ratio, and Cl = confidence interval. *Geometric means (mmol/l), adjusted for gender, body-weight index, pack-years, use of beta-blocker medication, race, gender and current diabetes. †Geometric standard error term. ‡Abnormal defined as a ≤ 5. §Adjusted for body-weight index, pack-years, and current diabetes.





Table 6.—Adjusted Linear Regression Model Coefficients for Triglyceride
Variable	Beta	SE
Whole weight TCDD     (per fg/g serum)*	2.3×10-5	1.2×10-5
Gender*	0.24	0.10
NJ plant	−0.04	0.05
MO plant*	0.19	0.08
Body-weight index†	0.04	5.5×10-3
Pack-years†	3.0×10-3	8.1×10-4
Use of beta-blockers†	0.33	0.09
Race‡	0.24	0.08
Current diabetes	0.14	0.10
R2	0.21
Notes: SE = standard error, fg/g = femtograms/gram, NJ plant = ever employed in New Jersey plant, MO plant = ever em- ployed in Missouri plant. Observations in the dataset = 530. *vp < .05. †p < .001. ‡p < .01.

among workers, compared with referents (workers = 406.6 fg/g serum [range: not detected-19 717 fg/g serum], referents = 36.9 fg/g serum [range: not detected-157.2 fg/g serum]; p < .001). Median lipid-adjusted serum TCDD levels, provided for the purpose of comparison with other studies, were 68 picograms/gram of lipid (pg/g lipid) for workers (with a range to 3 400 pg/g lipid) and 6 pg/g lipid for referents (with a range to 20 pg/g lipid). The mean alcohol years for workers was significantly less than the mean alcohol years for referents (workers = 41.4 alcohol-years, referents = 62.1 alcohol-years; p = .01). When we deleted 7 referents with alcohol-years well in excess of those observed for the other referents or workers, the mean alcohol-years of workers and referents was not significantly different. Given that exclusion of these 7 referents did not change the conclusions or change the exposure coefficients appreciably, the analytic results presented herein include these 7 referents. There were no other statistically significant differences or consistent pattern of differences between workers and referents for any other demographic characteristics (i.e., age, race, gender, education, and income). In addition, the mean total serum lipids were not significantly different between workers and referents (workers = 6.36 g/l [standard deviation {SD} = 1.74], referents = 6.18 [SD = 1.58]).

Total cholesterol, HDL cholesterol, and total cholesterol/HDL cholesterol ratio. The means for the lipid measures of interest among workers and referents are presented in Table 2. There were no statistically significant differences between workers and referents with regard to any of these lipid outcomes.

In logistic regression analyses, we found that the association between serum TCDD concentration and risk for an abnormal HDL cholesterol concentration approached statistical significance (p = .09), after controlling for body weight index, use of beta-blocker medication, age, diabetes, and employment at the New Jersey and Missouri plants. Concentration of TCDD was not associated with having either an abnormal total cholesterol concentration or an abnormal total cholesterol/HDL cholesterol ratio.

The analytic results after workers were stratified into quartiles, based on serum TCDD concentration, are presented in Tables 3–5. The data on the left side of the tables provide the adjusted means (based on linear regression models) for the various worker quartiles; on the right side of the tables are the risks for abnormal lipid concentrations, based on logistic regression models. Workers in the quartile with the highest serum TCDD concentrations had an elevated risk for an abnormal HDL cholesterol concentration (odds ratio [OR] = 2.2, 95% confidence interval [95% Cl] = 1.1, 4.7 [Table 4]), after controlling for important confounders.

Table 7.—Mean Triclyceride Concentration and Adjusted Odds Ratios (ORs) for an Abnormal Triglyceride Concentration, by Serum 2,3,7,8-TCDD Category
Serum 2,3,7,8-TCDD category (fg/g serum)	Triglyceride
	n	*	SE†	Abnormal (%)‡	OR	95% CI
Referents
< 158	259	1.15	1.03	7.4	1.0
Workers	273	1.20	1.03	7.5	1.0	0.5–2.0
< 158	87	1.04	1.06	5.7	0.7	0.2–1.9
158–520	62	1.26	1.07	8.1	1.1	0.4–3.2
521–1 515	62	1.23	1.07	6.5	0.9	0.3–2.9
1 516–19 717	62	1.35//	1.07	11.3	1.7	0.6–4.6
Test for trend		p = .05			p = .21
Notes: fg/g = femtograms/gram, OR = odds ratio, and Cl = confidence interval. *Geometric means (mmol/l), adjusted for body-weight index, pack-years, use of beta-blocker medication, race, gender, and current diabetes. †Geometric standard error term. ‡Abnormal defined as a triglyceride concentration ≥ 2.82 mmol/l. §Adjusted for body-weight index, pack-years, use of beta-blocker medication, age, and current diabetes. //p < .05, compared with the referent group.

No other significant differences in lipid means or risk of an abnormal lipid concentration were observed between any of the worker quartiles and the referent group.

Triglyceride concentration. Unlike the unadjusted analysis (Table 2), linear regression analysis resulted in the identification of a small but statistically significant association between triglyceride concentration and serum TCDD concentration (p = .05 [Table 6]), after controlling for gender, plant location (New Jersey versus Missouri), body-weight index, cumulative cigarette consumption, use of beta-blocker medication, race, and diabetes. Over the range of observed TCDD values (37–19 000 fg/g serum), triglyceride concentration increased only about 0.4 mmol/l (35 mg/dl). Cumulative cigarette consumption, race, gender, body-weight index, and the use of beta-blockers were found to have far more influence (based on the magnitudes of their regression coefficients) than TCDD concentration on triglyceride concentration (Table 6). In the logistic regression analyses, abnormal triglyceride concentration was not associated with serum TCDD concentration (p = .21).

The analytic results when workers were stratified into quartiles, based on serum TCDD concentration, are presented in Table 7. A statistically significant elevation was present in the mean triglyceride concentration of the worker quartile with the highest serum TCDD concentrations, compared with the referent group, after controlling for important confounders. There also was a statistically significant trend between mean triglyceride concentration and TCDD quartiles (p = .05). No trend was observed in the quartile analyses that evaluated risk for an abnormal triglyceride level.

Discussion

The purpose of this analysis was to investigate the possible association between TCDD exposure and disorders of lipid metabolism. This was achieved by comparing serum lipid concentrations between a group of workers with high body burdens of TCDD (but whose most recent occupational TCDD exposure was at least 15 y earlier) with an unexposed control group. We found no evidence of substantial lipid effects in our group of workers who had high exposure to TCDD. Although the linear regression analysis revealed a statistically significant association between triglyceride concentration and TCDD concentration, the paltry magnitude of this effect has almost no clinical relevance. According to our model, even a worker whose serum TCDD concentration was extremely elevated (i.e., 15 000 fg/g serum) would be expected to have a normal triglyceride concentration of 1.4 mmol/l. Triglyceride concentrations in excess of 2.82 mmol/l increase the risk for cardiovascular disease. That the results were essentially unchanged when we included a term for total serum lipids in the regression model suggests that the association we observed resulted from a small effect on triglyceride metabolism; however, the presence of a partitioning phenomenon, especially between serum triglyceride and serum TCDD, cannot be dismissed completely.

Two studies other than ours reported that TCDD-exposed individuals have higher mean triglyceride concentrations. One of the two studies was the U.S. Air Force Ranch Hand study. U.S. Air Force servicemen (i.e., Ranch Hands) were responsible for the aerial dissemination of herbicides in Vietnam from 1962 to 1971, and they had lower serum TCDD levels (median = 12.8 pg/g lipid, range 0 to 620 pg/g lipid>) than the workers we studied. The U.S. Air Force ranch hands with the highest (i.e., > 33.3 pg/g lipid) serum TCDD levels had elevated mean triglyceride levels, compared with the reference group, and they were 2.55 times more likely to have an abnormal triglyceride concentration. It should be noted that the ranch-hand investigators did not control for diabetes mellitus in their triglyceride analyses. Furthermore, the ranch-hand analyses involved the use of lipid-adjusted serum TCDD concentrations; therefore, the Ranch Hand findings cannot be compared directly with our findings. Martin found that workers, who were exposed 10 y earlier to TCDD-contaminated chemicals after a factory explosion, had an increased mean triglyceride concentration. Furthermore, TCDD-exposed workers at a chemical plant in Czechoslovakia had an elevated mean plasma very-low-density-lipoprotein (VLDL) concentration. The VLDLs are large, triglyceride-rich lipoproteins. Although the serum triglyceride concentrations of these Czechoslovakian workers were not reported, the elevation in VLDLs suggests that these workers may also have had elevated serum triglyceride concentrations. Several other studies did not find an association between TCDD exposure and triglyceride concentration, but some of these studies were limited by small sample sizes and low exposure to TCDD.

In our study, we found an association that approached—but did not reach—statistical significance between risk for an abnormally decreased HDL cholesterol concentration and serum TCDD concentration (Table 4). The risk in the worker quartile with the highest TCDD concentrations was more than twice that of the reference group, suggesting that the workers with the highest serum TCDD concentrations had an increased risk for abnormal HDL cholesterol concentrations. Nonetheless, the absence of a consistent rise in prevalence with increasing worker quartile, the failure of the trend statistic to reach statistical significance, and the absence of an association in the linear regression analysis all suggest that chance may also be an explanation. The literature provides inconsistent evidence for an association between TCDD exposure and decreased HDL cholesterol concentrations. In two studies an association between TCDD exposure and a decrease in HDL cholesterol concentration was found. Among chemical workers who had ceased production of 2,4,5-T approximately 10 y earlier, those with persistent chloracne had an elevated risk for an abnormal HDL cholesterol, compared with their TCDD-exposed colleagues who had no chloracne at the time of examination.In the U.S. Air Force Ranch Hand study, an association was found between half-life-extrapolated serum TCDD concentration and decreased HDL cholesterol concentration. The Ranch Hand authors suggested that the effect of TCDD on the HDL cholesterol concentration is subclinical and probably not clinically relevant. Evidence in support of the absence of an association was presented by Martinwho found no elevation in mean HDL cholesterol concentrations among workers exposed to TCDD after a factory explosion.

Although an association between elevated serum total cholesterol concentration and TCDD exposure has been reported consistently in animal studies,, human evidence is conflicting. In our study, we were unable to find such an association, which is consistent with the results of several other studies. In contrast, a statistically significant elevation in mean total cholesterol concentration among TCDD-exposed individuals was reported in two studies. In no study, however, has an association between TCDD exposure and an increased risk for an abnormal total cholesterol concentration been reported. If TCDD exposure has an effect on total cholesterol concentration in humans, it appears to be modest and of minimal relevance clinically.

In conclusion, in our study we found evidence for a small effect on triglyceride metabolism in a group of workers with high serum TCDD concentrations that resulted from occupational exposure that occurred at least 15 y earlier. In addition, an association that approached statistical significance was found between serum TCDD concentration and an abnormally decreased HDL cholesterol concentration. However, the association of serum TCDD concentration with these two lipid measures was found to be small when compared with the influence of many other factors. Exposure to TCDD may be associated with modest effects on serum lipid metabolism, as evidenced by our study and other studies of TCDD-exposed workers.

Funding for this study was provided by the Agency for Toxic Substances and Disease Registry. The authors are thankful for the assistance provided by Barbara Connally, Laurie Piacitelli, David Marlow, Mary Chung, and Fran Guerra. The authors also acknowledge Dr. James Pirkle, Dr. Don Patterson, and Dr. Larry Needham at the National Center for Environmental Health for their helpful suggestions and for performing the TCDD analyses on the serum samples.

Submitted for publication May 16, 1994; revised; accepted for publication July 20, 1995.

Requests for reprints should be sent to Geoffrey M. Calvert, M.D., National Institute for Occupational Safety & Health, 4676 Columbia Parkway, R-16, Cincinnati, OH 45226.

References

1. Neaton JD, Kuller LH, Wentworth D, et al. Total and cardiovascular mortality in relation to cigarette smoking, serum cholesterol concentration, and diastolic blood pressure among black and white males followed up for five years (MRFIT). Am Heart J 1984; 108:759–69.
2. Kannel WB, Castelli W, Gordon T, et al. Serum cholesterol, lipoproteins, and the risk of coronary heart disease. The Framingham study. Ann Intern Med 1971; 74:1–12.
3. Brunzell JD. Disorders of lipoprotein metabolism. In: Wyngaarden JB, Smith LH Jr., Bennett CJ, Plum F, Eds. Cecil Textbook of Medicine. 19th ed. Philadelphia, PA: Saunders, 1992: pp 1082–90.
4. Cooper GR, Myers GL, Smith SJ, et al. Blood lipid measurements: variations and practical utility. JAMA 1992; 267:1652–60.
5. DeCaprio AP, McMartin DN, O'Keefe PW, et al. Subchronic oral toxicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin in the guinea pig: comparisons with a PCB-containing transformer fluid pyrolysate. Fund Appl Tox 1986; 6:454–63.
6. Zinkl JG, Vos JG, Moore JA, Gupta BN. Hematologic and clinical chemistry effects of 2,3,7,8-tetrachlorodibenzo-p-dioxin in laboratory animals. Environ Health Perspect 1973; 5:111–18.
7. Buu-Hoi NP, Chanh P, Sesque G, et al. Enzymatic functions as targets of the toxicity of "dioxin" (2,3,7,8-tetrachlorodibenzo-p-dioxin). Naterwissenschaften 1972; 59:173–74.
8. Schiller CM, Adcock CM, Moore RA, et al. Effect of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and fasting on body weight and lipid parameters in rats. Toxicol Appl Pharmacol 1985; 81: 356–61.
9. Poli A, Francescieni G, Puglise L, et al. Increased total and high density lipoprotein cholesterol with apoprotein changes resembling streptozotocin diabetes in tetrachlorodibenzodioxin (TCDD) treated rats. Biochem Pharmacology 1980; 29:835–38.
10. Brewster DW, Bombick DW, Matsumura F. Rabbit serum triglyceridemia after administration of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). J Toxicol Environ Health 1988; 25:495–507.
11. Olson JR, Holscher MA, Neal RA. Toxicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin in the golden Syrian hamster. Tox Appl Pharmacol 1980; 55:67–78.
12. Roegner RH, Grubbs WD, Lustik MB, et al. The Air Force Health Study: An Epidemiologic Investigation of Health Effects in Air Force Personnel Following Exposure to Herbicides. Serum Dioxin Analysis of 1987 Examination Results. Springfield, VA: National Technical Information Service Publication; 1991; Publication AD A-237-519.
13. Martin JV. Lipid abnormalities in workers exposed to dioxin. Br J Ind Med 1984; 41:254–56.
14. Moses M, Lilis R, Crow KD, et al. Health status of workers with past exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin in the manufacture of 2,4,5-trichlorophenoxy-acetic acid. Comparison of findings with and without chloracne. Am J Ind Med 1984; 5:161–82.
15. Suskind RR, Hertzberg VS. Human health effects of 2,4,5-T and its toxic contaminants. JAMA 1984; 251:2372–80.
16. Hoffman RE, Stehr-Green PA, Webb KB, et al. Health effects of long-term exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin. JAMA 1986; 255:2031–38.
17. Mocarelli P, Marocchi A, Brambilla P, et al. Clinical laboratory manifestations of exposure to dioxin in children. A six-year study of the effects of an environmental disaster near Seveso, Italy. JAMA 1986; 256:2687–95.
18. Sweeney MH, Fingerhut MA, Connally LB, et. al. Progress of the NIOSH cross-sectional medical study of workers occupationally exposed to chemicals contaminated with 2,3,7,8-TCDD. Chemosphere 1989; 19:973–77.
19. Phillips DL, Pirkle JL, Burse VW, Bernert JT, Henderson LO, Needham LL. Chlorinated hydrocarbon levels in human serum: effects of fasting and feeding. Arch Environ Contam Toxicol 1989; 18:495–500.
20. Expert Panel. Report of the National Cholesterol Education Program Expert Panel on detection, evaluation, and treatment of high blood cholesterol in adults. Arch Intern Med 1988; 148:36–69.
21. Castelli WP. Epidemiology of coronary artery disease: the Framingham study. Am J Med 1984; 76(2A):4–12.
22. Calvert GM, Sweeney MH, Morris JA, Fingerhut MA, Hornung RW, Halperin WE. Evaluation of chronic bronchitis, chronic obstructive pulmonary disease (COPD) and ventilatory function among workers exposed to 2,3,7,8-tetrachlorodibenzo-p-dioxin. Am Rev Respir Dis 1991; 144:1302–06.
23. Russell RM. Nutritional assessment. In: Wyngaarden JB, Smith LH Jr., Bennett CJ, Plum F, Eds. Cecil Textbook of Medicine. 19th ed. Philadelphia, PA: Saunders, 1992; pp 1151–55.
24. Pazderova-Vejlupkova J, Nemicova M, Pickova J, et al. The development and prognosis of chronic intoxication by tetrachlorodibenzo-p-dioxin in man. Arch Environ Health 1981; 36: 5–11.
25. May G. Tetrachlorodibenzodioxin: a survey of subjects ten years after exposure. Br J Ind Med 1982; 39:128–35.
26. Webb KB, Evans RG, Knutsen AP, et al. Medical evaluation of subjects with known body levels of 2,3,7,8-tetrachlorodibenzo-p-dioxin. J Tox Environ Health 1989; 28:183–93.

• Home
• Privacy Policy
• Disclaimer
• Accessibility
• e-Government
• FOIA
• Contact Us