The role of neoadjuvant therapy in surgically resectable esophageal cancer.

OBJECTIVE: To determine the effect of neoadjuvant therapy (NT) (preoperative chemotherapy, radiation therapy, or both) in surgically resectable esophageal cancer. DESIGN: A retrospective review over a 20-year period. SETTING: A tertiary academic medical center. PARTICIPANTS: All patients undergoing surgical resection for esophageal cancer (N = 316) over this time period. MAIN OUTCOME MEASURES: Perioperative morbidity and mortality, local and distant recurrences, and overall survival. RESULTS: Patients undergoing NT (n = 106) had prognostic factors similar to those treated with surgery alone (n = 210). No increase was noted in surgical morbidity with NT (anastomotic leaks, reoperation rates, complications, or extended hospital stays). Overall survival was not improved by NT (median survival, 14 months) except in the subset of patients (11/83) who responded completely (100% histological necrosis) to preoperative chemotherapy (median survival, 79.2 months; P < .02). Complete response to radiation therapy alone was not associated with improved survival. Partial necrosis of the primary tumor was seen in 13 (15%) of 83 patients but conferred no survival advantage. Complete response to preoperative chemotherapy was associated with squamous cell pathological features and excellent performance status as measured by preanesthesia evaluation. CONCLUSIONS: The addition of NT did not increase perioperative morbidity or mortality. Only the subset of patients who had a complete response to preoperative chemotherapy showed a survival advantage. Excellent performance status and squamous cell pathological features were associated with an increased chance of complete pathological response following preoperative chemotherapy.

Factors that affect the performance of lipase on fat digestion and absorption in a canine model of pancreatic insufficiency.

In a previous experiment, absorption of [14C]triolein was poor under low lipase in the first postcibal hour during which luminal conditions change markedly. We wondered how low lipase might be affected by changing concentrations of fat, bile salts, titratable acid, pepsin, and food particles. Therefore, in dogs with duodenal and midintestinal fistulas, endogenous bile and pancreatic juice were excluded from the intestinal lumen and replaced with varied amounts of exogenous bile and pancreatic enzymes during steady perfusions. Oil emulsions contained [14C]triolein and [3H]glycerotriether. A double isotope ratio method and a double isotope, double extraction method were used to determine, respectively, the amount of [14C]triolein absorbed and hydrolyzed by the midgut. Lipolysis increased with both substrate and enzyme inflows, whether inflows were varied by changing concentrations or rates of volume flow. But at increasing rates of fat entry, the percent of fat hydrolyzed by the midgut declined. Neither pH 4 nor 5 citrate affected fat hydrolysis or absorption when titratable acid was infused at rates < or = 16 mEq/h; but pepsin reduced both. Whereas meat particles bound lipase, their presence augmented lipolysis. We speculate that rapid gastric emptying of fat and peptic deactivation of duodenal lipase were the main factors responsible for the previously poor performance of low lipase in the first postcibal hour.

Effect of replenished lipase on postcibal absorption of fat in a canine model of pancreatic insufficiency.

Clinical studies indicate that as little as 10% of pancreatic secretory capacity is needed to ensure normal digestion; but we found previously that supplying lipase to the postcibal duodenum at > or = 10% of normal rates did not normalize fat absorption in pancreatic insufficiency. Therefore, we examined the dose-response of endogenous lipase on fat absorption. Pancreatic juice was excluded and returned in varied amounts to the postcibal duodenum in dogs with pancreatic fistulas. Meals contained margarine labeled with digestible [14C]triolein and indigestible [3H]glyceroltriether. With an isotope ratio method, we estimated the amount of radiotriolein absorbed hourly from chyme collected for 6 h after a meal from midgut fistulas. When all pancreatic juice was excluded, there was almost no absorption. When 10 or 20% of pancreatic juice was returned, approximately 80% of triolein was absorbed by the midgut, compared with 90% absorption when all pancreatic juice was instilled. However, we observed that at 10 and 20% replenishments, the amount of triolein absorbed in the first hour was much less than in subsequent hours, and thus that absorptive efficiency varied with the fraction of fat emptied from the stomach during the first hour. At rates of 10 or 20% of normal, lipase was equally effective, whether from endogenous juice or exogenous pancreatin.

Disproportionate ileal digestion on canine food consumption. A possible model for satiety in pancreatic insufficiency.

In animals, ileal sensors of nutrients signal satiety more potently than similar sensors in jejunum. We postulated that inadequate food intake and weight loss in human pancreatic insufficiency might arise by the displacement of digestion to ileum, where excessive release of digestive products would enhance satiety. To test this idea, we studied dogs prepared with pancreatic fistulas, which allowed reversible switching of pancreatic juice from entry at duodenum to entry at mid-small intestine. Dogs were studied in a crossover design over successive eight-day periods. Food consumption and body weight were measured while the dogs had continuous access to food. Diversion of pancreatic juice to mid-intestine significantly (P < 0.01) depressed food intake by an average of 28%. Diversion also significantly (P < 0.01) reduced body weight. The findings support the idea that insufficient food intake in human pancreatic insufficiency may result from stimulation of ileal satiety mechanisms.

Erythromycin accelerates solid emptying at the expense of gastric sieving.

Erythromycin accelerates gastric emptying by inducing antral contractions similar to phase III of interdigestive MMC. These powerful contractions are capable of forcing coin-sized indigestibles out of the stomach. In contrast, fed motility is associated with submaximal contractions that fragment (trituration) and propel solids while retaining large (> 0.5 mm) pieces for further size reduction (gastric sieving). In this study, using dogs with duodenal fistulas, we tested the hypothesis that erythromycin-induced acceleration of gastric emptying resulted in the passage of inadequately triturated (> 0.05 mm) chunks of solids into the duodenum. We found that gastric emptying was accelerated by erythromycin (vs 0.15 M NaCl control, P < 0.05). However, the percentage of chyme collected in the > 0.5-mm fraction was much greater (P < 0.01) in the erythromycin-treated experiments (63 +/- 9%) than the controls (7 +/- 1%). Correspondingly, while a fine gruel was passed during controls, under erythromycin infusion, most of the solids were emptied as large chunks virtually unchanged from the swallowed pieces. We conclude that erythromycin accelerates gastric emptying at the expense of gastric sieving.

Sustained slowing effect of lentils on gastric emptying of solids in humans and dogs.

The distal small intestine is an especially potent site for carbohydrate-triggered intestinal inhibition of gastric emptying of solids. Poorly digestible carbohydrates, such as lentils, may escape proximal absorption, travel over time to reach these inhibitory mechanisms, and slow the gastric emptying of a later meal. A slowing effect on gastric emptying may be associated with a lowering effect on postprandial glucose. The aims of this study were to determine (a) whether lentils (a poorly digestible carbohydrate) vs. bread (an easily digestible carbohydrate) eaten as a premeal (with equal amounts of carbohydrates) slow the gastric emptying of a second solid meal taken 4.0-4.5 hours later and (b) whether a slowing effect on the gastric emptying of the second meal is associated with a lower postprandial glucose response. We found that in 7 dogs and 10 humans, gastric emptying of the second meal was delayed after a lentil premeal compared with a bread premeal. However, there was no difference in the glucose response to the second meal under the two conditions.

Gastric emptying of solid food is most potently inhibited by carbohydrate in the canine distal ileum.

Although glucose sensors regulating the gastric emptying of liquid meals are uniformly distributed throughout the canine small intestine, some data suggest that the distal small bowel more potently inhibits gastric emptying of solid foods. The aims of this study were to compare (a) the inhibition of gastric emptying by glucose sensors in the proximal intestine with the feedback from the distal intestine, (b) these effects on the gastric emptying of solids vs. liquids, and (c) the inhibitory effect of unhydrolyzed starch with glucose. In 7 dogs with chronic duodenal fistulas, the second, third, and fourth quarters of small bowel were perfused via chronically implanted transmural catheters. Gastric emptying of either solids or liquids was tracked by gamma camera while gastric output was diverted out the duodenal fistula and the small bowel perfused with test solutions of glucose (0.06-2.0 mol/L), 0.15 mol/L NaCl, or 8.5% soluble starch. It was found that (a) gastric emptying of solids but not liquids was approximately 3 times more potently inhibited by glucose in the fourth quarter vs. the first or second quarter of small bowel, and (b) only hydrolyzed starch inhibited gastric emptying of solids.

Acalculous candida cholecystitis: a complication of critical surgical illness.

Four patients with underlying diseases including multiple trauma, aortic graft infection, and complex fistulae developed acute acalculous cholecystitis with bile cultures positive only for Candida albicans. The primary site of the candida infection included urinary tract, gastrointestinal tract, and an aortic graft in one patient each and was undetermined in the trauma victim. All had received broad-spectrum antibiotics; three of the four were in the intensive care unit (ICU) with organ failure. Ultrasonography showed a thickened gallbladder wall in three patients and sludge in one. Hepato-iminodiacetic acid scans were nonvisualizing in these three patients. Operative findings included gangrenous cholecystitis in two patients and edematous cholecystitis in one. The fourth patient was treated with percutaneous cholecystostomy and interval cholecystectomy. The interval from the onset of symptoms to recognition of the need for operation was an average of 7 days. Two of the four patients died of ongoing sepsis. Candida cholecystitis is a life-threatening complication of critical surgical illness. Risk factors are similar to those for candida infection elsewhere and include antibacterial therapy, complex fistulae, disseminated malignancy, immunosuppression, and prolonged ICU stay. A high index of suspicion for this fungal pathogen and aggressive surgical therapy offer the only chance for a favorable outcome.

Inhibition of gastric emptying by acids depends on pH, titratable acidity, and length of intestine exposed to acid.

Exposure of the small intestine to acid inhibits gastric emptying in a dose-related fashion that depends on titratable acidity and pH. Little information is available on the location of this inhibitory mechanism or on the relative contribution of titratable acidity and pH to this feedback control. We hypothesized that the dependence on titratable acidity is related to the length of the intestine exposed to acid and that the dependence on pH is related to the region of the intestine exposed to acid. To test these ideas, we studied 11 dogs with duodenal and jejunal fistulas. The inhibitory effects were tested when different lengths of the small intestine were exposed to test solutions of 0.03, 0.06, and 0.12 meq/ml titratable acidities. pH as an independent covariable was separated from titratable acidity by comparing the inhibition of gastric emptying of lactic acid (pH fixed to 2.4) to HCl (pH 0.96-1.6). Maximal inhibition of gastric emptying by both acids depended on acid exposure of a length of small intestine that was greater than 65 but less than or equal to 150 cm long. When acid was confined to the proximal 15 cm, increasing concentration of HCl (decreasing pH) resulted in increasing inhibition, but this effect was absent with increasing concentration of lactic acid (fixed pH). Inhibition was absent when 0.06 meq/ml HCl was infused into the intestine beyond the midintestine.(ABSTRACT TRUNCATED AT 250 WORDS)

Inhibition of gastric emptying by sodium oleate depends on length of intestine exposed to nutrient.

Previously, we reported that inhibition of gastric emptying by glucose or acids depends on the length of gut exposed to the inhibitor [Gastroenterology 95: A877, 1988; Am. J. Physiol. 256 (Gastrointest. Liver Physiol. 19): G404-G411, 1989]. In this study, we hypothesized that feedback control by fat may be similarly regulated. In dogs with chronic intestinal fistulas, we compared the intensity of intestinal feedback when different lengths of the small intestine were exposed to meals of 3, 9, or 27 mM sodium oleate. We found that 1) inhibition of liquid emptying was dose dependent, 2) intensity of negative feedback was dependent on both the concentration of the oleate and the length of gut exposed to fat, 3) full inhibitory effect was achieved with exposure of fat to 150 cm of gut, 4) inhibition from the distal one-half of gut was less potent than that generated from the proximal one-half of gut, and 5) on a molar basis oleate was 20 times as effective as glucose at inhibition of gastric emptying and that this difference was related to the slower rate of fat absorption.

Reoperative surgery for the morbidly obese. A university experience.

Patients who undergo surgery for morbid obesity are often subjected to reoperation for a wide array of indications. To evaluate outcome following revisional procedures, we reviewed the records of 32 such patients treated at UCLA between April 1986 and May 1989. Twenty-five women (78%) and 7 men (22%) with a mean age of 44 years underwent 76 reoperations (2.4 per patient) for complications of prior obesity surgery. Indications for initial surgical revision consisted primarily of metabolic derangements (12 patients) and weight-related problems (11 patients). In contrast, indications for the patients' final surgical procedure were commonly for bowel obstruction (41%), intra-abdominal sepsis (12%), and gastrointestinal bleeding (6%). Following initial revision, 23 patients (71.8%) required further surgery for major complications and four patients died (12.5%). While initial revisions are frequently indicated for metabolic problems, final reoperations are more frequently undertaken for urgent, life-threatening complications. Revisional procedures for morbid obesity should be carefully considered, and the potential for major complications and/or death should be weighted heavily against proposed benefits.

Broad spectrum penicillin as an adequate therapy for acute cholangitis.

In a previous study of patients with acute cholecystitis, we demonstrated equal efficacy with a broad spectrum penicillin (piperacillin) and a penicillin plus amino-glycoside combination. Whether a single agent broad spectrum penicillin is adequate treatment for more severe infections, such as acute cholangitis, however, is still unclear. We, therefore, conducted a three center, prospective, randomized trial to determine whether or not a broad spectrum penicillin alone is adequate therapy for patients with acute cholangitis. During a 36 month period, 96 patients with sepsis and biliary obstruction were randomly assigned to receive either piperacillin (n = 49) or ampicillin plus tobramycin (n = 47). The two groups receiving antibiotics were similar with respect to all clinical and laboratory parameters. The incidence of blood cultures with positive results (20 versus 21 per cent) and underlying malignant lesions (51 versus 62 per cent) was also similar between the two groups. The percentage of patients with a clinical cure or significant improvement was the same in the two groups (69 versus 70 per cent). However, there was a significant difference in the cure rate between patients with benign and malignant biliary obstructions (83 versus 59 per cent, p less than 0.01). No significant differences were noted between the two antibiotic groups with respect to drug toxicity, but patients with malignant conditions were more prone to antibiotic related toxicities (2 versus 19 per cent, p less than 0.05). These data suggest that outcome of treatment in patients with acute cholangitis is similar with either a broad spectrum penicillin or a penicillin plus aminoglycoside combination and is dependent upon the nature of the biliary obstruction.

Predictive factors for bactibilia in acute cholecystitis.

Acute cholecystitis is well established as one of the high-risk factors bactibilla and wound infection. However, many patients with acute cholecystitis do not have bactibillia. Therefore, we analyzed 20 clinical and laboratory parameters in 49 patients with acute cholecystitis to determine which factor(s) predicted bactibilla. Twenty-one (42.9%) of 49 patients with pathologically proved acute cholecystitis had positive bile and/or gallbladder wall cultures. Univariate analysis suggested that a preoperative temperature greater than 37.3 degrees C, a total serum bilirubin level greater than 8.6 mumol/L, and a white blood cell count greater than 14.1 x 10(9)/L were the best predictors of bactibilia. Multifactorial analysis demonstrated that the 17 patients with zero or one predictive factor had a significantly lower chance of having bactibilia than the 32 patients with two or three predictive factors (6% vs 63%). We concluded that the culture status of patients with acute cholecystitis can be predicted preoperatively. We propose that patients with acute cholecystitis and zero or one of the predictive factors receive a single preoperative antibiotic dose. In patients with two or three predictive factors, antibiotics should be continued until culture data are available.

Emerging trends in medical practice: primary practice revisited.

Following the failure of the diversification era, hospitals are returning to the roots of medicine. The result is a new focus on primary care, a beneficial outcome for both the medical and hospital communities.

Alterations in digestive function caused by pancreatic disease.

Pancreatic exocrine secretion is regulated by a complex interaction of meal-stimulated neurohormonal reflexes. Pancreatic enzyme output must be reduced to less than 10 per cent of normal before fat absorption is appreciably impaired, proving that the pancreas secretes a large surplus of enzymes. Surgical therapy does not improve pancreatic exocrine insufficiency, and partial pancreatic resection frequently precipitates steatorrhea in patients with chronic pancreatitis. Therefore, pancreatic resection should be undertaken cautiously in patients who do not yet have clinically evident exocrine insufficiency. In most patients, oral pancreatic enzymes will control diarrhea secondary to steatorrhea. In others, concurrent administration of an H2 blocker is required to reduce gastric acidity and prevent enzyme inactivation in the stomach. Formulations with an acid-resistant coating are also effective in some patients. However, complete normalization of fat absorption with restoration of body weight and nutritional well-being requires careful management of multiple dietary and behavioral factors, as well as long-term follow-up. Unfortunately, this appears to be an elusive goal, as many patients with chronic pancreatitis continue to die of malnutrition.

Management of cystic disease of the liver.

The management of cystic diseases of the liver requires an understanding of their pathophysiology and natural history. Surgery for congenital solitary cysts and polycystic disease should be reserved for patients with significant symptoms. Caroli's disease requires careful preoperative evaluation and planning and long-term follow-up. Surgery for echinococcal liver cysts should be performed before complications of rupture and superinfection develop.

Inhibition of gastric emptying by glucose depends on length of intestine exposed to nutrient.

Nutrients inhibit gastric emptying in a dose-related fashion. We postulated that load-dependent gastric emptying results from the saturation of mucosal absorptive mechanisms, so that a longer length of the small intestine is exposed to unabsorbed nutrients as more nutrient enters the intestine to participate in this negative feedback. To test this idea, we limited exposure of 0.25 to 1.0 M glucose meals to various lengths of duodenum and jejunum in 17 dogs. The effects of these limited perfusions were then compared with experiments in which the whole gut (ALL) was exposed to the nutrient. Maximal inhibition was seen with 1.0 M meal and was similar with perfusions of 150 cm and ALL. By contrast, even with the 1.0 M load, no inhibition of gastric emptying was seen when glucose meal was confined to the first 15 cm of the proximal duodenum. Only 50-60% of maximal inhibition was observed during confinement of 1.0 M meal to the proximal 65 cm. We concluded that glucose sensors are present in both the proximal and the distal gut and the inhibition was related to the length of the small intestine exposed to glucose.

The effect of bile diversion on satiety and fat absorption from liquid and solid dietary sources.

In previous studies, liquid fat has been used to determine the effect of bile diversion on fat absorption. Since protein digests, in addition to bile salts, are capable of solubilizing lipids, we hypothesized that fat incorporated in the protein-rich matrix of solid food would be less sensitive to bile diversion than fat ingested as an oil or liquid. Using [3H]glycerol triether as a nonabsorbable fat recovery marker, we determined how much [14C]triolein was absorbed from solid (chicken liver) and liquid (margarine) dietary sources. After a standard liquid/solid meal with either the chicken liver or margarine labeled, midintestinal chyme was collected for 6 hr, extracted, and counted for 14C and 3H activity. Zero, eighty, or one hundred percent of endogenous bile was diverted. Fat absorption from both chicken liver and margarine was nearly complete by midintestine with 0% diversion and was little affected by diversion of 80% of bile. Complete biliary diversion significantly decreased fat absorption from margarine (87.9 +/- 4.4 to 37.2 +/- 9.2%, P less than 0.05) but reduced [14C]triolein absorption from chicken liver less consistently and insignificantly (78.8 +/- 6.9 to 43.9 +/- 10.6%). These data indicate that fat absorption is not solely dependent on bile and support the hypothesis that fat ingested in a cellular matrix is less dependent on bile than liquid fat. Using these same animals but with the midintestinal cannulas plugged to expose the distal intestine to unabsorbed luminal nutrients, we also demonstrated that bile diversion of an initial meal reduced food consumption at a meal offered 3 hr later.