Abstract 11954: Phenomapping: Hierarchical Cluster Analysis of Phenotypic Data for Novel Classification of Hypertension
Introduction: Improved phenotypic classification of hypertension (HTN), a heterogeneous syndrome, could improve understanding of HTN pathophysiology and lead to targeted treatment. We hypothesized that hierarchical cluster analysis, using repurposed gene expression software, would result in unique HTN categories that are associated with differential cardiac remodeling and mechanics.
Methods: We analyzed data from 1273 hypertensives from the HyperGEN study, a population- and family-based study of HTN. All participants underwent echocardiography and speckle-tracking for measurement of cardiac mechanics. 58 continuous phenotypic variables, including demographic, clinical, BP (at rest and with maneuvers), anthropomorphic, and laboratory data were standardized to a mean±SD of 0±1 to mimic gene expression data. Hierarchical clustering was performed using Pearson uncentered correlations and centroid linkage, agnostic to all 2D, Doppler, and speckle-tracking data. Hypertensives were classified into groups using the resulting dendrogram. Echocardiographic parameters (including indices of cardiac mehcanics) were compared between groups after adjustment for relatedness within families using linear mixed effects models.
Results: The mean age was 56±11 y (58% female) and comorbidities were common. A “phenomap” (see Figure) of hypertensive patients was successfully created, and 5 unique groups were identified. Global longitudinal strain (GLS) differed significantly between groups with average absolute GLS in groups 1-5 = 13.6%, 12.9%, 13.9%, 15.5%, and 13.8% respectively; P≤0.0001. The groups with the lowest and highest strain differed significantly across multiple domains (Table). The difference in GLS persisted after adjustment for age, sex, and type/number of anti-hypertensive medications.
Conclusions: Phenomapping is a novel technique for classification of clinical syndromes and may improve classification of systemic hypertension.
- © 2013 by American Heart Association, Inc.