If the lesion involves the large intestine, even including the appendix, the same rule should be applied to enemata for, although the injection is not directly deposited near the upper part of the colon, reflex peristalsis may carry it there or the entrance of fluid into the rectum may indirectly distend the upper colon or excite peristaltic waves. However, in the case of a small solution of continuity, as after a simple removal of the appendix, two or three days' delay is almost always sufficient and feeding by mouth may them be resumed. Even before this time, small quantities of beef juice, or innutritious beef tea, peptonized milk etc., which yield very little residue, may be given in most cases.

In operations on the salivary glands and ducts, it is desirable to feed by the bowel and by hypodermatic injection and inunction for several days. Thirst and hunger may be repressed by opiates and care should be taken not to stimulate the secretion of saliva by sapid medicine allowing the patient to see food or smell it in process of cooking etc. Even with such precautions, fistulae are liable to occur.

In oesophageal operations, it is frequently possible to introduce a tube so that feeding may be resumed almost immediately, without irritation of the wound. Otherwise, the general principles discussed should be followed, so as to allow a considerable time for healing.

Operations on the stomach and upper intestine, are directly amenable to the general principles applying to this group of cases.

After all operations requiring anaesthesia and in which the dorsal posture is to be maintained for some time, and the patient is unconscious or helpless, care should l>e taken that the patient does not drown in his own vomit or aspirate even small masses of mucus or food into the air passages. Vomiting is especially frequent in biliary lesions, perhaps most of all in cases of gall stones, but also frequently occurs when there is peritonitis or mechanic pinching of the peritoneum-covered viscera, such as the bowel, and when the female pelvic organs are involved. An important detail is the selection of a nurse vigilant and strong enough to handle a patient quickly in such an emergency.

When, for any reason, any considerable wound has been made through the abdominal wall, very much the same precautions must be taken against undue distention of the alimentary canal, increase of abdominal pressure by vomiting, retching, gagging, straining etc., as if there were a lesion of the alimentary canal itself. Indeed, the abdominal wall bears almost exactly the same relation to the alimentary canal as does the outer case of a double pneumatic tire to the inner tube. Aside from the danger of tearing out stitches - sepsis, haemorrhage, failure of primary union - such strains upon the abdominal wound must be avoided until firm cicatrization has occurred, and for this process, several months must be allowed.

In operations directly involving the rectum and the lower bowel generally, the diet should preferably be rather light and easily digested, so as to yield little residue for several days before the operation. The bowels should be thoroughly emptied, both by cathartic and enemata, the evening and the morning before the operation. Further passages should be avoided, if possible, for three or four days although it must be admitted that if sutures and ligatures are properly applied, earlier passages do not usually do mechanic injury and sepsis can be avoided by douching with physiologic salt solution or mild antiseptics. While the bowels may be locked up with opium, it is more rational to withhold nourishment for 24 hours and to give very light, digestible or pre-digested nourishment which yields little residue, for two or three days more, so that the first few passages are small.

As has been stated, operations involving direct lesion of the alimentary canal, its tributary ducts and its restraining wall, should generally have a post-operative period of 24 hours to a week - on the average, two to four days - in which no food or liquid enters the alimentary canal, hypodermatic injection and inunction being employed if necessary, though the rectum may also be used if the lesion is above the large intestine. Following this period, is a variable one of very light, digestible, practically residue-free diet and then one of more liberal but carefully selected diet before the patient is discharged by the dietetist.

The management of this third period is practically independent of the direct operative lesion and, obviously, too, of the lesion for which operation was undertaken, providing that the operation has been radically successful. This is merely another way of stating the fact that perfect union of the wound in the alimentary tube itself, is to be expected within a short time, say ten days to three weeks at the utmost. The qualification must be made that an intentional or adventitious fistulous opening requires more protracted care and that the abdominal wound does not become firmly cicatrized for several months.

The principles governing fistulae of the alimentary canal itself - gastrostomy, enterostomy, gastroenterostomy etc., - are largely due to interferences with the normal passage of contents and are discussed elsewhere.

Fistulae involving the conduction of the secretions of the tributary glands, salivary, hepatic and pancreatic, all demand for their closure the highest practicable degree of suppression of secretion during the process of healing. Salivary fistulae are practically always adventitious. While the secretion has a physiologic function in connection with digestion, it is not very important for reasons already discussed, and as it acts only on cooked starch, it has no corrosive action. Thus the fistula is objectionable rather on cosmetic grounds than physiologic.

Hepatic and biliary fistulae are usually intentional for the practical reason that drainage is more easily and more safely provided thus than by attempts at immediate discharge into the intestine. The hepatic secretion is not of great importance physiologically, the main function of bile being eliminative. Still, the permanent or prolonged drainage of so much water and alkali and of the biliary salts is physiologically detrimental. It is also inconvenient and mortifying to the patient. The fistula seems to have no more corrosive action on the skin than the discharge of so much saline solution.

Jaboulet, Tixier and von Chappert have called attention to the fact that the leakage from biliary fistulae is most troublesome late after meals. In their report, two cases are mentioned in which the flow began quite regularly 5 1/2 hours after meals. By giving a late evening meal, they succeeded in diverting the bile into the intestine and thus securing healing of the fistula. This method has been corroborated by other observers.

Pancreatic fistulae are not so common as biliary and are almost always either purely accidental or due to the impracticability of treating an abscess, cyst, tumor or calculus, or a direct traumatism, by immediate closure of the external wound and normal drainage into the intestine. The pancreatic secretion is of immense practical physiologic importance and. owing to its digestive action on tat and proteins. a pancreatic fistula or one into the upper intestine, is always highly corrosive and it is exceedingly difficult to protect the tissues by water-tight dressings. Sufficient operative and experimental experience has not been accumulated to establish rules tor feeding to expedite the healing of pancreatic fistulae but they are undoubtedly to be treated dietetically on the same general lines as biliary.

Without entering into a protracted discussion of the dietetic management of surgical eases involving the alimentary canal, its sustaining wall ami tributary glands, for the third period of more liberal but still carefully guarded diet, the following general principles may be stated: complicating diseases and such organic or functional digestive disorders as remain after operation, should receive due attention: the diet should be regulated with regard to the existence of fistulae. adhesion threatening obstruction, and other organic results of the operation; while the first period of star-vation and the second period of limited, residue-free diet suffice lor the healing of the direct lesion of the alimentarv tube, and the obliteration of blood vessels that have been cut, tie abdominal wound or any similar mechanically weak resulting lesion, requires the avoidance of all forms of strain for many months. Saprophy-tosis, with formation of gases; constipation, with subsequent increase of peristalsis and of intra-tubal pressure; straining at stool, retching, vomiting etc.. must be avoided.