As part of a longitudinal study of healthy children by the staff of the Child Research Council, roentgenograms of the chest have been made at frequent intervals. Three cardiac diameters (transverse, long, and broad) and the internal diameter of the chest were measured on each of 3205 of these roentgenograms, taken of 128 subjects over a period of years.
The size and shape of the heart are illustrated and discussed with emphasis on the range of variation that is seen in healthy individuals and on the inadequacy of one set of "normal standards" for evaluating the cardiac silhouette.
In spite of fluctuations in the growth curves for the cardiac diameters, a general pattern of agreement was found in the increases in the cardiac diameters and the increases in body height and weight during childhood and adolescence. It would seem that periods of rapid growth such as are usually seen in adolescence are frequently coincident with fairly rapid increases in the cardiac diameters, suggesting that cardiac demands are greater during such growth spurts.
The mean values for transverse diameter of the heart showed the same type of sex differentiation that is found in the mean values for height and weight in boys and girls. It seems logical to assume that changing cardiac size should be considered as part of the growth process rather than as an isolated physical and physiologic process.
The relations that seem apparent between transverse diameter of the heart and height, weight, and internal diameter of the chest could not be proved statistically by calculated coefficients of correlation. However, it was possible to show differences in the mean values for cardiac transverse diameter in three groups, classified as to height-weight relationships into overweight, medium-weight, and underweight individuals. The mean values were greatest for the fat group, least for the thin group and intermediate for the group that was of medium weight for height. Body build may therefore be a factor in determining cardiac size during childhood as well as during adolescence and adult life.
Since the width of the chest is increasing during childhood and adolescence in much the same manner that the transverse diameter of the heart is increasing, cardiothoracic ratios do not become progressively greater with advancing age. In fact, the successive ratios on the same individual show little regularity toward either increase or decrease although mean values for the different ages do decrease from a high of 0.44 at four years of age to a low of 0.40 in the post-adolescent age groups. Each individual showed considerable fluctuation in the cardio-thoracic ratios but no one person fluctuated as much as the range for the whole group. No ratios were found above 0.50 or below 0.32. No sex differences were found nor was there any significant difference in the cardio-thoracic ratios for the groups of different height-weight proportions. In evaluating the heart size of an individual from a single film, the cardio-thoracic ratio is probably as satisfactory as any other measurement if one recognizes the wide range of healthy variation. An increase in the cardio-thoracic ratio on successive roentgenograms might be more significant clinically than cardiac measurements which did not take into consideration the growth of the individual.
The nomogram constructed by Ungerleider based on height and weight for prediction of transverse diameter of the heart on teleoroentgenograms of adults was tested for its applicability to the later childhood, adolescent and early adult periods. Nearly half the predicted cardiac transverse diameters exceeded the measured values by 10% or more.
This study would seem to indicate, therefore, that one should not be discouraged by the range of variation or the fluctuations in cardiac measurements from routine roentgenograms of the chest. Valuable information regarding the significance of the size of the heart can be obtained from such roentgenograms if one relates those data to the basic process of growth and maturation of the individual.
Mixed convection in turbulent channels with unstable stratification