Racial (black-white) comparisons of the relationship. lipoproteins during male adolescence: the Bogalusa Heart Study

Transcription

1 PATHOPHYSIOLOGY AND NATURAL HISTORY ATHEROSCLEROSIS Racial (black-white) comparisons of the relationship of levels of endogenous sex hormones to serum lipoproteins during male adolescence: the Bogalusa Heart Study S. R. SRINIVASAN, PH.D., D. S. FREEDMAN, PH.D., G. S. SUNDARAM, PH.D., L. S. WEBBER, PH.D., AND GERALD S. BERENSON, M.D. ABSTRACT The cross-sectional relationship of endogenous androgens (testosterone, androstenedione, and dehydroepiandrosterone sulfate [DHEA-S]), estrogen (estradiol) and progestin (progesterone) to serum levels of lipoprotein cholesterol (very low-density [VLDL], low-density [LDL], and high-density lipoprotein [HDL]) and apolipoproteins (apo A-I and apo B) were studied in white (n 251) and black (n = 258) adolescent boys, ages 1 1 to 17 years, as part of the Bogalusa Heart Study. Black boys had significantly higher levels of estradiol, HDL cholesterol, and apo A-I, and lower levels of androstenedione and VLDL cholesterol than white boys, independent of age and adiposity. Age was correlated strongly with testosterone and androstenedione, and moderately with DHEA-S and estradiol levels in both races. However, only in white boys was age consistently related to VLDL cholesterol (positively), HDL cholesterol (negatively), and apo A-I (negatively). Overall, testosterone was associated inversely with HDL cholesterol and apo A-I in white boys. while progesterone was related positively to apo A-I in both races after adjusting for age and adiposity. However, these relationships were found to differ with age. Partial correlations between levels of sex hormones and lipoproteins adjusted for age and adiposity showed no associations in the 11 to 12 year age group in boys of either race. A significant positive relation of testosterone to VLD, 2holesterol, and inverse relations of testosterone to HDL cholesterol and apo A-I and DHEA-S to HDL cholesterol were apparent only in white boys in the 13 to 14 year age group. Among the older subjects (15 to 17 years old), the relationships of testosterone to HDL cholesterol and androstenedione to apo A-I were positive only in black boys. In addition, a significant positive association between progesterone and apo A-I was noted for both races among boys in the older age group. That the sex hormone-lipoprotein associations vary among different age groups suggests the influence of sexual maturation-related hormonal makeup on lipoproteins. Inherent metabolic differences between the races may account for some of the divergent sex hormone-lipoprotein associations. Circulation 74, No. 6, , 1986 THE INFLUENCE of sex hormones on serum lipid and lipoprotein levels is well known and is considered potentially associated with risk of development of coronary heart disease (CHD).`'2 The incidence of CHD is higher in men than in premenopausal women.3 From the Departments of Medicine, Biochemistry, Public Health and Preventive Medicine, and Biometry and Genetics, Louisiana State University Medical Center, New Orleans, and the Department of Obstetrics and Gynecology, Sinai Hospital and Johns Hopkins School of Medicine, Baltimore. Supported by research grants HL and the National Research and Demonstration Center Arteriosclerosis (HL-15103) from the National Heart, Lung, and Blood Institute of the U.S. Public Health Service. Address for correspondence: Gerald S. Berenson, M.D., Department of Medicine, LSU Medical Center, 1542 Tulane Ave., New Orleans, LA Received May 20, 1986; revision accepted Sept. 11, Women with normal ovarian function have lower levels of very low (VLDL) and low-density lipoproteins (LDL) and higher levels of high-density lipoproteins (HDL) than men, factors considered to be protective against CHD.4 Furthermore, it has been suggested that the relatively greater protection of black compared with white men from CHD is due partly to their higher HDL levels.5 However, it is not known whether levels of sex hormones have any role in determining the racerelated differences in lipoproteins. Profound changes in lipoproteins occur in children during sexual maturation before adult patterns are established.6 ' Both black and white girls have higher levels of VLDL and LDL cholesterol as compared with boys before sexual maturation. During sexual matura- CIRCULATION

2 PATHOPHYSIOLOGY AND NATURAL HISTORY-ATHEROSCLEROSIS tion, levels of HDL and LDL cholesterol decrease in both races, but the decrease in HDL cholesterol is much greater in white boys. Recent studies in children demonstrate that endogenous sex hormones relate to lipoprotein levels, 10 and may account for the sexrelated divergence of lipoprotein profiles after puberty. 10 However, the question arises whether the relation of levels of endogenous sex hormones to those of lipoproteins is the same in both races. The purpose of this report is to evaluate the relation of endogenous androgens (testosterone, androstenedione, and dehydroepiandrosterone sulfate [DHEA- S]), estrogen (estradiol), and progestin (progesterone) to serum levels of lipoprotein cholesterol (VLDL, LDL, and HDL) and apolipoproteins (apo A-I and apo B) in black and white adolescent boys living under similar environmental conditions. Materials and methods Population. The Bogalusa Heart Study is a long-term, population-based study of cardiovascular disease risk factors from birth through 26 years of age in the biracial community (64% white, 36% black), Bogalusa, LA.'1 Data collected from the fourth cross-sectional examination ( ) of school-age children (n = 3314, 80% participation) were used in the current analyses. All analyses were restricted to boys between the ages of 11 and 17 years. To assess the relationship between levels of sex hormones and serum lipoproteins during adolescence, a random sample of whites was selected to provide approximately equal numbers of blacks (n = 258) and whites (n = 251) in this age group. These samples represent 48% (white) and 90% (black) of all boys in this age range who were examined. Assessment of sexual maturation. Sexual maturation was determined during physical and anthropometric examinations by visual assessment of secondary sex characteristics according to the method of Tanner.'2 The ratings of sexual maturation ranged from 1 (no development) to 5 (complete development) according to the stages of male genitalia development. Collection of blood specimens. Children were instructed to fast for 12 to 14 hr before venipuncture. Antecubital venous blood was drawn separately for lipoprotein and sex hormone assays; serum samples were prepared in tubes containing thimerosal (Aldrich Chemical Co., Milwaukee, WI). Serum for lipoprotein measurements was kept at 40 C until analysis the following day in the core laboratory in New Orleans; the serum samples for determination of sex hormone levels were kept frozen (-20 C) and sent by air to Baltimore (Dr. G. S. Sundaram). Serum lipids, lipoprotein cholesterol, and apolipoprotein determinations. Cholesterol and triglycerides were determined in a Technicon AutoAnalyzer II (Technicon Instrument Corp., Tarrytown, NY) according to the Laboratory Manual of the Lipid Research Clinics Program.13 The laboratory has been designated as standardized by the Center for Disease Control, Atlanta, and currently is in the surveillance phase of its qualitycontrol program. Serum VLDL, LDL, and HDL cholesterol levels were measured by a combination of heparin-calcium precipitation and agar-agarose gel electrophoretic methods. A detailed description of this method, including measurement errors and its application in the Bogalusa Heart Study, has already been reported. 1'4 The levels of apo B and apo A-I in whole serum were assayed Vol. 74, No. 6, December 1986 by an electroimmunoassay procedure. Antibodies to LDL (d 1.03 to 1.05 g/ml) and apo A-I raised in goats were used. This method, including measurement errors, has been described elsewhere in detail.15 The laboratory is participating in the apo A-I and apo B standardization program of the International Union of Immunological Societies-CDC--NHLBI. Serum sex hormone determinations. Estradiol-17,8 was quantitated by the radioimmunoassay (RIA) method with rabbit antiserum against estradiol-17,3-3-carboxymethylether-bsa and a 1251-labeled antigen by a double antibody technique.'6 Testosterone, progesterone, and DHEA-S were quantitated with commercial RIA kits prepared by Immuchem Corp., Diagnostic Products, and Radio Assay System Laboratories, Inc., respectively. The levels of androstenedione were deternined by the RIA method. 17 The antiserum had moderate cross-reaction with 5-a-androstenedione, which has no significant serum levels and very low cross-reaction with DHEA, and generally less than 1% affinity for all related steroids. The laboratory participates in the endocrinology and metabolism quality-control program administered by the American Association of Clinical Chemistry. As in our previous study,'0 samples were analyzed in duplicate along with commercial lowand high-level quality-control samples. Overall measurement errors from these quality-control samples varied from 6% to 13%. Statistical analyses. Since several of the sex hormone and lipoprotein distributions were positively skewed, nonparametric tests were used in several analyses. Racial differences in levels of the sex hormones and lipoproteins were assessed by Wilcoxon tests'8 or analysis of covariance, adjusting for age and subscapular skinfold thickness. Because levels of several of the sex hormones and serum lipoproteins differed between the races, subsequent analyses were performed separately for black and white boys. Spearman correlations were used to examine the relationships of age, subscapular skinfold thickness, and ponderal index to levels of sex hormones. Within each race, associations between the serum lipoproteins and sex hormones were then examined with the use of correlation coefficients. Partial correlations were also calculated to assess whether the observed associations existed independently of age and subscapular skinfold thickness. Polynomial regression models,'9 containing both linear and quadratic terms, were also used to evaluate the relationships between the sex hormones and serum lipoproteins. Because of the nonlinearity of certain relationships, several of the sex hormone-serum lipoprotein associations were further examined within age groups. Results Sexual maturation and age. Breakdown of the study population by Tanner stage, age, and race is given in table 1. The majority of 1 1- and 12-year-old boys were distributed in Tanner stages 1 and 2, while boys between the ages of 15 and 17 years were distributed mainly in Tanner stages 4 and 5. Boys between the ages of 13 and 14 years, on the other hand, showed a broad distribution of sexual maturation levels (Tanner stages 1 to 5). Overall, black boys began maturing slightly earlier than did white boys (black vs white, mean Tanner stage: 1.7 vs 1.4 at age 11, p =.06; 2.5 vs 2.0 at age 12, p =.03). Relation of sex hormones to age and obesity. Relation- 1227

3 SRINIVASAN et al. TABLE 1 Sexual maturation status of study population by age and race - Bogalusa Heart Study Age (year) Total Tanner stage White Black White Black White Black White Black White Black White Black White Black White Black i Total SlA A 258 ATanner stage not available for one individual. ships of sex hormones to age and measures of obesity by race are shown in table 2. In both races age was strongly correlated with testosterone and androstenedione levels during puberty; correlations were moderate for DHEA-S and estradiol, and low for progesterone. In both races, DHEA-S was significantly associated with both subscapular skinfold and ponderal index. Testosterone, androstenedione, and estradiol levels correlated significantly with subscapular skinfold in black boys only. Subscapular skinfold thickness was more consistently related to levels of sex hormones than was ponderal index. Therefore, covariance adjustments were made for age and subscapular skinfold. Mean levels of study variables. Table 3 lists race-specific mean (and SD) values for study variables in the 11- to 17-year-old boys. Age, subscapular skinfold thickness, and ponderal index did not differ significantly between the races. Black boys had significantly higher estradiol (p <.01) and lower androstenedione (p <.05) levels than white boys. Testosterone, progesterone, and DHEA-S values did not differ significantly between the two groups. As noted in previous studies,6 14, 15 black children had lower VLDL cholesterol and higher levels of both HDL cholesterol and apo A-I than white children (p <.001) for each difference). The black-white differences in serum variables remained significant after adjusting for the confounding covariates of age and adiposity (subscapular skinfold thickness). (Adjusting for ponderal index gave essentially the same results as did adjusting for subscapular skinfold thickness.) Mean levels by age. Mean levels of selected variables (serum testosterone, VLDL cholesterol, HDL cholesterol, and apo A-I) by age and race are shown in figure 1. Testosterone level was positively associated with age in both white and black boys. Age was also strongly related to VLDL cholesterol (positively), HDL cho lesterol (negatively), and apo A-I (negatively) in white boys. In contrast, in black boys HDL cholesterol and apo A-I dropped only between the ages of 12 and 15 years and then increased to the 12 year level at ages 16 and 17 years. In addition, the magnitude of the rise in VLDL cholesterol level between 11 and 16 years of age was relatively less in black than in white boys. Interrelationships among sex hormones. Levels of endogenous sex hormones were interrelated in both races (table 4). Androstenedione and testosterone showed the strongest correlation, and estradiol was moderately correlated with both testosterone and androstenedione. Correlations among other sex hormones were relatively low. Sex hormone-lipoprotein relationship. The relation of sex hormone to lipoprotein cholesterol and apolipoprotein levels are shown in tables 5 and 6. Unadjusted simple correlations showed significant relationships between certain sex hormone and lipoprotein variables, especially in whites. Testosterone was associated TABLE 2 Relation (Spearman rank correlation) of sex hormones to age and obesity in adolescent white (n = 251) and black (n = 258) boys - Bogalusa Heart Study Andro- Testos- stenedi- Estra- Progesterone DHEA-S one diol terone Age White.76r.32c 59c 33c 19B Black.74c 35C.72c 54c.07 Subscapular skinfold White.09.25C Black.34c 36c 35c.32c.04 Ponderal indexd White A Black.02.17B Ap <.05; Bp <.01; Cp <.001. DWeightlheight3. CIRCULATION

4 PATHOPHYSIOLOGY AND NATURAL HISTORY-ATHEROSCLEROSIS TABLE 3 Mean levels of study variables by race for adolescent boys - Bogalusa Heart Study White (n = 251) Black (n = 258) Mean SD Mean SD Age (yr) Subscapular skinfold (mm) Ponderal index (Wtlht3) Testosterone (nglml) 3.7 (3.8) (3.7) 2.6 DHEA-S (ng/ml) 1426 (1445) (1442) 962 Androstenedione (ng/mnl) 1.03 (1.06)B (0.88) 0.54 Estradiol (pg/ml) 36.7 (37.3)A (40.8) 19.2 Progesterone (ng/ml) 0.18 (0.18) (0.18) 0.10 VLDL cholesterol (mgicdl) 8.7 (8.7)B (6.5) 5.1 LDL cholesterol (mgldl) 87.1 (86.4) (90. 1) 24.7 HDL cholesterol (mg/dl) 54.6 (54.6)B (63.0) 19.2 Apo A-1I(mg/dl) 79.7 (791 ) (81.5) 20.3 Apo B (mg/dl) (135.9) (143.9) 23.6 Values adjusted for age and subscapular skinfold in parentheses. Ap <.05 ; Bp <.001. positively with VLDL cholesterol (p <.00 1) and inversely with HDL cholesterol (p K.01), apo A-I (p K.001), and apo B (p K.05) in white boys only; these relationships remained significant for HDL cholesterol and apo A-I even after adjusting for age and adiposity. In both races DHEA-S correlated inversely with HDL cholesterol (p K.05), but the associations were not independent of age and adiposity. In white boys androstenedione was positively associated with VLDL cholesterol (p K.01) and negatively associated with LDL cholesterol (p K.05), HDL cholesterol (p K.05), and apo A-I (p K.0 1), and in black boys it was 0) c I /I Testosterone HDL-C 12 k :5 03 E r 145)- VLDL-C -0or -S related inversely to apo B (p <.05). However, none of these relationships persisted after adjustment for the covariates. In boys of both races the adjusted correlations showed a positive association between progesterone and apo A-I. Although estradiol was found to be inversely related to apo A-I in white boys, the relationship was not independent of age and adiposity. Changes in selected lipoprotein variables with changes in testosterone level estimated by a polynomial model indicated that, especially in blacks, the relationship of testosterone to lipoprotein levels was not linear. For example, VLDL cholesterol tended to de- /P White, N=251 AC Black, N ax APO A-I FIGURE 1. Changes in testosterone, VLDL cholesterol. HDL cholesterol, and apo A-I levels with age in adolescent black and white boys. E 501 I E ~~~~~~~~~~~~135 F1 40J- 130} AGE, years Vol. 74, No. 6, December

5 SRINIVASAN et al. TABLE 4 Interrelationships (Spearman correlation coefficients) among endogenous sex hormones in adolescent white (n = 251) and black (n = 258) boys - Bogalusa Heart Study Andro- Estradiol Progesterone stenedione DHEA-S Testosterone White.47c.25c.70c.29c Black.57c.19A.72c.35c Estradiol White.17A 41c 17A Black.08 55C 31C Progesterone White 28c.36c Black.20A 2 IB Androstenedione White.30c Black,40c Ap <.01; Bp <.001; Cp < crease at higher testosterone levels only in black boys (figure 2). The trend in the relationship between HDL cholesterol and testosterone was negative at lower testosterone levels in both races (figure 3), but at higher testosterone levels this trend became positive for black boys and negligible for white boys. Similar trends were noted for apo A-I (data not shown). Since the relationship of testosterone to the serum lipoprotein levels varied depending on hormone levels, the relation of sex hormones to lipoprotein variables was examined by age groups (1 1 to 12 years, 13 to 14 years, and 15 to 17 years) reflecting low, medium, and high testosterone levels (mean + SD levels [ng/ml] for the corresponding age groups were , , and for white boys and , , and for black boys). Partial correlations adjusted for age and adiposity are given in table 7; only data on sex hormones that were significantly related to lipoprotein variables in any age group are included in the table. None of the associations were significant in the 1 1 to 12 year age group, irrespective of race. Testosterone was related positively to VLDL cholesterol (p <.05) in white boys in the 13 to 14 year age group; no relationship was noted for black boys in any of the three age groups. In addition, the relationships of testosterone to HDL cholesterol (p <.05) and apo A-I (p <.05) and DHEA-S to HDL cholesterol (p <.05) were negative only in white boys in the 13 to 14 year age group. Among the older subjects (15 to 17 years), the relationships of testosterone to HDL cholesterol (p <.05) and androstenedione to apo A-I (p <.05) were positive only in black boys. In 13 to 14 year olds, progesterone was positively associated with apo A-I only in black boys, whereas these significant (p <.05) positive relationships were noted in the older boys (15 to 17 years) of both races. Discussion The interrelationships among sex hormone levels and their effect on lipoproteins are complex. The present data show positive associations among sex hormones in both races to varying degrees. The peripheral aromatization of testosterone to estradiol and androstenedione to estrone is well known.20 Peripheral interconversions of the androgens androstenedione and testosterone, and the estrogens estrone and estradiol, have also been reported.2' Adrenal androgens may contribute to the effects of sex hormones during progression of puberty.22 For example, DHEA-S can be TABLE 5 Relation (Spearman rank correlations) of endogenous androgen to serum lipoprotein levels in adolescent white (n = 251) and black (n = 258) boys - Bogalusa Heart Study White Testosterone Black DHEA-S Androstenedione White Black White Black VLDL cholesterol.25c B.14 (.09) (.06) LDL cholesterol -. la A -.08 (-.08) (-.04) HDL cholesterol -.20B A -.14A -.1 A -.02 Apo A-I (. 15)A (.07) (-.05) (-.06) -.29c B _.04 (-.22)c (_.04) Apo B -.13A A (-.05) (-.03) Values adjusted for age and subscapular skinfold are given in parentheses only if unadjusted or adjusted correlation was significant. Ap <.05; Bp <.01; Cp < CIRCULATION

6 PATHOPHYSIOLOGY AND NATURAL HISTORY-ATHEROSCLEROSIS TABLE 6 Relation (Spearman rank correlation) of endogenous progestin and estrogen to serum lipoproteins in adolescent white (n = 251) and black (n = 258) boys - Bogalusa Heart Study Progesterone Estradiol White Black White Black VLDL cholesterol LDL cholesterol HDL cholesterol Apo A-I.09.i5A - 19B -.05 (.03)A (.15)A (-.11) Apo B Values adjusted for age and subscapular skinfold are in parentheses. Partial correlations were included only if unadjusted or adjusted correlation was significant. Ap <.05; Bp <.01. converted to androstenedione and to biologically active testosterone, and DHEA-S can produce androgenic effects in males.23 Therefore, sex hormone-lipoprotein relationships during puberty may depend on the levels and metabolic interrelationships of gonadal and adrenal sex steroids. The present study demonstrates both similarities and differences in the levels of certain endogenous sex hormones and their relation to serum lipoproteins among white and black adolescent boys. Black boys consistently had higher estradiol and lower androstenedione levels than white boys, while the levels of testosterone, DHEA-S, and progesterone were similar in the races. In general it appears that androgens (testosterone, DHEA-S, and androstenedione) have diver gent influences on levels of VLDL and HDL in white and black adolescent boys, while progestin (progesterone) has a similar effect on HDL in both races. These cross-sectional observations are derived from an entire population of free-living children, and therefore should reasonably reflect the influence of endogenous sex hormones on lipoproteins in these racial groups. Exogenous androgens are known to decrease HDL and VLDL levels.2,'24 The spurt of endogenous testosterone and DHEA-S during progression of puberty seems to exert a similar effect of HDL (cholesterol and apo A-I) in white boys 13 to 14 years old. As one would expect given the metabolic interrelationships among lipoproteins,25 the relation of testosterone to VLDL was opposite to that seen for HDL in white boys of the same age group. It is of interest that no such association was seen for HDL or VLDL in black boys 13 to 14 years old. Earlier studies on the testosterone- HDL relationship showed a negative association in white adolescent boys`10 and a rather weak association in Pima Indian adolescent boys,26 suggesting divergence among racial groups. Testosterone level is known to relate inversely with VLDL and positively with HDL in adult white men Our present results show a positive association of testosterone and androstenedione with HDL cholesterol and apo A-I, respectively, only in older (15 to 17 years) black boys who are in the advanced stages of sexual maturation. It is likely that similar trends could also be seen in older white boys. Although exogenous progestational agents are known to decrease HDL levels in women,30' 31 in this study endogenous White, N=251 -D E ci > _J 81 4 Black, N = 258 FIGURE 2. boys. Vol. 74, No. 6, December TESTOSTERONE, ng/mi Relation, estimated by a polynomial model, of VLDL cholesterol to testosterone level in black and white adolescent ---j- 1231

7 SRINIVASAN et al. Black, N= 258 CD E (5 0 Ir White, N= TESTOSTERONE, ng/ml FIGURE 3. Relation, estimated by a polynomial model, of HDL cholesterol to testosterone level in black and white adolescent boys. progesterone was found to relate positively to HDL (apo A-I) in both white and black boys in the oldest age group (15 to 17 years). The progesterone effect on apo A-I was observed to appear even earlier in black boys (13 to 14 year age group), probably reflecting their early maturation. That the sex hormone-lipoprotein associations varied among different age groups suggests the influence of hormonal makeup on lipoproteins. After completion of sexual maturation the impact of endogenous testosterone on lipoprotein levels may be minimal because the levels may have exceeded a threshold. It has been suggested that the positive association in men is not casual but may be related to other factors, such as estradiol levels.9 27 We have previously shown that in pubertal girls endogenous estradiol correlates positively with HDL and negatively with VLDL level. The possibility that higher estradiol levels in blacks (vs whites) dampen the effect of the testosterone spurt on lipoproteins during progression of sexual maturation remains speculative but suggests that a teleologic basis for these hormonal differences be considered. Biochemical mechanisms involved in sex hormonelipoprotein relationships during sexual maturation are virtually unknown. The male-female differences in VLDL and HDL levels, with premenopausal women having lower VLDL and higher HDL than men, are attributed to an estrogen-induced stimulation of lipoprotein lipase and an efficient VLDL catabolism.32 Furthermore, the increased level of HDL in women (vs men) is considered to be due to an estrogen-related TABLE 7 Relation (correlation coeficients) of endogenous sex hormone to serum lipoprotein levels in adolescent boys by age group and race - Bogalusa Heart Study year olds year olds year olds White Black White Black White Black Testosterone vsa: VLDL cholesterol B HDL cholesterol B B Apo A-I C DHEA-S vs HDL cholesterol B Androstenedione vs apo A-I B Progesterone vs apo A-I B.26B.218 APartial correlations adjusted for age and subscapular skinfold. Bp <.05; Cp < CIRCULATION

8 PATHOPHYSIOLOGY AND NATURAL HISTORY-ATHEROSCLEROSIS increase in the apo A-I synthetic rate33 and a decrease in hepatic lipase.34 This implies that the decreased level of HDL in men may be due to an androgen-related decrease in apo A-I synthesis and an increase in hepatic lipase. Thus, in vivo, high levels of testosterone may promote low HDL, while high estradiol may favor low VLDL and high HDL. It should be noted that black boys who had similar levels of testosterone but increased levels of estradiol also had increased levels of HDL cholesterol and apo A-I and lower VLDL cholesterol than white boys. Furthermore, the racerelated differences became pronounced during puberty. Although estradiol did not relate directly to lipoproteins, it may modulate the biologic activities of other sex hormones. With respect to progesterone, the metabolic basis for the surprising positive association between this antiestrogenic hormone and HDL (apo A-I) in both races is not clear. A relationship between sex hormones, carbohydrate tolerance, obesity, and lipid metabolism has been suggested.9 35 However, the observed sex hormone-lipoprotein relationships appear to be independent of adiposity. It is likely that the effects of biologic determinants on lipoproteins vary between races due to inherent metabolic differences. For example, our earlier studies have shown that the association of adiposity with lipoprotein levels was most apparent in white boys and least evident in black boys, even at comparable levels of adiposity.36 Furthermore, with respect to measures of carbohydrate tolerance, black children showed significantly lower fasting plasma glucose levels and higher postglucose plasma insulin levels, insulin/glucose ratios, and insulin/free fatty acid ratios than white children, independent of obesity Alternatively, the divergence between the two racial groups may be related to differences in diet and physical activity. In the Bogalusa Heart Study, macronutrient intake of 10-year-olds was examined four times over the decade (1973 to 1982), including during the present cross-sectional study, and the results showed no significant differences between the races.39 Furthermore, longitudinal comparisons of a cohort of Bogalusa children examined at both 10 and 13 years showed no racial difference in intakes of fat and fatty acids.40 Glueck et al.4` have concluded that, overall, blackwhite differences in dietary intake and physical activity could not account for any race-related differences in HDL cholesterol levels observed in children and adults. Current observations in black and white adolescent boys have implications with respect to the cause of coronary heart disease. The divergent metabolic inter- Vol. 74, No. 6, December 1986 relationships in a common acculturated environment may be crucial for determining the atherogenic profile noted in white vs black adolescent boys. The Bogalusa Heart Study represents the collaborative efforts of many people. We thank the Bogalusa field staff for collecting the data. We are most grateful to the children of Bogalusa, LA and to their parents, without whose cooperation this study would not be possible. References 1. McGill HC Jr, Stem MP: Sex and atherosclerosis. In Paoletti R, Gotto AM Jr, editors: Atherosclerosis reviews. New York, 1979, Raven Press, p Furman RH, Alaupovic P, Howard RP: Effects of androgens and estrogens on serum lipids and the composition and concentration of serum lipoproteins in normolipemic and hyperlipemic states. Prog Biochem Pharmacol 2: 215, Kannel WB, Hjortland MC, McNamara PM, Gordon T: Menopause and risk of cardiovascular disease. Ann Intern Med 35: 447, Kannel WB, Castelli WP, Gordon T: Cholesterol in the prediction of atherosclerotic disease. New perspectives based on the Framingham Study. Ann Intern Med 90: 85, Tyroler HA, Hames CG, Krisan I, Heyden S, Cooper G, Cassel JC: Black-white differences in serum lipids and lipoproteins in Evans County. Prev Med 4: 541, Berenson GS, Srinivasan SR, Cresanta JL, Foster TA, Webber LS: Dynamic changes of serum lipoproteins in children during adolescence and sexual maturation. Am J Epidemiol 113: 157, Morrison JA, Laskarzewski PM, Rach JL, Brookman R, Mellies MJ, Frazer M, Khoury P, de Groot I, Kelly K, Glueck CJ: Lipids, lipoproteins, and sexual maturation during adolescence: The Princeton Maturation Study. Metabolism 28: 641, Insull W, Kirkland RT, Probstfield JL, Keenan BS, Clayton GW: Control of HDL-cholesterol plasma levels by testosterone in pubertal males. Circulation 64(suppl IV): IV-1 13, Laskarzewski PM, Morrison JA, Gutai J, Orchard T, Khoury PR, Glueck CJ: High and low density lipoprotein cholesterols in adolescent boys: Relationship with endogenous testosterone, estradiol, and Quetelet index. Metabolism 32: 262, Srinivasan SR, Sundaram GS, Williamson GD, Webber LS, Berenson GS: Serum lipoproteins and endogenous sex hormones in early life: observations in children with different lipoprotein profiles. Metabolism 34: 861, Berenson GS, McMahan CA, Voors AW, Webber LS, Srinivasan SR, Frank GC, Foster TA, Blonde CV: Cardiovascular risk factors in children. The early natural history of atherosclerosis and essential hypertension. New York, 1980, Oxford University Press, pp Tanner JM: Growth at adolescence, ed 2. Oxford, 1962, Blackwell Scientific Publications, pp National Institutes of Health: Manual of laboratory operations- Lipid Research Clinics Program. Volume I, Lipid and lipoprotein analysis. DHEW Publication No. (NIH) , Washington, D.C., 1974, U.S. Government Printing Office 14. Srinivasan SR, Frerichs RR, Webber LS, Berenson GS: Serum lipoprotein profile in children from a biracial community-the Bogalusa Heart Study. Circulation 54: 309, Srinivasan SR, Freedman DS, Sharma C, Webber LS, Berenson GS: Serum apolipoproteins A-I and B in 2854 children from a biracial community: The Bogalusa Heart Study. Pediatrics 78: 189, Abraham GE: The application of natural steroid radioimmunoassay to gynecologic endocrinology. New York, 1981, Marcel Dekker, p Friedrich E, Jaeger-Whitegiver ER, Bieden M: In Gupta D, editor: Radioimmunoassay of steroid hormones. Weinheim, Germany, 1975, Verlag Chemie, p Snedecor GW, Cochran WG: Statistical methods. Ames, IA, 1980, The Iowa State University, p

9 SRINIVASAN et al. 19. Neter J, Wasserman W: Applied linear statistical models. Homewood, IL, 1985, Richard D. Irwin, p Longscope C, Kato T, Horton R: Conversion of blood androgens to estrogens in normal adult men and women. J Clin Invest 48: 2191, Bleau G, Roberts KD, Chapdelaine A: The in vitro and in vivo uptake and metabolism of steroids in human adipose tissue. J Clin Endocrinol Metab 39: 236, Lee PA, Migeon CJ: Puberty in boys: Correlation of plasma levels of Gonadotropins (LH, FSH), androgens (testosterone, androstenedione, dehydroepiandrosterone and its' sulfate), estrogens (estrone and estradiol) and progestins (progesterone and 17-hydroxyprogesterone). J Clin Endocrinol Metab 41: Drucker WD, Blumberg JM, Gandy HM, David RR. Verde AL: Biologic activity of dehydroepiandrosterone in man. J Clin Endocrinol Metab 35: 48, Glueck CJ, Ford S Jr, Steiner P, Fallat R: Triglyceride removal efficiency and lipoprotein lipases: Effects of oxandrolone. Metabolism 22: 807, Eisenberg S: Plasma lipoprotein conversions: the origin of lowdensity and high-density lipoproteins. Ann NY Acad Sci 348: 30, Bennion LJ, Howard BV, Bennett PH: Pubertal changes in plasmalipids and lipoproteins: Lack of correlation with plasma estradioltestosterone. Clin Res 26: A126, Gutai J, LaPorte R, Kuller L, Wanju D, Falvo-Gerard L, Caggiula A: Plasma testosterone, high density lipoprotein cholesterol and other lipoprotein fractions. Am J Cardiol 48: 897, Nordoy A, Aakvaag A, Thelle D: Sex hormones and high density lipoproteins in healthy males. Atherosclerosis 34: 431, Mendoza S, Osuna A, Zerpa A, Gartside P, Glueck CJ: Hypertriglyceridemia and hypoalpha lipoproteinemia in azoospermic and oligospermic young men: relationships of endogenous testosterone to triglycerides and high density lipoprotein cholesterol metabolism. Metabolism 30: 481, Bradley DD, Wingerd J, Petitti DB, Krauss RM, Ramcharan S: Serum high-density-lipoprotein cholesterol in women using oral contraceptives, estrogen and progestins. N Engl J Med 299: 17, Tikkanen MJ, Nikkila EA, Kuusi T, Sipinen S: Reduction of plasma high-density-lipoprotein, cholesterol and increase of post heparin plasma hepatic lipase activity during progestin treatment. Clin Chim Acta 115: 63, Nikkila EA: Metabolic and endocrine control of plasma high density lipoprotein concentrations. Relation to catabolism of triglyceride-rich lipoproteins. In Gotto AM Jr, Miller NE, Oliver MF, editors: High density lipoproteins and atherosclerosis. New York, 1979, Elsevier/North Holland Biomedical Press, p Schaefer EJ, Zech LA, Jenkins LL, Bronzest TJ, Rubalcaba EA, Lindgren FT, Aamodt RL, Brewer HB Jr: Human apolipoprotein A-I and A-II metabolism. J Lipid Res 23: Kuusi T, Saarinen P, Nikkila EA: Evidence for the role of hepatic endothelial lipase in the metabolism of plasma high density lipoprotein2 in man. Atherosclerosis 36: 599, Orchard TJ, Becker DJ, Kuller LH, Wagener DK, LaPorte RE, Drash AL: Age and sex variations in glucose tolerance and insulin responses: parallels with cardiovascular risk. J Chron Dis 35: Frerichs RR, Webber LS, Srinivasan SR, Berenson GS: Relation of serum lipids and lipoproteins to obesity and sexual maturity in white and black children. Am J Epidemiol 108: 486, Voors AW, Harsha DW, Webber LS. Radhakrishnamurthy B, Srinivasan SR, Berenson GS: Clustering of anthropometric parameters, glucose tolerance, and serum lipids in children with high and low,b- and pre-13-lipoproteins. Bogalusa Heart Study. Arteriosclerosis 2: 346, Radhakrishnamurthy B, Srinivasan SR, Webber LS, Dalferes ER Jr, Berenson GS: Relationship of carbohydrate tolerance to serum lipoprotein profiles in childhood. The Bogalusa Heart Study. Metabolism 34: 850, Farris RP, Cresanta JL, Croft JB, Webber LS, Frank GC, Berenson GS: Macronutrient intakes of 10-year-old children, J Am Diet Assoc 86: 765, Farris RP, Cresanta JL, Frank GC, Webber LS, Berenson GS: Dietary studies of children from a biracial population: intakes of fat and fatty acids in 10- and 13-year-olds. Am J Clin Nutr 39: 114, Glueck CJ, Gartside P, Laskarzewski PM, Khoury P, Tyroler HA: High density lipoprotein cholesterol in blacks and whites: potential ramifications for coronary heart disease. Am Heart J 108: 815, CIRCULATION

10 Racial (black-white) comparisons of the relationship of levels of endogenous sex hormones to serum lipoproteins during male adolescence: the Bogalusa Heart Study. S R Srinivasan, D S Freedman, G S Sundaram, L S Webber and G S Berenson Circulation. 1986;74: doi: /01.CIR Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX Copyright 1986 American Heart Association, Inc. All rights reserved. Print ISSN: Online ISSN: The online version of this article, along with updated information and services, is located on the World Wide Web at: published Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally in Circulation can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for which permission is being requested is located, click Request Permissions in the middle column of the Web page under Services. Further information about this process is available in the Permissions and Rights Question and Answer document. Reprints: Information about reprints can be found online at: Subscriptions: Information about subscribing to Circulation is online at: