Echocardiographic left ventricular systolic function and volumes in young adults: Distribution and factors influencing variability

Echocardiographic left ventricular systolic function and volumes in young adults: Distribution and factors influencing variability Low left ventricular ejection fraction (LVEF), a measure of global systolic left ventricular dysfunction, is associated with an increased risk of recurrent coronary events or death in persons with cardiac disease. There are few data on the distribution of resting LVEF and component volumes in healthy young adults or on any association of LVEF with coronary risk factors. LVEF and left ventricular end-diastolic and end-systolic volumes (LVEDV and LVESV, respectively) were measured by two-dimensional echocardiography in 1782 men and women 23 to 35 years old without self-reported heart disease (other than mitral valve prolapse, n = 53) who were participants in the multicenter Coronary Artery Risk Development in Young Adults study. Factors analyzed as potential contributors to LVEF, LVEDV, and LVESV included age, gender, race, blood pressure, alcohol use, current smoking, family history of myocardial infarction, total and high-density lipoprotein cholesterol concentrations, obesity, reported physical activity, and fitness as assessed by treadmill exercise testing. LVEF was lower in men (mean 62.6%, SD 5.7%) than in women (mean 63.9%, SD 5.7%) (p < 0.01) but did not differ significantly between black and white subjects. Ninety percent of subjects had an LVEF between 53% and 71%. LVEDV and LVESV were >25% greater in men than in women. From multivariate analysis, male gender, history of hypertension, and current smoking were each positively and independently associated with an approximately 1% lower LVEF. Body surface area, a family history of premature myocardial infarction, and treadmill workload 150 time were positively related, whereas total skinfold thickness was negatively related to LVEDV and LVESV. In addition, systolic blood pressure was positively related and diastolic blood pressure and total cholesterol concentration negatively related to LVEDV, whereas treadmill performance measured in metabolic equivalents (METS) and current smoking were positively related and age negatively related to LVESV. These results suggest that LVEF in healthy young adults is maintained over a limited range and that its variability not substantially associated with coronary risk factors. Determination of whether differences in LVEF, LVESV, and LVEDV in young adults influence future cardiovascular risk and whether obesity and other factors express any effect on risk through differences in LVESV and LVEDV will require longitudinal follow-up studies. (AM HEARTJ 1995;129:571-7.)

N a t h a n D. Wong, PhD, a Julius M. Gardin, MD, a T o m Kurosaki, MS, b H o d a Anton-Culver, PhD, b S t e p h e n Sidney, MD, c Jeffrey Roseman, MD, P h D , M P H , d and Samuel Gidding, MD e

Irvine and Oakland, Calif., Birmingham, Ala., and Chicago, Ill.

Large-scale population-based studies have begun to use echocardiography as a means of noninvasively evaluating cardiac structure and function. Most published reports have dealt with risk factors for and prognostic implications of left ventricular mass as evaluated by M - m o d e echocardiography. 1-3

C o m p r o m i s e d left ventricular function, as reflected by a low left ventricular ejection fraction (LVEF), has been shown to be an i m p o r t a n t factor associated with increased short- and long-term mortality in patients with coronary heart disease (CHD). 4-8 These studies evaluating the significance of L V E F have been tim-

Fromthe aPreventiveCardiologyProgram,Divisionof Cardiology,and the bDivisionof Epidemiology,Department of Medicine, Universityof California, Irvine; the ¢Divisionof Research, Kaiser Permanente, Oakland; the HCARDIACoordinating Center, Birmingham; and eChildren's Memorial Hospital, Chicago. Supported by National Institutes of Health contracts N01-HC-48047 through N01-HC-84050,N01-HC-95095,and N01-HC-95100.

Received for publication Dec. 28, 1993;acceptedJune 27, 1994. Reprint requests: Nathan D. Wong,PhD, PreventiveCardiologyProgram, C240MedicalSciencesI, Universityof California, Irvine, CA92717. Copyright©1995by Mosby-YearBook,Inc. 0002-8703/95/$3.00 + 0 4/1/61196 571

572

wong et al.

ited to the a s s e s s m e n t of p a t i e n t s with known C H D or have used invasive a s s e s s m e n t s such as left ventriculography. No p u b l i s h e d studies h a v e used noninvasive a s s e s s m e n t of L V E F as a m e a s u r e of systolic function in a p o p u l a t i o n - b a s e d cohort. T h e r e exist no n o r m a t i v e d a t a on L V E F nor on its c o m p o n e n t volu m e s at end-diastole and end-systole ( L V E D V a n d L V E S V , respectively) in a s y m p t o m a t i c h e a l t h y persons except for a small s t u d y in elderly subjects. 9 In addition, a l t h o u g h chronic alcoholism has b e e n linked to i m p a i r e d systolic function as m e a s u r e d b y M - m o d e - d e t e r m i n e d p e r c e n t a g e fractional shortening, 1° it is u n k n o w n w h e t h e r k n o w n C H D risk factors are associated with differences in L V E F t h a t m a y be of clinical significance in young, otherwise h e a l t h y adults. C o n s e q u e n t l y we sought to evaluate the distribution of resting L V E F , L V E D V , a n d L V E S V in a norm a l a s y m p t o m a t i c p o p u l a t i o n a n d to e x a m i n e the e x t e n t to which selected coronary risk factors or physical conditioning m a y correlate with these indicators of systolic function and left v e n t r i c u l a r c h a m ber size in this population. N o n i n v a s i v e m e a s u r e m e n t s of L V E F , L V E S V , a n d L V E D V were done b y t w o - d i m e n s i o n a l echocardiography. C o r o n a r y Artery Risk D e v e l o p m e n t in Young Adults (CARDIA) is a m u l t i c e n t e r p o p u l a t i o n - b a s e d s t u d y of y o u n g m e n a n d w o m e n 23 to 35 years old w h e n an echocardiographic e x a m i n a t i o n was p e r f o r m e d .

METHODS Study population. The CARDIA cohort comprised 5115 participants 18 to 30 years old at the time of the baseline examinations (1985 to 1986). The CARDIA cohort was recruited and examined at four field centers (Birmingham, Ala.; Chicago, Ill.; Minneapolis, Minn.; and Oakland, Calif.). An echocardiography reading center was located at the University of California, Irvine, and a coordinating center at the University of Alabama, Birmingham. The overall design and objectives of the CARDIA study have been presented in detail elsewhere. 11 Echocardiography was performed at the 5-year follow-up examination (1990 to 1991), attended by 4352 participants (85% of the original cohort), who were fasting at the time of examination. Two-dimensional echocardiography was performed in approximately the first half of the examined cohort (n -- 1893). (The priority in accomplishing other echocardiographic measures in a required time precluded performance of two-dimensional echocardiography in the remainder of the cohort.) Data from 12 subjects with physiologically implausible measured values of LVEF (<0.35 or >0.90) were excluded from analysis as were data from subjects with any self-reported heart disease (n = 101) other than mitral valve prolapse (n = 53). (Subjects without reported heart disease and those with reported mitral valve

March 1995 American Heart Journal

prolapse were found to have a similar mean LVEF, 63.3 % vs 62.5 %, respectively.) These exclusions left a total of 1782 men and women with measurements of LVEF. Age, gender, and risk factor levels did not differ significantly between those with and those without LVEF measurements, suggesting that the cohort examined was representative of the entire CARDIA population. Two-dimensional echocardiography. The methods of two-dimensional echocardiographic examination were similar to those used in the Cardiovascular Health Study, described previously. 12 In brief, measurements were made from views in which >--80% of the endocardium and epicardium was visualized sufficiently well to be planimetered. Images were selected at end-diastole and end~systole for computation of LVEDV and LVESV. The video frame corresponding to end-diastole was identified as the frame with the largest visible left ventricular area recorded in early diastole. End-systole was determined by locating the frame having the smallest visible left ventricular area. LVEDV and LVESV were calculated from the twodimensional images (parasternal short-axis and apical two-chamber views) by using the modified Simpson's methods as described previously. 12-14 LVEF (percentage) was calculated as ([LVEDV - LVESV]/[LVEDV] x 100). Quality control for echocardiographic LVEF measurements involved periodic assessment of inter- and intratechnician and inter- and intrareader measurement variabilities. The coefficient of variability was 6% in studies (n = 34) read twice by the same reader (intrareader), 9% in studies (n -- 46) read by two readers (interreader), and 10% in studies performed twice by the same technician (n = 23) or by two technicians (n -- 11) but read by the same reader. Cardiovascular risk factors and medical history, Questionnaires regarding history of heart disease, including myocardial infarction, rheumatic heart disease, angina pectoris, and mitral valve prolapse, were administered to each participant. Each subject also was questioned regarding use (amount and length of time) of cigarettes and alcohol. 15 Smoking status, for the purposes of this report, was defined as current use versus no use or former use. Measures of tricep and subscapular skinfold thicknesses; weight (kilograms) and height (meters); body mass index (weight/height2), body surface area, resting seated blood pressure (mean of two measurements each of diastolic and systolic blood pressure), and lipid concentrations (fasting total and high-density lipoprotein cholesterol, and triglycerides) were obtained as described previously. 15 These factors were measured at the time of the echocardiographic examination. Physical activity intensity was estimated at the current examination, in which echocardiography was performed, and at the CARDIA baseline examination approximately 5 years earlier. Activity intensity was estimated from the weighted sum of the number of months of participation in a particular physical activity, accounting for the number of hours per week greater than a defined level, as previously described. 16

Volume 129, Number 3 American Heart Journal

To measure physical fitness at the baseline CARDIA examination, participants underwent a graded exercise treadmill test designed to assess maximal symptom-limited performance. 17 Pulse rate, standing blood pressure, and a 12-lead electrocardiogram were obtained for each subject before exercise. The exercise test involved as many as nine 2-minute stages for a total of ___18 minutes of exercise of progressively increasing difficulty. The estimated rate of energy expenditure at each stage was defined in metabolic equivalents (METS), ranging from 4.1 at stage 1 to 19.0 at stage 9 as defined by the protocol. (Although oxygen consumption was not measured in the current study, 1 M E T is defined as an oxygen consumption of 3.5 ml/kg body weight/min). M a x i m u m heart rate was obtained and blood pressures were monitored during all stages of t h e t e s t . Workload 150, the duration of exercise to a heart rate of 150 beats/min, was calculated as a measure of conditioning that is i n d e p e n d e n t of motivation. 17 In our cohort 860 (93 %) men and 793 (92 % ) women were able to reach this level on the treadmill. Statistics. Analysis of variance was used to compare mean LVEF, LVEDV, and LVESV values by gender-race subgroups among the 1782 black and white men and women included in the analysis. Extreme values and percentiles were defined in each stratum. T h e relation of age, gender, physical activity intensity, physical conditioning as assessed by M E T S or workload 150, and cardiac risk factors to L V E F were assessed. Gender-specific Pearson correlation coefficients were calculated, or in the case of binary covariates, t tests were performed to compare mean LVEF, LVEDV, or LVESV between subjects with and those without the factor of interest. Analyses were performed separately for men and women. In addition risk factor levels in individuals at the extremes (_+ 2 SD from the mean) of the L V E F distribution were compared by S t u d e n t ' s t test or chi square analysis. Stepwise multiple regression analyses were performed for the entire cohort to identify variables independently correlated with LVEF. Candidate variables included age, gender, cigarette and alcohol use, total skinfold thickness, body mass index, body surface area, blood pressure, lipid concentrations at examination 3, and physical performance at examination 1 (5 years before examination 3) as evaluated by M E T S and workload 150. An a (two-tailed) of 0.25 from univariate analysis (correlation or t test analysis) was required for a covariate to be considered a candidate in the regression analysis. Final models reflected variables achieving an a (two-tailed) value of <0.05.

RESULTS The distribution of LVEF among gender-race s u b g r o u p s is p r e s e n t e d in T a b l e I. O v e r a l l 5th, 50th, a n d 9 5 t h p e r c e n t i l e l e v e l s w e r e 54 %, 63 %, a n d 72 %, r e s p e c t i v e l y . T h e r e w e r e n o m a r k e d d i f f e r e n c e s in mean LVEF between black and white subjects. Mean L V E F , h o w e v e r , w a s s i g n i f i c a n t l y g r e a t e r in w o m e n

573

w o n g et al.

Table I. Distribution of LVEF, LVEDV, and LVESV Men

LVEF ( % ) Mean SD Minimum (0 percentile) 5th percentile 25th percentile Median (50th percentile) 75th percentile 95th percentile Maximum (100th percentile) LVEDV (ml) Mean SD Minimum (0 percentile) 5th percentile 25th percentile Median (50th percentile) 75th percentile 95th percentile Maximum (100th percentile) LVESV (ml) Mean SD Minimum (0 percentile) 5th percentile 25th percentile Median (50th percentile) 75th percentile 95th percentile Maximum (100th percentile)

Women

Black (n = 448)

White (n = 470)

Black (n = 455)

White (n = 409)

62.4* 5.8 41

62.8t 5.6 44

64.05 6.1 45

63.9 5.2 48

52 59 63

54 59 63

54 60 64

56 61 64

66 72 87

66 71 87

68 73 90

67 71 90

136.3" 29.8 62.4

136.4t 28.2 64.3

107.25 24.9 50.8

108.5 22.9 60.7

89.3 117.1 133.1

93.0 117.6 134.8

72.8 89.6 104.0

73.9 92.9 106.1

153.0 189.2 240.8

153.8 185.9 241.8

121.9 149.9 253.0

120.8 146.4 249.0

50.8* 13.8 18.8

50.4~ 13.3 16.5

38.45 11.6 8

38.9 9.9 9.2

29.7 41.1 50.0

30.3 41.7 48.8

22.1 30.8 36.9

24.4 32.4 38.2

58.7 74.8 104.7

58.6 74.8 89.8

45.3 59.4 101.3

44.3 56.5 93.9

*p < 0.01 compared with white women. t P < 0.01 compared with white women. Sp < 0.01 compared with black men a n d white men.

( 6 3 . 9 % ) t h a n in m e n ( 6 2 . 6 % ) (p < 0.01). L V E D V a n d L V E S V w e r e > 2 5 % h i g h e r in m e n t h a n in w o m e n . S i g n i f i c a n t d i f f e r e n c e s (p < 0.01) in all t h r e e f a c t o r s w e r e f o u n d in c o m p a r i s o n s o f b l a c k m e n to white women, white men to white women, and black w o m e n to b l a c k m e n a n d w h i t e m e n . In univariate analyses correlating LVEF with risk

March 1995 574

Wong et al.

,

Table [h Univariate analysis of cardiovascular risk factors with LVEF, LVEDV, and LVESV Risk factor

LVEF

LVEDV

LVESV

Pearson correlation coefficients

Age (yr) Systolic blood pressure (mm Hg) Diastolic blood pressure (ram Hg) Total cholesterol concentration (mg/dl) High-density lipoprotein cholesterol concentration (mg/dl) Body mass index (kg/m2) Body surface area (m2) Total skinfold thickness (mm) Daily alcohol consumption (ml) Physical activity intensity Maximum heart rate (beats/min) Workload 150 (min) Estimated METS

0.03 -0.01 -0.07*

-0.03 0.19"

-0.04 0.15"

0.06*

0.09*

-0.02

-0.05?

-0.02

0.04

-0.17"

-0.15"

0.01

0.20*

0.15"

-0.06?

0.60"

0.51"

0.06?

-0.10"

-0.12"

-0.05

0.08*

0.09*

-0.03

0.20*

0.18"

-0.03

0.01

0.02

-0.07*

0.31"

0.29*

-0.06*

0.24*

0.22*

Mean + SD

Current smoking Yes (n = 521) 62.6* _+ 5.6 124.0_+ 30.8 46.2* + 14.2 No (n = 1256) 63.5 -+ 5.8 121.9_+ 30.0 44.2 + 13.4 History of diabetes Yes (n = 21) 63.5 _+ 5.1 109.1? + 27.6 39.7 _+ 12.5 No (n = 1754) 63.2 _+ 5.8 122.7__ 30.3 44.9 _+ 13.7 History of premature myocardial infarction Yes (n = 214) 62.8 _+5.7 125.2 _+31.8 46.2 + 13.5 No (n = 1559) 63.3 + 5.8 122.2+_ 30.0 44.6 + 13.7 *p < 0.01for correlationor for comparisonwith those withoutrisk factor. tP < 0.05for correlationor for comparisonwith those withoutrisk factor.

factors, P e a r s o n correlation coefficients (Table II) were low, with nonsignificant correlations except for a weak positive correlation (r = 0.06, p < 0.05) with total skinfold thickness, and negative correlations (all p < 0.05 or p < 0.01) with diastolic blood pressure, b o d y surface area, workload 150, and estimated M E T S . Multiple risk factors, however, were more strongly associated (negatively or positively) with L V E D V and LVESV. N o t e w o r t h y were particularly strong correlations of L V E D V and L V E S V with body

American Heart Journal

surface area (r = 0.60 and 0.51, respectively; both p < 0.01), workload 150 (r = 0.31 and 0.29, respectively; b o t h p < 0.01), and estimated M E T S (r = 0.24 and 0.22, respectively; both p < 0.01). Smokers had a significantly lower (p < 0.01) L V E F and a greater L V E S V (p < 0.01). L V E D V also was significantly lower in subjects with t h a n in those without a history of diabetes (p < 0.05). W h e n comparing patients whose d a t a were at the extremes of the L V E F distrib u t i o n (approximately _+2 SD from the mean, or >76 % vs < 5 1 % , respectively), those at the lower end t e n d e d to be younger by 2 years (p < 0.05), b u t no other differences in risk factors were observed (results not tabulated). Risk factor variables analyzed (and ranges or prevalences) included age at e n t r y (17 to 34 years), body mass index (14.8 to 55.7 kg/m2), total skinfold thickness (8 to 95 mm), body surface area (1.14 to 2.62 m2), total physical activity score (0 to 2052), average systolic blood pressure (77 to 151 m m Hg), average diastolic blood pressure (30 to 111 m m Hg), total cholesterol concentration (70 to 364 mg/dl), average daily alcohol c o n s u m p t i o n (0 to 499 ml), current cigarette use (29%), family history of p r e m a t u r e myocardial infarction (10%), history of h y p e r t e n s i o n (7 %), and history of diabetes (1% ). Multiple stepwise regression analyses (Table III) revealed only male gender, c u r r e n t smoking, a history of high blood pressure, and m a x i m u m heart rate to be i n d e p e n d e n t l y associated (p <0.05) with lower LVEF. T h e s e effects were relatively small, however, with male gender, a history of high blood pressure, and c u r r e n t smoking associated with an approxim a t e l y 1% lower LVEF. Other variables considered, including age, blood pressure, lipids, physical activity, and body mass index or surface area did not enter into the final model. Left ventricular end-systolic stress, which m a y affect L V E F , was considered in an alternate model. Its inclusion did not substantially alter the relation with smoking, b u t the effect of gender was diminished. B y multiple regression analysis, L V E D V was positively associated with systolic blood pressure, a p r e m a t u r e family history of myocardial infarction, workload 150, and, in particular, body surface area, whereas total skinfold thickness, diastolic blood pressure, and total cholesterol concentration appeared to be negatively associated. Analysis of L V E S V revealed current smoking, family history of p r e m a t u r e myocardial infarction, workload 150, energy expenditure on treadmill estimated b y M E T S , and b o d y surface area to be positively associated, whereas total skinfold thickness and age were negatively associated. In separate analyses (results not shown), age, gender, and body weight were forced into the models for L V E S V and L V E D V to

Volume 129, Number 3 American Heart Journal

account for potential confounding of the effect of total skinfold thickness by these factors. Although the increase in the variance explained by adding body surface area and total skinfold thickness was reduced, the direction and strength of body surface area, total skinfold thickness, and other covariates associated with LVESV and LVEDV were not appreciably altered. DISCUSSION

Our data demonstrate that among relatively healthy young adults without reported disease, LVEF as determined by two-dimensional echocardiography is maintained over a limited range, with 90% of subjects having an LVEF of 53 % to 71%. This finding corresponds to previous results reported among smaller groups of healthy persons studied by M-mode echocardiography, is No previous reports, however, have described normative echocardiographic values for resting LVEF, LVESV, and LVEDV in a large, population-based sample of healthy young adults. Multivariate analyses revealed male gender, a history of high blood pressure, and current smoking to be independently associated with an approximately 1% lower LVEF. However, a weak, negative relation of maximum heart rate (on the treadmill test) with LVEF was observed. It has been shown that smoking in the CARDIA population 19 and long-term alcohol use in elderly patients 9 are negatively correlated with global left ventricular systolic function estimated by M-mode LVEF shortening. Possible mechanisms explaining the associations between substance abuse and compromised cardiac function need further evaluation. The relatively low proportion of the variance (r 2 = 0.02)in LVEF explained by the risk factors and other covariates investigated suggests that cardiovascular risk factors and physical activity or performance indicators are associated little with LVEF in healthy young adults at rest, in whom the heart works to maintain normal systolic function during various hemodynamic stresses. An adaptation effect of the heart to maintain normal LVEF over a relatively narrow range is a plausible explanation. In addition, LVEF may be a poor indication of early dysfunction or coronary risk, as suggested by its lack of association with the factors evaluated in this study. For example, subjects with higher blood pressures (but within the normal range) had an LVEF similar to that of subjects with lower blood pressures. These results differ from those of an intervention study in which blood pressure experimentally manipulated over a wider range does yield changes in systolic function. 2° In earlier reports, is, 21 LVEF was found to

T3FT

575

• o n g e t al.

Table I]1. Multiple stepwise regression: Final models pre-

dicting LVEF, LVEDV, LVESV

LVEF Intercept Female gender Current smoking High blood pressure history Maximum heart rate LVESV Intercept Total skinfold thickness (ram) Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) Total cholesterol concentration (mg/dl) Workload 150 (min) Body surface area (m 2) Family history of premature myocardial infarction LVESV Intercept Age (yr) Total skinfold thickness (mm) Current smoking Family history of premature myocardial infarction Workload 150 (min) Body surface area (m 2) Estimated METS

Coefficient

F

0.69 0.01 -0.01 -0.01 -0.0003

13.2" 12.3" 3.95 4.95

-28.1 -0.41 0.21

82.9* 7.35

-0.38

19.2"

-0.06

12.3t

0.02 93.13 3.80

12.4" 840.7* 4.6 $

-22.1 -0.16 -0.18 2,56 2.05

4.05 63.0* 15.2t 5.45

0.01 34.7 0.45

15.0" 526.3* 11.4t

*p < 0.001. ?p < 0.01. Sp < 0.05.

be relatively constant as age (in adults) or body surface area (regardless of age) increased, consistent with our findings in young adults. A recent study found no association between obesity and LVEF, despite significantly higher LVEDV and stroke volume in the obese group. 22 Thus an LVEF outside the normal range can be viewed as a phenomenon that represents end organ damage such as an ischemic event, although large changes in blood pressure also may affect systolic function. 2° The limited contribution of risk factors and lack of anthropometric measures for prediction of variation in LVEF is contrasted with an observed strong association of these factors to LVEDV and LVESV. Factors emerging in multivariate analysis explained >45% of the variation in LVEDV and 33% in LVESV. Body surface area was by far the strongest (positive) indicator of these volumes, followed by total skinfold thickness, which had negative associations with LVEDV and LVESV. However, there was only a weak univariate association between total

576

March 1995 American Heart Journal

Wong et al.

skinfold thickness and LVEF. A modest inverse correlation of age (despite its limited range in the CARDIA cohort) with LVESV is corroborated by others who have measured left ventricular dimension by M-mode echocardiography, ls, 21 These observations suggest that the strong direct known association of body surface area or size with LVESV and LVEDV may be counteracted in part by the relative amount of body fat as indicated by total skinfold thickness. Absolute blood volumes and therefore myocardial and peripheral perfusion therefore may be reduced relative to the degree of obesity. These findings, which persist after adjustment for body weight, may reflect body fat's lesser requirement for blood volume and hence the lower LVESV and LVEDV relative to body size. Our modest positive association of workload 150 with LVEDV and METS with LVESV suggests a possible role of physical conditioning in increasing cardiac volumes. This, however, needs further investigation through projective study. Although other techniques for measuring LVEF, such as left ventriculography, 48 may show greater accuracy in measurement, two-dimensional echocardiography provides a good method for close approximation of LVEF, LVESV, and LVEDV in healthy young adults (in whom invasive evaluation of LVEF would not have been possible). However, especially for LVEF, this technique involves inherent measurement variability that incorporates measurement error for LVEDV and LVESV. This variability may have obscured modest relations, in particular for LVEF, with coronary risk factors. Clinically significant differences, however, should have been observable given the large sample sizes studied and the adequate ranges in risk factor levels and left ventricular function observed. In addition, the cross-sectional nature of the current study did not allow testing for longitudinal effects of risk factors on subsequent changes in left ventricular function or LVESV or LVEDV. Reevaluation of risk factors and echocardiographic parameters during longitudinal follow-up of the CARDIA cohort will allow these effects to be tested. This report provides information on normative values and the distribution of LVEF, LVEDV, and LVESV not previously available in a large sample of normal healthy adults, The results suggest that in these persons LVEF has minimal association with cardiovascular risk factors. Whether current tobacco use, high blood pressure, or male gender may confer a negative effect on cardiac function or whether there may be any long-term consequences of obesity or other observed risk factors correlated with lower

LVESV and LVEDV requires further longitudinal assessment of this cohort. Moreover, it will be of interest to examine whether findings in the lower range of normal for LVEF in young adults or whether decreases in LVEF during early and middle adulthood may be associated with increased morbidity and "mortality from cardiac events later in life. W e t h a n k Diane Bild, MD, for her helpful c o m m e n t s in reviewing t h e m a n u s c r i p t a n d J u d i t h T o m l i n s o n for her assistance w i t h m a n u s c r i p t preparation.

REFERENCES

1. Levy D, Garrison RJ, Savage DD, Kannel WB, Castelli WP. Left ventricular mass and incidence of coronary heart disease in an elderly cohort: the Framingham Heart Study. Ann Intern Med 1989;110:101-7. 2. 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. 3. Burke GL, Arcilla RA, Culpepper WS, Webber LS, Chiang YK, Berenson GS. Blood pressure and echocardiographic measures in children: the Bogalusa Heart Study. Circulation 1987;75:106-14. 4. Ong L, Green S, Reiser P, Morrison J. Early prediction of mortality in patients with acute myocardial infarction: a prospective study of clinical and radionuclide risk factors. Am J Cardiol 1986;57:33-8. 5. Kelly MJ, Thompson PL, Quinlan MF. Prognostic significance of left ventricular ejection fraction after acute myocardial infarction: a bedside radionuclide study. Br Heart J 1985;53:16-24. 6. The Multicenter Postinfarction Research Group. Risk stratification and survival after myocardial infarction. N Engl J Med 1983;309:331-6. 7. Sanz G, Castafier A, Betriu A, Magrifia J, Roig E, Coll J, Par~ JC, Navarro-L6pez F. Determinants of prognosis in survivors of myocardial infarction: a prospective clinical angiographic study. N Engl J Med 1982;306:1065-70. 8. Taylor GJ, Humphries JO, Mellits ED, Pitt B, Schulze R, Griffith L, Achuff SC. Predictors of clinical course, coronary anatomy, and left ventricular function after recovery from acute myocardial infarCtion. Circulation 1980;62:960-70. 9. Aronow WS, Stein PD, Sabbah HN, Koenigsberg M. Resting left ventricular ejection fraction in elderly patients without evidence of heart disease. Am J Cardiol 1988;63:368-9. 10. MathewsEL, GardinJM, HenryWL, DelNegroAA, Fletcher RO, Snow JA, Epstein SE. Echocardiographic abnormalities in chronic alcoholics with and without overt congestive heart failure. Am J Cardiol 1981; 47:570-8. 11. Friedman GD, Cutter GR, Donahue RP, Hughes GH, Hulley SB, Jacobs DR Jr, Liu K, Savage PJ. CARDIA: study design, recruitment, and some characteristics of the examined subjects. J Clin Epidemio11988;41:110516. 12. Gardin JM, Wang ND, Bommer W, Klopfenstein HS, Smith VE, Tabatznik B, Siscovick D, Lobokzinski S, Anton-Culver H, Manolio TA. Echocardiographic design of a multicenter investigation of free-living elderly subjects: the Cardiovascular Health Study. J Am Soc Echocardiogr 1992;5:63-72. 13. Schiller NB, Skjoldebrand CG, Schiller EJ, Mavroudis CC, Silverman NH, Rahimtoola SH, Lipton MJ. Canine left ventricular mass estimation by two-dimensional echocardiography. Circulation 1983;68:210-6. 14. Helak JW, Reichek N. Quantitation of human left ventricular mass and volume by two-dimensional echocardiography: in vitro anatomic validation. Circulation 1981;63:1398-407. 15. Cutter GR, Burke GL, Over AR, Friedman GD, Hilner JE, Hughes GH, Hulley SB, Jacobs DR Jr, Liu K, Manolio TA. Cardiovascular risk factors in young adults: the CARDIA baseline monograph. Control Ctin Trials 1991;12(suppl 1):1S-77S. 16. Jacobs DR, Hahn LP, Haskell WL. Validity and reliability of short physical activity history: CARDIA and the Minnesota Heart Health Program. J Cardiopul Rehab 1989;9:448-59.

Volume 129, Number 3 American Heart Journal

17. Sidney S, Haskell WL, Crow R, Sternfeld B, Oberman A, Armstrong MA, Cutter GR, Jacobs DR, Savage PJ, VAn Horn L. Symptom-limited graded treadmill exercise testing in young adults in the CARDIA study. Med Sci Sports Exerc 1992;24:177-83. 18. Henry WL, Ware J, Gardin JM, Hepner SI, McKay J, Weiner M. Echocardiographicmeasures in normal subjects: growth-related changes that occur between infancy and early adulthood. Circulation 1978; 57:278-85. 19. Gidding S, Xie X, Manolio T, Flack J, Perkins L, Gardin J. Relation between smoking and resting cardiac function: the CARDIA study [Abstract]. Presented at the International Conference on the Prevention of Atherosclerosis and Hypertension Beginning in Youth; May 1992; Orlando, Fla.

Woi'~g et al.

577

20. Borow KM, Green LH, Grossman W, Braunwald E. Left ventricular end-systolic stress-shortening and stress-length relations in humans: normal values and sensitivity to inotropic state. Am J Cardiol 1982; 50:1301-8. 21. Gardin JM, Henry WL, Savage DD, Ware JH, Burn C, Borer JS. Ecbocardiographic measurements in normal subjects: evaluation of an adult population without clinically apparent heart disease. J Clin Ultrasound i979;7:439-47. 22. Stoddard MF, Tseda K, Thomas M, Dillon S, Kupersmith J. The influence of obesity on left ventricular filling and systolic function. AM HEART J 1992;124:694-9.