Intra-abdominal surgery in patients with advanced cirrhosis.

Between 1971 and 1984, intra-abdominal surgical procedures were required in 51 patients with cirrhosis who had advanced liver dysfunction and/or ascites. These included the following types of surgery: gastric, 17 patients; small bowel, two patients; colon, five patients; hepatic, nine patients; and pancreas, nine patients. Twelve patients also underwent exploratory celiotomy for an acute condition of the abdomen (six patients) and jaundice (six patients). Patients fell into two groups: (1) those with prothrombin time (PT) greater than 2.5 s over control (24 patients), and (2) those with PT within 2.5 s of control (27 patients). The 30-day mortality rate was 34 (67%) of 51 patients. Nineteen (83%) of 23 patients who had ascites died. Twenty-two (91%) of 24 patients with elevated PT greater than 2.5 s of control died. Twenty-five (86%) of 29 patients who underwent surgery under emergency conditions died. Intraabdominal surgery in decompensated patients with cirrhosis must be undertaken with great caution.

Intractable ascites treated with peritoneovenous shunts (LeVeen). A 24- to 64-month follow-up of results in 52 alcoholic cirrhotics.

Peritoneovenous shunts (LeVeen) were placed in 52 patients with intractable ascites due to alcoholic cirrhosis between March 1975 and June 1978. The ascites was defined as intractable if it persisted despite at least six weeks of in-hospital medical management. Only 4.5% of such patients seen during this interval met this criterion by failing to respond to intensive drug and diet therapy. Eight (15%) died five to 60 days postoperatively; six of these did not undergo diuresis. Hepatic encephalopathy and sepsis were important contributing factors. There were 21 late deaths (mean survival, 16 months), primarily related to complications characteristic of this group of cirrhotics: upper gastrointestinal hemorrhage and infection. Twenty-three patients were alive by June 25, 1980, with mean survival of 46 months. Four patients operated on five or more years ago survive without ascites. An accurately placed LeVeen shunt dramatically relieves intractable ascites with improvement in renal function, nutrition, and general health. The operation enables at least partial rehabilitation so that ambulatory care becomes feasible for many of these chronically and seriously disabled patients.

Combined pancreatic duct and upper gastrointestinal and biliary tract drainage in chronic pancreatitis.

In 55 patients undergoing pancreaticojejunostomy for intractable abdominal pain, common bile duct obstruction occurred in 29% (16/55) and duodenal obstruction occurred in 15% (8/55). Serum alkaline phosphatase and total and direct serum bilirubin levels were significantly higher in patients with intrapancreatic common bile duct stenosis. Transient upper gastrointestinal (UGI) tract obstruction was common with chronic pancreatitis; however, if symptoms persisted beyond 2 weeks, fixed duodenal obstruction was likely. Endoscopic retrograde cholangiopancreatography and UGI roentgenograms and endoscopy were useful in confirming mechanical obstruction to the biliary and UGI tracts, respectively. There was no difference in operative mortality and morbidity from combined drainage procedures compared with pancreaticojejunostomy alone. The biliary and UGI tracts should be investigated in symptomatic patients both before and after pancreaticojejunostomy. Combined drainage of the pancreatic duct and UGI and biliary tract is safe and effective treatment for obstructing complications of chronic pancreatitis.

Simultaneous treatment of chronic pancreatitis and pancreatic pseudocyst.

Records from 87 consecutive patients undergoing lateral pancreaticojejunostomy (LPJ) for chronic pancreatitis were reviewed to determine the incidence of pseudocyst and the safety of combined pancreatic duct and pseudocyst drainage. Twelve patients had undergone previous pancreatic pseudocyst drainage; four of them also had pancreatic pseudocysts present at the time of LPJ. In addition, 22 patients had pseudocysts identified preoperatively and/or confirmed at operation. The overall incidence of pseudocyst was 39%. Twenty-six patients (group 1) underwent pancreaticojejunostomy combined with pseudocyst drainage. Sixty-one patients (group 2) underwent pancreaticojejunostomy only. Operative morbidity and mortality results (19% and 8%, respectively, in group 1; 18% and 2%, respectively, in group 2) were similar. Patient outcome was also similar in the two groups (81% and 84% of patients obtained pain relief in groups 1 and 2, respectively). There were no pseudocyst recurrences in either group. Thus, there is a high incidence (39%) of pancreatic pseudocyst in patients undergoing LPJ for chronic pancreatitis. Combined drainage of the pancreatic duct and pseudocyst is safe and effective.

Umbilical herniorrhaphy in cirrhotic patients.

During a ten-year period, 92 patients underwent an umbilical herniorrhaphy. Patients were divided into three groups: group 1, cirrhotic patients with ascites with functioning peritoneovenous shunts (n = 15); group 2, cirrhotic patients with ascites with nonfunctioning or no peritoneovenous shunts (n = 24); and group 3, noncirrhotic patients (n = 53). The charts were analyzed for postoperative mortality and morbidity and recurrence of the umbilical hernia. Umbilical hernia in cirrhotic patients with uncontrolled ascites was associated with significant mortality (8.3%) and morbidity (16.6%) and a significantly greater incidence of recurrence (16.6%). Umbilical herniorrhaphy in patients with functioning peritoneovenous shunts was associated with minimal morbidity (7%). These data suggest that cirrhotic patients with ascites who require an umbilical herniorrhaphy preferably should undergo peritoneovenous shunting before repair of the hernia.

The spectrum of biliary tract obstruction from chronic pancreatitis.

Fifty-one patients with chronic pancreatitis manifested distal common duct obstruction from fibrosis of pancreatitis. The cause of the pancreatitis was alcohol in most patients. An elevated serum alkaline phosphatase level was the most frequent abnormal laboratory finding. The serum bilirubin level elevation was never progressive; a rising and falling pattern was most often encountered. Percutaneous transhepatic cholangiography and endoscopic retrograde cholangiopancreatography are the most useful diagnostic tests. An operation was performed on 47 patients and included choledochoduodenostomy in 16 patients, choledochojejunostomy in 19 patients, cholecystenteric anastomosis in seven patients, common bile duct exploration with T-tube drainage in three patients, and sphincteroplasty in two patients. Abdominal pain was lessened after operation in 40 of 44 patients who survived surgery. Two patients with T-tube drainage and two with cholecystenteric anastomosis required conversion operations to choledochoduodenostomies. Identification of associated pancreatic duct obstruction and dilatation, pseudocysts, and duodenal obstruction is important.

Gastric outlet and duodenal obstruction from inflammatory pancreatic disease.

During a ten-year period, 16 patients with gastric outlet and duodenal obstruction due to inflammatory pancreatic disease were seen. The cause of obstruction was chronic pancreatitis in ten patients, pseudocysts with associated pancreatitis in five patients, and pancreatic abscess in one patient. All patients had nausea and vomiting, 14 had abdominal pain, and five had weight loss greater than 4.5 kg. Diagnosis was made by plain abdominal film in one case, upper gastrointestinal tract roentgenographic series in 15 cases, and endoscopy in 11 cases. Mobilization of the duodenum relieved the obstruction in two patients. Fixed obstruction remained in 14 patients. This was relieved by gastrojejunostomy in 12 patients. Gastrojejunostomy was combined with drainage of a pseudocyst in three patients, a dilated pancreatic duct in three patients, and a dilated common bile duct in four patients. Obstruction was relieved by pseudocyst drainage in two patients. Associated common duct and pancreatic duct obstruction must be identified preoperatively.

Pancreaticojejunostomy for chronic pancreatitis. Two- to 21-year follow-up.

Between 1954 and 1975, 80 pancreaticojejunostomies were performed on 77 patients for intractable pain of chronic pancreatitis. All patients had a history of chronic alcoholism. Drainage operations done primarily for pseudocysts were excluded. Operative procedures included seven caudal pancreaticojejunostomies, 42 longitudinal pancreaticojejunostomies with splenectomy and implantation of the pancreas into the jejunum, and 31 side-to-side pancreaticojejunostomies. Eighty-one percent of the patients noted substantial improvement or complete resolution of their abdominal pain on follow-up that ranged up to 21 years. The operative mortality was 5%. Thirty-two patients died during the period of the follow-up. Continued alcohol abuse, carcinoma, and cardiovascular disease were the leading causes of mortality. Data from this review confirm the effectiveness of pancreaticojejunostomy in relieving the pain of chronic relapsing pancreatitis.

Evaluation of therapeutic options for pancreatic pseudocysts.

A review of 81 patients with pancreatic pseudocyst was conducted to assess the value of different treatment modalities. Resection was associated with 18% mortality (two of 11 patients) and 36% morbidity. In three of nine patients undergoing external drainage a recurrent pseudocyst developed, and in one additional patient, a pancreatic fistula persisted. Internal drainage by cystogastrostomy (21 patients) resulted in 9.5% mortality and 9.5% morbidity, whereas cystojejunostomy (33 patients) was associated with a 6% mortality and 6% morbidity. Endoscopic drainage through the posterior wall of the stomach was unsuccessful in the two patients in which it was used. Internal drainage into the stomach, duodenum, or jejunum is a safe and effective approach for most pseudocysts. Persistent symptoms following surgical treatment were primarily related to failure to recognize multiple cysts and/or pancreatic duct obstruction and dilation characteristic or chronic pancreatitis.

The nature and course of cystic pancreatic lesions diagnosed by ultrasound.

To determine the nature and course of cystic pancreatic lesions, we reviewed the records of 93 patients who were followed up with serial ultrasound until resolution occurred or definitive therapy was given. Initially, 105 cystic lesions were seen. Single cysts occurred in 83 patients and multiple cysts in ten. Eight patients had two cysts and two patients had three. Twenty-one patients had a single ultrasound examination and 72 had two or more examinations. Spontaneous resolution occurred in 29 of (28%) 105 cysts in 24 patients. In 46 patients, 53 pseudocysts were found at operation or autopsy. Five patients had pancreatic abscesses. Two patients became unavailable for follow-up and three refused surgery. The results of one examination were false negative. Twelve patients had false-positive ultrasound results. Factors influencing the likelihood of spontaneous resolution included initial cyst size, subsequent increase in size, multiple cysts, and the presence of calcifications. This study confirms the usefulness of serial ultrasound scans in determining the nature and necessity for treatment of cystic lesions of the pancreas.

Radiologic diagnosis of an intra-abdominal abscess. Do multiple tests help?

A review was made of the charts of 94 patients who underwent ultrasonography (US), computed tomography (CT), and gallium citrate Ga 67 (Gall) scan to rule out intra-abdominal abscesses. Of all the clinical and laboratory data, only the presence of pain and tenderness differentiated patients with and without abscesses. A review of radiologic data showed that CT was superior to US, and that US was superior to Gall scan with regard to sensitivity, specificity, accuracy, and positive and negative predictive values. When multiple radiologic tests were performed, results agreed in 72% of cases; therefore, the additional tests were essentially redundant. When radiologic test results disagreed, accuracy rates were CT, 0.86; US, 0.00; and Gall scan, 0.44. These findings suggest that, except to rule out pelvic abscesses in the presence of pelvic inflammatory disease, CT is usually the only special radiologic test that should be performed to localize a suspected intra-abdominal abscess.

Therapeutic options for biliary tract disease in advanced cirrhosis.

Between 1979 and 1984, 21 male cirrhotic patients with advanced liver disease, cholecystitis, and jaundice were seen. Eight patients had persistent symptoms of acute cholecystitis despite intense symptoms of acute cholecystitis despite intense medical management. Of these patients, five underwent cholecystostomy and survived. The other three patients had cholecystectomy and one died. Thirteen patients presented with jaundice. Twelve patients underwent endoscopic retrograde cholangiography which revealed gallbladder stones in four but no stones in the common bile duct. They did not undergo further surgical procedures. One patient presented with jaundice, cholangitis, and pancreatitis was found to have stones in the common bile duct and underwent endoscopic sphincterotomy with removal of multiple small, pigmented stones. This patient died from sepsis and renal failure 37 days after sphincterotomy. Endoscopic retrograde cholangiography was unsuccessful in four patients who later underwent percutaneous transhepatic cholangiography which revealed stones in one and cirrhotic ductal changes in three. The remaining jaundiced patient underwent cholecystectomy and common bile duct exploration which revealed no ductal stones. This patient died 21 days after operation from sepsis and multiple organ system failure. Three of five patients with gallstones on endoscopic retrograde cholangiography or percutaneous transhepatic cholangiography died, but none of the deaths were due to biliary tract disease. At last follow-up the two surviving patients were asymptomatic. The overall mortality rate was 14 percent (3 of 21 patients). Cholecystostomy in cirrhotic patients with advanced liver disease and acute cholecystitis is associated with minimal mortality and morbidity. Cirrhotic patients with jaundice are probably best evaluated initially by endoscopic retrograde cholangiopancreatography which is safe, diagnostic, and sometimes therapeutic.

Redrainage of the pancreatic duct in chronic pancreatitis.

Recurrent pain after a drainage procedure for chronic pancreatitis is considered an indication for pancreatectomy. To evaluate whether redrainage might be a better alternative, 14 patients who underwent redrainage after a failed pancreaticojejunostomy were reviewed. Patients with previous pseudocyst drainage were excluded. Initial operations included five caudal, three longitudinal, and six side-to-side pancreaticojejunostomies. Nine patients treated since 1974 had ERCP, which showed obstructed segments of pancreatic duct in the head of the gland. Two caudal pancreaticojejunostomies and one longitudinal pancreaticojejunostomy were revised to longitudinal pancreaticojejunostomies. The other 11 were revised to side-to-side pancreaticojejunostomies. Operative findings confirmed undrained segments of the pancreatic duct in the pancreatic head. Postoperatively, one patient died from hemorrhage and four patients had complications. At most recent follow-up from 6 months to 20 years postoperatively, three patients were pain free and six had substantial relief from pain (71 percent). Of eight patients who were not diabetic before redrainage, diabetes developed in only two. Only one of seven patients without pancreatic exocrine insufficiency required pancreatic enzymes after redrainage. Patients with recurrent pain after pancreaticojejunostomy should undergo ERCP. If segments of the pancreatic duct are obstructed, redrainage can provide satisfactory pain relief with a minimal loss of endocrine and exocrine function. This problem is best avoided by initial complete drainage of the major and minor pancreatic ducts.

Use of ultrasound to demonstrate gallstones in symptomatic patients with normal oral cholecystograms.

Seven patients who had typical and persistent symptoms of gallbladder disease but repeatedly normal oral cholecystograms were seen in an 18 month period. Ultrasound examination of the gallbladder was diagnostic of cholelithiasis in all seven patients. Six patients were found to have gallstones at operation, and the seventh had histologically proved chronic cholecystitis. Although oral cholecystrography is a highly accurate examination, false-negative results do occur. Gray scale ultrasonography is a safe and also highly accurate means of providing objective evidence of cholelithiasis in patients clinically thought to have gallstones despite normal oral cholecystograms.

Pancreatic abscess: an unresolved surgical problem.

Twenty patients with pancreatic abscesses were studied to determine if recent diagnostic and therapeutic advances have improved the outlook for those with the disease. An abscess developed as a complication of alcoholic pancreatitis in 10 patients and was due to previous surgery in 9. Ultrasonography and computerized tomographic scanning of the abdomen were helpful in the diagnosis and localization of the abscesses. All 20 patients were treated surgically. Operative mortality was 30 percent and was due to multiple system failure from continuing sepsis. Only 2 of 15 patients who had sump drainage died compared with 3 of 4 patients who were drained with Penrose drains alone. There were two deaths among 10 patients who received nutritional support and four deaths in 10 patients who did not receive hyperalimentation. Pancreatic abscess remains a life-threatening condition. Ultrasonography and computerized tomographic scanning have helped in diagnoses and localization. The addition of sump drainage has reduced the mortality rate from 75 ot 13 percent. Nutritional support also appears to be helpful in reducing mortality.

Cholecystectomy in cirrhotic patients: a formidable operation.

Cholecystectomy and common bile duct exploration in cirrhotic patients is associated with an 83 percent mortality if prothrombin time is prolonged 2.5 seconds over control. The causes of death are related to complications of liver disease such as hepatic encephalopathy, ascites, sepsis and hemorrhage. If the prothrombin time is prolonged, major intraoperative blood loss can be anticipated, and blood and plasma transfusion requirements may be massive. Jaundice in the presence of cirrhosis requires careful preoperative evaluation and is most frequently due to hepatocellular disease rather than extrahepatic biliary obstruction. Cholecystectomy and common duct exploration in cirrhotic patients should be performed only for life-threatening complications of biliary tract disease such as empyema, perforation and ascending cholangitis.

Neoplastic obstruction of the small and large intestine.

Obstructive symptoms due to small bowel tumors are the most common indication of primary malignant disease in the small intestine. Primary obstructing tumors of the small bowel are treated best by resection and primary anastomosis. Malignant lesions of the duodenum sometimes will require pancreaticoduodenectomy and those of the distal ileum, right hemicolectomy. Obstruction due to localized metastatic disease can be treated by resection and primary anastomosis but, more frequently, one or more side-to-side enteroenterostomies will be needed, especially in abdominal carcinomatosis. The complication of LBO due to colorectal cancer is an ominous sign. The less favorable prognosis is a result of the higher operative mortality, advanced stage of disease and lower resectability rate. Obstructing neoplasms of the right side of the colon are treated best by immediate resection and primary anastomosis. Left-sided colon obstruction due to malignancy traditionally is treated by preliminary diversion followed later by definitive resection. Insufficient data are available to evaluate any benefit on operative mortality and long-term survival with a more aggressive approach involving decompression and resection of the obstructing carcinoma at the initial operation. It is doubtful that any marked improvement in current mortality and survival figures will result from wide deviations of the current principles of operative managment. Early diagnosis of the cancer before obstruction occurs remains the primary means of improving survival rates. This involves not only primary means of improving survival rates. This involves not only patient education regarding presenting symptoms, but improvement of physician recognition and response to these complaints so that the appropriate tests are ordered and treatment is initiated.

Surgical management of carcinoid tumors of the gastrointestinal tract.

Between the years 1950 and 1976, 76 patients with carcinoid tumors of the gastrointestinal tract were seen at the Hines Veterans Administration Hospital, Hines, Illinois; of these, 48 tumors were recognized clinically. Five- and ten-year survivals were 56 and 44 per cent, respectively. The best survivals were seen in patients with carcinoids originating in the appendix, followed by the small bowel, rectum, and then the colon.

Biliary enteric bypass for benign and malignant disease.

Between the years of 1970 and 1984, a total of 96 patients underwent biliary enteric bypass to alleviate distal common bile duct obstruction from benign and all malignant disease. Cholecystoenterostomy (CCE) was performed in 13 patients (chronic pancreatitis 7, carcinoma 6), choledochoduodenostomy (CDD) was performed in 35 patients (stones 9, chronic pancreatitis 17, carcinoma 8, and fistula 1), cholecystojejunostomy (CDJ) was performed on 48 patients (stones 1, pancreatitis 21, carcinoma 25 and stricture 1). Operative mortality was 7 per cent and morbidity occurred in 12 per cent of the patients. Symptomatic improvement was measured by relief of pain and sepsis and decrease of bilirubin and alkaline phosphatase to normal. Overall improvement was seen in 73 per cent of patients (CCE 50%, CDD 8%, CDJ 65%), 27 per cent of the patients did not improve (CCE 50%, CDD 12%, CDJ 35%), 83 per cent of the poor results were in patients with advanced malignancy. Thirty-one per cent of patients undergoing CCE required conversion to CDD or CDJ. Cholecystoduodenostomy was associated with failure in 50 per cent of patients. CCD and CDJ are safe and reliable means of relieving distal common duct obstruction due to biliary or pancreatic disease. Cholecystojejunostomy may be performed in the terminal patient with advanced carcinoma requiring a short-term biliary bypass.

Catheter infection factors affecting total parenteral nutrition.

During the five-year interval from January 1, 1971 to January 1, 1976 118 seriously ill adults received 2916 patient days of TPN therapy with an average infection rate of 7.6%. All patients received the benefit of a well-defined TPN catheter care protocol which emphasized regular (every 48 hours) catheter dressing changes. The lowest risk of infection, 2.7%, was seen in 73 patients who received an amino acid-glucose solution through a silicone elastomer catheter protected by an iodophor dressing. When catheter-related sepsis occurred, Staphlylococcus aureus and Candida albicans were the most common organisms cultured.