Death from Hepatic Failure after Jejunoileal Anastomosis Frederick
P. Spin, MD,* Hanover, New Hampshire
Rodger E. Weismann,
MS, MD, Hanover, New Hampshire
Fatal hepatic failure as a compljcation of jejunocolic anastomosis is well documented in the literature [1-S]. To date, however, this complication after jejunoileal anastomosis has rarely been described [6]. We present two case reports of patients who underwent end to side jejunoileal anastomosis for massive obesity and subsequently died of hepatic failure. The clinical symptoms and liver function test results in these two patients are compared with those in 124 other patients who underwent similar surgical procedures but in whom serious liver disease has not been observed. The group consisted of 126 patients who underwent jejunoileal anastomosis for treatment of morbid obesity. In most patients, the ileal segment was 4 or 5 inches long, and the jejunal segment 14 or 15 inches. All patients had intraoperative liver biopsy, and most have had at least one percutaneous needle biopsy subsequently. Two patients have had the shunt taken down. One hundred patients have been followed up longer than one year, and the remainder longer than eight months. Fifty have been followed up longer than two years. Details of the series are reported elsewhere [ 71. Case Reports Case I. The patient (DS), a forty-one year old woman, was first seen because of lifelong obesity in January 1971; she weighed 350 pounds at the time. Later that month, she underwent dilatation and curettage, and a diagnosis of adenocarcinoma of the endometrium was made. She had radium implantation under nitrous oxide and oxygen anesthesia. On February 23,1972, she underwent total abdominal hysterectomy and jejunoileal anastomosis. The ileum, 4 inches proximal to the ileocecal valve, was anastomosed to the jejunum, 14 inches distal to the ligament of Treitz. Anesthesia was nitrous oxide and oxygen; 2 units of blood were given. Intraoperative liver biopsy showed From the Department of Surgery. the Hiichcock Clinic, Hanover, New Hampshire. * Present address and address for reprint requests: the Guthrie Clinic, Sayre. Pennsylvania 18840.
aa
moderate fatty change and slight increase in fibrous tissue. Postoperatively, she complained of nausea and vomiting. She was discharged on a regimen of chlorpromazine (Thorazine@), 150 mg, and promethazine (Phenergane), 25 mg, twice a day. Chlorpromazine was discontinued in May. In late June jaundice was noted, and she was rehospitalized. There was no history of alcoholism. On admission she weighed 260 pounds. The liver was palpable four fingerbreadths below the right costal margin and was firm and tender. The spleen was also enlarged. Australian antigen was negative. She received intravenous fluids and vitamin K. She continued to vomit and was lethargic and depressed. The jaundice increased and liver function test results became increasingly abnormal over the next three weeks. The transaminase levels were in the range of 200 to 306 units. The prothrombin time, initially corrected to fourteen seconds, increased to twenty-one. By July 21 hepatic coma, intestinal bleeding, and hypotension that did not respond to transfusion had developed. Blood pressure was maintained with pressor amines for three days, and she died on July 28, 1972. At autopsy, performed fifteen hours post mortem, the liver weighed 4,170 gm. Microscopically, there was massive necrosis with no remaining nuclear detail. There was marked fatty infiltration but no inflammatory infiltrate. The extrahepatic biliary system was patent and normal except for surgical absence of the gallbladder. The jejunoileal anastomosis was 8 inches from the ligament of Treitz and 2 inches proximal to the ileocecal valve. Case II. The patient (BS), a thirty-three year old housewife, underwent jejunoileal anastomosis on April 9, 1971, for massive obesity. She was 5 feet 2 inches tall and weighed 313 pounds. Anesthesia was thiopental sodium (PentothaF), nitrous oxide, and oxygen; 2 units of blood were given. A 14 inch segment of jejunum was anastomosed to the ileum, 4 inches from the ileocecal valve. Liver biopsy (Figure 1A) showed moderate fatty change. The postoperative course was unremarkable. On September 20, 1972, she was readmitted because of rectal bleeding. She had noted easy bruisability and bleeding from the gums. The sclerae were icteric. The liver had an overall vertical diameter of 15 cm. The spleen was not palpable and ascites was present. Several large hemorrhoids were bleeding profusely. White blood
The American Journal of Surgery
Hepatic
B
Failure after Jejunoileal
Bypass
C
Figure 1. Histologic sections of liver; case II. A, intraoperative liver biopsy showing moderate fatty infiltration (Masson trichrome stain; original magnification X 40, X 2.5). B, liver biopsy on final admission showing marked lobular disarray with complete loss of architecture and extensive necrosis. There is also moderate fatty change, some fibrosis, and round cell infiltrate (Masson trichrome stain; original magnification X 40, X 2.5). C, postmortem liver needle necropsy showing massive fibrosis with nests of regenerating cells (hematoxylin and eosin stain; original magnification X 40, X 2.5).
hepatic encephalopathy developed. The amino acids were deleted, and administration of prednisone was begun at 60 mg per day. She showed only transient improvement and died on November 2, 1972. At autopsy, the liver weighed 2,300 gm and was extremely firm. The extrahepatic biliary system was normal. The spleen weighed 500 gm. The jejunoileal anastomosis was as described at operation. Liver necropsy was performed twenty minutes post mortem. (Figure 1C.)
cell count was 1,300 per mm3, platelets 165,000 per mms, prothrombin time twenty-five seconds, and partial thromboplastin time greater than two minutes. Prothrombin time was corrected to twelve seconds with vitamin K and fresh frozen plasma administration, and liver biopsy was performed. (Figure 1B.) Australian antigen was negative. Parenteral hyperalimentation of amino acids, vitamins, electrolytes, and hypertonic glucose was begun, and oral cholestyramine (Questrang) was administered. Her course was complicated by sttiphylococcal sepsis, presumably from the intravenous catheters. After this, she became somnolent and progressive TABLE
I
Date
Laboratory
Weight (pounds)
Data
_ Serum Glutamic Oxalacetic Transaminase (units/ml)*
Prothrombin Time (set; control of
Albumin (gm/L plasma)
11 set) Case
l/20/72 3120172 418172 616172 7 128172
Laboratory data from patients in cases I and II are presented in Table I.
330 289 280 278 255
54 ... 300-t 372
313 214 162 160
... 126 82 220
...
Total/Direct Bilirubin (mg/lOOml)
Carotene (mg/lOO ml)t
Alkaline Phosphatase (units/ml)$
I
...
.,.
0.5/0.1
...
1.7 2.2
...
... ...
...
21
2.9
... 15
...
5 40
... 12.8
25.6/21.2
...
...
...
55 15 35 ...
5.3 12.4 9.1 8.9
...
7.211.9
6.5
Case II 4/20/71 12/6/71 g/20/72 10/13/72
...
...
17 28 15
... 2.5 3.1
0.5/0.3 5.512.5 31.612.7
*Normal: 0.5 to 45 units/ml. t Normal: 100 to 300 mg/lOO ml. $ Normal: 2 to 4.5 units/ml.
Volume130,July1975
89
Spin and Weismann
TABLE
II
Combined Laboratory Data from 126 Subjects Undergoing Jejunoileal Anastomosis
Carotene (mg/lOO ml) Preoperative Total tested Mean Minimum Maximum
112 129 25 247 < 25 (0) 25-50 (20)
Postoperative Total tested Mean Minimum Maximum
One month Total tested Mean Minimum Maximum
106 72 10 195 20 (0) 10 13.2 3.5 30.0
47 11 290
6 79 30 150
72 29 5 70 20 (2)
36 49 11 132 50-100 (16) >lOO (1)
24 53 186 50-100 (3) >lOO (4)
Comments
Clinical Features. In an attempt to delineate any significant features that might distinguish these two patients from the others, we compared the clinical symptoms. The patient in case I had severe anorexia and intractable vomiting. A majority of our patients have had various degrees of anorexia, at least transiently. This had been attributed to a metabolic ketoacidosis resulting from malabsorption and starvation and usually responded to increased carbohydrate intake. The degree of anorexia and vomiting in the patient in case I was more severe than that seen in most of our other patients. Conversely, the second patient manifested no loss of appetite and was able to maintain oral intake until well into the final hospi-
90
talization. Symptoms of fatigue, diarrhea, or weight loss did not distinguish these two patients from the majority. Laboratory Data. The cumulative serum carotene levels and liver function test results in the preoperative, immediate postoperative, and later follow-up periods are presented in Table II. Serum carotene measurements were made as a rough assessment of general nutritional status and fat absorption. Preoperative carotene measurements were made in one hundred twelve patients. Eighty-six of these were over 100 mg/lOO ml. Only two were under 50 mg/lOO ml. Postoperative measurements were made in one hundred and six patients. In thirty-two, values between 25 and 50 mg/lOO ml were recorded. Two had serum carotene levels of 20 and 10 mg/lOO ml. Additional measurements were made at one, six, and twelve months; nine, sixteen, and nine patients, respectively, had carotene levels of 15 mg/lOO ml or less. Thus, the extremely low carotene values in the patients in cases I and II do not seem to be predictive of incipient fatal hepatic disease. Values of more than 50 mg/lOO ml, however, were associated with insufficient weight loss. Alkaline phosphatase was measured preoperatively in one hundred and nine patients. Only thirty-six had values of less than 5 units/ml and ten had values of more than 10 units/ml. One year after operation, forty-nine patients were tested; nineteen had alkaline phosphatase levels of more than 10 units/ml. Two had values over 20 units/ml. Again, the mild degree of abnormality recorded in the two patients with fatal hepatic necrosis would not have separated them from the others. Serum glutamic oxalacetic transaminase was measured prior to jejunoileal anastomosis in eleven patients. In only one was an abnormal value found. However at follow-up study one year later, seven of twenty-four patients tested has serum glutamic oxalacetic transaminase values greater than 50 units/ml, two of which were over 180 units/ml. Prothrombin times were not measured routinely in our patients and therefore no conclusions can be drawn regarding the abnormal values in the two patients with hepatic failure. Fatty liver and mild fibrosis are common in massive obesity [8-121. Abnormal liver function test results have been reported by some and not found by others ]11,13-151. Starvation results in a reduction of steatosis, and focal necrosis may occur [12]. A recent study of twenty-three patients undergoing end to end bypass demonstrated a threefold increase in hepatic lipids during the pe-
The American Journal of Surgery
Hepatic
riod of rapid weight loss [16]. Drenic, Simmons, and Murphey [8] described massive fibrosis with virtual absence of hepatic tissue in a patient who died after jejunocolic anastomosis. Payne and DeWind [IO] found portal fibrosis in one patient after jejunocolic bypass. In the first patient reported on herein, the situation was complicated by a number of factors, including blood transfusion and chlorpromazine and promethazine for control of nausea. Postmortem examination showed the ileal segment to be shorter than that measured at operation. The autopsy specimen of the liver was obtained a number of hours after death, and postmortem autolysis had undoubtedly occurred. The absence of substantial cellular infiltrate in the liver makes fulminant viral hepatitis or an idiosyncratic reaction to phenothiazines unlikely. In the second patient, biopsy specimens clearly show a progression from steatosis with triaditis, to massive necrosis. and finallv to micronodular cirrhosis with hyaline degeneration and absence of significant inflammatory infiltrate not unlike that observed bv McGill et al 161. We have compared the clinical and laboratory data in these two patients with the findings in 124 others who underwent a similar nrocedure and found nothing distinctive in their courses prior to the onset of overt liver disease. In addition to carefully planned preoperative evaluation and meticulous long range follow-up study, repeated liver biopsy may prove to be the only accurate way of assessing liver function and morphologic features so that reanastomosis can be undertaken sufficiently early to prevent fatal hepatic necrosis. Because necrosis may occur at intervals that vary substantially after surgery, repeated biopsy should be timed to detect significant liver damage during this period. It may be necessary to evaluate the morphologic aspects of the liver every three to six months until the patient’s weight has stabilized. Other parameters that may prove to be of predictive value are the prothrombin time and, perhaps, a decrease in uptake of technetium sulfa colloid bv the liver. Additional data will have to be accumulated before this can be assessed finally [I 71. d
L
References
1. Maxwell JG, Richards RC, Alba D Jr: Fatty degeneration of 2. 3.
I
4. 5. 6.
7. 8.
9. 10. 11. 12.
13.
14.
15.
Summary 16. _I
Volume130,July 1975
Bypass
complication after the currently employed jejunoileal anastomosis has rarely been described before. We present two cases of hepatic necrosis and death after end to side jejunoileal anastomosis and compare the clinical data and liver function test results in these two patients with findings in 124 other patients who underwent a similar procedure for treatment of morbid obesity. No significant differences could be found that would allow early identification of those patients in whom massive hepatic necrosis was developing. Careful follow-up study and perhaps repeated liver biopsy may be necessary to monitor the hepatic status after jejunoileal anastomosis so that reanastomosis can be undertaken at a time when the damage may be reversible.
>
Severe liver dvsfunction and even death from kepatic failure after jejunocolic anastomosis has led to the virtual abandonment of that procedure for the surgical treatment of massive obesity. This
Failure after Jejunoileal
the liver after intestinal bypass for obesity. Am J Surg 116: 646, 1968. Wills CE Jr: Experiences with intestinal short circuiting procedures for obesity. J Med Assoc Ga 56: 365, 1967. Payne JH, DeWind LT, Commons RR: Metabolic observations in patients with jejunocolic shunts. Am J Surg 106: 273, 1963. Barrow J, Frame B, Bozalis JR: A shunt operation for obesity. Dis Co/on Rectum 11: 115, 1969. Bondar GF. Piseky W: Complications of small intestinal short-circuiting for obesity. Arch Surg 94: 707, 1967. McGill DB. Humohrevs SR. Baooenstoss AH, Dickson ER: Cirrhosis and death after.jejur%eal shunt. Gasfroeoferology 63: 872, 1972. Weismann RE: Surgical palliation of massive and severe obesity. Am J Surg 125: 437, 1973. Drenic EJ, Simmons F, Murphey JF: Effect on hepatic morphology of treatment of obesity by fasting, reducing diets and small-bowel bypass. Nfng/ J Med 282: 829, 1970. Meyerowitz BR: Surgical treatment of intractable morbid obesity. Sand J Gastroenterol7: 1, 1972. Payne HJ, DeWind LT: Surgical treatment of obesity. Am J Surg 118: 141. 1969. Westwater JO, Fainer D: Liver impairment in the obese. Gestroenterology 34: 686, 1958. Rozental P. Biava C, Spencer H, Zimmerman HJ: Liver morphology and function tests in obesity and during total starvation. Am J Dig Dis 12: 198, 1967. Juhl E, Christoffersen P, Baden H, Quaade F: Liver morphology and biochemistry in eight obese patients treated with jejunoileal anastomosis. N Errg/ J Med 285: 543, 197 1. Thompson RH Jr, Meyerowitz BR: Liver changes after jejunoileal shunting for massive obesity. Surg Forum 21: 366, 1970. Scott HW Jr, Law DH IV, Sanstead HH, Lanice VC Jr, Younger RK: Jejunoileal shunt in surgical treatment of morbid obesity. Ann Surg 171: 770, 1970. Holzbach RT, Wieland RG. Lieber CS, DeCarli LM, Koepke KR. Green SG: Hepatic lipid in morbid obesity. N Engl J Med290:296,1974.
17. Brown RG, O’Leary JP, Woodward ER: Hepatic effects of jejunoileal bypass for morbid obesity. Am J Surg 127: 53, 1974.
91
P. Spin, MD,* Hanover, New Hampshire
Rodger E. Weismann,
MS, MD, Hanover, New Hampshire
Fatal hepatic failure as a compljcation of jejunocolic anastomosis is well documented in the literature [1-S]. To date, however, this complication after jejunoileal anastomosis has rarely been described [6]. We present two case reports of patients who underwent end to side jejunoileal anastomosis for massive obesity and subsequently died of hepatic failure. The clinical symptoms and liver function test results in these two patients are compared with those in 124 other patients who underwent similar surgical procedures but in whom serious liver disease has not been observed. The group consisted of 126 patients who underwent jejunoileal anastomosis for treatment of morbid obesity. In most patients, the ileal segment was 4 or 5 inches long, and the jejunal segment 14 or 15 inches. All patients had intraoperative liver biopsy, and most have had at least one percutaneous needle biopsy subsequently. Two patients have had the shunt taken down. One hundred patients have been followed up longer than one year, and the remainder longer than eight months. Fifty have been followed up longer than two years. Details of the series are reported elsewhere [ 71. Case Reports Case I. The patient (DS), a forty-one year old woman, was first seen because of lifelong obesity in January 1971; she weighed 350 pounds at the time. Later that month, she underwent dilatation and curettage, and a diagnosis of adenocarcinoma of the endometrium was made. She had radium implantation under nitrous oxide and oxygen anesthesia. On February 23,1972, she underwent total abdominal hysterectomy and jejunoileal anastomosis. The ileum, 4 inches proximal to the ileocecal valve, was anastomosed to the jejunum, 14 inches distal to the ligament of Treitz. Anesthesia was nitrous oxide and oxygen; 2 units of blood were given. Intraoperative liver biopsy showed From the Department of Surgery. the Hiichcock Clinic, Hanover, New Hampshire. * Present address and address for reprint requests: the Guthrie Clinic, Sayre. Pennsylvania 18840.
aa
moderate fatty change and slight increase in fibrous tissue. Postoperatively, she complained of nausea and vomiting. She was discharged on a regimen of chlorpromazine (Thorazine@), 150 mg, and promethazine (Phenergane), 25 mg, twice a day. Chlorpromazine was discontinued in May. In late June jaundice was noted, and she was rehospitalized. There was no history of alcoholism. On admission she weighed 260 pounds. The liver was palpable four fingerbreadths below the right costal margin and was firm and tender. The spleen was also enlarged. Australian antigen was negative. She received intravenous fluids and vitamin K. She continued to vomit and was lethargic and depressed. The jaundice increased and liver function test results became increasingly abnormal over the next three weeks. The transaminase levels were in the range of 200 to 306 units. The prothrombin time, initially corrected to fourteen seconds, increased to twenty-one. By July 21 hepatic coma, intestinal bleeding, and hypotension that did not respond to transfusion had developed. Blood pressure was maintained with pressor amines for three days, and she died on July 28, 1972. At autopsy, performed fifteen hours post mortem, the liver weighed 4,170 gm. Microscopically, there was massive necrosis with no remaining nuclear detail. There was marked fatty infiltration but no inflammatory infiltrate. The extrahepatic biliary system was patent and normal except for surgical absence of the gallbladder. The jejunoileal anastomosis was 8 inches from the ligament of Treitz and 2 inches proximal to the ileocecal valve. Case II. The patient (BS), a thirty-three year old housewife, underwent jejunoileal anastomosis on April 9, 1971, for massive obesity. She was 5 feet 2 inches tall and weighed 313 pounds. Anesthesia was thiopental sodium (PentothaF), nitrous oxide, and oxygen; 2 units of blood were given. A 14 inch segment of jejunum was anastomosed to the ileum, 4 inches from the ileocecal valve. Liver biopsy (Figure 1A) showed moderate fatty change. The postoperative course was unremarkable. On September 20, 1972, she was readmitted because of rectal bleeding. She had noted easy bruisability and bleeding from the gums. The sclerae were icteric. The liver had an overall vertical diameter of 15 cm. The spleen was not palpable and ascites was present. Several large hemorrhoids were bleeding profusely. White blood
The American Journal of Surgery
Hepatic
B
Failure after Jejunoileal
Bypass
C
Figure 1. Histologic sections of liver; case II. A, intraoperative liver biopsy showing moderate fatty infiltration (Masson trichrome stain; original magnification X 40, X 2.5). B, liver biopsy on final admission showing marked lobular disarray with complete loss of architecture and extensive necrosis. There is also moderate fatty change, some fibrosis, and round cell infiltrate (Masson trichrome stain; original magnification X 40, X 2.5). C, postmortem liver needle necropsy showing massive fibrosis with nests of regenerating cells (hematoxylin and eosin stain; original magnification X 40, X 2.5).
hepatic encephalopathy developed. The amino acids were deleted, and administration of prednisone was begun at 60 mg per day. She showed only transient improvement and died on November 2, 1972. At autopsy, the liver weighed 2,300 gm and was extremely firm. The extrahepatic biliary system was normal. The spleen weighed 500 gm. The jejunoileal anastomosis was as described at operation. Liver necropsy was performed twenty minutes post mortem. (Figure 1C.)
cell count was 1,300 per mm3, platelets 165,000 per mms, prothrombin time twenty-five seconds, and partial thromboplastin time greater than two minutes. Prothrombin time was corrected to twelve seconds with vitamin K and fresh frozen plasma administration, and liver biopsy was performed. (Figure 1B.) Australian antigen was negative. Parenteral hyperalimentation of amino acids, vitamins, electrolytes, and hypertonic glucose was begun, and oral cholestyramine (Questrang) was administered. Her course was complicated by sttiphylococcal sepsis, presumably from the intravenous catheters. After this, she became somnolent and progressive TABLE
I
Date
Laboratory
Weight (pounds)
Data
_ Serum Glutamic Oxalacetic Transaminase (units/ml)*
Prothrombin Time (set; control of
Albumin (gm/L plasma)
11 set) Case
l/20/72 3120172 418172 616172 7 128172
Laboratory data from patients in cases I and II are presented in Table I.
330 289 280 278 255
54 ... 300-t 372
313 214 162 160
... 126 82 220
...
Total/Direct Bilirubin (mg/lOOml)
Carotene (mg/lOO ml)t
Alkaline Phosphatase (units/ml)$
I
...
.,.
0.5/0.1
...
1.7 2.2
...
... ...
...
21
2.9
... 15
...
5 40
... 12.8
25.6/21.2
...
...
...
55 15 35 ...
5.3 12.4 9.1 8.9
...
7.211.9
6.5
Case II 4/20/71 12/6/71 g/20/72 10/13/72
...
...
17 28 15
... 2.5 3.1
0.5/0.3 5.512.5 31.612.7
*Normal: 0.5 to 45 units/ml. t Normal: 100 to 300 mg/lOO ml. $ Normal: 2 to 4.5 units/ml.
Volume130,July1975
89
Spin and Weismann
TABLE
II
Combined Laboratory Data from 126 Subjects Undergoing Jejunoileal Anastomosis
Carotene (mg/lOO ml) Preoperative Total tested Mean Minimum Maximum
112 129 25 247 < 25 (0) 25-50 (20)
Postoperative Total tested Mean Minimum Maximum
One month Total tested Mean Minimum Maximum
106 72 10 195 20 (0) 10 13.2 3.5 30.0
47 11 290
6 79 30 150
72 29 5 70 20 (2)
36 49 11 132 50-100 (16) >lOO (1)
24 53 186 50-100 (3) >lOO (4)
Comments
Clinical Features. In an attempt to delineate any significant features that might distinguish these two patients from the others, we compared the clinical symptoms. The patient in case I had severe anorexia and intractable vomiting. A majority of our patients have had various degrees of anorexia, at least transiently. This had been attributed to a metabolic ketoacidosis resulting from malabsorption and starvation and usually responded to increased carbohydrate intake. The degree of anorexia and vomiting in the patient in case I was more severe than that seen in most of our other patients. Conversely, the second patient manifested no loss of appetite and was able to maintain oral intake until well into the final hospi-
90
talization. Symptoms of fatigue, diarrhea, or weight loss did not distinguish these two patients from the majority. Laboratory Data. The cumulative serum carotene levels and liver function test results in the preoperative, immediate postoperative, and later follow-up periods are presented in Table II. Serum carotene measurements were made as a rough assessment of general nutritional status and fat absorption. Preoperative carotene measurements were made in one hundred twelve patients. Eighty-six of these were over 100 mg/lOO ml. Only two were under 50 mg/lOO ml. Postoperative measurements were made in one hundred and six patients. In thirty-two, values between 25 and 50 mg/lOO ml were recorded. Two had serum carotene levels of 20 and 10 mg/lOO ml. Additional measurements were made at one, six, and twelve months; nine, sixteen, and nine patients, respectively, had carotene levels of 15 mg/lOO ml or less. Thus, the extremely low carotene values in the patients in cases I and II do not seem to be predictive of incipient fatal hepatic disease. Values of more than 50 mg/lOO ml, however, were associated with insufficient weight loss. Alkaline phosphatase was measured preoperatively in one hundred and nine patients. Only thirty-six had values of less than 5 units/ml and ten had values of more than 10 units/ml. One year after operation, forty-nine patients were tested; nineteen had alkaline phosphatase levels of more than 10 units/ml. Two had values over 20 units/ml. Again, the mild degree of abnormality recorded in the two patients with fatal hepatic necrosis would not have separated them from the others. Serum glutamic oxalacetic transaminase was measured prior to jejunoileal anastomosis in eleven patients. In only one was an abnormal value found. However at follow-up study one year later, seven of twenty-four patients tested has serum glutamic oxalacetic transaminase values greater than 50 units/ml, two of which were over 180 units/ml. Prothrombin times were not measured routinely in our patients and therefore no conclusions can be drawn regarding the abnormal values in the two patients with hepatic failure. Fatty liver and mild fibrosis are common in massive obesity [8-121. Abnormal liver function test results have been reported by some and not found by others ]11,13-151. Starvation results in a reduction of steatosis, and focal necrosis may occur [12]. A recent study of twenty-three patients undergoing end to end bypass demonstrated a threefold increase in hepatic lipids during the pe-
The American Journal of Surgery
Hepatic
riod of rapid weight loss [16]. Drenic, Simmons, and Murphey [8] described massive fibrosis with virtual absence of hepatic tissue in a patient who died after jejunocolic anastomosis. Payne and DeWind [IO] found portal fibrosis in one patient after jejunocolic bypass. In the first patient reported on herein, the situation was complicated by a number of factors, including blood transfusion and chlorpromazine and promethazine for control of nausea. Postmortem examination showed the ileal segment to be shorter than that measured at operation. The autopsy specimen of the liver was obtained a number of hours after death, and postmortem autolysis had undoubtedly occurred. The absence of substantial cellular infiltrate in the liver makes fulminant viral hepatitis or an idiosyncratic reaction to phenothiazines unlikely. In the second patient, biopsy specimens clearly show a progression from steatosis with triaditis, to massive necrosis. and finallv to micronodular cirrhosis with hyaline degeneration and absence of significant inflammatory infiltrate not unlike that observed bv McGill et al 161. We have compared the clinical and laboratory data in these two patients with the findings in 124 others who underwent a similar nrocedure and found nothing distinctive in their courses prior to the onset of overt liver disease. In addition to carefully planned preoperative evaluation and meticulous long range follow-up study, repeated liver biopsy may prove to be the only accurate way of assessing liver function and morphologic features so that reanastomosis can be undertaken sufficiently early to prevent fatal hepatic necrosis. Because necrosis may occur at intervals that vary substantially after surgery, repeated biopsy should be timed to detect significant liver damage during this period. It may be necessary to evaluate the morphologic aspects of the liver every three to six months until the patient’s weight has stabilized. Other parameters that may prove to be of predictive value are the prothrombin time and, perhaps, a decrease in uptake of technetium sulfa colloid bv the liver. Additional data will have to be accumulated before this can be assessed finally [I 71. d
L
References
1. Maxwell JG, Richards RC, Alba D Jr: Fatty degeneration of 2. 3.
I
4. 5. 6.
7. 8.
9. 10. 11. 12.
13.
14.
15.
Summary 16. _I
Volume130,July 1975
Bypass
complication after the currently employed jejunoileal anastomosis has rarely been described before. We present two cases of hepatic necrosis and death after end to side jejunoileal anastomosis and compare the clinical data and liver function test results in these two patients with findings in 124 other patients who underwent a similar procedure for treatment of morbid obesity. No significant differences could be found that would allow early identification of those patients in whom massive hepatic necrosis was developing. Careful follow-up study and perhaps repeated liver biopsy may be necessary to monitor the hepatic status after jejunoileal anastomosis so that reanastomosis can be undertaken at a time when the damage may be reversible.
>
Severe liver dvsfunction and even death from kepatic failure after jejunocolic anastomosis has led to the virtual abandonment of that procedure for the surgical treatment of massive obesity. This
Failure after Jejunoileal
the liver after intestinal bypass for obesity. Am J Surg 116: 646, 1968. Wills CE Jr: Experiences with intestinal short circuiting procedures for obesity. J Med Assoc Ga 56: 365, 1967. Payne JH, DeWind LT, Commons RR: Metabolic observations in patients with jejunocolic shunts. Am J Surg 106: 273, 1963. Barrow J, Frame B, Bozalis JR: A shunt operation for obesity. Dis Co/on Rectum 11: 115, 1969. Bondar GF. Piseky W: Complications of small intestinal short-circuiting for obesity. Arch Surg 94: 707, 1967. McGill DB. Humohrevs SR. Baooenstoss AH, Dickson ER: Cirrhosis and death after.jejur%eal shunt. Gasfroeoferology 63: 872, 1972. Weismann RE: Surgical palliation of massive and severe obesity. Am J Surg 125: 437, 1973. Drenic EJ, Simmons F, Murphey JF: Effect on hepatic morphology of treatment of obesity by fasting, reducing diets and small-bowel bypass. Nfng/ J Med 282: 829, 1970. Meyerowitz BR: Surgical treatment of intractable morbid obesity. Sand J Gastroenterol7: 1, 1972. Payne HJ, DeWind LT: Surgical treatment of obesity. Am J Surg 118: 141. 1969. Westwater JO, Fainer D: Liver impairment in the obese. Gestroenterology 34: 686, 1958. Rozental P. Biava C, Spencer H, Zimmerman HJ: Liver morphology and function tests in obesity and during total starvation. Am J Dig Dis 12: 198, 1967. Juhl E, Christoffersen P, Baden H, Quaade F: Liver morphology and biochemistry in eight obese patients treated with jejunoileal anastomosis. N Errg/ J Med 285: 543, 197 1. Thompson RH Jr, Meyerowitz BR: Liver changes after jejunoileal shunting for massive obesity. Surg Forum 21: 366, 1970. Scott HW Jr, Law DH IV, Sanstead HH, Lanice VC Jr, Younger RK: Jejunoileal shunt in surgical treatment of morbid obesity. Ann Surg 171: 770, 1970. Holzbach RT, Wieland RG. Lieber CS, DeCarli LM, Koepke KR. Green SG: Hepatic lipid in morbid obesity. N Engl J Med290:296,1974.
17. Brown RG, O’Leary JP, Woodward ER: Hepatic effects of jejunoileal bypass for morbid obesity. Am J Surg 127: 53, 1974.
91
