Physiologic and surgical experiment has shown that, while all the alimentary organs have their use, man can maintain a fair degree of health without mastication and insalivation, without the stomach, without the large intestine, with the bile diverted from the intestine; but the pancreas must discharge its secretion into the intestine and not more than half of the small intestine can be resected without causing gradual failure of nutrition.

Thus it must be understood at the outset that no adequate substitute for normal ingestion can be devised which does not pass the food through at least the greater part of the small intestine.

Feeding Through Fistulae

The principles of choice of the various nutritive fistulae are discussed elsewhere. When established, the principle of feeding is perfectly simple: Administer enough of the proper kinds of food at suitable intervals. Generally speaking, the food should be the same as would be chosen for anv other invalid, it should be as well and as neatly, though not necessarily as daintily, prepared. If it is at all feasible, the patient should masticate and insalivate his own food and spit it into a funnel connected with the fistula tube. Thus, even with regard to taste, the food should approximate that taken in the customary manner. Regarding frequency of feeding, the location of the fistula and the cause for making it must be considered. If a gastric fistula has been established on account of oesophageal obstruction, with no great gastric abnormality, the diet should be in practically all respects the same as in health, with due regard for ease of passage through the fistula tube and with attention to hygienic details which ought to be followed even by those not under medical supervision.

As is discussed elsewhere, a fistula ought not to be made into a stomach that is seriously diseased, especially if the disease is cancer, but a superior enterostomy should be performed. It sometimes happens, however, that the fistula has been made into a cancerous, dilated or degenerated catarrhal stomach, either through an error in locating and diagnosing the exact nature of the lesion, or an unpardonable failure to appreciate the rudimentary physiologic principles underlying the method of attacking the condition, or because of some obstacle to the performance of enterostomy. In such cases the intervals of feeding, the nature of the food, its preparation by predigestion, etc., the use of hydrochloric acid or other digestant, require appropriate modification, and the fistulous opening may even be used for lavage or local applications.

While, theoretically, food introduced directly into the duodenum or jejunum should be prepared to imitate the prior action of gastric digestion, as by artificial digestion with pepsin and hydrochloric acid for an hour or more, such preparation does not seem to be necessary or even advisable. It is, however, necessary to have the food warm and of soft consistency, and it should be introduced somewhat more gradually than through a gastric fistula, on account of the lesser lumen of the intestine and to imitate the gradual, intermittent entrance of food through the pylorus.

In many instances, patients have suffered after the establishment of a fistula, on account of a failure to realize the amount of food and of the various inorganic and organic nutrients required to maintain life. For example - and, unfortunately, these are not imaginary instances - a cancer patient is given two or three hundred c.c. of a proprietary peptone solution in a day, aggregating 10 - 15 grams of protein, about the same amount of sugar, and 50 c.c. of alcohol, less than 500 calories altogether, or a child on whom gastrostomy has been practiced for cicatrical closure of the oesophagus is given a mixture of milk and cod liver oil, theoretically adequate in protein and in calories, but containing far more fat than can be assimilated, so that diarrhoea is produced. Excepting that the appetite need not be consulted to any great degree, even more care should be exercised in feeding through fistulae than by mouth. Both to comfort the patient and to secure reflex stimulation of the digestive secretions, some food should usually be taken into the mouth. Whether this same food should be used in feeding or whether it should be spit out and other food substituted, should be determined by carefully balancing the advantage of salivary digestion against the inevitable introduction of bacteria even after the most careful cleansing of the mouth.; While, as a matter of convenience, milk, eggs, cereal gruels, meat juice, etc., will usually be employed, the diet should be carefully measured, compared with analytic tables, and criticized with regard to adequacy in water, sodium chlorid, iron, protein, fat and carbohydrate. The water and salt may be introduced by enema, or subcutaneously, if preferred; and it may even be that water can be introduced by mouth and will be absorbed in sufficient quantity when the oesophagus will not permit the passage even of sufficient quantities of milk, or when the pylorus is almost absolutely closed so that an enterostomy has been required. The general state of nutrition, the examination of the stools, and direct tests of gastric or intestinal juices withdrawn through the fistula, will determine whether predigestion is necessary or not. The following sample dietary is suggested, and may be modified in various ways:

The above diet is scarcely sufficient, even for a patient in bed, but it is about all that can be given at first. One thousand five hundred c.c. of physiologic salt solution should also be given by bowel or otherwise. The meat juice and bovinine will probably contain an adequate ration of iron. The latter contains some alcohol, not included in the caloric estimate. In a day or two, 1500 c.c. of milk can be given, which will increase the calories to nearly 1900, and various cereals may be employed, provided that they make a soft mass with the milk.