Subclinical, hemodynamic, and echocardiographic abnormalities of high pulse pressure in hypertensive and non-hypertensive adults.

BACKGROUND: High pulse pressure (PP) is associated with cardiovascular events, but subclinical abnormalities in cardiac structure and function in relation to high pulse pressure are not well described. METHODS AND RESULTS: 2225 hypertensive and 1380 non-hypertensive participants with adequate echocardiographic left ventricular measurements were evaluated. Non-hypertensives in the highest PP tertile (compared to the lower tertiles) were older (44 years vs. 40 years, p<0.009), had higher systolic pressure [(SBP) 136 mmHg vs. 108 mmHg] and lower diastolic pressure [(DBP) 54 vs. 71 mmHg (p=.0001)], greater BMI (27 vs. 25 kg/m2, p<.001) and more diabetes (4% vs. 2.25%, p<.001). In the hypertensive group, subjects in the highest PP tertile were older (52 vs 42 years), had higher SBP (157 vs. 116 mmHg) but lower DBP (65 vs. 83 mmHg). In the non-hypertensive group, higher PP (>60 mmHG) was associated with a higher frequency of echocardiographic structural and functional abnormalities, specifically, greater posterior and relative wall thickness, longer isovolumic relaxation time, and concentric left ventricular (LV) hypertrophy. CONCLUSION: In a population-based sample of hypertensive and non-hypertensive participants, higher PP was associated with subclinical abnormalities of cardiac structure and function, which exist even in the absence of hypertension and/or the use of antihypertensive treatment.

Adducin 1 (alpha) Gly460Trp variant is associated with left ventricular geometry in Caucasians and African Americans: The HyperGEN Study.

Normal left ventricular (LV) mass and geometry is required for optimal LV functioning. Abnormalities in either result in increased morbidity and mortality. The adducing 1 (alpha) gene (ADD1) Gly460Trp polymorphism has been associated with high blood pressure and increased plasma volume, both predictors of LV mass and function. In this cross-sectional study, we evaluate the association between this polymorphism and LV mass and geometry. LV mass, relative wall thickness (RWT), and systolic and diastolic parameters were measured using echocardiography in 3483 African American and Caucasian subjects from the Hypertension Genetic Epidemiology Network (HyperGEN). Analysis of covariance was used to estimate the polymorphism's association with echocardiograph parameters, stratified by race. The model was adjusted for age, diastolic and systolic blood pressure, glomerular filtration rate, smoking, low and high density lipoprotein cholesterol, urinary sodium, and body mass index. In Caucasians, the Trp allele was associated with higher ejection fraction (EF) (P = .02), fractional shortening (FS) (P = .02), and RWT (P = .03). In African Americans, the Trp allele was negatively associated with RWT (P = .02), but no association was found with EF (P= .08) or FS (P= .09). The polymorphism was not associated with diastolic function parameters in either racial group. We found no association of ADD1 Gly460Trp with LV mass in Caucasians or African Americans; however, it was associated with unfavorable LV geometry (higher RWT) in Caucasians and favorable LV geometry (lower RWT) in African Americans after controlling for factors that would affect plasma volume.

Longitudinal trends in the association of metabolic syndrome with 550 k single-nucleotide polymorphisms in the Framingham Heart Study.

We investigated the association of metabolic syndrome (MetS) with a 500 k and a 50 k single-nucleotide polymorphism (SNP) gene chip in the Framingham Heart Study. We cross-sectionally evaluated the MetS longitudinal trends. Data analyzed were from the Offspring Cohort (four exams: first (n = 2,441), third (n = 2,185), fifth (n = 2,308), and seventh (n = 2,328)) and the Generation 3 Cohort (one exam: the first exam (n = 3,997)). The prevalence of MetS was determined using the National Cholesterol Education Program Adult Treatment Panel III diagnostic criteria, modified with a newly developed correction for medication use. The association test between an SNP and MetS was performed with a generalized estimating equations method under the additive genetic model. Multiple-testing corrections were also performed. The prevalence of MetS in the offspring cohort increased from one visit to the next, and reached the highest point by the seventh exam comparable with the prevalence reported for the general US population. The pattern of the MetS prevalence over time also reflected itself in the association tests, in which the highest significances were seen in the fifth and seventh exams. The association tests showed that SNPs within genes PRDM16, CETP, PTHB1, PAPPA, and FBN3, and also some SNPs not in genes were significant or close to significance at the genome-wide thresholds. These findings are important in terms of eventually identifying with the causal loci for MetS.

Comparison between single-marker analysis using Merlin and multi-marker analysis using LASSO for Framingham simulated data.

We compared family-based single-marker association analysis using Merlin and multi-marker analysis using LASSO (least absolute shrinkage and selection operator) for the low-density lipoprotein phenotype at the first visit for all 200 replicates of the Genetic Analysis Workshop 16 Framingham simulated data sets. Using "answers," we selected single-nucleotide polymorphisms (SNPs) on chromosome 22 for comparison of results between single-marker and multi-marker analyses. For the major causal SNP rs2294207 on chromosome 22, both single-marker and multi-marker analyses provided similar results, indicating the importance of this SNP. For the 12 polygenic SNPs on the same chromosome, both single-marker and multi-marker analyses failed to provide statistically significant associations, indicating that their effects were too weak to be detected by either method. The main difference between the two methods was that for the 14 SNPs near the causal SNPs, p-values from Merlin were the next smallest, whereas LASSO often excluded these non-causal neighboring SNPs entirely from the first 10,000 models.

Genome-wide association analysis of Framingham Heart Study data for the Genetics Analysis Workshop 16: effects due to medication use.

Problems associated with medication use and the consequent effects on genome-wide association analyses were explored using the Genetic Analysis Workshop 16 Problem 3 data. Lipid phenotypes were simulated in the Framingham Heart Study using several measured variables including causal genes (based on a 500 k SNP panel), smoking, dietary intake, and medication usage. We report a sensitivity analysis of how medication use (which artificially alters lipid values) affects association inferences. Associations were performed for LDL-c under seven different correction schemes: 1) ignore medication use entirely (no correction) and adjust for covariates; 2) delete medicated subjects then adjust for covariates; 3) include medication use (Yes/No) as a covariate during covariate adjustments; 4) correct raw values using clinical trials information then adjust for covariates; 5) correct raw values using the actual simulation protocol ("truth") then adjust for covariates; and 6-7) over-corrections (add arbitrary values) then adjust for covariates. Results indicate that failure to properly correct for medication usage can profoundly affect the heritability, and hence the association results. The empirical results yielded one genome-wide significant locus on chromosome 22 (RS2294207), consistent with the simulation protocol. This signal was detected under all schemes that corrected the raw values (clinical trials, simulation protocol, or over corrections), but was not detected under the first three adjustment schemes (ignore medication use, delete medicated individuals, use medication status as covariate). In summary, we confirm that failure to properly account for medication usage can have a profound impact on genetic associations.

Combining least absolute shrinkage and selection operator (LASSO) and principal-components analysis for detection of gene-gene interactions in genome-wide association studies.

Variable selection in genome-wide association studies can be a daunting task and statistically challenging because there are more variables than subjects. We propose an approach that uses principal-component analysis (PCA) and least absolute shrinkage and selection operator (LASSO) to identify gene-gene interaction in genome-wide association studies. A PCA was used to first reduce the dimension of the single-nucleotide polymorphisms (SNPs) within each gene. The interaction of the gene PCA scores were placed into LASSO to determine whether any gene-gene signals exist. We have extended the PCA-LASSO approach using the bootstrap to estimate the standard errors and confidence intervals of the LASSO coefficient estimates. This method was compared to placing the raw SNP values into the LASSO and the logistic model with individual gene-gene interaction. We demonstrated these methods with the Genetic Analysis Workshop 16 rheumatoid arthritis genome-wide association study data and our results identified a few gene-gene signals. Based on our results, the PCA-LASSO method shows promise in identifying gene-gene interactions, and, at this time we suggest using it with other conventional approaches, such as generalized linear models, to narrow down genetic signals.

An evaluation of the metabolic syndrome in the HyperGEN study.

BACKGROUND: In 2001 the National Cholesterol Education Program (NCEP) provided a categorical definition for metabolic syndrome (c-MetS). We studied the extent to which two ethnic groups, Blacks and Whites were affected by c-MetS. The groups were members of the Hypertension Genetic Epidemiology Network (HyperGEN), a part of the Family Blood Pressure Program, supported by the NHLBI. Although the c-MetS definition is of special interest in particular to the clinicians, the quantitative latent traits of the metabolic syndrome (MetS) are also important in order to gain further understanding of its etiology. In this study, quantitative evaluation of the MetS latent traits (q-MetS) was based on the statistical multivariate method factor analysis (FA). RESULTS: The prevalence of the c-MetS was 34% in Blacks and 39% in Whites. c-MetS showed predominance of obesity, hypertension, and dyslipidemia. Three and four factor domains were identified through FA, classified as "Obesity," "Blood pressure," "Lipids," and "Central obesity." They explained approximately 60% of the variance in the 11 original variables. Two factors classified as "Obesity" and "Central Obesity" overlapped when FA was performed without rotation. All four factors in FA with Varimax rotation were consistent between Blacks and Whites, between genders and also after excluding type 2 diabetes (T2D) participants. Fasting insulin (INS) associated mainly with obesity and lipids factors. CONCLUSIONS: MetS in the HyperGEN study has a compound phenotype with separate domains for obesity, blood pressure, and lipids. Obesity and its relationship to lipids and insulin is clearly the dominant factor in MetS. Linkage analysis on factor scores for components of MetS, in familial studies such as HyperGEN, can assist in understanding the genetic pathways for MetS and their interactions with the environment, as a first step in identifying the underlying pathophysiological causes of this syndrome.

The effects of exercise training on abdominal visceral fat, body composition, and indicators of the metabolic syndrome in postmenopausal women with and without estrogen replacement therapy: the HERITAGE family study.

The purpose of this research was to investigate the effects of apriori estrogen replacement therapy (ERT) and endurance exercise training in postmenopausal women on abdominal visceral fat (AFV) and other selected variables related to body composition and the metabolic syndrome (MS). Forty-eight healthy and previously sedentary postmenopausal women (mean age, 54.3 years) who were enrolled in the HERITAGE Family Study (HFS) served as subjects. Of these 48 women, 18 were currently taking ERT and the remaining 30 were taking no supplemental estrogen (NHRT). Computed tomography (CT) scans were used to assess AVF as well as total abdominal fat (TAF) and abdominal subcutaneous fat (ASF). Body mass index (BMI) and waist-to-hip ratios (WHR) were calculated while body fat percentage (%FAT) and total fat mass (FATM) was assessed using underwater weighing. Blood assays for HDL-cholesterol (HDL-C), LDL-cholesterol (LDL-C), and triglycerides (TG) were conducted at a Centers for Disease Control (CDC) certified laboratory, while blood pressure measurements were assessed using an automated system. All measurements were obtained in duplicate before and after a regimen of endurance exercise training. Analysis of variance (ANOVA) showed AVF to be an average of 31.6 cm(2) less in the women receiving ERT, but lost statistical significance when AVF was adjusted for FATM. Mean values for TAF, ASF, and waist girth were also less in the women receiving ERT, but only waist girth achieved statistical significance. No differences were found in BMI or %FAT, but mean WHR was 5% smaller in the ERT group. Baseline values for HDL-C was higher and LDL-C lower in the ERT group. Prevalence of the MS tended to be greater in the NHRT group, but did not achieve statistical significance. There were no differences in training responses in any of the body composition variables between groups, however, in the ERT group LDL-C decreased with training while TG increased. It was concluded that postmenopausal women taking ERT tended to have lower values of AVF and other indicators of body composition, a more favorable lipid profile, and a slightly reduced risk of the MS when compared with women not taking supplemental hormones. Also exercise training did not improve the overall MS status of either group, as LDL-C status improved in the ERT group while TG decreased in the NHRT group.

Cross-trait familial resemblance for resting blood pressure and body composition and fat distribution: The HERITAGE family study.

Cross-trait familial resemblance between resting blood pressure (BP) and body composition and fat distribution was examined in 98 Caucasian families participating in the HERITAGE Family Study by using a multivariate familial correlation model assessing both intraindividual and interindividual cross-trait correlations. The 520 family members were sedentary at baseline examination, and both resting systolic (SBP) and diastolic (DBP) BP were cross-analyzed with each of the following 10 indications of body composition and fat distribution: percent body fat (%BF), abdominal visceral fat (AVF), body mass index (BMI), fat-free mass (FFM), fat mass (FM), sum of eight skinfolds (SF), total abdominal fat (TAF), ratio of trunk-to-extremity skinfolds (TER), waist circumference (WAIST), ratio of waist-to-hip circumferences (WHR). Five of the variables were also corrected for fat mass (AVFf, TAFf, TERf, WAISTf, WHRf) to index these measures independent of total degree of adiposity. In general, the results suggested strictly intraindividual cross-trait resemblance, with occasional spouse cross-trait resemblance, but few or no sibling or parent-offspring cross-trait correlations. This pattern is largely consistent with nongenetic specific environmental determinants for the BP-body composition and fat distribution covariation, with possibly some common environmental influence between spouses and negligible genetic effects. The only findings suggesting any familial cross-trait resemblance were significant sibling correlations for DBP-FFM and DBP-WHR, although the parent-offspring correlation was not significant. These findings suggest that the observed BP-body composition and fat distribution cross-trait correlations in these sedentary families are probably not due to multifactorial effects such as polygenic and/or common familial environmental effects. Whether or not other factors such as nonadditive effects are involved warrants further investigation using other methods. Am. J. Hum. Biol. 12:32-41, 2000. Copyright 2000 Wiley-Liss, Inc.

Familial resemblance in fatness and fat distribution.

The purpose of the study was to estimate the degree of familial resemblance in anthropometric indicators of fatness and fat distribution. The sample consisted of 327 Caucasian participants from 102 nuclear families. Indicators of fatness included the body mass index (BMI), the sum of six skinfolds (SF6: triceps + biceps + medial calf + subscapular + suprailiac + abdominal), and waist circumference (WAIST), while indicators of fat distribution included WAIST adjusted for BMI (WAIST(ADJ)), the trunk-to-extremity skinfold ratio, adjusted for SF6 (TER(ADJ)), and the first principal component of skinfolds, adjusted for the mean skinfold of the individual (PC1). A general familial correlation model was fit to the data, and a series of nested reduced models were also fit so as to test hypotheses about familial resemblance. The hypothesis of no familial resemblance (all familial correlations are zero) was rejected for all phenotypes, indicating that fatness and fat distribution aggregate within families. For the three indicators of fatness (BMI, SF6, and WAIST), the sibling and parent-offspring correlations were significant. Further, there were no sex or generation differences in the familial correlations. For the three indicators of fat distribution (TER(ADJ), WAIST(ADJ), and PC1), there was no parent-offspring resemblance; sibling resemblance was significant for TER(ADJ) and PC1. Further, spouse resemblance was not significant for WAIST(ADJ), but was for TER(ADJ) and PC1. For both WAIST(ADJ) and PC1 there were significant sex differences in the familial correlations. A combination of models including no sex or generation differences and no spouse resemblance was the most parsimonious model for BMI, SF6, and TER(ADJ). The environmental model (all correlations equal) was the most parsimonious for WAIST, the model of no sibling resemblance was the most parsimonious for WAIST(ADJ), and the model of no spousal resemblance was the most parsimonious for PC1. Estimates of maximal heritability range from 46-60% for fatness and from 29-48% for fat distribution, independent of overall fatness, suggesting that in this sample the heritability of fatness is greater than that for fat distribution. Further, the pattern of correlations, which generally includes no spousal resemblance but significant parent-offspring and sibling correlations, suggests the role of genes in explaining at least part of the heritability. Am. J. Hum. Biol. 12:395-404, 2000. Copyright 2000 Wiley-Liss, Inc.

Major gene effect on subcutaneous fat distribution in a sedentary population and its response to exercise training: The HERITAGE Family Study.

Complex segregation analysis of baseline subcutaneous fat distribution and the change in response to exercise training (post-training minus baseline indices) was performed in a sample of 482 individuals from 99 Caucasian families who participated in the HERITAGE Family Study. The sum of skinfold (SF) thicknesses at eight sites, and the waist and hip circumferences were measured at baseline and after completing a 20-week exercise training program. The trunk-to-extremity ratio (TER) was calculated by dividing the sum of skinfold thicknesses at four trunk sites (subscapular + suprailiac + abdominal + midaxillary) by the sum of skinfold thicknesses at four extremity sites (triceps + biceps + medial calf + thigh). While SF was used to assess total subcutaneous adiposity, TER and the ratio of the waist-to-hip circumferences (WHR) were used to characterize subcutaneous fat distribution. Baseline TER and WHR were age-adjusted and age-SF-adjusted within four sex-by-generation groups. The changes of SF, TER, and WHR in response to training were adjusted for age effects alone and for the effects of age and baseline values. Baseline SF was influenced by a multifactorial component (30%) plus a major effect that may be environmental in origin accounting for 47% of the variance. Baseline TER was influenced by a multifactorial component (18%) and a major codominant gene (q(2) = 0.10), which accounted for 56% of the variance. The major gene effect was independent of total subcutaneous adiposity. Baseline WHR was regulated by a major codominant gene (q(2) = 0.15), which accounted for 48% of the variance. However, this major gene effect for baseline WHR should be interpreted with caution, given the estimates of the tau's under the general model. No familial effect was found for the changes in response to training for these subcutaneous adiposity and fat distribution phenotypes. Am. J. Hum. Biol. 12:600-609, 2000. Copyright 2000 Wiley-Liss, Inc.