Journal of the American College of Cardiology Vol. 41, No. 6, 2003 2003 by the American College of Cardiology Foundation ISSN 0735-1097/03/$30.00 Published by Elsevier Science Inc. doi:10.1016/s0735-1097(03)00052-4 Influence of Systolic Blood Pressure and Body Mass Index on Left Ventricular Structure in Healthy African-American and White Young Adults: The CARDIA Study Richard Lorber, MD, Samuel S. Gidding, MD, FACC, Martha L. Daviglus, MD, PHD,* Laura A. Colangelo, MS,* Kiang Liu, PHD,* Julius M. Gardin, MD, FACC Chicago, Illinois; Wilmington, Delaware; Cleveland, Ohio; and Irvine, California OBJECTIVES BACKGROUND METHODS RESULTS CONCLUSIONS In the Coronary Artery Risk Development in Young Adults (CARDIA) Study comprised of a generally healthy, biracial cohort of 28- to 40-year-old adults, we sought to characterize the distribution of left ventricular (LV) mass and LV geometry and the relationship of systolic blood pressure (SBP), body mass index (BMI), and fasting insulin to LV mass and geometry. Left ventricular mass is a risk factor for cardiovascular morbidity and mortality. Two-dimensionally guided M-mode echocardiograms were used to calculate LV mass index (g/height 2.7 ) and geometry. Black men had highest LV mass index followed by white men, black women, and white women. Blacks had higher LV wall thickness/diameter ratios than whites. Left ventricular hypertrophy was present in 2% of the cohort. Going from highest to lowest quartile for LV mass index and LV wall thickness/diameter ratio, SBP and BMI were highest in those with the highest LV mass index and LV wall thickness/diameter ratio. Increasing BMI and SBP over a 10-year interval was also strongly related to LV structure in most race/gender groups. In a generally healthy young adult cohort, LV structure as defined by LV mass and geometry is associated with SBP and BMI at levels generally considered normal. (J Am Coll Cardiol 2003;41:955 60) 2003 by the American College of Cardiology Foundation From the *Department of Preventive Medicine, Northwestern University Feinberg School of Medicine, Chicago, Illinois; Nemours Cardiac Center, dupont Hospital for Children, Wilmington, Delaware; Department of Pediatrics at the Cleveland Clinic, Cleveland, Ohio; and Division of Cardiology, Department of Medicine, University of California, Irvine, California. Supported by contracts N01-HC-48047 through 48050 and N01-HC-95095 from the National Heart, Lung, and Blood Institute, National Institutes of Health. Manuscript received June 28, 2002; revised manuscript received December 10, 2002, accepted December 12, 2002. Blood pressure in the high normal range and excessive body weight (often associated with hyperinsulinemia) are cardiovascular risk factors with long-term effects on left ventricular (LV) structure and function (1 3). It is known that LV mass assessed by echocardiogram is an independent risk factor for cardiovascular morbidity and mortality (4,5). Concentric hypertrophy (increased LV mass with higher LV wall thickness/diameter ratio) confers higher risk than eccentric hypertrophy, whereas concentric remodeling (normal LV mass with increased thickness/diameter ratio) confers higher risk than normal LV mass with a normal wall thickness to diameter ratio (6). However, there is little information in the literature on LV hypertrophy and geometry in young adults and the distribution of LV geometry in generally healthy young adults. The Coronary Artery Risk Development in Young Adults (CARDIA) Study is a prospective, multi-center, epidemiologic study of cardiovascular risk in a biracial cohort of young adults age 18 to 30 years at the initial examination (7). The goals of this report are: 1) to characterize the distribution of LV mass and LV geometry (LV wall thickness/diameter ratio) in a generally healthy cohort of young adults; 2) to examine the relationship of LV mass and geometry to systolic blood pressure (SBP), body mass index (BMI), and fasting serum insulin levels, both in cross-sectional and longitudinal models; and 3) to determine if significant racial differences exist in these relationships. METHODS In the year 10 examination of the CARDIA study, echocardiograms were obtained in adults 28 to 40 years old (n 1,618; 457 white men and 280 black men; 492 white and 389 black women) in Minneapolis and Chicago in 1995 through 1996. Protocols have been approved by the institutional review boards of participating institutions. Participants were seated quietly for 5 min, and arm blood pressure was measured three times using a random-zero cuff sphygmomanometer, with the average of the second and third readings used in this analysis. While the participants were wearing light clothing and no shoes, height was measured to the nearest 0.5 cm, and weight was measured to the nearest 0.5 lb. Baseline fasting insulin levels were measured at Linco Research Inc. (St. Charles, Missouri) by radioimmunoassay using an overnight equilibrium incubation (8). Other cardiovascular risk factors, including family history, were not considered in this analysis, because previous reports and other analyses of the data set have shown small or insignificant correlations of these factors with LV mass and LV remodeling (3,9). Two-dimensionally directed M-mode echocardiograms were performed on 1,618 participants using a protocol
956 Lorber et al. JACC Vol. 41, No. 6, 2003 Effect of SBP and BMI on Left Ventricular Structure March 19, 2003:955 60 Abbreviations and Acronyms BMI body mass index CARDIA Coronary Artery Risk Development in Young Adults LV left ventricle, left ventricular LVIDd left ventricular internal dimension at enddiastole PWTd left ventricular posterior wall thickness at end-diastole SBP systolic blood pressure VSTd ventricular septal thickness at end-diastole similar to the protocol used in year 5 of the CARDIA examination (9). Echocardiograms were recorded onto super-vhs tape using an Acuson cardiac ultrasound machine and a standardized recording protocol. Measurements were made from digitized images using a Tom Tec/ Freeland off-line image analysis system (Boulder, Colorado). Strict quality control measures were applied. Left ventricular mass was derived from the formula: LV mass (g) 0.80 1.04 [(VSTd LVIDd PWTd) 3 (LVIDd) 3 ] 0.6, where VSTd is ventricular septal thickness at end-diastole, LVIDd is LV internal dimension at end-diastole, and PWTd is LV posterior wall thickness at end-diastole (9,10). Left ventricular mass was indexed to body size by dividing raw LV mass by height 2.7 (g/ht 2.7 ) (10). The intra- and inter-reader coefficient of variation for LV mass calculation was 10%. All analyses of LV mass were performed using the index of g/ht 2.7. Relative wall thickness was calculated as the ratio PWTd VSTd/LVIDd and was used as the index for LV geometry. From the 1,618 participants from the Chicago and Minneapolis field centers on whom echocardiograms were performed, 260 were excluded from analyses for the following: missing data on variables used to define relative wall thickness or LV mass index (n 65), missing fasting time or fasting time of 8 h(n 88), sex change procedure during the follow-up period (n 1), and having poor scores on echocardiography (n 106), leaving 1,358 participants (3,9). Data analysis. Descriptive statistics were calculated for all study variables. Histograms were developed for the distribution of indexed LV mass and LV geometry, and the coefficient of variation was computed for LV mass and LV geometry within each race/gender group. A cut point of 51 g/m 2.7 was established for the diagnosis of LV hypertrophy (10). The relationships among LV mass, LV geometry, SBP, BMI, and insulin were examined in three ways. First, for both LV mass and LV geometry, quartiles of the distribution for the entire cohort were established. Then, for each variable and race/gender group, three subgroups were established based on the cohort-wide cut points: the lowest quartile, middle half, and upper quartile. Mean values of SBP, BMI, and fasting serum insulin were calculated for each subgroup. Group comparisons were made using oneway analysis of variance. Finally, the trends of BMI and SBP over the years 0, 2, 5, 7, and 10 were assessed for associations with year 10 LV mass index and LV geometry by first fitting for BMI and SBP a separate regression line to the 10-year follow-up for each person and then computing the Pearson correlation coefficient between the personspecific slopes estimated from the regression models and LV mass index and LV geometry. A linear random coefficients model was used to obtain the person-specific slopes. RESULTS Baseline characteristics for the cohort by race and gender are presented in Table 1. Black men had the highest SBP; black women had significantly higher values for BMI and fasting insulin. Among echocardiographic variables used to calculate LV mass, men had higher LV diameters in diastole than did women. Black men had higher LV mass index than all other groups, followed by white men, black women, and white women. Blacks had higher LV wall thickness/ diameter ratios than whites; white men had higher ratios than white women. Figure 1 shows the distribution of LV mass/height 2.7. Men had higher LV mass than women, and blacks higher LV mass than whites. Note that the range of the distribution within the cohort is broad, with 82% between 20 and 40 g/ht 2.7. Only 2% of the cohort had LV mass above 51 g/m 2.7 : 5% (11/243) of black men, 1% (4/376) of white men, 3% (11/323) of black women, and 1% (3/416) of white women. The 97.5% cut points were 56.7, 46.9, 52.9, and Table 1. Baseline Data From Year 10 of the CARDIA Study on All Participants (n 1,358) by Race and Gender Black Men (n 243) White Men (n 376) Black Women (n 323) White Women (n 416) p Value Age (yrs) 34.3 (3.8) 35.6 (3.2) 34.4 (4.0) 35.6 (3.3) 0.0001 1,3,4,6 SBP (mm Hg) 114 (10) 110 (9) 109 (14) 102 (10) 0.0001 1,2,3,5,6 BMI (kg/m 2 ) 27.3 (5.1) 26.3 (3.8) 29.3 (7.3) 25.2 (5.6) 0.0001 2,3,4,6 Insulin (uu/ml) 12.9 (7.7) 11.8 (6.7) 15.1 (10.2) 11.0 (6.6) 0.0001 2,3,4,6 LV mass (g/m 2.7 ) 36.5 (8.6) 33.3 (6.7) 33.3 (8.8) 29.4 (6.7) 0.0001 1,2,3,5,6 Relative wall thickness 0.33 (0.05) 0.31 (0.05) 0.33 (0.06) 0.30 (0.05) 0.0001 1,3,4,5,6 LVIDd (cm) 5.3 (0.5) 5.3 (0.4) 4.8 (0.4) 4.8 (0.4) 0.0001 2,3,4,5 Data are expressed as mean (SD). In pairwise comparisons, p 0.008 (p 0.05 overall using Bonferroni s method) for: 1: black men versus white men; 2: black men versus black women; 3: black men versus white women; 4: white men versus black women; 5: white men versus white women; 6: black women versus white women. BMI body mass index; CARDIA Coronary Artery Risk Development in Young Adults; LV left ventricular; LVIDd left ventricular internal dimension, diastole; SBP systolic blood pressure.
JACC Vol. 41, No. 6, 2003 March 19, 2003:955 60 Lorber et al. Effect of SBP and BMI on Left Ventricular Structure 957 Figure 1. Distribution of left ventricular mass index (g/ht 2.7 ) by race and gender. 47.1 g/m 2.7 for black men, white men, black women, and white women, respectively. Figure 2 shows the distribution of the LV geometry. The 97.5% cut points for LV geometry were 0.452, 0.418, 0.456, and 0.420, respectively. The distributions for LV geometry have less dispersion than do those for LV mass, with coefficients of variation for LV geometry ranging from 15.7% to 18.7% in the four race/ gender groups. In comparison, the coefficients of variation were larger for LV mass, indicating greater variability; they ranged from 20.2% to 26.5%. Tables 2 and 3 show race- and gender-specific mean values for SBP, BMI, and fasting serum insulin for the LV mass and relative wall thickness subgroups. In general, there was a consistent and graded relationship between risk factors and echocardiographic variables, with the highest values for each risk factor coinciding with the highest values for LV mass and relative wall thickness. These results were statistically significant for all race/gender groups for LV mass (with the exception of insulin for men and SBP for white men) and for LV geometry (with the exception of insulin for men and BMI for black men). The cumulative 10-year effects of SBP and BMI were tested in a longitudinal model by assessing a unique regression line for each person for each variable over the course of the study (Table 4) and correlating the slopes from the regression lines with LV mass index and LV geometry. This showed that black men and women with increasing SBP and BMI over time (indicated by positive slope coefficients) had higher LV mass index. White women with increasing BMI over time also had higher LV mass index. In black women, the same was true for LV geometry. DISCUSSION Left ventricular mass and geometry occupy an unusual position in the assessment of cardiovascular risk in that these factors are not only in part determined by conventional cardiac risk factors such as BMI and SBP but also function independently as intermediate risk factors themselves. Blacks and men generally had distributions skewed toward increased risk. Left ventricular mass indexed for height 2.7 has a relatively broad distribution in the general population. Extremes of the distribution were uncommon; only a few individuals were identified with LV hypertrophy, and this was more common in African Americans than whites. Most important, increasing SBP and BMI over time correlated with higher LV mass in black men and women and with LV geometry in black women (3). There was a relatively broad range for LV mass, suggesting that LV mass (and its response to physiologic stress) may be controlled by genetic factors which create interindividual variability. Genes responsible for LV hypertrophy are being actively investigated, both as causes of various forms of cardiomyopathy and as being responsible for
958 Lorber et al. JACC Vol. 41, No. 6, 2003 Effect of SBP and BMI on Left Ventricular Structure March 19, 2003:955 60 Figure 2. Distribution of left ventricular geometry by race and gender. variation in LV structure within the normal population. Future studies of LV mass and geometry may wish to consider whether current cut points for LV hypertrophy are too liberal and whether increased morbidity is observed in the upper quintile of the distribution as well as in the extreme upper end of the distribution. There is relatively little information available on LV geometry in a generally healthy population. It is known that patients with hypertension can have concentric remodeling (increased LV wall thickness/diameter ratio), and the presence of this condition may be associated with increased cardiovascular risk independent of other risk factors Table 2. Relationship of Risk Factors (SBP, BMI, and Insulin) to LV Mass Index Group by Race and Gender at Year 10 High (>37.37 g/ht 2.7 ) Middle (27.05 37.33 g/ht 2.7 ) Low (<27.04 g/ht 2.7 ) p Value Black men SBP (mm Hg) 116 (1.0) 113 (0.9) 108 (2.0) 0.0007 BMI (kg/m 2 ) 29.1 (0.48) 26.6 (0.44) 23.5 (0.96) 0.0001 Insulin (uu/ml) 13.8 (0.77) 12.6 (0.70) 10.8 (1.53) 0.18 White men SBP (mm Hg) 112 (0.9) 110 (0.6) 109 (1.2) 0.11 BMI (kg/m 2 ) 28.3 (0.35) 25.8 (0.25) 24.4 (0.44) 0.0001 Insulin (uu/ml) 12.1 (0.67) 11.6 (0.47) 11.7 (0.83) 0.82 Black women SBP (mm Hg) 116 (1.4) 108 (1.1) 104 (1.5) 0.0001 BMI (kg/m 2 ) 34.0 (0.66) 29.7 (0.50) 23.2 (0.70) 0.0001 Insulin (uu/ml) 18.2 (1.05) 14.9 (0.80) 12.0 (1.13) 0.0003 White women SBP (mm Hg) 105 (1.3) 103 (0.7) 99 (0.7) 0.0001 BMI (kg/m 2 ) 31.4 (0.70) 25.9 (0.35) 22.7 (0.37) 0.0001 Insulin (uu/ml) 16.1 (0.89) 11.0 (0.45) 9.4 (0.48) 0.0001 Data are expressed as mean (SE). Abbreviations as in Table 1.
JACC Vol. 41, No. 6, 2003 March 19, 2003:955 60 Lorber et al. Effect of SBP and BMI on Left Ventricular Structure 959 Table 3. Relationship of Risk Factors (SBP, BMI, and Insulin) to Relative Wall Thickness Subgroup by Race and Gender at Year 10 High (>0.346) Middle (0.280 0.345) Low (<0.279) p Value Black men SBP (mm Hg) 116 (1.1) 113 (0.9) 111 (1.8) 0.04 BMI (kg/m 2 ) 27.6 (0.55) 27.4 (0.46) 26.1 (0.89) 0.36 Insulin (uu/ml) 14.3 (0.82) 12.0 (0.68) 12.7 (1.33) 0.09 White men SBP (mm Hg) 112 (1.0) 110 (0.7) 108 (0.9) 0.008 BMI (kg/m 2 ) 27.6 (0.40) 26.2 (0.27) 25.2 (0.37) 0.0001 Insulin (uu/ml) 13.0 (0.73) 11.5 (0.48) 11.1 (0.68) 0.13 Black women SBP (mm Hg) 115 (1.4) 108 (1.0) 106 (1.6) 0.0001 BMI (kg/m 2 ) 31.1 (0.74) 29.4 (0.56) 26.5 (0.89) 0.0003 Insulin (uu/ml) 17.7 (1.03) 14.9 (0.79) 12.0 (1.24) 0.002 White women SBP (mm Hg) 106 (1.1) 102 (0.7) 100 (0.8) 0.0001 BMI (kg/m 2 ) 28.3 (0.65) 24.7 (0.38) 24.5 (0.46) 0.0001 Insulin (uu/ml) 13.2 (0.78) 10.7 (0.45) 10.3 (0.55) 0.007 Data are expressed as mean (SE). Abbreviations as in Table 1. (6,11,12). The presence of concentric remodeling in the absence of hypertrophy may confer an intermediate level of risk compared with the risk conferred by concentric hypertrophy (12). However, Krumholz et al. (13) did not confirm these findings; in contrast, they found that in a population of subjects 40 years of age without cardiovascular disease, little prognostic value of LV geometry was added to that available from LV mass. Thus, the overall clinical significance of LV geometry requires further study (14). The relationship of elevated insulin levels (or insulin resistance) to increased LV mass is not completely understood. Verdecchia et al. (15) found that insulin and insulinlike growth factor-1 are important determinants of LV mass and geometry in patients with essential hypertension and that the relationship was independent of gender, obesity, and blood pressure. In the Bogalusa Heart Study, after stratifying for both insulin level and obesity, insulin was independently related to LV mass only in those with the most severe obesity (16). In the current study, fasting insulin was not independent of the effects of SBP and BMI. Left ventricular hypertrophy was more prevalent in the black population in this study, as was elevated relative wall thickness. This may be due in part to higher SBP, BMI, and positive family history. In longitudinal analyses of the Table 4. Pearson Correlation Coefficients of LV Mass Index and Relative Wall Thickness With Slope From Individual Regressions of Longitudinal BMI and SBP Variable Black Men White Men Black Women White Women LV mass index BMI 0.14 0.06 0.39 0.22 SBP 0.07 0.05 0.26 0.11 Relative wall thickness BMI 0.06 0.10* 0.13 0.10* SBP 0.08 0.04 0.22 0.05 *p 0.10; p 0.05; p 0.0001. Abbreviations as in Table 1. CARDIA cohort, black women were the only race/gender group to have an increase in LV mass over a five-year follow-up interval (3). This study has shown that, at a young age, LV structure as defined by LV mass and geometry is associated with SBP and body size at levels seen in a generally healthy population of young adults (12,17). Sustained exposure over time, that is, increasing SBP and BMI, increases these effects. These findings are consistent with the knowledge that acquired cardiovascular disease evolves over a considerable period of time. The relationship of cardiovascular mortality to risk factors is continuous and graded with increases in risk observed well within the normal range (18). Though individuals at the adverse extreme of the population distribution of risk factors have substantial increases in coronary disease rates, the majority of events occur at older ages in those with either average or slightly increased risk profiles. Our data suggest that healthy individuals with favorable changes in SBP and BMI over time will have favorable changes in LV structure, further emphasizing the importance of public health measures to lower SBP and BMI in the general population and in African Americans in particular. Reprint requests and correspondence: Dr. Martha L. Daviglus, 680 N. Lake Shore Drive, Suite 1102, Chicago, Illinois 60611. E-mail: daviglus@northwestern.edu. REFERENCES 1. Stamler J, Stamler R, Neaton JD. Blood pressure, systolic and diastolic, and cardiovascular risks. U.S. population data. Arch Intern Med 1993;153:598 615. 2. Hubert HH, Feinlieb M, McNamara PM, Castelli WP. Obesity as an independent risk factor for cardiovascular disease: a 26-year follow-up of participants in the Framingham Heart Study. Circulation 1983;67: 968 77. 3. Gardin JM, Brunner D, Schreiner P, et al. Demographics and correlates of 5-year change in echocardiographic left ventricular mass
960 Lorber et al. JACC Vol. 41, No. 6, 2003 Effect of SBP and BMI on Left Ventricular Structure March 19, 2003:955 60 in young black and white adult men and women: the CARDIA Study. J Am Coll Cardiol 2002;40:529 35. 4. Casale PN, Devereux RB, Milner M, et al. Value of echocardiographic measurement of left ventricular mass in predicting cardiovascular morbid events in hypertensive men. Ann Intern Med 1986;105:173 8. 5. Levy D, Garrison RJ, Savage DD, Kannel WB, Castelli WP. Prognostic implications of echocardiographically determined left ventricular mass in the Framingham Heart Study. N Engl J Med 1990;322: 1561 6. 6. Koren MJ, Devereux RB, Casale PN, Savage DD, Laragh JH. Relation of left ventricular mass and geometry to morbidity and mortality in uncomplicated essential hypertension. Ann Intern Med 1991;114:345 52. 7. Friedman GD, Cutter GR, Donahue RP, et al. CARDIA: study design, recruitment, and some characteristics of examined subjects. J Clin Epidemiol 1988;41:1105 16. 8. Haffner SM, Bowsher RR, Mykkanen L, et al. Proinsulin and specific insulin concentration in high- and low-risk populations for NIDDM. Diabetes 1994;43:1490 3. 9. Gardin JM, Wagenknecht LE, Anton-Culver H, et al. Relationship of cardiovascular risk factors to echocardiographic left ventricular mass in healthy young black and white adult men and women. The CARDIA Study. Circulation 1995;92:380 7. 10. de Simone G, Devereux RB, Daniels SR, Koren MJ, Meyer RA. Effect of growth on variability of left ventricular mass: assessment of the allometric signals in adults and children and their capacity to predict cardiovascular risk. J Am Coll Cardiol 1995;25:1056 62. 11. Daniels SR, Loggie JMH, Khoury P, Kimball TR. Left ventricular geometry and severe left ventricular hypertrophy in children and adolescents with essential hypertension. Circulation 1998;97:1907 11. 12. Ganau A, Devereux RB, Roman MJ, et al. Patterns of left ventricular hypertrophy and geometric remodeling in essential hypertension. J Am Coll Cardiol 1992;19:1550 8. 13. Krumholz HM, Larson M, Levy D. Prognosis of left ventricular geometric patterns in the Framingham Heart Study. J Am Coll Cardiol 1995;25:879 84. 14. Devereux RB. Left ventricular geometry, pathophysiology and prognosis. J Am Coll Cardiol 1995;25:885 7. 15. Verdecchia P, Reboldi G, Schillaci G, et al. Circulating insulin and insulin growth factor-1 are independent determinants of left ventricular mass and geometry in essential hypertension. Circulation 1999; 100:1802 7. 16. Urbina EM, Gidding SS, Bao W, Berenson GS. Correlation of fasting blood sugar and insulin with left ventricular mass in healthy children and adolescents: the Bogalusa Heart Study. Am Heart J 1999;138: 122 7. 17. Lauer MS, Anderson KM, Kannel WB, Levy D. The impact of obesity on left ventricular mass and geometry. The Framingham Heart Study. JAMA 1991;266:231 6. 18. Neaton JD, Wentworth D, for the Multiple Risk Factor Intervention Trial Research Group. Serum cholesterol, blood pressure, cigarette smoking, and death from coronary heart disease. Arch Intern Med 1992;152:56 64.