A Statistical Study of Normal Male Adult Human Hypophysis.

Abstract

Due to the uncertainty that still exists with reference to the involvement of particular parts of the hypophysis in certain diseases, and the general lack of quantitative data on the normal proportion of the different subdivisions of this organ, it seemed desirable to study by recognized statistical methods enough normal hypophyses to give standards which would be sufficiently accurate for evaluating abnormal proportions. Clinical and pathological literature contain numerous statements to the effect that one or more parts of the hypophysis are enlarged or are too small, etc., without the author knowing the range of normal variability. In many cases unwarranted clinical deductions are based on such statements. This condition can lead only to further confusion. The material used consisted of 111 carefully selected normal male hypophyses, 20 to 76 years of age, from cases of relatively sudden accidental death. After getting the fresh weight of the organ serial sections were made. Every 20th section was projected and the different parts of the gland outlined at a magnification of 20 diameters, in general. In certain cases pars intermedia was projected at a higher magnification (up to 100 diameters). The different portions of the outline were cut out and weighed. This gave the proportion that any particular part constitutes of the whole. The weight of the paper was always more than 100 times the weight of the corresponding part of the fresh organ. The method is more fully explained in a previous note. The routine method of obtaining the hypophysis at autopsy made it necessary to cut off and discard the stalk and as much of the surrounding connective tissue as possible. The frequency distribution of the weight of the whole gland, of the anterior lobe, and of the posterior lobe, is of the common fairly symmetrical unimodal type; while that of the epithelium of pars intermedia and of the colloid is greatly skewed in the positive direction, i. e., towards the small values.