Amencan Jounal of Eptdenmotogy Copyright O 1992 by The Johns Hopkins University School of Hygene and Pubic Health Al rights reserved
Vol. 136, No. 9 Printed in U S A
Steven M. Haffner, Katherine K. Gaiber, Philip A. Morales, Helen P. Hazuda, Rodolfo A. Valdez, Braxton D. Mitchell, and Michael P. Stern
There is considerable evidence that lipoprotein(a) (Lp(a)) is a strong independent risk factor for coronary heart disease. Based on their risk factor profile, Mexican Americans have an increased risk of coronary heart disease, yet Mexican Americans have coronary heart disease mortality similar to or lower than that of non-Hispanic whites. The authors therefore attempted to determine whether Mexican Americans had decreased Lp(a) concentrations relative to non-Hispanic whites in the San Antonio Heart Study, a population-based study of diabetes and cardiovascular disease. Lp(a) concentrations (mg/dl) were significantly lower in Mexican Americans (n = 316) than in non-Hispanic whites (n = 242) (men: 10.4 vs. 16.3; women: 11.5 vs. 16.4). In addition, the proportion of persons with Lp(a) concentrations of >30 mg/dl (the threshold at which increased risk of coronary heart disease is believed to occur) was significantly higher in nonHispanic whites than in Mexican Americans (18.6% vs. 7.6%; Mantei-Haenszel odds ratio (adjusted for sex) = 2.79). Age, obesity, body fat distribution, cigarette smoking, alcohol consumption, and glucose and insulin concentrations were not significantly related to Lp(a) levels. Decreased Lp(a) concentrations may account in part for Mexican Americans' relative protection from coronary heart disease mortality. Am J Epidemiol 1992;136:1060-8. coronary disease; diabetes metlitus; Hispanic Americans; lipoproteins
Recently, evidence has been presented suggesting that lipoprotein(a) (Lp(a)) is a strong independent risk factor for coronary heart disease (1-5). Determinants of Lp(a) concentrations are not completely understood, but genetic factors are believed to play a major role (6, 7). Subjects with larger apoprotein(a) isoforms generally have lower concentrations of Lp(a) (5-7). Thus far, Received for publication November 21, 1991, and in final form April 1, 1992 Abbreviations: Cl, confidence interval, Lp(a), lipoprotem(a), OR, odds ratio. From the Division of Clinical Epidemiology, Department of Medicine, University of Texas Health Science Center at San Antono, San Antonio, TX Repnnt requests to Dr. Steven M Haffner, Division of Clinical Epidemiology, Department of Medicine, University of Texas Health Science Center at San Antonio, 7703 Floyd Curl Drive, San Antonio, TX 78284-7873. This work was supported by grants R01HL24799 and R37HL36820 from the National Heart, Lung, and Blood Institute
there has been little study of Lp(a) concentrations in different ethnic groups. Recent data have suggested that blacks have increased levels of Lp(a) relative to Caucasians (8-11). To our knowledge, no previous study has examined whether Mexican Americans have different Lp(a) concentrations than non-Hispanic whites. This issue is of considerable interest, since Mexican Americans have excess non-insulindependent diabetes mellitus (12), greater obesity (13), an unfavorable body fat distribution (14), more hyperinsulinemia (15), and greater insulin resistance (16) relative to non-Hispanic whites. Mexican-American men also have lower cardiovascular mortality (17) and a lower prevalence of myocardial infarction (18) than do non-Hispanic white men. A possible explanation for the decreased rate of myocardial infarction in spite of increased risk factors for coronary
1060
Downloaded from https://academic.oup.com/aje/article-abstract/136/9/1060/81543 by University of Western Sydney user on 10 January 2019
Lipoprotein(a) Concentrations in Mexican Americans and NonHispanic Whites: The San Antonio Heart Study
Lipoprotein(a) Levels in Mexican Americans
MATERIALS AND METHODS
The San Antonio Heart Study is a population-based study of diabetes and cardiovascular disease in Mexican Americans and non-Hispanic whites. From 1979 to 1982, we randomly selected households from several San Antonio, Texas, census tracts: two low-income (barrio) census tracts; two middle-income (transitional) census tracts (60 percent Mexican American/40 percent non-Hispanic white); and a cluster of highincome (suburban) census tracts (10 percent Mexican American/90 percent nonHispanic white) (13). All men and nonpregnant women aged 25-64 years who resided in the randomly selected households were eligible for study participation. Only Mexican Americans were sampled in the barrio. Mexican Americans were defined as individuals whose ancestry and cultural traditions were derived from a Mexican national origin (21). A detailed description of the 19791982 survey has been published previously (13). The study was approved by the Institutional Review Board of the University of Texas Health Science Center at San Antonio. All subjects gave informed consent.
In October 1987, we began an 8-year follow-up study to determine the incidence of non-insulin-dependent diabetes mellitus and cardiovascular disease in the participants. At the follow-up examination, blood specimens were obtained following a 12- to 14-hour fast, and a second specimen was obtained 2 hours after administration of a 75-g glucose equivalent load (Orangedex; Custom Laboratories, Baltimore, Maryland). Plasma glucose concentrations were measured with an Abbott Bichromatic Analyzer (Abbott Laboratories, North Chicago, Illinois). Insulin was measured by a commercial radioimmunoassay (Diagnostic Products Corporation, Los Angeles, California) (15). Diabetes mellitus was diagnosed according to the criteria of the World Health Organization (a fasting plasma glucose level of >140 mg/dl and/or a 2-hour glucose level of >200 mg/dl) (22). Subjects who did not meet the criteria but were being treated with oral antidiabetic agents or insulin were also considered diabetic. Fasting lipids and lipoproteins were measured using methods described previously (13). Anthropometric measurements (height, weight, and subscapular and triceps skinfolds) were obtained after participants had removed their shoes and upper garments and donned an examining gown (14). Body mass index was calculated as weight (in kilograms) divided by height (in meters) squared. The ratio of subscapular skinfold to triceps skinfold (the centrality index) was used as a measure of central adiposity. Lp(a) was measured in serum specimens that had been frozen for an average of 12 months at -70°C using a monoclonal antiLp(a) antibody technique (Terumo Medical Corporation, Elkton, Maryland). The intraand interassay coefficients of variation for this assay were 4 percent and 8 percent, respectively (23). In a recent report using this monoclonal antibody, no crossreactivity with plasminogen, low density lipoprotein cholesterol, very low density lipoprotein cholesterol, or high density lipoprotein cholesterol was observed (24). Statistical analyses included parametric analysis of variance, multiple linear regres-
Downloaded from https://academic.oup.com/aje/article-abstract/136/9/1060/81543 by University of Western Sydney user on 10 January 2019
heart disease (19) is that Mexican Americans have decreased Lp(a) concentrations relative to non-Hispanic whites. Preliminary data from the Strong Heart Study suggest that Native Americans in the US Southwest have lower Lp(a) concentrations than Caucasians (Dr. B. V. Howard, Medlantic Corporation (Washington, DC), personal communication, 1991). Since Mexican Americans have considerable Native American genetic admixture (20), it is plausible that they too may have lower Lp(a) concentrations. In this report, we examine Lp(a) concentrations in Mexican Americans and nonHispanic whites in the San Antonio Heart Study, a population-based study of diabetes and cardiovascular disease. We also examine the relation of Lp(a) concentrations to certain demographic, behavioral, and metabolic variables, including age, sex, obesity, body fat distribution, and glucose and insulin concentrations.
1061
1062
Haffneretal.
The Mantel-Haenszel test was used to evaluate the association of ethnicity and Lp(a) in both sexes simultaneously (table 2). The subjects described in this report were Mexican Americans and non-Hispanic whites who originally resided in one of the two middle-income census tracts and a 50 percent sample of Mexican Americans who originally resided in one of the low-income census tracts. All had attended the 8-year follow-up examination. RESULTS
Table I presents anthropometric, behavioral, and metabolic characteristics of the study sample by ethnic group and sex. Mexican Americans were more obese and had an unfavorable body fat distribution and higher glucose and insulin concentrations than non-Hispanic whites. In both men and women, Mexican-American smokers smoked fewer cigarettes per day than nonHispanic white smokers. Mean Lp(a) concentrations (mg/dl) were lower in Mexican Americans than in non-Hispanic whites (men: 10.4 vs. 16.3; women: 11.5 vs. 16.4;
TABLE 1. Mean values for anthropometric, behavioral, and metabolic variables, by ethnicity and sex, San Antonio Heart Study follow-up (phase I), 1987-1990 Men MA*
No. of subjects Age (years) Body mass index}: Centrality index§ Cigarettes/day (among current smokers) Alcohol consumption (g/week) % diabetic Fasting glucose (mg/dl) 2-rtour glucose (mg/dl) Fasting insulin (^U/ml) 2-hour insulin (MLI/IDI) Triglyceride (mg/dl) HDL* cholesterol (mg/dl) LDL* cholesterol (mg/dl) Total cholesterol (mg/dl) UpoproteirKa) (mg/dl)
Women Ethnicity
Sex
182 51 ±1.2 29.6 ± 0.69 1.16 ±0.03
135 55 ± 0.95 27.8 ± 0.7 0.92 ± 0.03
0.01
Vol. 136, No. 9 Printed in U S A
Steven M. Haffner, Katherine K. Gaiber, Philip A. Morales, Helen P. Hazuda, Rodolfo A. Valdez, Braxton D. Mitchell, and Michael P. Stern
There is considerable evidence that lipoprotein(a) (Lp(a)) is a strong independent risk factor for coronary heart disease. Based on their risk factor profile, Mexican Americans have an increased risk of coronary heart disease, yet Mexican Americans have coronary heart disease mortality similar to or lower than that of non-Hispanic whites. The authors therefore attempted to determine whether Mexican Americans had decreased Lp(a) concentrations relative to non-Hispanic whites in the San Antonio Heart Study, a population-based study of diabetes and cardiovascular disease. Lp(a) concentrations (mg/dl) were significantly lower in Mexican Americans (n = 316) than in non-Hispanic whites (n = 242) (men: 10.4 vs. 16.3; women: 11.5 vs. 16.4). In addition, the proportion of persons with Lp(a) concentrations of >30 mg/dl (the threshold at which increased risk of coronary heart disease is believed to occur) was significantly higher in nonHispanic whites than in Mexican Americans (18.6% vs. 7.6%; Mantei-Haenszel odds ratio (adjusted for sex) = 2.79). Age, obesity, body fat distribution, cigarette smoking, alcohol consumption, and glucose and insulin concentrations were not significantly related to Lp(a) levels. Decreased Lp(a) concentrations may account in part for Mexican Americans' relative protection from coronary heart disease mortality. Am J Epidemiol 1992;136:1060-8. coronary disease; diabetes metlitus; Hispanic Americans; lipoproteins
Recently, evidence has been presented suggesting that lipoprotein(a) (Lp(a)) is a strong independent risk factor for coronary heart disease (1-5). Determinants of Lp(a) concentrations are not completely understood, but genetic factors are believed to play a major role (6, 7). Subjects with larger apoprotein(a) isoforms generally have lower concentrations of Lp(a) (5-7). Thus far, Received for publication November 21, 1991, and in final form April 1, 1992 Abbreviations: Cl, confidence interval, Lp(a), lipoprotem(a), OR, odds ratio. From the Division of Clinical Epidemiology, Department of Medicine, University of Texas Health Science Center at San Antono, San Antonio, TX Repnnt requests to Dr. Steven M Haffner, Division of Clinical Epidemiology, Department of Medicine, University of Texas Health Science Center at San Antonio, 7703 Floyd Curl Drive, San Antonio, TX 78284-7873. This work was supported by grants R01HL24799 and R37HL36820 from the National Heart, Lung, and Blood Institute
there has been little study of Lp(a) concentrations in different ethnic groups. Recent data have suggested that blacks have increased levels of Lp(a) relative to Caucasians (8-11). To our knowledge, no previous study has examined whether Mexican Americans have different Lp(a) concentrations than non-Hispanic whites. This issue is of considerable interest, since Mexican Americans have excess non-insulindependent diabetes mellitus (12), greater obesity (13), an unfavorable body fat distribution (14), more hyperinsulinemia (15), and greater insulin resistance (16) relative to non-Hispanic whites. Mexican-American men also have lower cardiovascular mortality (17) and a lower prevalence of myocardial infarction (18) than do non-Hispanic white men. A possible explanation for the decreased rate of myocardial infarction in spite of increased risk factors for coronary
1060
Downloaded from https://academic.oup.com/aje/article-abstract/136/9/1060/81543 by University of Western Sydney user on 10 January 2019
Lipoprotein(a) Concentrations in Mexican Americans and NonHispanic Whites: The San Antonio Heart Study
Lipoprotein(a) Levels in Mexican Americans
MATERIALS AND METHODS
The San Antonio Heart Study is a population-based study of diabetes and cardiovascular disease in Mexican Americans and non-Hispanic whites. From 1979 to 1982, we randomly selected households from several San Antonio, Texas, census tracts: two low-income (barrio) census tracts; two middle-income (transitional) census tracts (60 percent Mexican American/40 percent non-Hispanic white); and a cluster of highincome (suburban) census tracts (10 percent Mexican American/90 percent nonHispanic white) (13). All men and nonpregnant women aged 25-64 years who resided in the randomly selected households were eligible for study participation. Only Mexican Americans were sampled in the barrio. Mexican Americans were defined as individuals whose ancestry and cultural traditions were derived from a Mexican national origin (21). A detailed description of the 19791982 survey has been published previously (13). The study was approved by the Institutional Review Board of the University of Texas Health Science Center at San Antonio. All subjects gave informed consent.
In October 1987, we began an 8-year follow-up study to determine the incidence of non-insulin-dependent diabetes mellitus and cardiovascular disease in the participants. At the follow-up examination, blood specimens were obtained following a 12- to 14-hour fast, and a second specimen was obtained 2 hours after administration of a 75-g glucose equivalent load (Orangedex; Custom Laboratories, Baltimore, Maryland). Plasma glucose concentrations were measured with an Abbott Bichromatic Analyzer (Abbott Laboratories, North Chicago, Illinois). Insulin was measured by a commercial radioimmunoassay (Diagnostic Products Corporation, Los Angeles, California) (15). Diabetes mellitus was diagnosed according to the criteria of the World Health Organization (a fasting plasma glucose level of >140 mg/dl and/or a 2-hour glucose level of >200 mg/dl) (22). Subjects who did not meet the criteria but were being treated with oral antidiabetic agents or insulin were also considered diabetic. Fasting lipids and lipoproteins were measured using methods described previously (13). Anthropometric measurements (height, weight, and subscapular and triceps skinfolds) were obtained after participants had removed their shoes and upper garments and donned an examining gown (14). Body mass index was calculated as weight (in kilograms) divided by height (in meters) squared. The ratio of subscapular skinfold to triceps skinfold (the centrality index) was used as a measure of central adiposity. Lp(a) was measured in serum specimens that had been frozen for an average of 12 months at -70°C using a monoclonal antiLp(a) antibody technique (Terumo Medical Corporation, Elkton, Maryland). The intraand interassay coefficients of variation for this assay were 4 percent and 8 percent, respectively (23). In a recent report using this monoclonal antibody, no crossreactivity with plasminogen, low density lipoprotein cholesterol, very low density lipoprotein cholesterol, or high density lipoprotein cholesterol was observed (24). Statistical analyses included parametric analysis of variance, multiple linear regres-
Downloaded from https://academic.oup.com/aje/article-abstract/136/9/1060/81543 by University of Western Sydney user on 10 January 2019
heart disease (19) is that Mexican Americans have decreased Lp(a) concentrations relative to non-Hispanic whites. Preliminary data from the Strong Heart Study suggest that Native Americans in the US Southwest have lower Lp(a) concentrations than Caucasians (Dr. B. V. Howard, Medlantic Corporation (Washington, DC), personal communication, 1991). Since Mexican Americans have considerable Native American genetic admixture (20), it is plausible that they too may have lower Lp(a) concentrations. In this report, we examine Lp(a) concentrations in Mexican Americans and nonHispanic whites in the San Antonio Heart Study, a population-based study of diabetes and cardiovascular disease. We also examine the relation of Lp(a) concentrations to certain demographic, behavioral, and metabolic variables, including age, sex, obesity, body fat distribution, and glucose and insulin concentrations.
1061
1062
Haffneretal.
The Mantel-Haenszel test was used to evaluate the association of ethnicity and Lp(a) in both sexes simultaneously (table 2). The subjects described in this report were Mexican Americans and non-Hispanic whites who originally resided in one of the two middle-income census tracts and a 50 percent sample of Mexican Americans who originally resided in one of the low-income census tracts. All had attended the 8-year follow-up examination. RESULTS
Table I presents anthropometric, behavioral, and metabolic characteristics of the study sample by ethnic group and sex. Mexican Americans were more obese and had an unfavorable body fat distribution and higher glucose and insulin concentrations than non-Hispanic whites. In both men and women, Mexican-American smokers smoked fewer cigarettes per day than nonHispanic white smokers. Mean Lp(a) concentrations (mg/dl) were lower in Mexican Americans than in non-Hispanic whites (men: 10.4 vs. 16.3; women: 11.5 vs. 16.4;
TABLE 1. Mean values for anthropometric, behavioral, and metabolic variables, by ethnicity and sex, San Antonio Heart Study follow-up (phase I), 1987-1990 Men MA*
No. of subjects Age (years) Body mass index}: Centrality index§ Cigarettes/day (among current smokers) Alcohol consumption (g/week) % diabetic Fasting glucose (mg/dl) 2-rtour glucose (mg/dl) Fasting insulin (^U/ml) 2-hour insulin (MLI/IDI) Triglyceride (mg/dl) HDL* cholesterol (mg/dl) LDL* cholesterol (mg/dl) Total cholesterol (mg/dl) UpoproteirKa) (mg/dl)
Women Ethnicity
Sex
182 51 ±1.2 29.6 ± 0.69 1.16 ±0.03
135 55 ± 0.95 27.8 ± 0.7 0.92 ± 0.03
0.01
