very distinctly one finger's breadth below the left costal cartilages. Both right and left lobes were hardened and slightly nodular. There was no indication of malignancy and no indication of syphilis. The pelvic organs were fairly healthy. The lower portion of the rectum showed extensive hemorrhoidal engorgement. The urine averaged from 40 to 45 ounces in 24 hours, with a trace of albumen, but no casts. The urea, estimated every day, averaged from one and eight-tenths to two per cent, or practically

normal. The abdominal veins about the umbilicus (caput Menusæ) were enlarged. There was no anasarca of the limbs, as will be seen from the illustrations and no oedema of the lungs at any time.

Operation.-The operation of venous anastomosis was determined upon, as there was clearly nothing else to do. The patient could not live in her present condition; accordingly the operation was performed on November 5th. I was assisted by my staff, Dr. Day, Dr. Stark, Dr. Eklund, Dr. Gregory, and my corps of trained nurses; Dr. Hodghead administering the anaesthetic. The abdominal wall

was opened in the median line from the enciform cartilage to the umbilicus. The veins of the great omentum and portal circulation were found to be very much enlarged and engorged. Ritzius' veins, the portal branches of the intestines and mesentery were engorged. Schiff's para-umbilical veins anastomizing with the abdominal subcutaneous veins were also much engorged. Sappey's accessory portal system, consisting of numerous small veins formed around the liver and gall-bladder and the suspensory ligament of the liver with the gastro-epiploic omentum were also engorged. The liver was found to be of a deep-red color, very hard (cirrhotic) and nodular; the nodules studded all over the liver being about the size and the elevation of a split pea. Both the right and left lobes were enlarged, but evidently undergoing atrophy. It is to be regretted that this cirrhosis of the liver had become so extensive, although it may be remarked that practically the normal amount of urea was secreted. The operation consisted in abrading the peritoneal covering of the under surface of the diaphragm and of the peritoneal covering of the convex surface of the liver by means of a dry gauze sponge. The convex surface of the spleen was also rubbed, as was its adjacent surface of the diaphragm. The rubbing of these peritoneal surfaces with gauze was done with great gentleness, although sufficiently hard to cause a small amount of sanguinous oozing from their surfaces. The peritoneum covering the anterior abdominal wall for a space of two inches each side. of the incision was next rubbed until it showed signs of diapadesis. To this raw surface was attached, by means of chromosized, cumulized catgut sutures, the gastrocolic and folds of the great omentum. The abdominal wound was closed with a layer to layer stitch. The dressings were then applied, consisting of three-inch adhesive rubber plaster strips united in the center by means of buckles in such a manner as to force the liver and spleen up to the diaphragm. These dressings were retained for fourteen days. A perforated 3-inch rubber tube was inserted through the abdominal wall, between the umbilicus and the pubes into the cul-de-sac of Douglass. The outer end of this tube, some three feet in length, was placed beneath a sterilized carbolic solution, 5 per cent, to admit of free drainage. The whole abdominal wall was then strapped tightly from the enciform cartilage to the crests of the ilium.

The patient made an uninterrupted recovery from the operation. The subcuticular stitches were removed in about two weeks, the drainage tube by this time having been displaced, was left out altogether. During these two weeks the ascitic fluid removed from the abdominal cavity amounted to less than three gallons, instead of from twelve to fifteen gallons, the amount withdrawn from the peritoneal cavity prior to the operation for the same length of time, showing the enormous benefit derived from the operation in the way of preventing peritoneal exudation. The liver could be plainly felt attached to the diaphragm, and the omentum could easily be felt attached to the parietal portion of the abdominal peritoneum beneath the incision. During the first week after the operation the patient complained of some little diaphragmatic pains, showing conclusively that adhesions were forming and tightening assisting the enormous reduction in the quantity of peritoneal fluid exuded. After the first week following the operation the patient began to pick up, the appetite improved and she remained cheerful. The pulse became more rapid, ranging from 90 to 100 per minute with a little less volume than before the operation. During the second week after the operation she excreted about 40 ounces of urine a day with a little more albumen, about half of one per cent-the urea remaining about one and eight-tenths per cent each 24 hours. The patient was allowed to sit up in bed on the 16th day, and her progress towards recovery was very satisfactory. On the 18th day the urea suddenly diminished to nine-tenths of one per cent. The pulse reached 120. On the nineteenth day she excreted 35 ounces of urine, and the urea diminished to five-tenths of one per cent, pulse remaining 120 with no elevation of temperature. On the 20th day the patient passed 30 ounces of urine containing two-tenths of one per cent of urea, the pulse reaching 130 per minute. Respirations became more rapid and her general condition became grave. On the 21st day she excreted about 30 ounces of urine with scarcely a trace of urea, albumen remaining about half of one per cent. Muscular twitchings were observed in the morning. A few hours subsequently she had a decided uremic convulsion and died of uremic coma at the close of the 21st day. Three weeks after the operation and one week after the

abdominal drainage tube had been removed paracentesis was performed once, removing a gallon and a half of fluids showing that the amount of ascitic fluid had diminished to less than a quarter of the former quantity. The advanced stage of the cirrhosis of the liver and the diminution in the amount of urea were undoubtedly the immediate causes of her death.

Collateral Circulation.-It is interesting to note the avenues through which collateral circulation was established after the liver and omentum had been attached to the diaphragm and parietal peritoneum. Through the iliocolic, right colic, middle colic and veins of the great omentum, collateral circulation was established by means of the circulation in the abdominal walls anastomosing with the superior epigastric vein, internal mammary, long thoracic or external mammary, anterior intercostals, infracostals, the four lumbar veins, superficial and deep circumflex iliac, superficial and deep epigastric, and through them with the anterior and posterior vertebral, ascending lumbar and vena azygos entering the superior cava and right subclavian. The anastomosis between the veins of the convex surface of the liver and the diaphragm, was carried on by the musculo-phrenic, superior phrenic, sternal, mediastinal, pericardial, œsophageal, veins of the thoracic duct, infracostals, intercostals, mammary, vertebral, upper lumbar and vena azygos. I am quite satisfied that in the early stages of ascites from cirrhosis of the liver, the operation of venous anastomosis is of great practical benefit.

MEDICAL ASPECT OF VENOUS ANASTOMOSIS OF THE LIVER.

By H. D'ARCY POWER, L. R. C. P. Ire, L. S. A. Lond. Professor of Principles and Practice of Medicine, College of Physicians and Surgeons of San Francisco.

The increasing tendency of surgical methods to invade the domain of internal medicine is exemplified by the recent introduction of the operation of forming extra hepatic anastomoses for the relief of portal obstruction; one of the last instances of which is reported by Dr. Winslow Anderson in this number. It is natural and right that the physician should examine all such proposals with a critical eye, knowing well that surgical success and restored health are not

always synonymous terms. In submitting the operation in question to such a criticism, we must consider both the relief offered and the cost, and finally the prospect as to length of life with or without the operation.

The symptoms of liver disease naturally group themselves. into two classes: Those due to interference with the functions of the liver, and those having their origin in obstruction of the passage of blood from the abdominal viscera to the vena cava by the portal vein. The first group of symptoms is always present in all liver disease; the second group only under certain conditions of fibrosis of the organ, or gross obstruction of the portal vein. It is not contended. that the operation in question can in anyway benefit the first group, but that it can and does relieve the second.

Portal obstruction manifests itself most obviously by the development of a progressive ascites, and further by a condition of congestion throughout the alimentary tract, that not only interferes with the proper secretory activity of these organs, but seriously affects the processes of assimilation. It may, therefore, be taken as granted that portal obstruction, in itself, and without any help from the coexistent hepatic insufficiency can, and often does, lead to a directly fatal result. Such being the case, any means taken to relieve or prevent the condition is justifiable, providing it does not substitute a greater danger. This latter proviso is the crux of the question.

The liver is the largest organ of the body and is engaged in many and essentially vital functions. It secretes bile, which latter is largely an excretion. Its glycogenic activity is of first importance. The power possessed by the hepatic cells of depurating the blood and, at least temporarily, fixing poisons absorbed from the alimentary canal is our chief protection from auto-intoxication; while as the chief factor in the production of urea, the cessation of its function immediately threatens life. An immense body of clinical evidence, as well as abundant experimental data, establish the fact that the liver cannot be dispensed with. Now, Talma's operation, as a matter of fact, does practically dispense with the liver. It is used for the relief of ascites, but ascites does not occur until there is considerable obstruction to the passage of the blood through the organ, and the only reason that the liver receives any quantity of blood at