Background Little is known about the determinants of atrial size, and no study has analyzed whether genetic factors are involved in the pathogenesis of LA enlargement.Materials and methods We studied the heritability of echocardiographic left atrial size in 290 parents from the Tecumseh Blood Pressure Study and 25 1 children from the Tecumseh Offspring Study. All data from the parents and children were obtained at the same field office in Tecumseh, USA. Left atrial dimension was determined echocardiographically in accordance with American Society of Echocardiography guidelines with the use of leading-edge-to-leading-edge measurements of the maximal distance between the posterior aortic root wall and the posterior left atrial wall at end systole.Results For correlation between the left atrial dimensions of the parents and their offspring, several models were generated to adjust the atrial dimensions in both groups for an increasing number of clinical variables. After removing the effect of age, gender, height, weight, skinfold thickness, and systolic blood pressure, parent-child correlation for left atrial size was 0.19 (P=0.007). Further adjustment for left ventricular mass and for measuring left ventricular diastolic function increased the correlation to 0.25 (P=0.001).Conclusions The present data indicate that heredity can explain a small but definite proportion of the variance in left atrial dimension.

Objective To estimate the contribution of heredity to the variance in left ventricular mass (LVM), and to ascertain whether genetic factors may interact with non-genetic factors in promoting LVM growth. Subjects and setting The study population consisted of 290 healthy parents and 251 healthy children living in Tecumseh, Michigan, USA. Main outcome measure Correlation of parents’ LVM with offspring's LVM adjusting for a number of clinical variables. Methods LVM in parents and offspring was measured with M-mode echocardiography by the same investigators. Results Parents unadjusted LVM was unrelated to offspring unadjusted LVM , but after removing the confounding effect of age, sex, anthropometric measurements, systolic blood pressure, plasma insulin and urinary sodium excretion, parent–child correlation for LVM was 0.28 (P = 0.006). The relative contribution of parental-adjusted LVM and of several offspring phenotypic and environmental variables on offspring LVM was evaluated by multivariable regression analysis. When age, gender, anthropometric measurements and systolic blood pressure were accounted for, adjusted LVM of parents explained only 1.6% of the total variance in offspring LVM. However, after inclusion of insulin and urinary sodium in the model heredity explained 7.6% of the total variance in offspring LVM, and its predictive power was second only to that of child's height. Furthermore, an interactive effect of parental LVM with offspring systolic blood pressure was found on child's left ventricular mass. Conclusion Heredity can explain a small, but definite proportion of the variance in LVM. Higher blood pressure favors the phenotypic expression of the genes that regulate LVM growth.

Objective In this study, we tested the hypothesis that sympathetic over-activity may cause metabolic abnormalities and affect left ventricular (LV) structure and mass early in life. Subjects and setting The study population consisted of 111 healthy adolescents and young adults living in Tecumseh, Michigan (USA). Main outcome measures Correlations of LV mass and structure with several clinical variables in relation to the activity of the sympathetic nervous system. Methods Power spectrum density estimates of heart rate variability were calculated with an auto-regressive method, and subjects were divided by cluster analysis into two groups according to low-frequency and high-frequency components. LV data were obtained by echocardiographic assessment. Results Subjects with signs of sympathetic over-activity (n = 38, group 1) had higher heart rate, blood pressure (BP), waist/hip ratio and cholesterol levels than the rest of the group (n = 73, group 2). In group 1 subjects, insulin emerged as the strongest univariate correlate of interventricular septum and posterior wall thicknesses (P <0.001 for both) and of LV mass (P = 0.009). These relationships remained significant when body mass index was accounted for. By contrast, the marginal univariate relationship with diastolic BP did not remain significant in multivariate analysis. In group 2 subjects, BP was strongly correlated with LV wall thickness and mass both in univariate (P values from 0.03 to <0.001) and multivariate analyses, while insulin was not. The interactive effect of sympathetic activity and insulin on echocardiographic data was confirmed by multivariate analyses performed in the subjects grouped together (P values from 0.02 to 0.001 for the sympathetic activity × insulin interaction term). Conclusions In young subjects with heightened sympathetic activity and initial metabolic abnormalities, insulin is a strong determinant of LV wall thickness and geometry, while in subjects with normal autonomic nervous system activity, the main determinant of left ventricular size is the haemodynamic load.

Background Tobacco smoking is associated with a higher prevalence of atherosclerosis and respiratory disease. Objective To investigate differences between hemodynamic and biochemical findings in smokers and nonsmokers in the two sexes separately in the Tecumseh population. Methods We studied 851 subjects. They were divided according to smoking habits into group 1, nonsmokers (258 men and 234 women); and group 2, smokers (185 men and 174 women). Results Unpaired Student's t-tests and nonparametric tests were performed to determine the between-group P-values. Only hematocrit differed significantly between smokers and nonsmokers in both sexes (43.9 ± 0.2 and 44.6 ± 9.3%, P < 0.05 in men; 39.2 ± 0.3 and 40.3 ± 0.3%, P = 0.007 in women, respectively in nonsmokers and smokers). Triglycerides (80.6 ± 3.8 and 99.6 ± 4.3 mg/dl, P < 0.001), left ventricular mass index (95.4 ± 1.9 and 100.0 ± 1.2 g/m2, P = 0.008), and posterior wall thickness (9.5 ± 0.1 and 9.71 ± 0.01 mm, P = 0.044) were elevated and high-density lipoproteins were decreased (48.7 ± 0.8 and 44.5 ± 0.9 mg/dl, P < 0.01) only in women smokers. After adjustment for home systolic blood pressure and body mass index the differences in women remained significant except for posterior wall thickness. Conclusion Tobacco smoking is deleterious to both sexes but it appears to be particularly harmful to women. Our data can, in part, explain why the relative risk of myocardial infarction is higher in women than it is in men.

BACKGROUND:Because of population stability, Tecumseh, a rural community in Michigan, was the site of health surveys over a period of three decades.METHODS:In the recent survey anthropometrics, blood pressure (BP), blood chemistry, echo/Doppler cardiac exam, personality and exercise questionnaires were collected on site.RESULTS:In this rural community 70% of subjects (447 men, 410 women, average age 30 years) do not engage in recreational exercise. Sedentary subjects were heavier (4 kg), had higher BP (2.3/2 mm Hg), faster heart rate (4 beats/min), and lower stroke volume (2 ml/m2) than physically more active subjects (p = 0.02 to 0.00001). Measures of cardiac structure and function were more favorable in exercising subjects. Cholesterol (+8 mg/dl), triglycerides (+15 mg/dl) and insulin (+2.4 microU/ml) were higher and HDL cholesterol was lower (-2 mg/dl) in the sedentary group (p = 0.04 to 0.003). Being sedentary was associated with more anxiety, anger and feeling time pressure (p = 0.001 to 0.00001). Exercise demands at work had no effect while even once a week recreational exercise was associated with a more favorable cardiovascular risk status. A difference in cardiovascular status between the sedentary and exercising subjects was not apparent throughout childhood, adolescence or early adult life, suggesting that subjects who exercise presently were not a priori healthier than presently sedentary subjects.CONCLUSIONS:Seventy percent of the residents studied in Tecumseh are physically inactive and have a less favorable cardiac risk profile. Enhancement of exercise habits may beneficially affect cardiovascular status and, presumably, the prognosis.

BACKGROUND Left ventricular hypertrophy (LVH) defined by either ECG or echocardiographic criteria is a risk factor for cardiovascular morbidity and mortality. A number of determinants of LVH have been described in previous studies, principally male sex, hypertension, obesity, and aortic valvular stenosis. We examined the distribution of LV mass (LVM) in a population of 18- to 42-year-old normotensive men and women who were free of valvular heart disease to establish sex-specific normal values for LVM index (LVMI) and to determine the correlates of LVMI. METHODS AND RESULTS LVM was derived from measurements obtained by M-mode echocardiography. Average LVMI is significantly greater in men (102.9 +/- 0.7 g/m2) than women (88.2 +/- 0.7 g/m2). By defining LVH as an LVMI greater than the 90th percentile, we developed sex-specific criteria for LVH: men, > 125.4 g/m2; women, > 110 g/m2. We found that LVH in men is associated with indices of enhanced sympathetic nervous system reactivity and with elevated fasting insulin and triglyceride levels, which may be caused by insulin resistance. In women, LVH was associated with higher body weight and obesity. CONCLUSIONS Before the onset of hypertension, increased LVMI appears to have different determinants in men and women. We suggest that early LVH in young men is a manifestation of hyperkinetic borderline hypertension, a state previously shown to be associated with increased sympathetic nervous system activity and insulin resistance. The hyperkinetic state is less prevalent in young women, in whom increased adiposity seems to be the predominant factor associated with LVH.

1. Left ventricular (LV) hypertrophy has been implicated in the reduction of baroreflex sensitivity present in hypertension. The aim of the current study was to investigate the mean arterial pressure-heart rate reflex (MAP-HR) in a model which induced left ventricular hypertrophy but no sustained blood pressure elevation. 2. Five mongrel dogs were exposed to transient blood pressure elevation of between 20 and 30 mmHg, through hindlimb compression using a pneumatic pressure suit, for 7 h per day, 6 days per week for 6 weeks. Resting blood pressure was not altered by the 6 week hindlimb compression intervention. 3. Echocardiographically determined LV mass (mean +/- s.e.m.) was 116.0 +/- 7.4 g prior to hindlimb compression (baseline) and elevated to 125.4 +/- 8.1 g (P = 0.003) after 6 weeks of compression. A reduction in the early (E) to late (A) transmitral diastolic flow ratio (E/A) from 1.80 +/- 0.06 at baseline to 1.54 +/- 0.09 (P = 0.037) after the 6 week intervention suggested that cardiac compliance was reduced. 4. The maximum gain of the MAP-HR reflex, studied using the 'steady-state' drug technique, when blood pressure was normal, showed a trend for reduction from 3.85 +/- 0.43 beats/min per mmHg at baseline to 3.10 +/- 0.45 beats/min per mmHg (P = 0.067) after 6 weeks of compression. This gain reduction became significant after beta-adrenoceptor blockade with propranolol (3.13 +/- 0.55 vs 2.32 +/- 0.25 beats/min per mmHg; P = 0.039). Covariant analysis showed a significant inverse correlation between LV mass and maximum gain (r = 0.96; P < 0.001) during the 6 week compression period.(ABSTRACT TRUNCATED AT 250 WORDS)

An estimate of arterial compliance--the stroke volume/pulse pressure ratio (SV/PP)--was studied in 801 participants in the Tecumseh Blood Pressure Study, a population-based ongoing cardiovascular, epidemiological investigation. The subjects were normal young adults (mean age 30 +/- 5.6 years) and 373 were females. Cardiac anatomy and function were studied by echo-Doppler methods, blood pressure being measured at the same time by the indirect method. The distribution of the SV/PP ratio was skewed toward higher values in larger subjects. After statistical adjustments for body surface area, this estimate of arterial compliance was found to be higher in females (2.00 +/- 0.62) than in males (1.90 +/- 0.58) (p < 0.05). When subjects were divided into tertiles, the group with the lowest estimated arterial compliance was normotensive but had higher systolic pressure, lower diastolic pressure, a similar mean arterial pressure, higher heart rate and higher left ventricular wall thickness ratio compared to subjects with higher arterial compliance. Indices of systolic ejection decreased and diastolic function was altered in the low compliance subjects. In addition, low compliance subjects also had higher fasting insulin levels. These findings suggest that the low arterial compliance in an otherwise normal population has negative cardiovascular correlates. The early association of decreased arterial compliance with anatomic, functional and biochemical aberrations suggests that estimates of arterial compliance might prove useful for prediction of cardiovascular complications.

Human essential hypertension has more than one cause, but to dissect out subtypes, markers are required. The maximal activity of red blood cell lithium-sodium countertransport has been shown to be increased in hypertensive patients in case-control and population-based studies; in the latter, its distribution is a mixture of two overlapping but distinguishable subpopulations. In the present study, we classified 705 participants in the Tecumseh Blood Pressure Study as having either normal (mean, 0.234 mmol/l cells/hr; n = 614) or high (mean, 0.463 mmol/l cells/hr; n = 91) red blood cell lithium-sodium countertransport to determine if the red blood cell marker is associated with distinctive physiological characteristics. We found that subjects with elevated lithium-sodium countertransport have higher average blood pressure and a greater prevalence of hypertension than those with normal countertransport and that elevated blood pressure had been present since youth. Hemodynamically, the high countertransport group is characterized by elevated vascular resistance, whereas sympathetic nervous system activity appears to be slightly depressed. Subjects with increased lithium-sodium countertransport, compared with those with normal countertransport, have significantly lower average left ventricular mass index and only very infrequently demonstrate left ventricular hypertrophy. Our results support the usefulness of measurements of the maximal activity of red blood cell lithium-sodium countertransport as a way of distinguishing subgroups in the population. Our data are consistent with the idea that subjects with an elevated maximal activity for red blood cell lithium-sodium countertransport are a subset of the population with a genetic lesion that predisposes them to the development of essential hypertension.

The relation between blood pressure level and reactivity to mental arithmetic and isometric exercise was investigated in 169 men and 120 women (average age, 32.3 years) from the village of Tecumseh, Mich. In the entire population, the correlation between baseline blood pressure and blood pressure response to both stressors was not significant. Blood pressure reactivity to both stressors was not increased in participants with borderline hypertension (one clinic reading of more than 140 mm Hg systolic and/or 90 mm Hg diastolic). When subjects were classified according to blood pressure response (below and above the 80th percentile), the hyperreactors to mental and physical stress had normal baseline blood pressure values. The hyperreactors also had clinic-to-home blood pressure differences similar to those of the rest of the population. Participants who had borderline hypertension at age 32 years had significantly elevated blood pressures at ages 5, 8, 12, 21, and 22 years. Those who were hyperreactors at age 32 years had normal blood pressures as children and young adults. Results of the present study lend no support to an association between higher blood pressures and blood pressure hyperreactivity. Study participants in Tecumseh will be recalled for future examinations. The independence of blood pressure levels from blood pressure reactivity offers a unique opportunity to prospectively evaluate their separate effects on cardiovascular morbidity.

During a survey of young subjects not receiving treatment for hypertension in Tecumseh, Michigan, clinic and self-monitored blood pressures taken at home (14 readings in 7 days) were obtained in 737 subjects (387 men, 350 women, average age 31.5 years). Hypertension in the clinic was diagnosed if the clinic blood pressure exceeded 140 mm Hg systolic or 90 mm Hg diastolic. In the absence of firm criteria for what constitutes hypertension at home, subjects whose average home blood pressure was in the upper decile of the whole population were considered to have hypertension at home. By these criteria, 7.1% of the whole population had "white coat" hypertension (i.e., high clinic but not elevated home readings). The prevalence of "sustained" hypertension (i.e., high readings in the clinic and at home) was 5.1%. Subjects with white coat and sustained borderline hypertension in Tecumseh were very similar. Both groups showed, at previous examinations (at ages 5, 8, 21, and 23 years), significantly higher blood pressure readings than the normotensive subjects. As young adults (average age 33.3 years), the parents of both hypertensive groups had significantly higher blood pressure readings than the parents of normotensive subjects. Both hypertensive groups had faster heart rates, higher systemic vascular resistance, and higher minimal forearm vascular resistance. Both hypertensive groups were more overweight, had higher plasma triglycerides, insulin, and insulin/glucose ratios than normotensive subjects. The white coat hypertensive group also had lower values of high density lipoprotein than the normotensive group. White coat hypertension is a frequent condition.(ABSTRACT TRUNCATED AT 250 WORDS)

The Tecumseh project investigates the evolution of hypertension in a healthy population. Of 946 subjects aged 18 through 38 years, 124 had clinic blood pressure readings higher than 140/90 mm Hg (the mean for borderline hypertensive subjects was 130/94 mm Hg). Compared with normotensive subjects, borderline hypertensive subjects had higher home blood pressures (mean, 12/7 mm Hg higher). Their childhood and postpubertal blood pressures were elevated (6/4 mm Hg higher than normal at age 6 years and 12/7 mm Hg higher than normal at age 21 years), and hypertensive target organ changes were detected. Borderline hypertensive subjects also had elevated minimal forearm resistance (0.22 U higher than normal), decreased stroke index (1.8 mL/m2 lower than normal), and impaired ventricular diastolic relaxation (mitral Doppler peak early diastolic blood flow [E] to peak late diastolic blood flow [A] ratio 0.13 lower than normal). Borderline hypertensive subjects had significant abnormalities in other coronary risk factors (cholesterol levels were 0.39 mmol/L higher, triglyceride levels were 0.45 mmol/L higher, high-density lipoprotein levels were 0.08 mmol/L lower, insulin levels were 38 pmol/L higher, and 16.5% more of them were overweight). Borderline hypertension is neither transient nor innocuous. Its association with other predictors of atherosclerosis calls for clinical attention.