Showing papers in "Archives of Surgery in 1995"

Perioperative Plasma Volume Expansion Reduces the Incidence of Gut Mucosal Hypoperfusion During Cardiac Surgery

Abstract: Objective: To test the hypothesis that perioperative plasma volume expansion would preserve gut mucosal perfusion during elective cardiac surgery. Design: Prospective randomized open study. Setting: Teaching hospital. Patients: Sixty American Society of Anesthesiology grade III patients with a preoperative left ventricular ejection fraction of 50% or greater undergoing elective cardiac surgery. Interventions: Patients were allocated randomly to a control or protocol group. The control group was treated according to standard practices. After induction of general anesthesia, the protocol group received, in addition, 200-mL boluses of a 6% hydroxyethyl starch solution to obtain a maximum stroke volume. This procedure was repeated every 15 minutes until the end of surgery, except when the patient underwent cardiopulmonary bypass. Measurements and Results: Cardiac stroke volume was estimated by an esophageal Doppler system, and gastric mucosal perfusion was measured by tonometric assessment of gastric intramucosal pH in all patients. Patients were followed up postoperatively until discharge from the hospital or death. The incidence of gut mucosal hypoperfusion(gastric intramucosal pH P P =.01), mean number of days spent in the hospital (6.4 [range, 5 to 9] vs 10.1 [range, 5 to 48]) ( P =.011), and mean number of days spent in the intensive care unit (1 [range, 1 to 1] vs 1.7 [range 1 to 11] days) ( P =.023). Conclusions: Perioperative plasma volume expansion with colloid during cardiac surgery, guided by esophageal Doppler measurement of cardiac stroke volume, reduced the incidence of gut mucosal hypoperfusion. This group of patients also had an improved outcome when compared with controls. (Arch Surg. 1995;130:423-429)

Standards for pancreatic resection in the 1990s.

Abstract: Objective: To describe the current indications and operative outcomes of pancreatic resection. Design: Retrospective case series. Setting: Referral practice in a university hospital. Patients: Two hundred thirty-one consecutive patients undergoing pancreatoduodenectomy (PD), distal pancreatectomy (DP), or total pancreatectomy (TP) over a 44-month period. Their ages ranged from 16 to 85 years, with a mean of 54 years; 20% of the patients were 70 years old or older. Main Outcome Measures: Mortality, complications, and length of hospital stay. Results: Operative mortality was 0.4% (one death following DP); there were no deaths in 142 PDs or in 18 TPs. The most common complication following PD was delayed gastric emptying. Pancreatic fistula occurred in 6.3% of PD and in 9.8% of DP patients. Overall, 58% of PD, 80% of DP, and 78% of TP patients had no complications. The mean±SD length of hospital stay was 15±7, 10±5, and 15±6 days for PD, DP, and TP, respectively. Reoperation for any cause was necessary in only 1.2% (3/231). The most frequent indication for PD was pancreatic cancer (36%) followed by chronic pancreatitis (26%); for DP it was chronic pancreatitis (28%) and cystic neoplasms (27%); and for TP, chronic pancreatitis (55%). Newer indications for pancreatic resection included mucinous ductal ectasia and intraductal papillary tumors (eight cases, 4%) and metastatic tumors (eight cases, 4%). Conclusions: Current indications for pancreatic resection have expanded. These procedures are associated with a low risk for death and postoperative complications when performed in a high-volume setting. (Arch Surg. 1995;130:295-300)

Minimal-access surgery for staging of malignant melanoma.

Abstract: Objective: To develop a simple, minimally invasive technique of determining whether regional node metastasis has occurred in patients with melanoma. Setting: Teaching hospital tertiary care and private practice settings. Patients: Between February 1993 and October 1994, 121 patients with invasive malignant melanoma and clinically negative lymph nodes were enrolled in this clinical trial. Design: Consecutive sample clinical trial. Within 24 hours prior to lymph node resection, a radioactive tracer was injected into the dermis around the site of the primary melanoma. Forty-four patients also had blue dye injected immediately prior to surgical resection. Measurement of radioactivity in the lymph nodes and surgical localization were made using a handheld gamma detector. Radiolabeled nodes were selectively removed with the least dissection possible. In patients with pathologically positive radiolabeled nodes, regional lymphadenectomy was performed. Outcome Measures: Successful identification of radiolabeled sentinel lymph nodes, correlation of radiolabeling with injection of blue dye, and regional node recurrence rate. Results: Surgeons successfully resected the radiolabeled sentinel lymph nodes in 118 (98%) of 121 patients. One hundred percent of blue-stained lymph nodes were successfully radiolabeled. Fifteen patients had pathologically positive sentinel lymph nodes. In 10 patients, the sentinel node was the only node with metastasis. Two systemic and one regional node recurrences occurred during a mean follow-up of 220 days. Conclusions: Selective gamma probe—guided resection of the radiolabeled sentinel lymph node is possible in over 95% of patients with melanoma. This technique offers a simple and reliable method of staging of regional lymph nodes in these patients without performing a regional lymphadenectomy. (Arch Surg. 1995;130:654-658)

Bile duct injuries during laparoscopic cholecystectomy. Factors that influence the results of treatment.

Abstract: Objective: To analyze the treatment of bile duct injuries during laparoscopic cholecystectomy to discern the factors affecting outcome. Design: An analysis of the treatment of 88 patients with laparoscopic bile duct injuries. Setting: A university hospital. Patients: Eighty-eight patients with major bile duct injuries following laparoscopic cholecystectomy. Main Outcome Measures: Success of treatment, morbidity rate, mortality rate, and length of illness. Results: Operations to repair bile duct injuries were unsuccessful in 27 (96%) of 28 procedures when cholangiograms were not obtained preoperatively, and they were unsuccessful in 69% when cholangiographic data were incomplete. In some cases, lack of complete cholangiographic information led to an inappropriate and harmful operation. When cholangiographic data were complete, the first repair was successful in 16 (84%) of 19 patients. A primary end-to-end repair over a T tube (13 patients) was unsuccessful in every case in which the duct had been divided. Direct closure of a partial defect in the duct was successful in four of seven patients. Fifty-four (63%) of 84 Roux-en-Y hepaticojejunostomies were successful. Factors responsible for the unsuccessful outcomes were the following: incomplete excision of the scarred duct, use of nonabsorbable suture material, use of two-layer anastomosis, and failure to eradicate subhepatic infection before the attempted repair. Dilatation and stenting was uniformly unsuccessful as primary treatment (three patients) and was successful in only seven of 26 patients following a previous operative repair. Patients first treated by the primary surgeon had an average length of illness of 222 days (P Conclusions: Surgeons who specialize in the repair of bile duct injuries achieve much better results than those with less experience. The worse results of other surgeons could be attributed in many instances to specific correctable errors. Nonsurgical treatment was usually unsuccessful and substantially increased the duration of disability. (Arch Surg. 1995;130:1123-1129)

Five-year follow-up after radical surgery for colorectal cancer. Results of a prospective randomized trial.

Abstract: Objective: To determine whether an intensified follow-up of patients with colorectal cancer can lead to improved reresectability and a better long-term survival. Design: A prospective randomized trial of 106 patients. Setting: Oulu University Hospital, a referral center in northern Finland. Patients: A total of 106 consecutive patients who underwent radical resection for colorectal cancer, 54 of whom were randomized into a conventional follow-up group and 52 into an intensified follow-up group. Main Outcome Measures: After a 5-year follow-up, the time of detection of recurrence, the recurrence rates, the first method showing recurrence, the mode of recurrence, reresectability, and survival were compared between the groups. Results: The recurrences were identified earlier in the intensified follow-up group than in the conventional follow-up group (mean±SD, 10±5 months vs 15±10 months). The overall recurrence rate was 41%, with 39% in the conventional group and 42% in the intensified group. Carcinoembryonic antigen determination was the most common method showing recurrence in both groups. Endoscopy and ultrasound were beneficial in the intensified follow-up group, but computed tomography failed to improve the diagnostics. The mode of recurrence did not differ between the groups. Radical reresections were performed on 19% (8/43) of the patients, 14% (3/21) in the conventional group and 22% (5/22) in the intensified group. The cumulative 5-year survival was 54% in the conventional group and 59% in the intensified group. Conclusion: Earlier detection of recurrent colorectal cancer by intensified follow-up does not lead to either significantly increased reresectability or improved 5-year survival. (Arch Surg. 1995;130:1062-1067)

A Comparison of Laparoscopic and Open Adrenalectomies

Abstract: Objective: To compare the relative merits of conventional transabdominal and posterior methods with a laparoscopic approach for adrenalectomy. Design: A retrospective cohort study of consecutive series of patients having unilateral adrenalectomy for lesions less than 10 cm in diameter. Setting: University hospital. Patients: Ten patients who underwent laparoscopic adrenalectomy; 11, transabdominal adrenalectomy; and 13, posterior adrenalectomy. Main Outcome Measures: Operative time, estimated blood loss, length of hospital stay, and postoperative parenteral analgesic need. Results: There was no significant difference in the operative time for laparoscopic and anterior adrenalectomy (mean±SD, 212±77 minutes vs 174±41 minutes), but the time for posterior adrenalectomy was significantly shorter (139±36 minutes) ( P P Conclusions: Laparoscopic adrenalectomy may take longer to perform than conventional open approaches but it has clear-cut advantages in shortening postoperative hospital stay and lessening postoperative analgesic requirements. It may be the preferred method for most patients requiring adrenalectomy. (Arch Surg. 1995;130:489-494)

Analysis of morbidity and mortality in 227 cases of esophageal atresia and/or tracheoesophageal fistula over two decades.

Abstract: Objective: This report analyzes the morbidity and mortality in 227 infants (127 boys and 100 girls) with variants of esophageal atresia and/or tracheoesophageal fistula who were treated from 1971 to 1993. Design: Data were collected retrospectively from hospital and office records. Mean follow-up was 76 months, ranging from 1 month to 22 years. Setting: Patients were treated at a tertiary care children's hospital. Results: The mean birth weight was 2557 g (range, 1100 to 4460 g), and the mean gestational age was 38 weeks (range, 28 to 42 weeks). Classification included 29 cases of type A esophageal atresia (13%); two cases of type B (1%), 178 cases of type C (78%), five cases of type D (2%), and 13 cases of type E (6%). Associated anomalies occurred in 146 infants (64%), including cardiac defects in 86 (38%), skeletal defects in 44 (19%), neurological defects in 34 (15%), renal defects in 35 (15%), anorectal defects in 18 (8%), and other abnormalities in 30 (13%). A single-layer anastomosis was performed in 81%, and a two-layer repair, in 17%. Esophagomyotomy was necessary in 9% of the patients. Anastomotic complications included leakage (16%), symptomatic stricture (35%), and recurrent tracheoesophageal fistula (3%). Gastroesophageal reflux was present in 127 cases (58%), with 56 (44%) requiring an antireflux procedure. Tracheomalacia occurred in 32 cases (15%), and 13 required operative treatment. Postoperative esophageal dysmotility was documented in 56 children (30%). The overall survival rate was 95%. The cause of death in 12 patients included severe cardiac anomalies (n=3), fatal sleep apnea (n=1), renal failure (n=1), trisomy 18 (n=2), accidental decannulation of tracheostomy (n=1), pulmonary failure (n=1), and unknown causes (n=3). Conclusions: Early diagnosis, improved surgical technique, neonatal anesthesia, sophisticated ventilatory support, advanced intensive care management, early treatment of associated anomalies, responsiveness of anastomotic strictures to dilatation, and aggressive treatment of gastroesophageal reflux have influenced survival positively. Improved survival rates were noted irrespective of the traditional Waterston criteria, which now seem outdated. With few exceptions, most infants with esophageal atresia and/or tracheoesophageal fistula should survive in the current era. (Arch Surg. 1995;130:502-508)

Tumor cells in blood shed from the surgical field.

Abstract: Objectives: To analyze blood shed from the surgical field during oncologic surgery for tumor cells and to assess functional characteristics of these cells. Design and Patients: Series of 61 patients with cancer who underwent surgery for an abdominal, orthopedic, urological, gynecological, or head and neck malignant tumor, and blinded comparison with 15 patients with benign diseases undergoing surgery. Setting: A 500-bed tumor center and a tertiary care hospital. Main Outcome Measures: Tumor cells were isolated from intraoperatively salvaged and washed blood by density gradient centrifugation. They were identified in cytospin specimens by their content of cytokeratins and nucleolar organizer regions with a sensitivity of 10 cells in 500 mL of blood. Clonogenicity was tested in a cell colony assay; invasiveness, in Boyden chambers; and tumorigenicity, in nude mice. Results: In 57 of 61 patients, tumor cells were detected in the blood shed during oncologic surgery. They demonstrated proliferation capacity, invasiveness, and tumorigenicity. The total number of tumor cells identified ranged from 1×10 1 to 7×10 6 , with no close correlation to the amount of blood loss. Circulating tumor cells were demonstrated in only 26% of these patients and in small numbers. Conclusions: Malignant cells identified regularly in the blood shed during tumor surgery and different from circulating tumor cells are of concern, since at the surgical site they may cause local tumor recurrence, or in the salvaged blood they may cause hematogenic metastasis after retransfusion. Therefore, the contraindication of intraoperative autotransfusion in tumor surgery is strongly supported, and a review of surgical procedures and adjuvant therapy may be indicated, as the passage of the identified cells to the shed blood is yet unknown. (Arch Surg. 1995;130:387-393)

Hospital Mortality of Major Hepatectomy for Hepatocellular Carcinoma Associated With Cirrhosis

Abstract: Objective: To define the safety of major hepatectomy for hepatocellular carcinoma (HCC) associated with cirrhosis and the selection criteria for surgery in terms of hospital mortality. Design: Major hepatectomy for HCC in the presence of cirrhosis is considered to be contraindicated by many surgeons because the reported mortality rate is high (26% to 50%). Previous workers recommended that only selected patients with Child's A status or indocyanine green (ICG) retention at 15 minutes of less than 10% undergo major hepatectomy. A survey was made, therefore, of our patients with HCC and cirrhosis undergoing major hepatectomy between 1989 and 1994. Setting: A tertiary referral center. Patients: The preoperative, intraoperative, and postoperative data of 54 patients with cirrhosis who had major hepatectomy were compared with those of 25 patients with underlying chronic active hepatitis and 22 patients with normal livers undergoing major hepatectomy for HCC. The data had been prospectively collected. Intervention: Major hepatectomy, defined as resection of two or more liver segments by Goldsmith and Woodburn nomenclature, was performed on all the patients. Main Outcome Measure: Hospital mortality, which was defined as death within the same hospital admission for the hepatectomy. Results: Preoperative liver function in patients with cirrhosis was worse than in those with normal livers. The intraoperative blood loss was also higher (P=.01), but for patients with cirrhosis, chronic active hepatitis, and normal livers, the hospital mortality rates (13%, 16%, and 14%, respectively) were similar. The hospital mortality rate for patients with cirrhosis in the last 2 years of the study was only 5%. Patients with cirrhosis could tolerate up to 10 L of blood loss and survive the major hepatectomy. By discriminant analysis, an ICG retention of 14% at 15 minutes was the cutoff level that could maximally separate the patients with cirrhosis with and without mortality. Conclusion: Major hepatectomy for HCC in the presence of cirrhosis is associated with a mortality rate that is not different from the rate for patients with normal livers. An ICG retention of 14% at 15 minutes would serve as a better selection criterion than the 10% previously used. (Arch Surg. 1995;130:198-203)

Increased Tumor Establishment and Growth After Laparotomy vs Laparoscopy in a Murine Model

Abstract: Objective: To test our hypothesis that tumors would be more easily established and grow more aggressively after laparotomy than after laparoscopy. This hypothesis was based on studies that have demonstrated that surgery can suppress immune function and facilitate tumor growth and that have shown preservation of immune function after laparoscopic procedures. Design: Double-blinded, randomized, control trial. Setting: Research laboratory and animal care facility. Animals: One hundred forty 5- to 6-week-old C3H/He female mice. Interventions: Three experiments with three groups each: laparotomy, insufflation, and anesthesia controls. All animals received an intradermal inoculation of tumor cells in the dorsal skin. The anesthesia control cohort underwent no procedure. The laparotomy cohort underwent a midline laparotomy from the xiphoid process to the pubis, which was closed after 30 minutes. The insufflation cohort underwent peritoneal insufflation with carbon dioxide for 30 minutes. Main Outcome Measures: Tumor volume, tumor mass, and incidence of tumor establishment. Results: In the first experiment, the tumor volumes of the anesthesia control and insufflation groups followed a similar pattern of plateau and regression. The tumor volumes of the laparotomy group followed a different pattern and were significantly larger than those of the control and insufflation groups on postoperative days 6 and 12 ( P P P P P Conclusions: Tumors were more easily established and grew more aggressively after laparotomy than after insufflation. These results, coupled with those that demonstrate an immune advantage to laparoscopy over laparotomy, suggest that the difference in observed tumor growth may be related to immune function. While much work remains to be done, we believe these data provide evidence of a previously undemonstrated benefit of laparoscopic intervention. (Arch Surg. 1995;130:649-653)

Tube Thoracostomy: Factors Related to Complications

Abstract: Objective: To determine the complication rate and risk factors associated with tube thoracostomy (TT) in the trauma patient. Design: Retrospective hospital chart review. Setting: Level I trauma center. Patients: Four hundred twenty-six consecutive patients who underwent TT were initially reviewed; 47 deaths occurred unrelated to TT placement. The remaining 379 patients required 599 tubes and composed the study population. Main Outcome Measures: The determination of adverse outcomes related to TT, including thoracic empyema, undrained hemothorax or pneumothorax, improper tube positioning, post–tube removal complications, and direct injuries to the lung. Results: The overall complication rate was 21% per patient. Although complications were not related to the Injury Severity Score, the presence of shock, admission to the intensive care unit, and the need for mechanical ventilation were associated with the increased incidence of complications. There were fewer complications (6%) when the TT was performed by a surgeon compared with TT performed by an emergency physician (13%, P P Conclusions: Tube thoracostomy is associated with significant morbidity. The striking difference in the complication rate between surgeons and other physicians who perform this procedure suggests that additional training may be indicated. (Arch Surg. 1995;130:521-526)

Acute Pancreatitis Induces Intrapancreatic Tumor Necrosis Factor Gene Expression

Abstract: Objective: To examine the intrapancreatic production of tumor necrosis factor (TNF) α and define its cell of origin during acute pancreatitis. Design: Acute necrotizing pancreatitis was induced in adult male mice by administering cerulein (50 μg/kg intraperitoneally four times over 3 hours). Animals were killed at 0, 0.5, 1, 2, 4, 6, and 8 hours, with the severity of pancreatitis established by blind histologic grading and serum amylase, lipase, and TNF levels. The expression of TNF messenger RNA within the pancreas was established by the reverse transcription polymerase chain reaction. Intrapancreatic TNF protein was analyzed by enzyme-linked immunosorbent assay, Western blot, and immunohistochemical methods. Results: Acute pancreatitis was manifest within 1 hour of the first cerulein injection and increased in severity through 8 hours. There was no constitutive expression of TNF messenger RNA within the pancreas, but transcripts were induced within 30 minutes following the onset of pancreatitis, increasing through 4 hours. Intrapancreatic and serum TNF peptide levels became detectable at 1 hour and increased over 6 hours (both P P P Conclusions: Tumor necrosis factor gene expression is induced locally during acute pancreatitis, resulting in large amounts of intrapancreatic TNF with levels consistently higher than those found in the serum. The overall rise in both tissue and serum TNF concentrations correlates directly with the severity of pancreatic damage and inflammation. The infiltrating macrophage appears to contribute most to this process. (Arch Surg. 1995;130:966-970)

Primary vs Secondary Iliopsoas Abscess: Presentation, Microbiology, and Treatment

Abstract: Objective: To review the characteristics of patient presentation, microbiology, and treatment of primary iliopsoas abscess. Design: A case series of patients with iliopsoas abscess diagnosed on computed tomographic scans from 1987 to 1994. Setting: Tertiary care inner-city university hospital. Patients: Eleven patients with secondary iliopsoas abscess, defined as being secondary to gastrointestinal or genitourinary causes or trauma, and seven patients with primary abscess, defined as the absence of the above causes. Main Outcome Measures: Patient characteristics, presenting symptoms and signs, microbiologic characteristics, treatment, and clinical course of patients with primary iliopsoas abscesses compared with those in patients with secondary abscesses. Results: In the primary group, six patients (86%) were intravenous drug users and four (57%) were positive for human immunodeficiency virus.Staphylococcus aureusgrew from cultures from five of seven patients with primary abscesses, whereas secondary abscesses had enteric flora. The typical patient presentation included fever, with complaints of pain in the flank, hip, or abdomen. Comparison of abscess drainage options showed shorter hospitalizations for surgical drainage than for percutaneous drainage (15.9 vs 28.5 days;P≤.01). Conclusions: A patient who presents with pain in the flank, hip, or abdomen may have a primary iliopsoas abscess. Computed tomography is the standard method of diagnosis. Antibiotic regimens for patients with primary iliopsoas abscess should include coverage forS aureus, and patients with secondary abscesses should have antibiotic regimens tailored for enteric bacteria. Drainage of abscess is essential for appropriate treatment, and surgical drainage is superior to percutaneous drainage in achieving prompt recovery. (Arch Surg. 1995;130:1309-1313)

Aggressive Surgical Resection for Cholangiocarcinoma

Abstract: Objectives: To review the spectrum of cholangiocarcinoma in patients treated by a single team of hepatobiliary surgeons over an 8-year period, to evaluate the predictors of survival, and to assess the results of an aggressive approach to surgical resection. Design: Retrospective review of all clinical records of patients referred for treatment of cholangiocarcinoma, with univariate analysis of clinical and pathologic factors in relation to patient survival. Setting: New England Deaconess Hospital, Boston, Mass. Patients: Eighty-eight consecutive patients referred with the established diagnosis of cholangiocarcinoma, from December 31, 1985, to April 15, 1994. Interventions: Seventy-five of 88 patients were treated surgically, with 59 undergoing major resection for cure. Of the 29 patients treated palliatively, 16 had operations and 13 had wire mesh stents placed nonoperatively. Main Outcome Measures: Morbidity, mortality, and patient survival. Results: Survival correlates directly with the pathologic stage (TNM). Tumor location had no impact on survival. Patients undergoing resection survived significantly longer (median, 23.2 months) than palliated patients (median, 7.7 months; P=.0015). Nonoperative palliation resulted in better survival than surgical palliation (P=.045). Major hepatic resection was used alone in eight patients with predominating intrahepatic lesions, while 18 patients with hilar lesions underwent en bloc skeletonization in conjunction with major hepatic resection. Resection with microscopically free margins significantly improved survival. Only patients undergoing major resection enjoyed survival greater than 2 years. Conclusions: Patient survival can be significantly improved by aggressive surgical resection. Hepatic resection should be used aggressively to achieve disease-free margins to optimize survival. Hepatic resection can be performed with low morbidity and mortality. Liver transplantation should be avoided as a treatment for cholangiocarcinoma. The best palliation for unresectable disease remains debatable. We advocate nonoperative treatment with endobiliary expandable wire mesh stents for patients with unresectable disease. (Arch Surg. 1995;130:270-276)

Do Prehospital Trauma Center Triage Criteria Identify Major Trauma Victims

Abstract: Objective: To evaluate anatomic, physiologic, and mechanism-of-injury prehospital triage criteria as well as the subjective criterion of provider "gut feeling." Design: Prospective analysis. Setting: A state without a trauma system or official trauma center designation. Patients: Patients treated by emergency medical services personnel statewide over a 1-year period who were injured and met at least one prehospital triage criterion for treatment at a trauma center. Main Outcome Measures: Outcome was analyzed for injury severity using the Injury Severity Score and mortality rates. A major trauma victim (MTV) was defined as a patient having an Injury Severity Score of 16 or greater. The yield of MTV and mortality associated with each criterion was determined. Results: Of 5028 patients entered into the study, 3006 exhibited a singular entry criterion. Triage criteria tended to stratify into high-, intermediate-, and low-yield groups for MTV identification. Physiologic criteria were high yield and anatomic criteria were intermediate yield. Provider gut feeling alone was a low-yield criterion but served to enhance the yield of mechanism of injury criteria when the two criteria were applied in the same patient. Conclusions: A limited set of high-yield prehospital criteria are acceptable indicators of MTV. Isolated low- and intermediate-yield criteria may not be useful for initiating trauma center triage or full activation of hospital trauma teams. (Arch Surg. 1995;130:171-176)

Fibrin glue sealing for the prevention of pancreatic fistulas following distal pancreatectomy

Abstract: Objective: To evaluate the use of fibrin glue sealing of the pancreatic stump for the prevention of postoperative pancreatic fistulas. Design: A prospective, randomized clinical trial. Patients and Methods: Fibrin glue is a biologic adhesive consisting of highly concentrated human fibrinogen, thrombin, and factor VIII. Twenty-six of 56 patients who underwent distal pancreatectomy for gastric cancer or pancreatic disease were randomly assigned to the fibrin glue group. Fibrin glue was applied to the suture line of the pancreatic stump with the ligated main pancreatic duct. Pancreatic fistula was defined as a pancreatic fluid discharge for over 7 postoperative days diagnosed by local findings, with amylase concentration in the discharge fluid more than three times the serum amylase concentration, a level low enough that even a small pancreatic leakage could be diagnosed. Results: The overall incidence of pancreatic fistula was 28.6%. Postoperative pancreatic fistulas occurred in four patients (15.4%) in the fibrin glue group and 12 (40.0%) in the control group ( P =.04). The lower pancreatic fistula rate was seen in the fibrin glue group also when analyzing patients with gastric cancer or pancreatic disease only, although there was no statistically significant difference. Conclusions: Intraoperative use of fibrin glue following distal pancreatectomy could prevent pancreatic fistula formation. This method was feasible, safe, and reliable and will complement other prophylactic methods. (Arch Surg. 1995;130:952-955)

Familial Nonmedullary Thyroid Cancer: An Emerging Entity That Warrants Aggressive Treatment

Abstract: Objective: To determine whether familial nonmedullary thyroid carcinoma behaves like sporadic carcinoma of follicular cell origin. Design: Retrospective review. Setting: University medical center. Patients: Fourteen patients were treated for familial nonmedullary thyroid carcinoma between 1980 and 1994. Thirteen families were identified, with 30 affected individuals. Interventions: Patients were treated with total or completion total thyroidectomy. Thirteen additional operations were performed to control recurrent disease. Main Outcome Measures: Stage, recurrence, and survival. Patients were followed up for a mean of 6.5 years. Results: In our 14 patients, 13 tumors were multifocal, and six of these were bilateral. The incidences of lymph node metastasis and local invasion were both 57% (n=8). Seven patients (50%) had recurrences during follow-up. Conclusions: Familial nonmedullary thyroid carcinoma has a high incidence of multifocality and invasion and a high rate of local recurrence. Aggressive initial treatment and careful follow-up seem to be indicated. (Arch Surg. 1995;130:892-899)

Postcholecystectomy bile duct strictures. Management and outcome in 130 patients.

Abstract: Objective: To evaluate management strategies for the treatment of patients with postcholecystectomy bile duct strictures. Design: Retrospective study. Setting: The Hepatobiliary Unit of Hammersmith Hospital, London, England. Patients: One hundred thirty consecutive patients referred for treatment of postcholecystectomy bile duct strictures. The majority (80 patients [61.5%]) had undergone multiple operative procedures before referral, and 81 (62.3%) had undergone at least one previous stricture repair. At referral, more than half of the patients had a stricture involving the confluence of the bile ducts (n=78 [60%]), and 23 (17.7%) had evidence of portal hypertension. Main Outcome Measures: Perioperative mortality, stricture recurrence, and long-term outcome. Results: One hundred twenty-two patients (94%) underwent operative treatment: 110, stricture repair alone; four, portosystemic shunt and stricture repair; and eight, miscellaneous operative procedures. Among the 110 patients treated by stricture repair alone, there was an operative mortality rate of 1.8% (n=2), and 79 patients (76%) had a good result, with no biliary symptoms and no need for intervention during mean follow-up of 7.2 years (range, 1 to 13 years). Twenty-two patients (21%) required either radiological intervention or operative revision of the biliary-enteric anastomosis, but 11 (50%) of these patients subsequently did well and had no biliary symptoms. Thus, 90 patients (87%) had a good or excellent long-term result after initial or follow-up treatment. There were no deaths among the 108 patients who underwent stricture repair alone by direct suture techniques. Factors influencing mortality included hypoalbuminemia, an elevated serum bilirubin level, and the presence of liver disease and portal hypertension. Preoperative factors influencing failure of the stricture repair in long-term follow-up included discontinuity of the right and left ducts at the time of stricture repair (Bismuth grade 4) and three or more previous attempts at operative repair before referral to our center. Conclusions: Operative repair of bile duct strictures using direct sutured techniques remains the procedure with which alternative methods will need to be compared, with close attention to long-term outcome. (Arch Surg. 1995;130:597-604)

Biliary tract complications after liver transplantation

Abstract: Objective: To assess the incidence, type, and treatment of biliary complications after orthotopic liver transplantation. Design: Case series. Setting: Tertiary referral center. Patients: One hundred ninety consecutive adults who underwent 220 orthotopic liver transplantations with biliary reconstruction between January 1, 1989, and December 31, 1993, with follow-up of all survivors to May 1994. Main Outcome Measures: Incidence, type, and treatment of biliary complications. Results: Biliary complications were identified in 65 of the 190 patients who underwent biliary reconstruction (49 of 147 with choledochocholedochostomy and 16 of 43 with Roux-en-Y choledochojejunostomy). The group with complications who had choledochocholedochostomy had 32 biliary leaks (22 T-tube related), 11 strictures or obstructions, and six cases of choledocholithiasis. Twelve percent of choledochocholedochostomies were converted to choledochojejunostomies, while 26 of 49 biliary complications in patients who had choledochocholedochostomies were treated nonoperatively. Elective removal of T tubes resulted in biliary leak in 15 of 89 patients, treated nonoperatively in 12. Leaks (unrelated to scheduled removal of the tube) occurred earlier than strictures (choledochocholedochostomy, mean±SEM 25.6±5.8 vs 184.7±61.0 days; choledochojejunostomy, 13.4±4.4 vs 521.0±142.0 days) and were more often treated operatively (choledochocholedochostomy, 14 of 17 vs three of seven; choledochojejunostomy, four of five vs three of eight). Three deaths were associated with early biliary leaks, all in patients with preexisting multiorgan dysfunction. There was no significant difference in the incidence of biliary complications by type of reconstruction, year of transplantation, age, UNOS (United Network for Organ Sharing) status, preservation time, or indication for transplantation. Conclusions: Biliary complications are common after orthotopic liver transplantation but are rarely an isolated cause of death. Stenting of the choledochocholedochostomy or choledochojejunostomy anastomosis does not prevent strictures, and T tubes are associated with a high incidence of biliary leakage on removal. Nonoperative interventions have an increasing role in the treatment of biliary complications. (Arch Surg. 1995;130:312-317)

Increases in Intra-abdominal Pressure Affect Pulmonary Compliance

Abstract: Objectives: To determine the effect of increased intra-abdominal pressure (IAP) on pulmonary compliance and to determine an effective means to measure IAP. Design: A prospective study. Setting: An urban tertiary care hospital. Patients: Twenty-six adult patients undergoing laparoscopic cholecystectomy. Interventions: Intra-operative management of laparoscopic cholecystectomy requiring endotracheal intubation with general anesthesia, nasogastric and urinary bladder catheters, and position changes. Additional interventions included use of a rectal manometer and a respiratory pressure module inserted within the ventilator circuit. Main Outcome Measures: Correlation of changes in IAP with changes in dynamic pulmonary compliance, measured as tidal volume/(end inspiratory pressure—end expiratory pressure) and comparison of three different measurement techniques (bladder, rectal, and gastric) with a standard technique (insufflation pressure) in three different positions (supine, Trendelenburg's, and reverse Trendelenburg's Results: Compliance was significantly related to insufflation pressure (P Conclusions: Increased IAP has a major influence on pulmonary compliance (50% decrease at 16 mm Hg). Measurements of IAP by intraorgan manometry are position dependent and may not accurately reflect the intraperitoneal pressure. (Arch Surg. 1995;130:544-548)

Incidence, Complications, and Management of Meckel's Diverticulum

Abstract: Objectives: To elucidate the incidence of Meckel's diverticulum; to determine the correlation between the histologic type of the diverticulum's mucosa and its clinical presentation; and to review our experience with Meckel's diverticulum. Design: Case-control study. Setting: University hospital in loannina, Greece. Patients: A total of 2074 patients undergoing an appendectomy were examined intraoperatively for Meckel's diverticulum (subgroup A 1 ). In addition, Meckel's diverticulum was found incidentally in 15 patients who were undergoing intra-abdominal surgery (subgroup A 2 ). An operation for a complicated Meckel diverticulum was performed in an additional 15 patients (group B). Results: Thirty-three (1.59%) Meckel diverticulae were found incidentally among 2074 patients in subgroup A 1 . A positive correlation between Meckel's diverticulum and male sex was found ( P 1 ( P P Conclusion: Resection of the unexpected Meckel diverticulum can be performed safely with a low complication rate, regardless of the patient's age. (Arch Surg. 1995;130:143-146)

Patient satisfaction following laparoscopic and open antireflux surgery.

Abstract: Objective: To compare laparoscopic (LNF) with open Nissen fundoplication (ONF) in terms of hospital charges, efficacy, and patient satisfaction Design: A prospective, nonrandomized study with a median follow-up of 370 days Setting: Two tertiary care university hospitals Patients: Eighty-six patients with complications of gastroesophageal reflux who had not had previous antireflux surgery were studied Patients chose ONF or LNF following discussion with the surgeon; 12 underwent ONF and 74 underwent LNF, of whom eight required conversion to laparotomy Main Outcome Measures: Hospital charges, disability, satisfaction, and side effects of fundoplication Results: Patients were demographically similar Total charges (mean±SD) for LNF ($11 673±$4723) were significantly less than for ONF ($18 394±$17 264) Patients who underwent LNF returned to work sooner (10±3 days) than those who underwent ONF (28±1 days) Bloating, dysphagia, and recurrent heartburn occurred with equal frequency in both groups Recurrent reflux occurred in four of 74 LNF patients and one of 12 ONF patients Overall satisfaction scores were similar, irrespective of operative technique (LNF, 335±087; ONF, 350±094 Conclusions: Laparoscopic Nissen fundoplication is as effective as ONF in the treatment of complications of gastroesophageal reflux disease and appears to cost less and lead to faster recovery from surgery, but does not result in higher patient satisfaction than ONF The most important factor in patient satisfaction is the abolition of preoperative symptoms rather than the type of operation (Arch Surg 1995;130:289-294)

Postoperative Intestinal Motility Following Conventional and Laparoscopic Intestinal Surgery

Abstract: Objectives: To evaluate the influence of conventional vs laparoscopic right-sided colectomy on postoperative motility of the stomach, small bowel, and large bowel, and to study the effects on postoperative intestinal motility of anesthesia only and of laparotomy vs laparoscopy only Design: Prospective randomized study in a canine model Setting: Animal research laboratory Animals: Twelve mongrel dogs weighing 234 to 296 kg Interventions: (1) Laparotomy or laparoscopy and (2) 1 week after complete recovery from the first procedure, right-sided colectomy with ileocolic anastomosis using conventional or laparoscopic techniques End points of the study were (1) the postoperative time until the electrical response activity of the stomach, small intestine, and large intestine returned to a normal interdigestive pattern and (2) the time until first postoperative bowel movement occurred Results: Median time until return to normal interdigestive pattern of myoelectrical activity after laparoscopic colectomy was about 40% less than after conventional colectomy (P Conclusions: Recovery from postoperative ileus following laparoscopic surgery is more rapid than after conventional surgery in the canine model Confirmatory human studies should be performed (Arch Surg 1995;130:415-419)

Incidence and Management of Pancreatic and Enteric Fistulas After Surgical Management of Severe Necrotizing Pancreatitis

Abstract: Objective: To determine the incidence, type, and outcome of complications of necrotizing pancreatitis. Setting: Major tertiary referral center (Mayo Clinic, Rochester, Minn). Patients: Sixty-one patients seen from 1985 to 1994 who underwent surgical management of severe necrotizing pancreatitis and who developed pancreatic or gastrointestinal fistulas. Main Outcome Measures: Incidence, management, and outcome of pancreatic and gastrointestinal fistulas. Results: Twenty-five patients (41%) developed pancreatic (14 patients) and/or gastrointestinal tract cutaneous (19 patients) fistulas. While three duodenal fistulas and one colonic fistula were recognized at the initial operation for pancreatic necrosectomy, the remainder developed 4 to 60 days after the initial operation. Spontaneous closure occurred in nine of 14 pancreatic, two of two gastric, two of four enteric, two of eight colonic, and four of five duodenal fistulas. Mortality of the group with fistulas was 24% (6/25) and was not different from the mortality of the patients with necrotizing pancreatitis without fistulas (28% [10/36]). Conclusions: Pancreatic and gastrointestinal tract fistulas are common complications of surgical treatment of severe necrotizing pancreatitis. Well-controlled gastric, pancreatic, and enteric fistulas have the greatest likelihood of spontaneous closure. Duodenal and colonic fistulas may need surgical intervention for control or repair. Mortality in these patients parallels the mortality for severe necrotizing pancreatitis. (Arch Surg. 1995;130:48-52)

Long-term Morbidity Following Jejunoileal Bypass: The Continuing Potential Need for Surgical Reversal

Abstract: Objective: To review the late sequelae of jejunoileal bypass (JIB) and the potential role of late surgical reversal in ameliorating morbidity and mortality following JIB. Design: Patients who underwent JIB between 1965 and 1977 were contacted and pertinent health-event information was gathered. Early sequelae were defined as disorders occurring within the first 2 years after JIB; late sequelae were those occurring after 2 years. Health events occurring between () and 23 years after JIB were documented. Setting: A private, tertiary referral center. Patients: Patients underwent JIB for morbid obesity that had failed medical and/or psychiatric interventions. Main Outcome Measures: Body mass index (BMI) (weight in kilograms divided by the square of the height in meters), diarrhea, electrolyte imbalance, acute and chronic liver disease, renal disease, JIB reversal, reason for JIB reversal, death, and cause of death. Results: A total of 453 morbidly obese patients underwent JIB. By 2 years following JIB, the mean (±SD) BMI dropped from 49.3±8.1 to 31.1±0.8 and remained at this level until year 15, after which weight gradually increased (BMI, 35.4±3.1). The most severe early complication was acute liver failure, which occurred in 7% of patients and caused seven deaths. At 15 years, the actuarial probability of the most common serious late complications related to JIB were renal disease (37%), with two deaths; diarrhea (29%); and liver disease (10%), with three deaths. One hundred thirty-eight patients (31%) had a bypass reversal. The most common indications for reversal were diarrhea and electrolyte disturbance (29%), renal disease (19%), and liver disease (17%). Fifty-six patients died more than 30 days after JIB: 64% before JIB reversal, 13% at the time of reversal, and 23% subsequently. Conclusions: Jejunoileal bypass is associated with progressive accrual of serious, sometimes life-threatening complications. Lifelong follow-up for early diagnosis and surgical reversal before life is threatened should reduce the morbidity and mortality associated with this procedure. (Arch Surg. 1995;130:318-325)

Inflammatory response in peritoneal exudate and plasma of patients undergoing planned relaparotomy for severe secondary peritonitis

Abstract: Objective: To study the pattern of intraperitoneal cytokine release in secondary peritonitis and its correlation with plasma levels and prognosis. Design: Noncomparative descriptive case series. Setting: Department of surgery in a university hospital. Patients: Seventeen consecutive patients undergoing planned relaparotomy for severe intra-abdominal infection (Acute Physiological and Chronic Health Evaluation [APACHE II] score >10; mean score, 17.5). Interventions: The following were measured at the first and last serial operations in the peritoneal exudate and plasma: endotoxin, tumor necrosis factor α (TNF-α), interleukin-1 (IL-1), interleukin-6 (IL-6), elastase, and neopterin. Main Outcome Measures: Survival and death. Results: Six patients died. Peritoneal endotoxin levels were significantly higher than in the plasma and were significantly higher in the nonsurvivors. Plasma TNF-α, IL-6, elastase, and neopterin levels remained elevated in the nonsurvivors prior to death. Levels of TNF-α, IL-6, elastase, and endotoxin were 19, 993, 239, and 7 times higher, respectively, in the peritoneal exudate than in plasma, all significant differences. Elastase and TNF-α levels decreased in survivors during the operative treatment but remained elevated in the nonsurvivors. Conclusions: Secondary peritonitis is associated with a significant cytokine-mediated inflammatory response that is compartmentalized in the peritoneal cavity and indicates an adverse prognosis. Levels of cytokines in the exudate of peritonitis may be used to better stratify the severity of peritonitis and, in future, to guide local therapy. (Arch Surg. 1995;130:1314-1320)

Pylorus-Preserving Whipple Resection for Pancreatic Cancer: Is It Any Better?

Abstract: Objective: To compare the short- and long-term morbidity and mortality rates of the standard Whipple pancreatoduodenectomy (SW) and its pylorus-preserving modification (PPW) in patients with malignant periampullary disease. Design: Retrospective medical record review and quality of life assessment by telephone interview. Setting: University medical center. Study Participants: Sixty-seven patients who underwent pancreatoduodenectomy (52 SW and 15 PPW) from June 1988 to January 1994. Intervention: The SW and PPW. Main Outcome Measures: Operative features and short- and long-term complications were analyzed with respect to the type and stage of cancer and the kind of pancreatic resection. Mean follow-up was 32 months (range, 1 to 5 years). Results: The operative mortality rate for all patients who had a pancreatic resection was 1.5%. The diagnoses in the PPW vs SW groups were pancreatic cancer (four vs 27 patients), ampullary cancer (six vs seven patients), duodenal cancer (zero vs six patients), and bile duct cancer (five vs one patient). Operative mortality rates (0% vs 1.55%) and operative times (2 minutes longer for SW) were similar. Delayed gastric emptying (61% vs 41%) was more common in the PPW group, resulting in a longer hospitalization (24 vs 18 days) and a greater cost in the PPW group (P=.04). In the PPW group, a mean of five lymph nodes was removed compared with 10 in the SW group (P=.04). Conclusions: The data provided no evidence of any advantage for the PPW in patients with malignant periampullary tumors. We continue to advocate the SW for pancreatic cancer. (Arch Surg. 1995;130:838-843)

A 15-Year Review of Esophagectomy for Carcinoma of the Esophagus and Cardia

Abstract: Objective: To evaluate the effect of surgical approach and adjuvant therapy on patients with carcinoma of the esophagus and/or cardia. Design: Retrospective analysis of 157 consecutive patients who underwent esophagectomy. Setting: A private university medical center and its affiliated community hospital. Patients: One hundred twenty men and 37 women (mean age, 61.7 years) with carcinoma of the esophagus and/or cardia that was surgically treated between 1978 and 1993. Interventions: Three approaches were used for resection: Transhiatal esophagectomy (THE) (n=67), transthoracic esophagectomy (TTE) (n=71), and abdominalonly esophagectomy (AOE) (n=19). Sixty-five patients received adjuvant radiotherapy and chemotherapy. Main Outcome Measures: Surgical mortality, morbidity, and survival and the effect of adjuvant therapy. Results: The overall surgical mortality rate was 7.6%: 12.7% with the TTE, 4.5% with the THE, and 0% with the AOE approach. A significantly increased incidence of adult respiratory distress syndrome (P Conclusions: The THE and AOE approaches have fewer early complications than does TTE. Both THE and TTE have equal long-term survival rates. Adjuvant therapy provides increased survival to node-positive patients with carcinoma of the esophagus and/or cardia. (Arch Surg. 1995;130:617-624)

Comparison of Medical and Minimally Invasive Surgical Therapy for Primary Esophageal Motility Disorders

Abstract: Objective: To compare medical with minimally invasive surgical therapy in the treatment of primary esophageal motility disorders. Design: Prospective study. Setting: University-based tertiary care center. Patients: Eighty-nine patients (46 men and 43 women) with either achalasia or nutcracker esophagus and diffuse esophageal spasm (DES). Choice of treatment was based not on randomization but on the preference of the referring physician, the patient's choice, and/or the patient's eligibility to access the University of California, San Francisco, for treatment. Interventions: Nineteen patients with achalasia and 30 patients with nutcracker esophagus and DES were treated with dilatations and/or medications. Thirty patients with achalasia and 10 with nutcracker esophagus and DES underwent a thoracoscopic myotomy. Main Outcome Measures: Dysphagia, pain, and overall quality of life. Results: In the surgical group, 80% of the patients with nutcracker esophagus and DES and 87% of the patients with achalasia had good or excellent results. In contrast, in the medical group, 26% of the patients with nutcracker esophagus and DES and 26% of the patients with achalasia had good or excellent results. Conclusions: Surgery by minimally invasive techniques offers a better chance than does medical therapy or dilatation of rendering the patient with achalasia, nutcracker esophagus, and DES asymptomatic. (Arch Surg. 1995;130:609-616)

Planned relaparotomy vs relaparotomy on demand in the treatment of intra-abdominal infections. The Peritonitis Study Group of the Surgical Infection Society-Europe.

Abstract: Objective: To define the role of planned relaparotomy (PR) in the treatment of intraperitoneal infection, compared with that of relaparotomy on demand (RD). Design: Case-control study on the basis of a prospective multicenter cohort analytic study. Statistical evaluation was done by the McNemar test for qualitative data and the Wilcoxon matched-pairs signed rank test for qualitative data. Setting: Eighteen hospitals of different care levels in Austria, Germany, and Switzerland. Patients: Thirty-eight of 42 patients with intraabdominal infections who underwent PR were matched for APACHE II (Acute Physiology and Chronic Health Evaluation II) score, age, cause of infection, site of origin of peritonitis, and the ability of the surgeon to securely eliminate the source of infection with 38 patients taken from a cohort of 278 undergoing RD. Interventions: Planned relaparotomy was defined as at least one relaparotomy decided on at the time of the first surgical intervention; RD, relaparotomy indicated by clinical findings. Main Outcome Measures: Mortality and incidence of postoperative multiple organ failure and infectious complications. Results: There was no significant difference in mortality between patients treated with PR (21%) or RD (13%). Postoperative multiple organ failure as defined by a Goris score of more than 5 was more frequent in the group of patients undergoing PR (50%), compared with the group undergoing RD (24%) ( P =.01), as were infectious complications (68% vs 39% [ P =.01]). Infectious complications were due to more frequent suture leaks (16% vs 0% [ P =.05]), recurrent intra-abdominal sepsis (16% vs 0% [ P =.05]), and septecemia (45% vs 18% [ P =.05]) in the PR vs the RD groups. The incidence of other complications was not different in the two groups. Conclusions: Until larger prospective studies are available, the indication for PR should be evaluated with caution. (Arch Surg. 1995;130:1193-1197)