LICHTMAN--GASTRO-INTESTINAL

B L E E D I N G I N D I S E A S E OF T H E L I V E R A N D B I L I A R Y T R A C T

439

SUMMARY Three cases of regional ileitis are reported. Credit for a real advance in g a s t r o - i n t e s t i n a l disease - - t h e recognition of t e r m i n a l or regional e n t e r i t i s - is due Drs. Crohn, Ginsburg and Oppenheim. The cause is, as yet, unknown but with f u r t h e r study probably will be discovered. The condition must be t h o u g h t of in all cases of chronic d i a r r h e a or constipation and in atypical abdominal disease. In X - r a y b a r i u m " p r o g r e s s " meals, more frequent examinations, at s h o r t e r intervals, and at different angles, may reveal a b n o r m a l i t i e s in the small bowel t h a t would otherwise be overlooked. B a r i u m meal should always be supplemented by b a r i u m enema and vice versa, enema checked by " p r o g r e s s " meal. L a r g e incision and careful examination of t e r m i n a l small bowel are indicated in all laparotomies for chronic abdominal symptoms, especially in cases operated upon for so-called "chronic appendicitis." " T u m o r s " of the lower small bowel are not necess a r i l y cancer or "carcinoids." Spontaneous fistulae between abdominal or pelvic viscera may not be malignant. Extensive resection, into the normal bowel, is usually successful.

inches of terminal ileum. Followed by fairly satisfactory condition to June 4, 1934. June 4, 1934. Third hospitalization. Tender mass baseball size in right iliac fossa. Nausea and vomiting, diarrhea, loss of weight. June 6, 1934. F i r s t stage of Mikulicz's operation. Recurrence of tumor at site of anastomosis in terminal ileum. June 13, 1934. Second stage of Mikulicz's resection. July 24, 1934. Discharged from hospital in good condition, but with small intestinal fistula. December, 1934. Present condition satisfactory, but has a persistent fistula. July, 1935. Closure of fistula. Case 3. Mrs. N. B., Female, married, age 29 years. 4 children, 2 miscarriages. History: This case has been unclassified in my files since 1912, until the subject of "terminal ileitis" now serves as an excuse for including it with the two previous cases. 1905. F i r s t hospitalization (elsewhere). Appendectomy in 1905 for "chronic appendicitis." Five or six attacks in two years before operation. Was symptom-free for one year after operation. For the next six years attacks of intestinal obstruction three to five times a year, accompanied by pain and swelling in right iliac fossa. June, 1912. Second hospitalization. Operation: Terminal ileum thickened and enclosed in fibrous sac. This parchment-like membrane was removed and ileum straightened out and omentum placed over bowel. September, 1912. Third hospitalization. Symptoms of chronic obstruction. Operation: Ileo-sigmoidostomy. October 15, 1912. Fourth hospitalization. Acute intestinal obstruction. Operation: Enterostomy. June, 1913. Fifth hospitalization. Lower ileum markedly improved. Localized jejunitis, the most striking finding at operation. Operation: Gastro-enterostomy, excluding the thickened convoluted jejunum. Improved to November 13, 1913. November, 1913. Sixth hospitalization. Operation: Resection of terminal ileum (serosa red, congested and rough, wall thickened, mucosa not ulcerated) and resection of ascending and half of transverse colon. The previous gastro-enterostomy and ileosigmoidostomy were functioning normally. Up to 1925, Mrs. B. was quite comfortable with occasional digestive upsets after which she was lost track of.

REFERENCES Crohn. Gingsburg and Oppenheim: Regional Ileitis. J. A. M. A., Oct. 15, 1932. Harris, Bell and Brunn." Continue the discussion and report cases under the heading, Chronic Cicatrizing Enteritis. S. G. O., p. 937, Nov., 1933. Stout, Frantz, Haagensen and Smith: Regional Ileitis. (Presbyterian Hospital Reports, N. Y., 1934). Brown, P. W., and Bargen, J. A.: Chronic Inflammatory Lesions of the Small Intestine, Regional Ileitis. Amer. Jour. Digest. Dis. and Nutrit., Sept., 1934. DeCourcy, J. L . : Terminal Ileitis Simulating Acute Appendicitis. A Case Report. Jour. Med. Cincinnati, Vol. XV, p. 216, 1934. Anschutz, G.: Inflammatory Tumors of L a r g e Intestine of Non-specific Nature. Deut Leib. fur Chit.. Berlin., J u n e 20, 1934. Itomans and H a s s : Regional Ileitis. N e w Eng. Med. Jour., Dec. 28, 1933. Kantor, J. L.: Regional, Terminal Ileitis, the Roentgen Diagnosis. J. A. M. A., p. 2016, Dec. 29, 1934. Probstein a ~ l Gruenfeld: Acute Regional Ileitis with Bibliography. Annals of Surgery, p. 273, Vol. 103, No. 2.

Gastro-lntestinal Bleeding in Disease oF the Liver and Biliary Tract By S. S. LICHTMAN, M.D. N E W YORK, N E W YORK

T E M E S I S or melena, occurring in the course H EofM Aliver and b i l i a r y t r a c t disease, is usually associated with a r u p t u r e d oesophageal v a r i x or the bleeding tendency of jaundiced subjects. Less commonly h e m o r r h a g e s from the g a s t r o - i n t e s t i n a l t r a c t occur on a different basis and these instances deserve consideration. A classification of the types of g a s t r o - i n t e s t i n a l bleeding associated with liver and b i l i a r y t r a c t disease has t h e r e f o r e been attempted. Bleeding from a varix, peptic ulcer or m a l i g n a n t ~ F r o m the Medical Division, Service of Dr. George Baehr, of The Mount Sinai Hospital, New York, N. Y. Submitted February 6, 1936.

disease of the b i l i a r y t r a c t has been ruled out in the cases here reported. Our cases emphasize the fact t h a t occult and gross bleeding in the i n t e s t i n a l t r a c t may occur in acute gall bladder disease w i t h o u t j a u n dice and in damage of the liver p a r e n c h y m a without a general bleeding tendency. CASE R E P O R T S Case No. 1. R . B . Subacute Yellow Atrophy of the Liver. Melena. A female, aged 57 years, with a history of recurrent myalgias and arthralgias, recently developed severe pain. in the left sacro-iliac region radiating to the left thigh

440

AMERICAN JOURNAL OF DIGESTIVE DISEASES AND NUTRITION

and leg. F o r four weeks she received intravenous injections of sodium iodide and intramuscular injections of Aolan, a milk preparation. She also received coal t a r analgesics and one five grain tablet of cinchophen. One week before admission, seven weeks after receiving the therapy, she noted icterus, dark urine, anorexia, epigastric distress, and pale yellow stools. Physical examination revealed an intense icterus. The liver was enlarged two fingers breadth below the costal margin. The spleen was not palpable. There was marked tenderness over both sacro-iliac regions. Bile was present in the stool and urine. X-ray examinations of the gall bladder and gastro-intestinat tract proved negative. Leucocytosis of 17 and 22,000 cells per cubid millimeter of blood was present despite clinical improvement. Tyrosine was demonstrated in the urine by the author's tyrosinase method. Biliary drainage indicated the presence of f r a n k blood with some clots without trauma. Chemical examination of the blood showed a prompt positive Van den Bergh reaction and bilirubinemia of 23 rag. per 100 c.c. The icterus index was 200. The bilirubinemia dropped to 16 and finally 14 mg., the icterus index to 140. The total cholesterol was 210, the ester fraction being 125 rag. per 100 c.c. The total, eventually dropped to 125, the ester fraction to traces. Despite intravenous and oral administration of large amounts of carbohydrate the patient lapsed, into coma. Hyperglycemia (350 rag.) and glycosuria (5 per cent) developed and were controlled by insulin therapy. The size of the liver decreased terminally. Edema and ascites appeared. She vomited coffee ground material and died in coma one month after admission to the hospital. Necropsy showed the presence of a subacute yellow atrophy of the liver, subacute pancreatitis, ascites, hydrothorax and superficial erosions of the stomach. Histopathological examination of the liver revealed widespread destruction of liver tissue with a large amount of necrotic g r a n u l a r debris, pyknotic nuclei, and large areas of hemorrhage. Islands of liver cells adjoining the necrotic areas showed similar but less pronounced changes. These islands of relatively less severely injured cells were enveloped by bands of stroma containing plasma cells, round cells and polymorphonuclear cells, and by areas of hemorrhage and by portal spaces. The presence of f r a n k blood in the biliary drainage fluid prompted the diagnosis of an ulcerating malignant lesion of the papilla of Vater. The histological demonstration of pools of blood in the necrotic liver parenchyma established the possibility of this local source of bleeding in this type of case as severe parenchymal liver damage. Case No. 2. A. G. Chronic Cholecystitis; Cholelithiasis; Choledocholithiasis; Melena; Hematemesis; Cholecystectomy; Choledochostomy. A female, aged 40, a huusewife, entered the hospital for relief from attacks of pain in the right hypochondrium, 20 years in duration. 20 years ago, an attack had been followed by t r a n s i e n t jaundice. In the preceding 6 months, there had been increasingly severe attacks of pain in the right hypochondrium radiating to the back and shoulders and accompanied by vomiting. Two months before admission, icterus, dark urine and pruritus developed. 24 hours before admission she vomited a large amount of black fluid and had t a r r y stools thereafter. Examination showed the presence of icterus and tenderness in the right upper quadrant of the abdomen. The rectal temperature was 100.4 degrees Fahrenheit. Chemical study of the blood indicated an icterus index of 22, a bilirubinemia of 0.9 rag. per 100 c.c.. The Van den Bergh reaction was delayed, direct, positive. The total cholesterol was 190 and the ester fraction decreased to 36 mg. per 100 c.c. The intestinal bleeding prompted several diagnoses: ulcerating malignant disease of the papilla of Vater, ulcerating choledocholithiasis, and bleeding peptic ulcer. A plate of t h e abdomen revealed gall stones by X-ray. The

Graham test failed to visualize the gall bladder. X-ray examination of the intestinal tract disclosed no abnormality. Many gall stones were found at laparotomy. The bladder was removed and a single stone, the size of an olive pit was removed from the common bile duct together with much detritus. The papilla and the ducts were freely patent. A source for the bleeding was not uncovered. The stomach and duodenum were normal to palpation. The common duct was drained for nine days. The bleeding did not recur. The patient was observed for one year following operation. Peptic ulcer was ruled out by X-ray and laparotomy. There was no hemorrhage in the gall bladder or bile ducts. There was no evidence of bleeding tendency before or after operation. There was no violent retching or vomiting. The vomiting was probably caused by the accumulation of blood in the stomach. The cause of bleeding in this case must be recorded as undetermined. Case No. 3. Cholelithiasis, Chronic Cholecystitis, Acute Hemorrhagic Cholecystitis, Hematemesis. S . S . A female, aged 64, a housewife, suffered from arthritis for 15 years. For a year prior to admission to the hospital the patient suffered attacks of pain in the right hypochondrium, radiating to the back and right shoulders accompanied by nausea, vomiting and chills, usually at midnight. The attack preceding the present illness was the severest. There was transient jaundice and acholic stools. Several weeks prior to admission the patient had a biliary colic associated with the vomiting of a small amount of dark blood. The physical examination disclosed nothing significant. There was no icterus. Laboratory findings showed an icterus index of 10, bilirubinemia 0.3 mg. per 100 c.c. The Van den Bergh reaction was faintly delayed positive. The cholesterol content of the blood was 220 mg. per 100 c.c., the ester fraction 75 rag. The leucocytes were counted 11,100 per cubic millimeter. An edematous hemorrhagic gall bladder containing one large walnut shaped stone was removed at operation. The patient ran a febrile course, 100 to 101 degrees Fahrenheit for 17 days following operation. She was finally discharged, relieved of her symptoms. The gall bladder was assumed to be the source of bleeding. The history of hematemesis several weeks prior, indicated that the hemorrhagic process was probably dated then. The edematous state of the gall bladder interfered with the patency of the cystic duct preventing continued passage of blood into the bowel. Case No. 4. Chronic Cholecystitis, Cholelithiasis, Hematemesis, Acute Pancreatitis, Transient Jaundice, Adiposity. I . M . A female, aged 46, suffered post-prandial distention, belching and occasionally vomiting directly after meals. About eleven days before admission to the hospital she began to complain of vague generalized aches and three days later had a sudden onset of severe cramp-like periumbilical pains. She vomited some coffee ground material. The symptoms lasted for several hours. Vomiting recurred. Soreness about the navel persisted. The morning of admission she had a recurrence of pain and again vomited coffee ground material. The physical examination indicated an extreme degree of adiposity. There was some tenderness in the periumbilical region. The icterus was slight. The blood indicated showed only a latent jaundice. The urea content of the blood was normal, the glucose elevated to 150 rag. per 100 c.c. The cholesterol was elevated to 300 rag. per 100 c.c. and the ester fraction was markedly depressed to 45 rag. I n the next six days the total and ester fraction returned to normal. The blood amylase was definitely elevated to 16.4 units on admission with r e t u r n to normal five and seven days after admission. Acetone was present in the urine, and disappeared on administration of fluids. The Graham test failed to visualize the gall bladder on

LICHTMAN--GASTRO-INTESTINAL BLEEDING IN DISEASE OF THE LIVER AND BILIARY TRACT

three repeated attempts. The barium meal indicated no abnormal findings. The patient ran a febrile course despite the cessation of vomiting and abdominal pain. This was attributed to a subsiding pancreatitis. Owing to the extreme adiposity, the favorable course of events and the reluctance of the patient to undergo immediate exploration, surgical intervention was deferred until considerable weight reduction was achieved. For six months symptoms did not recur. Thereafter the patient escaped observation. The cause of the hematemesis was not demonstrated. Case No. 5. Acute Gangrenous Cholecystitis with Sloughing, Chronic Cholecystitis, Cholelithiasis, Pericholecystic Abscess, Thrombosis and Erosion of the Cystic Artery, Hematemesis, Post-Operative Bleeding. T.N. A female, aged 45 years, had suffered two attacks of severe pressing epigastric pain radiating to the right hypochondrium 15 years previously with nausea and vomiting and once accompanied by dark urine and jaundice. She was in good health up to two days before admission to the hospital when she developed excruciating epigastric and hypochondriac pain radiating to the back and right shoulder, accompanied by nausea and vomiting and a constant dull ache in the right hypochondrium. The vomitus contained a large amount of blood, the urine was dark, the stools tarry. Icterus and pruritus were marked on admission. A somewhat tender mass was palpable in the right hypochondriac region. Due to the fact that the patient was a poor operative risk the operation was necessarily delayed for 18 days despite fever and chills. Then a huge abscess cavity containing gall stones was found in the region of the gall bladder. Aspiration of pus was followed by brisk bleeding. Transfusion was promptly performed. Hemorrhages again occurred on the 9th and llth days after operation. Again transfusions were performed. Death occurred on the 12th day with hyperpyrexia. Following the operation the biIirubinemia which had been 3 mg. per 100 c.c. dropped to 0.5 rag. Necropsy disclosed an acute gangrenous cholecystitis with sloughing, and fat necrosis of the pancreas. The cystic artery was thrombosed and eroded. An acute serosanguinous peritonitis was also present. The hematemesis early in the history probably originated from the gangrenous gall bladder or eroded cystic artery which subsequently thrombosed. The post-operative bleeding occurred from the same eroded artery. Case No. 6. Chronic and Acute Cholecystitis, Cholecysto-Duodenal Fistula, Gall Stone Ileus, Melena. C . L . A female, aged 47 years, experienced digestive disturbances, gaseous eructations, epigastric distress, abdominal distention, and occasional vomiting over a period of seven years. Seven years before, she had an episode of sharp pain in the right hypochondrium radiating to her shoulders and back with nausea, vomiting, fever and definite jaundice. Tarry stools occurred with certainty at this time. Pain and vomiting lasted for a fortnight, jaundice for a month. During the seven months preceding admission she had frequently noted tarry stools and had lost both weight and strength. The examination exhibited diffuse abdominal distention, and fecal vomitus. An abdominal mass was not palpable. Enterotomy was performed and a gall stone removed. The distention and vomiting were promptly relieved. However, restlessness and hyperpyrexia set in wtih fatal outcome.

Necropsy demonstrated the status after ileostomy for impacted gall stone ileus. A cholecysto-duodenal fistula was present. A small shrunken gall bladder with ulcerated interior and containing a blood clot was found. The liver was ulcerated and perforated in the region of the gall bladder bed. The early history of an acute episode with tarry stools seven years previously fixes the probable date of establish-

441

ment of the choIecysto-duodenal fistula. The recent bleeding arose from the ulcerated areas in the gall bladder and liver secondary to the recent acute inflammatory flare-up. CLASSIFICATION Naunyn (1892) classified the types of bleeding associated with gall stones: 1. Cholemic, with hemorrhage into the gastro-intestinal tract. 2. Portal vein thrombosis complicating cholelithiasis with bleeding into the bowel. 3. Perforation of a gall stone into the lumen of the bowel or into the liver with gastro-intestinal bleeding. 4. Hemorrhage from the bile passages into the bowel. Budinger (1925) developed a more comprehensive classification: I.

II.

Hemorrhages, Non-Dyscrasic in Nature. A. Spontaneous Bleeding from Bile Passages and Adjoining Vessels. 1. Spontaneous massive bleeding from known blood vessel. 2. Spontaneous massive bleeding from unknown blood vessel. 3. Spontaneous bleeding from ulcerated areas and superficial lesions. B. Hemorrhage from Neighboring Structures (including liver) with Drainage of Blood through Bile Passages. 1. From aneurysms. 2. From the liver. C. Operative Hemorrhages. 1. F r o m known blood vessel. 2. F r o m unknown blood vessel. 3. F r o m liver bed or operative field. Cholemic-Dyscrasic Bleeding. Multiple hemorrhages in various organs including gastro-intestinal tract.

The Author has modified this classification: 1. Hemorrhages of Local Origin. A. Portal in Origin. 1. Secondary to liver disease, i.e. portal hypertension. Ruptured varix. 2. Secondary to portal vein disease, i.e. thrombosis, cavernomatous transformation. B. Hepatic in Origin. 1. Hepatic congestion and inflammation, i.e. acute and subacute liver atrophy. 2. Traumatic. C. Origin in Biliary Tract. 1. Vascular. Rupture of aneurysm. Erosion of blood vessel by gall stone decubitus ulcer, or inflammation and ulceration of gall bladder wall. Post-operative bleeding vessel. Perforation of gall bladder into bowel, eroding vessel. 2. Diffuse inflammatory disease of gall bladder with hemorrhage into lumen and no demonstrable bleeding vessel. Infarction and hematoma of gall bladder wall. Hemorrhagic cholecystitis. 3. Neoplastic disease of passages with ulceration.

442

AMERICAN JOURNAL OF DIGESTIVE DISEASES AND NUTRITION D.

II.

Gastro-intestinal in Origin. 1. Ulcer complicating obstructive jaundice. 2. Jaundice complicating peptic ulcer (Catarrhal Jaundice). Cicatricial contraction of the papilla of Vater. Hemorrhages of Systemic Origin. Associated with hepatic insufficiency and hemorrhagic tendency. Gastro-intestinal erosions, toxic in origin.

C I T A T I O N S FROM THE L I T E R A T U R E Hemorrhage from the Gall Bladder. A. Bleeding Source Demonstrable. Schmidt (1893) described a patient with four massive hemorrhages in the course of five weeks, one accompanied by icterus. Despite gastric symptoms the presence of gall stones was suspected. Necropsy demonstrated a gall bladder with three areas of perforation into the duodenum each obturated by a gall stone. These were not bleeding. A false aneurysmal thrombotic mass was found between the neck of the gall bladder and the hepatic duct with pressure on the hepatic artery. The cystic duct was intact. Blood passed freely from the cystic into common hepatic duct and then into the bowel. It was difficult to determine whether a stone had ulcerated through the common duct and the artery with secondary aneurysmal formation or whether a marked cholangitis had weakened the vessel wall. Chiari's case, similarly, bled from the cystic artery. The gall bladder was involved in an inflammatory process. Schnyder (1915) reported a case of fatal bleeding into the peritoneal cavity from a perforated gall bladder. Ulcerations had produced bleeding into the gall bladder and thrombosis of a blood vessel in its wall. Some blood had escaped into the common duct but recent bleeding into this duct had been prevented by blood clot formation and occlusion of the cystic duct. Ulceration had extended through the gall bladder wall, eroded a vessel and caused a fatal hemorrhage. No gall stones were present. Lobstein noted a case with recovery following perforation of the gall bladder with bleeding into the gall bladder and the peritoneal cavity. The bladder and stones were removed. Gjellerup, Huguenin and Leaved have recorded similar instances. Esau (1925) described a patient with abdominal colic followed by occult blood in the stool for 14 days. Local tenderness in the gall bladder region persisted. Peptic ulcer was diagnosed because of the bleeding, although the possibility of diseased gall bladder was also entertained. There was no icterus. No X-ray examination was made. The stomach and duodenum were found to be normal at the operation, 17 days after blood was first observed in the stool. A thickened gall bladder contained several stones. One was found in the cystic duct. The common bile duct was free. The entire inner aspect of the gall bladder wall was infarcted. The patient made a complete recovery. Heusser's case (1925) recovered following cholecystectomy. Cystic duct occlusion prevented bleeding into the bowel. The gall bladder was filled with blood clots, its wall ulcerated and infarcted. The mucosa was completely denuded. Hemorrhage occurred by seepage and not by vessel erosion. Carnioley (1927) recorded a case of hemorrhagic cholecystitis with cystic duct occlusion and without intestinal bleeding. I.

Junghanns (1930) recorded a necropsy finding of decubitus gall stone ulcer eroding a vessel in the neck of the gall bladder. The cystic duct was patent. Death was caused by hemorrhage. Kehr has also referred to erosion of the portal vein and hepatic artery by common duct stone with profuse bleeding into the common duct. Naunyn recalled two cases with hematemesis caused by pyelthrombophlebitis secondary to pressure of a stone in the cystic duct. Reports of gastro-intestinal bleeding in the course of the formation of cholecystoduodenal fistulae by passage of gall stones into the bowel are too common to enumerate in this brief account. B. Bleeding Source Not Demonstrable. Vysin (1923) reported a patient with attacks of colic and blood in the stools. Peptic ulcer was diagnosed, Ulcer was not demonstrable at laparotomy. The gall bladder was chronically thickened and shrunke~ and contained no stones. A bleeding source could not be found. Healing of an ulcer in the lapse of six weeks before operation was considered possible. Bleeding associated with gall bladder disease was apparently unfamiliar to the author. Budinger (1925) recorded four cases. The first patient bled severely from a decubitus ulcer of the common duct. There was recent icterus. A large amount of blood was lost through the stomach and bowel. The gall bladder was full of pus, stones, and blood. The common bile duct was ulcerated by a stone, dilated and filled with pus and blood. Following cholecystectomy the symptoms vanished. The second case was diagnosed as gastric ulcer by X-ray. Symptoms suggested ulcer but accompanied by slight icterus existed for two months. The patient had vomited a large amount of blood before admission to the hospital. Following cholecystectomy, periodic bleeding took place from the wound and the bowel. The gall bladder wall was thickened, infected and its mucosa ulcerated. A stone was found in the cystic duct. Blood was present in the lumen of the gall bladder. A recent abscess was found. The gall bladder was adherent to the abdominal wall and the jejunum. No ulcers could be found in the intestinal tract. Fatal hemorrhage occurred from the hepatic artery. The source of early bleeding in this case was not demonstrated, unless it be assumed that the cystic duct occlusion occurred afterwards. The third case was also suspected of suffering from duodenal ulcer for one year. Attacks of acute pain occurred. The stools were often tarry. Lately the symptoms of gall bladder inflammation came to the foreground. Fever, t a r r y stools and jaundice developed. The X-ray indicated the presence of periduodenal adhesions. An abscess was found under the liver border. The gall bladder was filled with pus and blood, its mucosa gangrenous. The cystic duct though involved, was patent. Stones were not present. The main ducts were not involved and contained clean bile. Recovery followed cholecystectomy. In a fourth case, with gall stones and two bouts of jaundice, the patient again developed icterus. A thickened distended gall bladder containing about 100 tiny stones and blood was found at operation. A small stone occluded the cystic duct. A bleeding source was not demonstrable. Blood was not demonstrable in the stool a f t e r operation. I f the cystic duct occlusion is relieved, blood in the gall bladder may reach the intestinal tract sub-

LICHTMAN--GASTRO-INTESTINALBLEEDING IN DISEASE OF THE LIVER AND BILIARY TRACT sequently. This event appears then not to be related to any immediate clinical episode. Frick and Irland (1923) noted a case with profuse hematemesis and t a r r y stools at the ages of 18, 20 and 36 and slight jaundice several times. At operation a distended gall bladder containing one large mulberry stone was removed. A bleeding source was not demonstrable. X-ray examination of the bowel had been negative. The stomach and duodenum was also explored at laparotomy. DeCourcy (1928) described three cases in this category. They suffered from gall stones. They suffered severe hematemesis one to five months before laparotomy. One had three attacks of severe hematemesis in the two years prior to operation. In two cases the stomach was incised and searched for ulcer. The gall bladders were removed in the three cases. Bleeding never recurred thereafter. DeCourcy attributed the bleeding to marked hepatitis. Polichetti (1930) described a patient with a history of a profuse hematemesis five months previously and continuous indigestion despite ulcer therapy. A bladder containing 82 gall stones was removed. The stomach and duodenum were normal. Bleeding did not recur. Chronic intermittent intestinal bleeding was cured in a second patient with identical findings in the gall bladder. The stomach and bowel also showed no source of bleeding. Bleeding did not recur following operation. White and Jankelson (1931) recounted six cases. The first patient had suffered three intestinal hemorrhages with gross blood in the stools in three years. Barium meals on three occasions revealed a slight irregularity on the lesser curvature side of the duodenum. This was interpreted as due to active or healed ulcer, adhesions, or gall bladder pressure. A thorough exploration was made of the interior of the stomach. No adhesions were present. The small intestine and colon were normal. A slightly inflamed gall bladder containing several stones was removed. There was no recurrence of bleeding for five years after operation, the period of observation. A second case with symptoms typical of gall stones developed hematemesis and melena. Barium meal two months later, indicated no ulcer. Gall stones were demonstrable by X-ray. Operation was not performed. A third patient gave a history of recurrent slight jaundice over a period of ten years. The present attack of colic lasted 24 hours. The patient then vomited blood and coffee ground material. Jaundice set in and the pain subsided. A pathological gall bladder without stones was removed. Jaundice recurred after freedom from symptoms for three years. A barium meal revealed no abnormality. A fourth patient developed colic and hematemesis following a cholecystostomy two years previously. A stone was now found in the cystic duct. No lesions were demonstrable in the stomach or. duodenum. Bleeding did not recur following cholecystectomy over an observed period of five and one-half years. A fifth patient with a long standing history of gall stones developed a profuse hematemesis, t a r r y stools, and mild icterus. The patient died cholemic. There was no necropsy. A sixth subject with long standing symptoms of gall stones also developed hematemesis. A week later barium meal and X-ray of the gall bladder proved negative. Two stones were found in the gall bladder

which was removed. negative findings.

443

The ducts were explored with

Bleeding from Known Blood Vessel Without Gall Bladder Disease. A.

Aneurysm of the hepatic artery. Aneurysm of the hepatic artery has been noted by Stokes, Lebert, Quincke, Borchers, Chiari. Schmidt (1893) reported a case associated with gall stones. Trauma or infection were incriminated. Kading (1919) reviewed the matter, referred to a case, and presented a case report. Hematemesis occurred in both cases. Duodenal ulcer was suspected but not found at operation in one case. It was considered a late result of some previous trauma. In the second case, the liver was injured by a wound and hemothorax was present. In both instances the aneurysm had ruptured into the bile passages. Kading reviewed four intrahepatic and 42 extrahepatic aneurysms of the hepatic artery. Perforation occurred into the gall bladder three times, into the cystic duct three times and into the common bile duct once. Friedman (1921) reported a case from this hospital, with jaundice, and melena seven years following cholecystectomy. Carcinoma of the hepatic ducts was suspected at laparotomy. At necropsy the source of bleeding was found to be an aneurysm of the hepatic artery which had ruptured into the biliary passages. B. Erosion of the Cystic Artery. Jaff~e reported a fatal hemorrhage from cystic artery erosion by a peptic ulcer adherent to the liver and the lower edge of the liver. Bleeding following operations for gangrenous gall bladder are often caused by erosion of the cystic artery.

Bleeding from the Liver Parenchyma Into Bile Passages. Quincke has referred to hemorrhage from the liver secondary to infarction and embolization. Rolleston described a case of acute yellow atrophy with intestinal bleeding the predominant symptom from the outset. Strauss (1929) reported a patient with traumatic subphrenic hematoma bleeding into the bile passages as well as intraperitoneally. Roessle demonstrated a fistulous tract from the hematoma into the bile passages postmortem.

Chronic Portal Vein Disease. Fleischhauer (1932) and Scheidegger (1933) have recently reviewed the literature on cavernomatous transformation of the portal vein. Klempeper (1928) has reported a case in the American literature.

Occult Bleeding. Eusterman (1913) reported occult blood in the stools of 20 per cent of patients with chronic gall bladder disease and in the gastric analyses of 43 per cent. Kehr maintained that occult blood occurred more often in the stools of patients with cholecystoduodenal fistula. Peiper (1921) found occult blood in the stool of 2 of 10 cases of cholecystitis. Wohlgemuth (1928) reported a patient with gall stones and hematomas of the gall bladder wall and occult blood in the stool. The blood disappeared following removal of the gall bladder. Statistics on the incidence of occult blood in the stool of patients with gall bladder disease are not complete since most cases are admitted to surgical wards

444

/ k M E R I C A N J O U R N A L OF D I G E S T I V E D I S E A S E S A N D N U T R I T I O N

directly and are promptly operated upon without adequate routine study. ETIOLOGICAL CONSIDERATIONS Gastro-intestinal hemorrhages which can be traced to a ruptured aneurysm of the hepatic or cystic a r t e r y or varix, or a vessel eroded by a decubitus ulcer or an ulcerating lesion in the gall bladder require no comment. The same is true of bleeding based on a hemorrhagic tendency due to jaundice, or malignant disease of the biliary tract. Several hypotheses have been advanced to explain the occurrence of bleeding in gall bladder disease where a bleeding point is not demonstrable. I t has been claimed that venous thrombosis m a y cause hemorrhage, t h a t gastric ulcers m a y be embolic in origin, t h a t toxins in the blood f r o m chronic infection in the gall bladder m a y cause bleeding f r o m the stomach and upper bowel as it does in essential h e m a t u r i a or epistaxis, t h a t hepatitis m a y be the underlying factor. The failure of recurrence of bleeding following removal of the gall bladder points to the gall bladder itself. The effects of vomiting and retching cannot be ruled out. However, cases have occurred with no retching. Violent vomiting in other conditions is not attended by massive hemorrhages f r o m the bowel. In some cases the vomiting is probably induced by the presence of blood in the stomach: Bleeding into the.gall bladder may occur by seepage in a diffuse phlegmonous or hemorrhagic process. Infarction of the entire wall may occur. Bleeding into the bowel occurs if the cystic dust is patent. Bleeding attending the formation of a eholecystoduodenal fistula may be a minor incident in the history, m a y even be entirely unnoticed. The symptomatology if cavernomatous t r a n s f o r m a tion of the portal vein m a y simulate cirrhosis of the liver or peptic ulcer. Jaundice may supervene. Jaundice m a y complicate duodenal ulcer. The causes may be intereurrent catarrhal jaundice or mechanical cicatricial closure of the papilla of Vater. Obstructive jaundice m a y become associated with ulceration of the upper bowel. DIAGNOSTIC FEATURES In the presence of jaundice differentiation m u s t be made between bleeding due to a general bleeding tendency or a local cause. Without jaundice, the cause is usually a local one. I t is important to remember that gall stones may be accompanied by a massive hemorrhage. Though the bleeding source m a y not be demonstrable, removal of the stone-containing gall bladder

promptly relieves the bleeding, as a symptom. This also rules against the probability of a missed peptic ulcer. Parenchymal liver degeneration may be associated with f r a n k bleeding into the duodenum. Blood was obtained by transduodenal drainage in such a case. Malignant disease of the biliary passages may thus be simulated and laparotomy performed in a case of subacute yellow atrophy. Occult bleeding may occur in gall bladder and liver disease, though the presence of blood in the stool usually favors the diagnosis of peptic ulcer or malignant disease. The incidence of gross bleeding in one to five per cent of our patients with gall bladder disease, in a carefully observed material confirms the estimates of other authors. COMMENT The massive hemorrhages occurring in individuals suffering f r o m gall stones, without duodenal ulcer demonstrable by X-ray or laparotomy are remarkable. In some instances a history of previous hematemesis or melena may be obtained yet no cholecystoduodenal fistula can be demonstrated. L a p a r o t o m y performed soon a f t e r a hemorrhage m a y reveal no ulceration due to passage of a stone. I t is conceivable t h a t a superficial gastric mucosal erosion may escape the most careful search. Vomiting or a toxic factor may be responsible. Appendicitis or peritonitis has been known to be complicated by massive hemorrhage on the basis of gastric erosion, toxic in origin. SUMMARY Local types of gastro-intestinal bleeding associated with disease of the liver and biliary t r a c t have been distinguished f r o m the systemic. Hematemesis and melena may occur in conditions other than peptic ulcer, ruptured oesophageal varix or ulcerating malignant disease of the biliary tract. F r a n k hemorrhage into the intestinal t r a c t m a y occur in degenerative lesions of the liver parenchyma (liver atrophy), in aneurysm of the hepatic artery, and in hemorrhagic disease of the gall bladder if the cystic duct remains patent. Bleeding into the upper bowel may occur in gall stones without jaundice and without a demonstrable bleeding point. This bleeding does not recur following cholecystectomy. Occult bleeding may occur in disease of the liver parenchyma as well as in non-malignant disease of the biliary tract.

REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9.

N a u n y n : Kllnik der Cholelithiasls, Leipzig, 1892. Budlnger, K.: Ueber Blutungen nach Gallenoperationen und bei E r k r a n k u n g e n der Gallenwege. Arch. f. klin. Chlr., 137:199, 1925. Sehmidt, M. B.: Toedliehe Blutung aus einem A n e u r y s m a der Leberarterie bei Gallensteinen. Dtsch. Arch. F. klin. Med., 52:536, 1893-94. Schnyder, K. : Toedliehe Gallenblasenblutung in die freie Bauchhoehle. Centralbl. f. aug. Path. u. path. Anat., 26:361, 1915. Heusser, H. : Die Blutende Gallenblase. Mueneh. meal. Wchnechr., 72:2007, 1925. Gjellerup, O. : Ein Fall yon R u p t u r der Gallenblase mit maechtiger intraperitonealer Blutung. Hoapitalatidende, 64:826, 1921. Huguenin : Ueber einen Fall yon Gallenblasenruptur m i t toedIicher Blutung infolge eines Carcinoma haematodes an der vereinigung der drei Hauptgallengaenge. Vlrch. Arch., 173:552, 1903. Junghanns, H.: Blutendes Ulcus in der Gallenblase. Dtsche. Ztschr. f. Chit., 224:93, 1930. Vysin : Blutige Stuehle bei Gallensteinen. Ca~opic lekaruv ceskych., 62:30, 1923.

10.

Frick, W. J., and IrIand, R. D. : Gastric and Intestinal Bleeding in Appendix and Gall Bladder Disease. Surg. Clinics N o r t h A m e r i ~ , 3:1553, 1923. 11. DeCourey, J. L . : Gastric H e m o r r h a g e Associated with Cholellthiasis. Cincinnati J. Med., 7:618, 1926-27. 12. Poliehetti, E. : Osservazioni di cinrlue casi di ematemesi e melene alte senza ulceri gastroduodenali. Annali italiani de Chirurgica, 9:245, 1930. 13. White, F. W., and Jankelson, I. R.: Gastro-intestinal H e m o r r h a g e in Disease of the Gall Bladder. N e w England J. Med., 205:793. 1931. 14. Kading, K. : Ein geheilter Fall yon ir~trahepatlschem Aneurysma mit besonderer Beruecksichtlgung der traumatisehen Leberarterienaneurysmen. Dtech. Ztsch. 5. Chit., 150:82, 1919. 15. Friedman, W . : A n e u r y s m of the Hepatic A r t e r y . Proc. N e w Y o r k Pathol. Soc., 21:177, 1921. 16. Jaffee, R. H . : Fatal H e m o r r h a g e from Eroded A r t e r i a Cystica of the Gall Bladder. Trans. Chic. Path. Sot., 11:394, 1919-23. 17. Rolleston, H. : Diseases of the Liver, Gall Bladder and Bile Ducts. Page 587, 2nd Ed. W. B. Saunders. 18. Strauss, A. : Ueber Verbluntung aus den Gallenwegen. Monttschrift, f. Unfallheilkunde u. Vers~cherungemedlzin, 36:438, 1929.

ABSTRACTS 19.

Fleischhauer, H. : U b e r den chronischen Pfortaderverschluss. I. D. Friedrich Wilhelms Univ., 1932. Scheidegger, S.: Ueber zwei seltene F o r m e n yon Blutungen aus Speiseroehre u n d Magen. F r a n k f u r t . Ztschr. f. Pathol., 44:527, 1933. 21. Klemperer, P . : Cavern~)rnatous T r a n s f o r m a t i o n of the P o r t a l Vein. Arch. P~th., 6:353, 1928. 22. E u s t e r m a n , G. B. : Incidence and Diagnostic Value of Blood or Hemorrhage in Gastric and Intestinal Lesions. Clinical a n d Statistical Study. St. Paul Mcd. J., 15:587, 1913.

20.

445

23.

Carnioley, C. : U n e Complication R a r e de la Lithiase Biliaire. Arch. des Mat. de l'App. Digest. et de la Nutrltlon, 17:294, 1927. Peiper : Cir. by Rudinger. Wohlgemuth, K. : Cholelithiasis m i t okkulten Blutungen. Arch. f. klln. Chlr., 151:438, 1928. 26. E s a u : Hamorrhagischer wandinfarkt der chrontsehen Steingallenblase m l t B l u t u n g e n in tier D a r m . Arch. f. klan. Chit., 35:782, 1925. 27. Aschoff, L. : cir. Esau. 28. Judd, E, S. : Relation of the Liver and Pancreas to Infection of the Gall Bladder. J . A. M. A., 77:197, 1921. 24. 25.

ABSTRACTS CLINICAL MEDICINE G O L D M A N , J O E L , AND C O H E N , A B R A H A M .

Abdominal Distention in Lobar Pneumonia. Ann. Int. Med., I X , 1222, March, 1936. I n discussing the origin of gases in the intestinal tract the authors make clear that they are interested in those gases which are not readily absorbed, such as hydrogen, nitrogen and methane, though admitting that other gases may accumulate in the intestinal tract iu a rate of formation in excess of their rate of absorption. In discussing the origin of intestinal gas, Cutting is quoted, in which he held that the origins were (1) composition of food stuffs, (2) diffusion of gas from the blood stream, (3) swallowed atmospheric air. He went on to discuss their production by stating that one source was from the action of bacteria on sugars in the lower portion of the small intestine; (2) the action of bacteria on cellulose residue remaining undigested; that free hydrogen occurred in the blood stream under pressure of 4/5 of an atmosphere, while oxygen and carbon dioxide occurred in chemical combinations only. The nitrogen is left free to leave the blood stream to replace other gases in the intestine when pressure relationships are favorable, whereas in order for oxygen or carbon dioxide to leave the blood stream there must not only pressure relationships be disturbed, but also must conditions be favorable for the liberation of these gases from their chemical combination. Alvarez is quoted as stating that little or no trouble was experienced from flatulence in a series of patients at the Rochester Clinic suffering from pneumonia when treated in a chamber containing 50 per cent of oxygen, because in such an atmosphere more of the nitrogen in the bowel could diffuse out. The consensus of opinion of authors consulted in the preparation of the paper appeared to w a r r a n t the conclusion that putrefactive and fermentative changes in the intestines were of minor importance as compared to disturbances in the diffusion of gases and that an additional embarrassment of the criculation resulted from chemical interference with the splancbnic circulation or perhaps also from

compression of the heart through the diaphragm; that in abdominal distention it was thought that gaseous distention depended chiefly upon alterations in the interchange of gases between the gastro-intestinal tract and the blood. Thus the normal equilibrium of these gases could be upset by disturbances in the toxicity and motility of the gastro-intestinal musculature or by interference with the local fixation. L1 pneumonia possibly the diminished capacity of the lung may thus be in part responsible for the distention commonly associated with that disease. A discussion of the drugs that have been recommended to be used to combat distention of this nature follows, with the conclusion that none of them was svtisfactory in all cases. Impressed by the report in 1933 of Paine, Carlson and Wangensteen, describing a form of continuous lavage of the duodenum by means of a nasal tube to relieve postoperative distention, nausea and vomitbig, the authors introduced a modification of this apparatus to be used in the management of abdominal distention in lobar pneumonia and other conditions producing distention. The modification of the continuous lavage of the duodenum consisted in its application to rectal suction siphonage. The only difference in the apparatus used by the authors and t h a t described by Paine and his associates was the substitution of a soft rubber rectal tube for the Levine duodenal tube. The technique of its use consisted in giving a cleansing soapsuds enema before the apparatus was-set in action and an attempt to keep the stools in a liquid state to prevent block of the tubes. This latter desideratum is obtained by the use of saturated magnesium sulphate solution. The use of the apparatus was continued as long as necessary to reduce the distention and discontinued so soon as the distention was relieved, and was repeated whenever necessary. The only untoward effect was tenesmus. The authors treated 100 cases and report control of distention in less than 24 hours in most of the cases, only 18 per cent requiring a longer period for control. Rectal irritation occurred in 13 per cent; in 34 per cent the distention was controlled notwithstanding a fatal

termination. It was concluded the procedure gave added rest to the patients, that the need of bed-pan disturbance was reduced, the n u r s i n g problem was simplified, that it could be carried out in the home as well as in the hospital, that incontinence was infrequent and that it is safe, inexpensive and effective. Virgil E. Simpson, Louisville. RATTNER, H.

Stomatitis Due to Sensitization to Dental Plates. J. A. M. A., Vol. 106, pp. 2230-2232, June 27, 1936. Stomatitis due to sensitization to material in dental plates is a condition not generally recognized by dentists or physicians. The author presents three cases who developed stomatitis after changing their dental plates or having their old ones repaired with new materials. The severest of the cases was characterized by b u r n i n g tongue, excessive salivation, soreness of the mucous membranes of the entire oral cavity, nausea, eructations and a thick discharge from the mouth occurring towards the end of the day. Examination revealed a markedly inflamed mucous membrane with serous exudation. The other cases were of less severity. In two of the cases there was a positive patch test where the plate was applied to the arm. The composition of the material from which dental plates are made is a trade secret, but enough is known about them to indicate that they contain substances which may be i r r i t a t i n g to susceptible cases. Francis D. Murphy, Milwaukee. I~[ENTZER, S T A N L E Y I-I.

Obstructive Cholecystitis. S. G. 0., Vol. 62, No. 5, May, 1936. The author discusses Obstructive Cholecystitis with particular reference to acute obstructive cholecystitis and its sequelae. He discusses the progressive steps in the pathology of acute obstructive chotecystitis to empyema which is the eventual outcome. In this connection he points out that the mortality in the treatment of acute cholecystitis will be decreased if the obstructive lesions are differentiated, clinically, from the non obstructive. Dr. Mentzer