Considerations in the use of echocardiography in epidemiology. The Framingham Study.

Abstract

To quantify potential biases in the use of echocardiography in epidemiologic studies, we assessed the relationship of sex, age, vital capacity, body fatness, and overt cardiovascular disease to prevalence of acceptable echocardiograms in 6148 Framingham men and women aged 17 to 90 years. Echocardiograms adequate to assess left ventricular chamber dimensions and wall thickness, aortic root, and left atrial dimensions as well as mitral and aortic valves were obtained in 4947 (80%) of the 6148 subjects. The prevalence of acceptable echocardiograms ranged from less than 50% for those more than 80 years of age to more than 96% for subjects under 30 years of age. A significant learning curve was apparent, particularly in the older subjects (more than 60 years of age) for whom prevalence of acceptable echocardiograms rose from a minimum of 28% during the first 5 months of studies to a maximum of 74 to 81% during studies 2 years later. The likelihood of unacceptable echocardiograms was slightly greater in men. Obesity (in subjects younger than 60 years of age), lower vital capacity, and overt cardiovascular disease were associated with unacceptable echocardiograms, independent of age. Consideration of these biases should aid in the interpretation and planning of epidemiologic and other studies using echocardiography.