Longitudinal study of heart size in older people.

Transverse cardiac diameter and transverse thoracic diameter were measured in a longitudinal study of older men and women at the original examination and after five years. The cardiothoracic ratio overestimated 5-year changes in heart size, because of significant decreases with age in transverse thoracic diameter. A regression equation to predict transverse cardiac diameter from age and weight had been previously computed from the data obtained at the initial examination. This was a satisfactory predictor of recorded 5-year changes except in women of 70 years and over at entry to the study in whom the predicted change was significantly larger than the recorded change. This had resulted from the death during the 5 years of women in that age group with larger transverse cardiac diameters.