Distribution and Categorization of Echocardiographic Measurements in Relation to Reference Limits

BACKGROUND: Despite widespread categorization of echocardiographic measurements, there are no standardized guidelines for partitioning values exceeding reference limits. METHODS AND RESULTS: We used regression analyses to develop sex- and height-specific reference limits for cardiac M-mode measurements (left ventricular LV mass, LV wall thickness, and LV and left atrial dimensions) in a healthy reference sample (n=1099) from the Framingham Heart Study. We then examined the distribution of measurements in a broad sample (n=4957) and classified the measurements according to increasing deviation from the height- and sex-specific reference limits and the 95th, 98th, and 99th percentile values for the broad sample (categories 0 through 4, respectively). To validate the categorization scheme, we used multivariable proportional-hazards regression to assess the relations of LV mass and LV wall thickness categories to risk of cardiovascular events and the relations of left atrial size to risk of atrial fibrillation. During a mean follow-up period of 7.7 years, 587 subjects developed new cardiovascular disease events, and 166 subjects developed new-onset atrial fibrillation. After adjustment for known risk factors, there was a 1.2- and 1.3-fold risk of cardiovascular disease events per category of LV wall thickness and LV mass, respectively, and a 1.6-fold risk of atrial fibrillation per category of left atrial size. CONCLUSIONS: Using a large community-based study sample, we propose a classification scheme that provides a standardized and validated framework for partitioning echocardiographic measurements. If adopted, the categorization scheme should promote uniformity in describing measurements among echocardiographic laboratories and enhance the comprehensibility of measurements to clinicians.