Small The patient made an uninterrupted recovery. The temperature at no time was above 99, the pulse at no time above 90. quantities of water were administered by mouth 12 hours after the operation. The patient received saline injections. Soft food in small quantities was allowed on the seventh day. The button passed on the eighteenth day. The patient walked out of the hospital on the twenty-sixth day. -At the present time, ten months after the operation, the patient is well. He has gained about twenty-five pounds in weight and has not suffered from digestive disturbance. CONCLUSION. It has seemed to the writer, in reviewing the literature on the subject of intestinal anastomosis, that the mistake has been made of advancing a particular method as the best for general use. At the present time, both the Murphy button and the suture have their advantages and disadvantages, both their limitations. The Murphy button is unquestionably the simplest method of effecting anastomosis in most cases, in the hands of the majority of surgeons. It is certainly the quickest. Axial anastomosis of the small intestine, cholecyst-enterostomy and many cases of gastroenterostomy afford special indications for the Murphy button and here it can be advocated. In anastomoses of the large intestine, particularly when complicated by malignant growths where cachexia has impaired vitality, the suture should be regularly used. Sentiment at present favors one of the varieties of continuous sutures. In preparing this paper, reports on about twenty-two hundred cases of intestinal anastomosis have been examined, including the sixteen hundred and twenty cases collected by Murphy. The writer has been thoroughly impressed with the difficulty of drawing very satisfactory conclusions from these statistics. It is evident that cases where results have been unfortunate, have not been reported, a fact that may account for the opposition to the button method as being out of proportion to the results from recorded cases. When it is taken into consideration that the Murphy button has been regularly advocated in operations requiring great haste, that the suture has been undoubtedly used in more favorable cases, that the mortality in the two methods, even under these circumstances differs but little, it would seem that the Murphy button should be regarded today as one of the valuable practical discoveries of the age. The writer wishes to acknowledge the assistance of Dr. F. B. Standish in the preparation of this paper. DISCUSSION. Dr. William H. Carmalt (New Haven): Mr. President, I am disappointed that Dr. Sullivan is not here. I have but little to say in regard to the subject, for Dr. Sanford's paper goes over the matter so completely that it is hardly worth while to take the time of the society in attempting to add to it. There are, however, two or three points I might refer to. In his paper he says that an end-to-end anastomosis is only adapted to the small intestine. You might say it is preferably adapted to the small intestine, but to say it is only adapted to the small intestine is too broad. It depends on the condition of the two parts of the intestinal tract that you are operating upon. Some fifteen years ago I operated upon a case of cancer of the cæcum, in which the lumen of the intestine had been very much narrowed, and I found, on cutting into the abdomen to remove the diseased mass, that the large intestine was atrophied from disuse, and contracted to a small size. The small intestine was hypertrophied to a very considerable amount, so when I brought the ends together they were practically of the same size, and there was no difficulty at all in making a union. I made that union with the suture method. I think the question of time as being a reason for selecting the Murphy button in preference to a suture, depends very much upon the skill of the individual operator. Certainly a foreign body in the intestines is objectionable in itself. I do not deny that in the hands of many, perhaps you might say the majority of surgeons who are not operating like the Mayos, daily, or several times a day, the Murphy button can be put in quicker than an anastomosis by suture; at the same time, the anastomosis by suture can be done very quickly in skilled hands. I must, however, say my preference for speed and accuracy of application is still in favor of the McGraw ligature. I am in favor of that by reason of the facility with which it can be done. The McGraw ligature, you know, is an elastic cord, and I want to insist upon one's using the cord, and not attempting to do the operation with a tube: it must be the solid cord that is to be used, which you can stretch to a very considerable extent, so small that it will draw through, as McGraw says (it is not worth while to go over the technique of the operation again, as it has been described by Dr. Sanford), but, with a round needle, which perforates the intestine or the stomach, as the case may be, it crowds the fibers of the cut apart. It doesn't cut, but it crowds them apart, and when the ligature is drawn through tight, it makes a small opening. When the ligature is relaxed it fills the opening entirely and there is no leakage. Then it is tied together, the knot again tied with a piece of silk in order to secure against slipping, and finally the ends drawn in and a Lembert suture put on outside. You make no opening into the intestine at all except what you fill up with the ligature, and there is no chance of leakage. The only objection to it is in cases where we can'not wait for an opening to slough through. It is particularly adaptable to cases of gastro-enterostomy, performed for carcinoma of the stomach, associated with stenosis, when the radical operation is inadmissible. Dr. Oliver C. Smith (Hartford): Just one word. The end-to-end anastomosis, we should remember, should be done (as was brought out in Dr. Sanford's paper), in chronic conditions, not in acute conditions. An end-to-end anastomosis is not performed in acute conditions, for in order to stand the tension and strain, you want a larger caliber which a lateral anastomosis will give. Therefore, we require a lateral anastomosis in acute conditions, and an end-to-end in chronic conditions. In cases of hernia, as Dr. Sanford states, it seems to me the second incision is better, doing the anastomosis first, the hernia can be done afterwards. You have more room and can draw up the intestine. You can put down a drainage tube, but it is not wise to put gauze down to the suture point. If you put gauze down to the suture point it absorbs the plastic lymph and invites leakage, the very thing you don't want. Dr. Leonard W. Bacon (New Haven): Mr. President. With regard to the elastic suture in gastro-enterostomy, it seems to me it fills one important particular. The tendency in gastro-enterostomy, especially where the pylorus is patent (the authority quoted by Dr. Brown to the contrary, nevertheless), the tendency in those cases, I say, is for the anastomotic opening to become insufficient. This was brought out, I remember, . by Dr. Moynihan when he was discussing this subject in Washington, that it was a matter of considerable importance, where it was possible, in these cases of gastro-enterostomy, to remove a portion of the gastric wall (which is around the margin of your anastomosis), in order to cut away a certain amount of the superabundant mucosa, and also the whole thickness of the gastric wall. You cannot cut away much of the intestinal wall, because there is not enough of it, but you can cut away quite a generous amount of the gastric wall, and I think that helps very decidedly in a gastro-enterostomy. But the anastomotic opening through a linear incision, which is caused by pressure from one of these devices, either with twine or with a rubber ligature to get a symmetrical incision from one organ to another, I don't think is adequate. Another point in Dr. Sanford's paper which was brought out very distinctly by Dr. Lee in his paper, is that the only gastric surgery which amounted to very much was within the last five years. I strongly sus pect the statistics Dr. Sanford has gathered with regard to the relative advantages of anastomosis (and particularly gastro-enterostomy, and anastomosis by suture and anastomosis by mechanical appliances), do not apply to the last five years. Dr. D. Chester Brown (Danbury): There seems to be a great discrepancy in the statistics reported in the use of the Murphy button. Personally I don't like it. I find a great difference in the strength and the spring of the Murphy button, as much difference as there is in the tensile strength of any elastic. If you take time to test a number of Murphy buttons as to their power of compression, test the same button that you have had in your instrument case for a number of weeks, or months, or years, you will find that the individual button varies in the amount of pressure. It seems to me here is a fact we have not appreciated in the results we get from the button. Tuberculous Peritonitis. JOHN B. BOUCHER, M.D., Hartford. Tuberculous peritonitis is dependent upon an infection by means of the bacilli circulating in the blood, or upon an extension of tuberculous inflammation, or ulceration from adjacent organs. All agree that the tubercle bacilli are the invariable cause of the disease, but the source of invasion admits of more and various opinions; the route by which they reach the peritoneum is frequently difficult or impossible to determine, but is evidently by many different sources. Dieulafoy believes that the most frequent source is through the intestines where the bacilli have been introduced by sputum or food, infected milk or meat. Clinically, I do not believe that we can eliminate bovine tuberculosis as a source of infection in man. The work of Salmon, Ernst, and a host of other accurate observers, has shown the possibility of infection through the milk and food supply. The structure of the large follicles of the intestines and of Peyers patches corresponds closely to the follicles of the tonsils. We know that next to the lungs, the most susceptible localities to tuberculous invasion are the tonsils, where the lymphatic tissue lies directly exposed on the mucous surface; in the intestines and appendix, practically the same condition exists. The bacilli may attack the intestines primarily and the peritoneum next, or being absorbed by the superficial lymphatics of these structures, may attack the peritoneum first; this hypothesis is confirmed by the experience of Wesner and Cornil. Unquestionably the bacilli are capable of passing through the walls of the intestines without a primary lesion of the bowel and thus reaching the peritoneum without leaving any trace of trouble at the atrium of invasion, as it so often does in the pharynx when the cervical glands become infected. Primary tuberculosis of the intestines occurs in three forms: (1) Multiple ulcers due to inoculation from sputum; (2) single ulcera |