Observations on Human Lactation

The purpose of this research has been to study individual variations in the yield and composition of human milk and the factors related to these differences, and to assess their practical effects in terms of breast feeding. Although breast milk is acknowledged to be the best food for the young baby, surprisingly little is known about the variations in yield and composition which occur. That there are wide variations in yield is clinically obvious, but variations of composition can only be discovered by chemical analysis. Since it is difficult to collect milk for analysis under standard sampling conditions, chemical data reported in the literature are often unreliable and even contradictory. The management of breast feeding by many doctors and nurses in this country seems to rest largely on the assumption that virtually all women are capable of adequate lactation. No biochemical data are given to support this assertion, and the findings reported here support the opposite view, namely that many women do produce milk of poor quality. Part I of the thesis deals with the fundamental and difficult problem of removing milk from the breasts. A recently developed milking machine was used throughout this study. Part II deals with the physiological variations in milk which occur (a) within a nursing, that is during the emptying of the breast (b) during the course of 24 hours, and (c) during the course of lactation. The changes in composition which occur during nursing were studied in greater detail than hitherto. The fat content was found to increase progressively throughout the feed, slowly at first and rapidly towards the end; there was an associated decrease in the content of the water soluble constituents. It was concluded that adsorption of fat globules in the alveolar and duct surfaces is responsible for this phenomenon in the breast. During the first week of lactation, the protein content of the milk falls steeply, and the lactose content tends to rise. By obtaining serial samples from a number of women, the fat content was shown to have a haphazard variation during the first few days, but to stabilize by the end of the first week. During the following two or three weeks it rises to its definitive level, but the rise is small, and a fat content which is low on the 7th day will be low at any subsequent point in lactation. Thus, in the first parts of the thesis, it is established that the minimum sample which will give a reliable picture of milk composition is an entire 24 hours secretion, and that by the 7th day of lactation this picture is sufficiently stable to give a reliable estimation of lactation as a whole. 24 hour samples from the 7th day of lactation are examined. The fat content was not related to the content of the other constituents, and seemed to be an entirely individual characteristic. There was a very low correlation between the size of the breast and the milk yield. Of very much greater predictive value was the amount of enlargement which occurred in the breast during pregnancy. This was established in a number of women by measuring the breast early in pregnancy and again in the puerperium, the difference probably being due almost entirely to an increase in the functional tissue. The wide variation in the response of the breast to pregnancy was not related to the initial size of the breast. In primaparae, the milk yield was negatively correlated with age and with net weight gain during pregnancy, at a high level of significance. That is, the milk yield tends to be low in older primaparae and in those who have stored body fat during pregnancy. There was no relationship between the milk yield and composition and the mother's diet during pregnancy. The duration of breast feeding and the weight gain of the baby were shown to be strongly related to the 7th day yield of milk and to its fat content. About one-third of women examined were so deficient in one or both of these respects, that maintenance of breast feeding for more than a few weeks would seem to be impossible. If this finding—that many women are incapable of satisfactory lactation—is generally applicable, and there is no reason to believe that the sample of women studied is unrepresentative or that local conditions are unique, then some modifications of existing attitudes to breast feeding and its management are called for.