Facilitated vertical gastric pouch construction for gastric bypass.

BACKGROUND: Efforts to simplify and reduce surgical risk of gastric pouch construction in super obese patients have led to technical changes which are described. METHODS: Operative technique of vertical gastric pouch construction and hand-sewn Roux gastrojejunostomy are described. RESULTS: 171 patients with average BMI = 55 have undergone gastric bypass with the modified technique, a majority with double stapling of the gastric pouch. Staple-line dehiscence is reduced and operative morbidity is limited. CONCLUSIONS: Technical changes described have shortened operative time, reduced intraoperative frustrations, and produced limited postoperative morbidity.

Economics of pancreatoduodenectomy in the elderly.

BACKGROUND: Managed care and the increasing percentage of surgical procedures performed in the elderly have renewed the focus on hospital charges and expenditures. The objective of this study was to determine whether septuagenarians and octogenarians accrue more hospital charges or have a higher risk of morbidity and death. METHODS: We retrospectively reviewed the charges and pertinent clinical outcomes data that were available on 70 of the last 100 pancreatoduodenectomies performed at our institution (1989 to 1994). Charges from four cost centers were analyzed and normalized to 1995 dollars by using the Consumer Price Index and Wilcoxon rank sum test. Patients were divided into two groups: group 1, 70 years of age or older (n = 21); group 2, younger than 70 years of age (n = 49). RESULTS: Anesthetic charges were $2657 +/- $835 for group 1 versus $2815 +/- $826 for group 2, which was not a statistically significant difference. Laboratory charges were $4650 +/- $3284 for group 1 versus $5969 +/- $5169 for group 2, which was not a significant difference. Pharmaceutical charges were $5424 +/- $4435 for group 1 versus $9243 +/- $9695 for group 2, which was not a significant difference. Charges for operative units were $6198 +/- $1671 for group 1 versus $7469 +/- $2116 for group 2, p < 0.02. Total charges were $41,180 +/- $20,635 for group 1 versus $50,968 +/- $33,783 for group 2, which was not a significant difference. No difference was noted in morbidity, mortality, length of stay, or survival. CONCLUSIONS: Pancreatoduodenectomy in the elderly can be performed safely without accruing higher cost, increased morbidity, or increased mortality.

Results of surgery: long-term effects on hyperlipidemia.

To define frequency of lipid abnormalities and to monitor improvement or correction of those abnormalities postoperatively, 66 patients with chronic morbid obesity had total cholesterol, high-density-lipoprotein (HDL) cholesterol, and triglyceride determinations preoperatively and at staged intervals up to 5-7 y after Roux-Y gastric bypass. Preoperative abnormal HDL-cholesterol and triglyceride concentrations were frequent. Major improvements occurred in these lipid concentrations by 6 mo postoperatively, and some further improvements occurred with additional weight loss at 1 y. At 5-7 y, among 33 patients, raised concentrations of HDL cholesterol were upheld in women (P less than 0.01); reductions in triglycerides were maintained in men (P less than 0.025); and reduced total cholesterol:HDL-cholesterol, which was achieved by 6 mo, was sustained in both men and women (P less than 0.01). In comparing lipid profiles of gastric surgery through 5 y with recent data from the surgical arm of the Program on Surgical Control of the Hyperlipidemias (POSCH), postulates are made of anticipated reduction in morbid, and even fatal, cardiac events in the operated morbidly obese population.

Perianal Paget's disease. Classification and review of management: report of two cases.

Fewer than 100 cases of Paget's disease located in the perianal skin have been reported since the extramammary location was first described in 1893. Two patients seen and treated in the authors' institutions with disease limited to the epidermis and its adnexae are described to illustrate the usual presentation and pathobiology of the disease. A staging classification based on the disease pathology has been developed from the cases reported in the literature and correlated with appropriate surgical treatment. Stage I disease treated with wide local excision promises unlimited survival, whereas the prognosis worsens for Stage II, with synchronous localized malignancies, and for Stages III and IV, with regional and distant metastatic disease, respectively.

Sustained coronary-risk-factor reduction after gastric bypass for morbid obesity.

Lipids and clinical changes including diabetes and hypertension were monitored in morbidly obese patients after Roux-Y gastric bypass. Total cholesterol (Chol), high-density-lipoprotein (HDL) cholesterol, and triglycerides at 1 and at 5-7 y postoperatively in 33 patients and at 1 y in 23 patients (including apolipoproteins A-I and B) were compared with preoperative concentrations. Mean concentrations of Chol and both apolipoproteins were unchanged. Elevated serum triglycerides became normal, and reduced concentrations persisted at 5-7 y in men (p less than 0.025). HDL-cholesterol concentrations increased at 1 y (p less than 0.01) and remained higher at 5-7 y in women. Ratios of Chol to HDL cholesterol were lower at 1 y (p less than 0.01) in both men and women. Diabetes (9 patients) and hypertension (22 patients) also were reduced at 1 y (p less than 0.01) and remained lower at 5-7 y. A mean 61% of excess weight was lost in 1 y whereas a 12% weight gain occurred by 5-7 y. The beneficial changes in most coronary risk factors lasted 5-7 y after surgery.

The cause of coagulopathy after peritoneovenous shunt for malignant ascites.

Thirty-five patients with malignant ascites who received a peritoneovenous shunt were studied to determine the type and duration of postoperative coagulopathy. Coagulation factors were measured before and on the first and third day after the placement of a Denver peritoneovenous shunt; 1 to 10 L of ascites was removed at operation. Levels of platelets, antithrombin III, plasminogen, antiplasmin, fibrinogen, and factors V and VIII decreased by the first postoperative day but did not change further through the third day. The levels of fibrinolytic split products increased on day 1 but were lower by day 3. The platelet count reduction by the third day correlated with the hematocrit change (-0.031). The prothrombin and activated partial thromboplastin times remained normal postoperatively. The patterns of change were similar for patients with positive (n = 18) and negative (n = 17) ascites cytologic findings, with elevated (n = 24) and normal (n = 11) preoperative fibrinolytic split product levels, and elevated bilirubin value (greater than 25 mumol/L; n = 9), and no jaundice (n = 26). Bleeding did not occur. The data indicated that plasminogen-rather than thromboplastin-activated fibrinolysis occurred and that platelet reduction was largely dilutional. The reactions were not progressive when ascites was removed operatively.

Effects of vagal cryo-interruption on colon contractions in monkeys.

The influence of the vagus nerve on normal colon contractions, and the distal extent of this influence along the colon, is unknown. It is generally thought that the vagus nerve has little, if any, influence on normal colon contractions, and if there is any effect, it is limited to the proximal colon. The technique of reversible truncal vagotomy by cooling (vagal cryo-interruption) was used to explore the influence, and extent of any influence, of the vagus nerve on contractions of the colon in unanesthetized monkeys. Each monkey was subjected to vagal cryo-interruption in the fasted and fed states. In both states this caused a significant decrease in contractions in the proximal, middle, and distal portions of the colon, compared with baseline control and recovery periods. These studies indicate that the vagus nerve plays a role in normal contractions of the intra-abdominal colon in both the fasted and the fed states and that it also plays a role in the gastrocolic response.

Preoperative coagulation indices in patients with morbid obesity.

Preventive measures against venous thromboembolism are used frequently in surgical patients with morbid obesity because the risk has been claimed to be higher than usual. We addressed this risk in 81 morbidly obese patients by measuring preoperative plasma proteins associated with coagulation and by correlating variations in these indices to excess weight, liver histology, patient sex, tobacco use and serum triglycerides. Plasma levels of antithrombin III and plasminogen were normal in these patients. Plasma fibrinogen concentrations were elevated in 34 percent of patients; the values did not relate to excess weight and they correlated negatively (P less than 0.04) with increasing serum triglycerides and grades of liver fat content. Premenopausal women had higher plasminogen and fibrinogen levels than menopausal women or men, although their mean levels were normal. Smokers had lower plasminogen levels than nonsmokers (P less than 0.01). However, none of the measured levels of preoperative coagulation proteins identified increased risk for venous thrombosis in cohorts of this population, and the coagulation indices were not different from those reported in normal weight patients.

Expedient surgical treatment of chronic ulcer stenosis. A case for proximal gastric vagotomy.

Forty-nine patients with chronic pyloroduodenal ulcer stenosis were treated surgically in our hospitals between 1977 and 1985. Three operations were used: 16 patients had a proximal gastric vagotomy-pyloroplasty (PGV-P); 26 had a vagotomy-antrectomy (V-A); and seven had a truncal vagotomy-pyloroplasty (TV-P). Historical data and preparations for operation were similar in each group. We compared the early clinical outcome in these patients. Two patients died after V-A and one died after TV-P. Five patients were reoperated: V-A (three patients), PGV-P and TV-P (one patient each). Delayed gastric emptying with nonoperative resolution (DGE) occurred in nine patients after V-A and in two patients after TV-P. Preoperative gastric suction and parenteral nutrition did not avoid DGE or affect timing of return to a general diet. A solid diet was resumed earlier (p less than 0.01) after PGV-P (5 +/- 1 days) than after V-A (13 +/- 8 days) or TV-P (9 +/- 2 days). Ninety-four percent of patients after PGV-P had an uneventful recovery, compared with 46 and 43% after V-A and TV-P, respectively. Among the variables measured, operative choice of PGV-P expedited early surgical recovery of patients with chronic ulcer stenosis.

Truncal vagotomy as a possible potentiator of gastric atony.

To stimulate ulcer patients undergoing operation for gastric outlet stenosis, pyloric obstruction was created in dogs and repaired with pyloroplasty to which was added truncal vagotomy, proximal gastric vagotomy, or no vagotomy. Gastric antral contractile activity after feeding a solid meal was studied before and after repair (2 week period of study). This activity was correlated with the initial lag and regulated phases of solid meal emptying. Five quantified indices of contractile activity measured during the first postprandial hour indicated variable and inconclusive results in the antrum during the lag phase (first 20 minutes). Consistent percentage changes in these indices after obstruction repair were seen during the subsequent regulated phase. Gastric work was reduced 28 to 35 percent, but not work capability (mean area), by pyloric obstruction in the no vagotomy dogs. Reductions seen in proximal gastric vagotomy dogs were not different from those in the no vagotomy dogs. Higher percentages of reduction in amplitude (70 percent) and mean area of contractions (53 percent) occurred after truncal vagotomy compared with what occurred in the no vagotomy dogs. Mean area was also reduced more compared with what occurred in the proximal gastric vagotomy dogs. These data indicate that the reduced gastric work after feeding and impaired work capability caused by truncal vagotomy when superimposed on that produced by pyloric obstruction may exaggerate gastric atony and contribute to the delayed recovery of gastric emptying seen in the clinical setting.

Four-year weight loss Roux-Y gastric bypass: anastomotic reinforcement not additive.

There were 133 morbidly obese patients who received a Roux-Y gastric bypass in which a 12-mm gastrojejunostomy was sutured with multifilament steel (72 patients) for reinforcement or with interrupted silk (61 patients). Weight loss through 4 years was not different between the two patient groups. Moreover, weight loss in patients whose gastric stomas enlarged up to 16 mm was not different from those in whom the stomas remained at 12 mm. Stoma reinforcement with noncompliant steel did not add to long-term weight loss.

Infiltrative fibrosis with acute hemorrhagic intestinal infarction: association with a peritoneovenous shunt.

This article describes an unusual complication of acute hemorrhagic infarction of the gastrointestinal tract that developed in a 53-year-old man 3 years after a Denver peritoneovenous shunt was implanted for control of cirrhotic ascites. Infiltrative fibrosis, which involved the bowel and mesentery and surrounded its blood vessels, was the cause of the infarction. A spectrum of this fibroproliferative disease in patients with cirrhosis with ascites and peritoneovenous shunts is described, and possible pathogenetic mechanisms are discussed.

Canine cyclic motor activity of stomach and small bowel: the vagus is not the governor.

The influence of the vagus nerve on gastric cyclic motor activity and small bowel migrating motor complexes is controversial. Diaphragmatic vagesection does not alter their occurrence or periodicity, but cervical vagal cooling inhibits both gastric cyclic motor activity and duodenal phase II activity. We have clarified this contradictory data by reversible vagal cooling at the diaphragm level in 5 dogs implanted with strain gages and bipolar electrodes to record gastric and small intestinal cyclic activities. Circulation of coolant through an implanted convectional jacket surrounding the vagal trunks lowered jacket temperature to 2 degrees-8 degrees C and maintained this temperature for 3-5 h in each experiment. Vagal denervation during cooling was proven at the end of each trial by abolition of intravenous insulin-stimulated gastric contractions, which promptly appeared with warming. More than 90% of gastric motor cycles persisted during vagal cooling. The mean duration of gastric phase III activity was reduced during cooling but the mean period of gastric motor cycles was unchanged. Duodenal phase II and III activities were unchanged and migrated normally through the small bowel. We conclude that the vagus nerve may modulate the duration of gastric phase III activity but does not govern the initiation of gastric cyclic motor activity or the duration, period, and migration of small intestinal migrating myoelectric complexes. These findings concur with those after truncal vagectomy but are different from observations made with cervical vagal cooling.

Peritoneovenous shunts: predictive factors of early treatment failure.

Death or uncorrectable shunt malfunction occurred within 4 months in a majority (23) of 39 consecutive alcoholics who had peritoneovenous shunts. Clinical, biochemical, and operative variables in these patients were reviewed to detect factors that might predict early failure. Hepatorenal disease, expressed as the sum of total bilirubin and creatinine greater than 4 mg/dl, characterized 15 of 18 early deaths and was a reliable preoperative criterion of clinical outcome (p less than 0.001). Except for prothrombin elevation, abnormal coagulation indices did not predict early failure. Encephalopathy was more prevalent in patients who died (p less than 0.05). Among operative variables, failure to drain ascites was associated with more hospital deaths (p less than 0.05) and documentation of venous catheter placement in a central location reduced likelihood of early shunt malfunction (p less than 0.001). Preference for the LeVeen or Denver shunt did not affect the rate of early shunt malfunction. The overriding predictive factor was hepatorenal disease, both obvious and occult, by which judicious patient selection may avoid early death in surgical palliation for intractable alcoholic ascites.

Mortality and rebleeding after hypertensive variceal disconnections.

Forty-five patients had operative disconnections of portoazygos venous collaterals for variceal hemorrhage. Nineteen alcoholic patients had urgent partial azygos disconnection (PAD) to control bleeding; this involved intragastric variceal and coronary and distal gastric-vein ligation. A complete azygos disconnection (CAD), which also included splenectomy, was done urgently in 14 patients and electively in 12 patients; 15 were alcoholics. All patients were assessed for operative risk by a modified Child's classification. Hospital mortality after urgent PAD or CAD in alcoholics was 67%, largely due to intraperitoneal sepsis or hepatorenal failure with recurrent hemorrhage. Operative modified Child's classification of survivors was better (lower) than in nonsurvivors. Eleven nonalcoholic patients had CAD; two died of intraperitoneal sepsis. Nine survivors did not rebleed nor have encephalopathy develop during an average follow-up of 41 months. Complete azygos disconnection was a good alternative, particularly in the elective setting, for patients with nonalcoholic portal hypertension.

Danger of fundoplication after selective vagotomy and antrectomy.

Antireflux procedures are required in some patients at the same time or later after operations for chronic duodenal ulcer. The consequences to gastric blood supply are different between the three vagotomies usually performed to treat duodenal ulcer. A serious ischemic complication, incurred when a patient underwent fundoplication several years after a selective vagotomy and antrectomy, is reported to emphasize that the stomach relies on greater curvature arterial blood supply after this operation. Additional arteries ligated during fundoplication may surpass the capacity of remaining gastric arterial collaterals and produce ischemia. The same danger exists with fundoplication after proximal gastric vagotomy, if antrectomy has been added to treat recurrent ulcer. The technical differences of these two vagotomies from truncal vagotomy and the potential danger of fundoplication in these clinical situations must be recognized by the general surgeon.

Reversible truncal vagotomy in conscious dogs.

Prior attempts at in vivo reversible vagal denervation of the gastrointestinal tract have been limited to cervical cooling techniques that also denervate both sympathetic and vagal pulmonary and cardiac branches. Denervation of vagal efferent fibers at this level has produced results that are inconsistent with those obtained after surgical truncal vagotomy. We have, therefore, developed a technique to provide reversible vagal denervation below the pulmonary and cardiac branches for the study of gastric motility. Five dogs, previously equipped with gastric strain gages and electrodes, underwent implantation of a tubular cooling jacket around a distal thoracic vagal trunk with contralateral vagotomy (4 dogs), or around both vagal trunks (1 dog). The jacket was made of stainless steel tubing in a "J" design. Its inside channel was lined with a sterling silver sheet, and a thermistor was attached to record temperature change. Silicone tubing coursed externally to a pump and flask to which 95% ethanol at -70 degrees C was circulated at variable speeds. Thoracic vagal cooling, extending up to 5 h, reversibly blocked gastric contractions induced by insulin hypoglycemia. Contractile waves were terminated at device temperatures of 2 degrees-6 degrees C but promptly returned with warming. Dogs were tranquil during denervation, and enclosed nerves remained functional for greater than 40 days.

Prospective trial of proximal gastric vagotomy.

Forty men who were to have elective operation for nonobstructive duodenal or pyloric channel ulcer were randomized prospectively to undergo either proximal gastric vagotomy without drainage (PGV, n = 18) or selective vagotomy, antrectomy, and gastroduodenostomy (SVA, n = 22). Gastric acid analyses were accomplished before and 3 and 12 months after operation. Clinical interviews were conducted yearly. Thirty-nine patients were evaluable at 2 years, 25 at 4 years, and 15 at 5 years. No operative deaths occurred. Recovery was more rapid and the incidence of serious operative morbidity was lower after PGV than after SVA. Reduction of basal and stimulated gastric secretion was greater after SVA than PGV. Significant long-term sequelae other than recurrent ulcer were less frequent after PGV compared to SVA. Recurrent ulcer may occur more often after PGV; 3-month gastric secretory studies may be helpful in anticipating recurrence. Patients who undergo PGV have a particularly increased risk of developing pyloric channel ulcer disease, and low secretory values indicating an adequate vagotomy do not assure future protection from pyloric channel ulcer recurrence. Long-term sequelae after SVA, particularly dumping, do not have dependable reoperative options, whereas antrectomy should be a reliable reoperative solution to ulcer recurrence after PGV. PGV, performed correctly with a 5 to 7 cm vagal-esophageal separation, is preferable to vagotomy and resection for elective treatment of nonobstructing duodenal ulcer disease.

Improvement in heart disease risk factors after gastric bypass.

Fasting blood levels of three lipid classes (high-density lipoprotein-cholesterol [HDL-C], total cholesterol, and triglycerides) were measured in morbidly obese patients before, and six and 12 months after, Roux-en-Y gastric bypass. Weight above ideal in 31 women averaged 72 kg, and in 12 men, 78 kg. Excess weight lost after six months was 49.7%. Total cholesterol levels and abnormally high triglyceride levels (183 +/- 106 mg/dL [SD]) decreased and became normal after six months. Low-normal HDL-C levels in 31 women (45 +/- 13 mg/dL) improved by six months to 54 +/- 19 mg/dL; similarly, HDL-C levels in 12 men (38 +/- 9 mg/dL) increased to 57 +/- 17 mg/dL. One of eight diabetic patients remained hyperglycemic; ten of 15 patients who were hypertensive, requiring medication preoperatively, became normotensive by six months. Lipid levels measured again at one year showed little change. Risk factors of coronary heart disease were improved by gastric bypass in these morbidly obese patients, and the changes were clearly established by six months after operation.

Hepatic free fatty acids in alcoholic liver disease and morbid obesity.

Alcoholic liver disease is characterized by the accumulation of fat and inflammatory changes in the liver. Because free fatty acids, the precursors of triglycerides, can damage biological membranes, accumulation of free fatty acids in the liver might be in part responsible for the functional and morphological changes seen in alcoholic liver disease. We, therefore, determined the hepatic lipid composition in biopsies from 31 patients with alcoholic liver disease, 18 patients with morbid obesity, and 5 patients without evidence of liver disease. Free fatty acids were found in all liver biopsies. Patients with morbid obesity or alcoholic liver disease had significantly higher fatty acid and triglyceride levels than did controls (p less than 0.01). Patients with alcoholic liver disease had significantly higher fatty acid levels than did patients with morbid obesity (p less than 0.05), while there was no difference in the triglyceride concentrations between these two groups. The distribution of the fatty acids in the free fatty acid fraction differed significantly from that in the triglyceride fraction indicating a preferential incorporation of unsaturated fatty acids into triglycerides. This difference in the distribution pattern was lost in patients with the most severe forms of alcoholic liver disease. The data are consistent with the hypothesis that accumulation of free fatty acids in patients with alcoholic liver disease may be responsible for or contribute to the observed functional and morphological damages.