The illness entailed by an ulcer of the stomach is usually a long one, and patients come before us at different stages in this long career. At all stages the dietetic treatment is of the greatest importance. There is good reason to believe that the number of cases of "chronic" ulcer and "recurring" ulcer might be reduced if a more rigid diet were maintained for a much greater length of time than is now customary. But it must be allowed that such a plan is often found to be difficult or impossible, considering that the bulk of the cases of gastric ulcer are drawn from the class of working girls, who have little choice as regards food, and no opportunity for rest.

For the purpose of dietetic treatment cases of gastric ulcer may be arbitrarily divided into several classes : (A) Very commonly the patient comes under treatment within a few hours after an attack of haematemesis; (B) sometimes days or weeks have elapsed since the haematemesis; (C) there is often a long antecedent stage during which (in the absence of haematemesis) an ulcer is suspected but cannot be recognized with certainty; (D) often haematemesis has occurred on several occasions in the past, or with one previous haemorrhage there has been continual pain after food, so that the story may run back for years. Though gastro-enterostomy should always be performed in such cases, dietetic treatment still remains a necessity. One cannot help thinking that in these cases the chronicity or recurrence of the disease is often due to defective treatment at their outset; (E) the patient may come before us suffering entirely from one or more sequelae consequent on the healing of an ulcer, such as pyloric stenosis, perigastric adhesion and hour-glass contraction; (F) in a small proportion of cases (about 10 per cent of hospital admissions) perforation occurs without previous symptoms of ulcer, and after recovery from operation (with or without gastro-enterostomy) dietetic treatment is as important as in any of the other classes.

The dietetic treatment in all these classes is directed to the affording of rest to the stomach. Movements of the stomach-wall delay cicatrization of the ulcer, and in the few days following haematemesis movements may disturb the thrombus upon which safety depends. But rest to the secretory as well as to the motor function of the stomach is needed, and the food given (especially in the later stages of treatment) must be such as will reduce the secretory requirements to as low a level as possible, and this choice of food will at the same time tend to keep down the hyperacidity (q.v.) which is such an important accompaniment of gastric ulcer. These principles are matter of general agreement, but in points of detail it will be seen that there is room for difference of opinion.

(A) Immediately After Haematemesis

I do not doubt that for the first few days after haematemesis no food or fluid should be given by the mouth and that strict rectal feeding should be adopted. I feel sure that even the giving of small pieces of ice is pernicious. During these few days we are not concerned with the healing of the ulcer. Our object is to avoid mechanical disturbance of the thrombus during the few days necessary for the beginning of its organization.

Opinions differ, however, as to the duration that should be prescribed for this strict rectal feeding. The principle of rest to the stomach is good, and there are those who carry it out to the extreme by employing it for two or three weeks. On the other hand it must be recognized that rectal feeding is always a state of semi-starvation, which is not calculated to promote the healing of an ulcer. Moreover, during prolonged rectal feeding it is probable that further disadvantages occur, which militate against cicatrization. The oral and gastric secretions are absent or small. The mouth is apt to become dry, and the tongue foul, so much so that parotitis occasionally sets in. It is probable that owing to the lack of acidity there is an abnormal growth of bacteria in the stomach, as there certainly is on the tongue. Presumably also, as in the case of all disused organs, there is some degree of bloodlessness of the stomach during deprivation of food. So that even if we possessed a method of maintaining the general nutrition, while keeping the stomach empty and at rest, I doubt if it would be wise to employ such a method for long. It is a matter of opinion and admits of no proof, but I believe not only that the gastric ulcer is formed suddenly by necrosis to a varying depth of the stomach-wall and the immediate digestion of the necrosed area, but also that there is from the first a strong natural tendency to rapid healing of the ulcer, if during this early stage the local and general conditions are favourable.

Impressed by the necessity of procuring stomach rest at all costs, some physicians maintain strict rectal feeding for long periods, two or three weeks in some cases. But there is an opposite extreme, and Lenhartz, influenced by the malnutrition and other evils which attend rectal feeding has discarded it. Even on the day of haemorrhage he allows 7-10 oz. of iced milk in spoonfuls, and from 2-4 beaten eggs in the 24 hours. For my part I think that strict rectal feeding for a certain time is necessary after haemorrhage and that the advantage gained, if it is not too prolonged, outweighs any possible malnutrition. As regards the duration of this stage of treatment, each case must be considered on its merits, and no definite rule can be laid down. In all cases I think that at least three or four days should elapse after haemorrhage before any food or fluid can be safely given by the mouth. As a matter of fact after an interval of four clear days haemorrhage seldom recurs. During the actual occurrence of haematemesis, the rule of course must be absolute. In cases of recurring haemorrhage, this practice of allowing three or four days to elapse without food or fluid by the mouth after the last appearance of blood, necessitates in some cases a much longer period of rectal feeding than is desirable. But it is a choice of evils and for the moment stoppage of bleeding and organization of clot are more important objects than healing of the ulcer.