Influence of subtotal gastrectomy on bile acid concentration, total bacterial counts in the gastric juice, and histologic changes in the gastric mucosa

The concentrations of bile acids, PH and total bacterial counts in the gastric juice were measured among 44 patients with peptic ulcer before and after subtotal gastrectomy (Billroth II in 14 cases, Billroth I in 10 cases, and PAFPG in 20 cases). Thirty three patients were studied by endoscopy and gastric mucosa biopsy one year after the surgery. The results showed that the fasting gastric bile acids concentrations in both B-II and B-I groups were significantly higher than in PAFPG group, when examined 3 weeks and 1 year postoperatively. The PH and total bacterial counts in gastric juice were increased significantly after B-II and B-I reconstruction compared with PAFPG. The abnormal histology of gastric remnant mucosa was more common in B-II or B-I groups than in PAFPG (P less than 0.05). Our data demonstrated that B-II and B-I gastrectomy caused considerably enterogastric reflux, while PAFPG prevented it effectively by keeping the gastric physiology in a relatively stable status. It is the authors' belief that the results provide objective basis for selecting surgical procedure and evaluating operative effects.     

Reconstructive Procedure after Distal Gastrectomy for Gastric Cancer that Best Prevents Duodenogastroesophageal Reflux

Billroth I and II reconstructions are commonly performed after distal gastrectomy. Both may cause duodenogastric and duodenogastroesophageal reflux, conditions reported to have carcinogenetic potential. The aim of this study was to investigate which reconstructive procedure would most effectively prevent bile reflux into the gastric remnant and esophagus after distal gastrectomy. A group of 92 patients who underwent curative distal gastrectomy for gastric cancer were subjected and classified into three groups retrospectively by the reconstructive procedure undertaken: group A, Roux-en-Y (Roux-Y) reconstruction (n = 29); group B, Billroth I reconstruction (n = 41); group C, Billroth II reconstruction (n = 22). The bile reflux periods (percent time) for the gastric remnant and esophagus were measured with the Bilitec 2000 under standardized conditions. The percent time for the gastric remnant was significantly less in group A than in group B or C. In 61% of all patients, bile reflux into the esophagus was found to be more than 5.0% of the time; it was less in group A than in group B or C (p = 0.057). A questionnaire revealed a good correlation between the incidence of reflux symptoms and the percent time for the gastric remnant and esophagus. Roux-Y reconstruction is superior to either Billroth I or II reconstruction for preventing bile reflux into the gastric remnant and esophagus after distal gastrectomy.     

Quantification of duodenogastric reflux in gastroduodenal peptic ulcer and in gastric operation patients, using a 24-h gastric pH measurement as a quantification technique

Twenty-four-hour gastric pH measurement was used to study duodenogastric reflux. To differentiate between gastric hyposecretion and duodenogastric reflux, we also measured bile acid concentrations in the gastric juice, and regarded pH increases to above 4 as possible episodes of reflux. The procedure was used in 60 patients, divided into the following groups: (1) control group (ten patients); (2) duodenal ulcer (ten patients); (3) type 1 gastric ulcer (five patients); (4) type 3 gastric ulcer (five patients); (5) bilateral truncal vagotomy plus pyloroplasty (ten patients); (6) truncal vagotomy plus Billroth I partial gastrectomy (ten patients); and (7) truncal vagotomy plus Billroth II partial gastrectomy (ten patients). The amount of reflux (areas of pH greater than 4) in the type 1 gastric ulcer and Billroth I and Billroth II groups was significantly greater than that found in the control, duodenal ulcer, type 3 gastric ulcer and truncal vagotomy plus pyloroplasty groups. The mean concentration of total bile acids was also greater in the gastrectomized patients than in the rest of the groups studied. In the type 1 gastric ulcer group the mean bile acid concentration was similar to that of the control group.     

Reconstructive procedure after distal gastrectomy for gastric cancer that best prevents duodenogastroesophageal reflux

Billroth I and II reconstructions are commonly performed after distal gastrectomy. Both may cause duodenogastric and duodenogastroesophageal reflux, conditions reported to have carcinogenetic potential. The aim of this study was to investigate which reconstructive procedure would most effectively prevent bile reflux into the gastric remnant and esophagus after distal gastrectomy. A group of 92 patients who underwent curative distal gastrectomy for gastric cancer were subjected and classified into three groups retrospectively by the reconstructive procedure undertaken: group A, Roux-en-Y (Roux-Y) reconstruction (n = 29); group B, Billroth I reconstruction (n = 41); group C, Billroth II reconstruction (n = 22). The bile reflux periods (percent time) for the gastric remnant and esophagus were measured with the Bilitec 2000 under standardized conditions. The percent time for the gastric remnant was significantly less in group A than in group B or C. In 61% of all patients, bile reflux into the esophagus was found to be more than 5.0% of the time; it was less in group A than in group B or C (p = 0.057). A questionnaire revealed a good correlation between the incidence of reflux symptoms and the percent time for the gastric remnant and esophagus. Roux-Y reconstruction is superior to either Billroth I or II reconstruction for preventing bile reflux into the gastric remnant and esophagus after distal gastrectomy.     

Experimental gastric carcinogenesis in the rat: Effects of hypergastrinemia and acid secretion

The present experiments examined in rats the influence of chronic hypergastrinemia and the resulting gastric hyperacidity on induction of gastric adenocarcinoma by N-methyl-N′-nitro-N-nitrosoguanidine (MNNG). Three groups of animals were studied: (A) control rats (no operation, normogastrinemia); (B) rats with antrectomy and Billroth II (normogastrinemia); and (C) rats with implantation of the antrum into the colon and Billroth II (hypergastrinemia). All rats were fed MNNG, 83 μg/ml, in drinking water for 6 months and were observed for 6 more months before being killed. The number, location, and histology of malignant gastric neoplasms were recorded.The incidence of tumors in the proximal gastric remnant of the Billroth II antrectomy rats was 88 percent, significantly greater than the incidence in the proximal gastric remnant (excluding the implanted antrum) of Billroth II antral implant rats (50 percent) or in control rats (50 percent). In addition, 19 gastric neoplasms were found in the 12 antra implanted into the transverse colon.Thus antrectomy Billroth II predisposed the animals to tumors, which was at least partially offset by hypergastrinemia. This suggests that gastric mucosa is more vulnerable to carcinogens after antrectomy and Billroth II, and the effects of hypergastrinemia (or acid secretion and mucosal growth) reduce this vulnerability. Hypergastrinemia did not protect the antrum from tumor induction when the antrum was removed from exposure to acid. These data suggest that in rats (1) antrectomy Billroth II predisposes to carcinogenically induced gastric cancer, (2) hypergastrinemia directly or indirectly decreases this predisposition, and (3) in the presence of hypergastrinemia and after removal from the acid stream, the antrum is highly vulnerable to the carcinogenic effects of MNNG.     

Early detection of gastric remnant carcinoma. The role of gastroscopic screening

Partial gastrectomy for benign ulcer disease has been associated with carcinoma in the gastric remnant. To detect formation of this cancer in patients having undergone this operation, we initiated a screening protocol using barium contrast studies, flexible gastroscopy, and biopsy. Patients were selected from a group of 233 patients who had undergone partial gastrectomy for benign disease between 1960 and 1975. In this group, operations for duodenal ulcer had been performed in 156 patients (83 Billroth I and 73 Billroth II reconstructions) and subtotal gastrectomy in 77 patients with gastric ulcer (17 Billroth I and 60 Billroth II reconstructions). From July 1980 to July 1985, 163 patients underwent gastroscopy and biopsy with a median postoperative interval of 14.6 years. Through screening, three resectable remnant carcinomas were found. We conclude that routine gastroscopy leads to earlier detection and a higher rate of resectability if gastric remnant carcinoma is found; yearly screening should be performed after a ten-year postresection interval; and gastroscopic biopsy is more accurate than upper gastrointestinal tract barium contrast studies and should be used preferentially to identify gastric remnant carcinoma.     

Quantitative determinations of duodenogastric reflux,prevalence of Helicobacter pylori infection,and concentrations of interleukin-8

Billroth I or II reconstruction after distal gastrectomy often is associated with inflammation in the gastric remnant. We sought to determine which reconstructive procedure was most effective in preventing such remnant gastritis. Patients undergoing curative distal gastrectomy for cancer ( n = 82) were classified as group A (Roux-en- Y, n = 22); group B (Billroth I, n = 40); or group C (Billroth II, n = 20). Interleukin (IL)-8 concentrations in gastric mucosa were measured 3 months after surgery. In the absence of Helicobacter pylori infection, IL-8 concentrations were 13, 56, and 87 pg/mg protein in groups A, B, and C, respectively ( p < 0.05). In the presence of H. pylori infection, IL-8 concentrations were 61, 161, and 234 pg/mg protein in groups A, B, and C ( p < 0.01). Roux-en- Y reconstruction is better able to prevent remnant gastritis than either the Billroth I or II procedure as judged from IL-8 concentrations in gastric remnant mucosa.     

Microflora and deconjugation of bile acids in alkaline reflux after partial gastrectomy

It has been postulated that reflux of bile into the stomach promotes gastric carcinogenesis. Bilestained aspirates from 50 asymptomatic patients, partially gastrectomized more than 10 years earlier, were examined bacteriologically and with regard to conjugated and deconjugated bile acids. Endoscopic biopsies showed atrophic gastritis in all patients, cancer in two and severe dysplasia in another two. pH in the reflux aspirates was 7.3 ± 0.4 (mean ± standard deviation). Bacterial cultures were positive in all patients studied. Fecal type flora, mostly E. coli, klebsiella and Clostridium perfringens, was found in 85 percent of the patients. Total bile acids were found to be 2.6 ± 2.0 mg/ml, 23 percent of which were deconjugated. Deoxycholic acid, known to promote carcinogenesis in animals, amounted to 27 percent of total bile acids and deconjugated deoxycholic acid was 5 percent of total bile acids. The mostly anaerobic microflora and the presence of mainly free secondary and primary bile acids may contribute to the high incidence of cancer in the gastric remnant observed after Billroth I or II operations.     

高选迷切加保留胃窦浆肌层的半胃切除对防止胆汁返流的价值

胃大部分切除由于切除了幽门可导致术后胆汁返流、倾倒现象等，而胆汁的返流是导致返流性胃炎、溃疡复发、癌前病变和死胃癌的主要原因为了防止这些并发症，我院于１９８８年设计了高迷切加保留胃窦浆肌胃窦浆肌层的半胃工发除治疗消化性溃病５０例，并与Ｂ－Ⅰ、Ⅱ式各１０例，ＨＳＶ２０例进行对比，结果显示，我们设计的术式可以防止胆返流。     

Prospective comparison of gastric secretory function after gastrectomy with either Billroth II or Roux-en-Y anastomosis

In order to delineate the role of enterogastric reflux in changes of postoperative gastric secretory functions, 22 patients with peptic ulcers, who were randomly assigned to partial gastrectomy without vagotomy with either Billroth II or Roux-en-Y anastomosis, were prospectively studied before and 6 months after surgery. Preoperatively, there were no significant differences in gastric secretory functions between the two groups of 11 patients. Postoperatively, median fasting bile acids in the stomach increased in the Billroth II patients from 0.35 to 16.10 mumol/hr (p less than 0.01), but significantly decreased in the Roux-en-Y patients from 0.30 to 0.10 mumol/hr (p less than 0.05), which indicated adequate prevention of enterogastric reflux after the Roux-en-Y procedure. Gastrectomy resulted in significant reductions of median values of basal acid output (4.6 vs 0.6 mmol/hr, p less than 0.01, and 4.2 vs 0.4 mmol/hr, p = 0.02), peak acid output (31.6 vs 4.2 mmol/hr, p less than 0.01, and 38.7 vs 4.5 mmol/hr, p less than 0.01), serum pepsinogen A (121 vs 86 micrograms/L, p less than 0.01, and 92 vs 45 micrograms/L, p less than 0.01), meal-stimulated serum gastrin secretion (1472 vs 199 pM.60 min, p less than 0.0001, and 1017 vs 199 pM.60 min, p less than 0.0001) in the patients with Billroth II and Roux-en-Y anastomosis, respectively. There were, however, no significant differences in gastric secretory parameters between the two groups when studied 6 months after surgery. Therefore it is concluded that after gastrectomy, enterogastric reflux does not affect the secretory function of the gastric remnant within the first 6 months after surgery.     

Gastric juice factors after Roux-Y reconstruction compared with Billroth II partial gastrectomy

Gastric juice was studied in five patients after Billroth II partial gastrectomy, in six patients after Roux-Y reconstruction, and in eight healthy control subjects. Juice was obtained over a 24 hour period by hourly nasogastric aspiration prior to measurement of pH, total and stable N nitrosocompounds, nitrites, bile acids (total and free), and bacterial count. Large variations in almost all compounds were seen during the 24 hour period in individual patients, and in addition, there were large variations between patients within the clinical groups. The gastric juice pH concentration was higher in the Billroth II group, as was the bacterial flora count (median 3 X 10(6) organisms/ml). Although Roux-Y bile diversion reduced the exposure to bile acids, it did not prevent it, and bacterial proliferation was increased (1 X 10(7) organisms/ml). No significant differences or sizeable trends were seen in N nitrosocompounds or nitrite concentrations.     

Roux-en-Y Reconstruction after Distal Gastrectomy to Reduce Enterogastric Reflux and <Emphasis Type="Italic">Helicobacter pylori</Emphasis> Infection

Enterogastric reflux (EGR) is regarded as an unavoidable consequence of distal gastrectomy. We evaluated the efficacy of Roux-en-Y (RY) gastrojejunostomy and Braun enteroenterostomy (BEE) for preventing EGR. Between January 2002 and January 2005, 60 patients who underwent distal gastrectomy for gastric cancer or peptic ulcers were divided into RY, Billroth II reconstruction (BII) without or with BEE (BII+B) according to reconstructive method. After 12 months, EGR and mucosal alterations of the remnant stomach were evaluated using biliary scintigraphy, endoscopy, and histology. Scintigraphy showed fasting and postprandial EGR into the remnant stomach occurred in 5.3% and 21.1% of the RY group, 62.1% and 93.1% of the BII group, and 50.0% and 91.7% of the BII+B group, respectively. Endoscopy showed bile reflux occurred in 15.8% of the RY group, 75.9% of the BII group, and 83.3% of the BII+B group. In addition, the prevalence of Helicobacter pylori (HP) infection in the RY group was less than in the other groups (P < 0.02). Therefore, RY after distal gastrectomy was effective in reducing EGR and HP infection. BEE was ineffective in diverting bile flow away from the gastric remnant. The study was supported by a grant from the Research Foundation of The Tri-Service General Hospital (TSGH-C96-14-S05).     

Duodenogastric reflux and gastric histology after cholecystectomy with or without sphincteroplasty

Sixteen patients who had undergone cholecystectomy plus sphincteroplasty, 14 cholecystectomized patients and ten control patients were studied to evaluate whether differences existed in duodenogastric reflux and whether these were related to morphological damage of the gastric mucosa. Duodenogastric bile reflux during fasting was evaluated by measuring the concentration of total bile acids (by an enzymatic method) and single bile acids in the gastric juice by high performance liquid chromatography. The damage was evaluated histologically by systematic endoscopic biopsy of the antrum and body of the stomach. There was a statistically significant difference in fasting bile reflux between the three groups (Kruskal-Wallis test, P less than 0.001), and the group that underwent cholecystectomy plus sphincteroplasty had a significantly higher median value than the cholecystectomized group (P less than 0.05) and the control group (P less than 0.01). The distribution of chronic antral atrophic and superficial gastritis was different in the three groups (chi 2 test, P less than 0.005). Chronic atrophic gastritis was associated with cholecystectomy plus sphincteroplasty (P less than 0.01), while chronic superficial gastritis was more frequent in cholecystectomized patients. These results suggest that there may be more duodenogastric reflux after cholecystectomy plus sphincteroplasty than after cholecystectomy alone, and that there may be a correlation between the amount of duodenogastric reflux and the severity of mucosal damage.     

Reflux und Refluxkrankheit im Resektionsmagen

Zusammenfassung An 112 Patienten mit verschiedenen Formen der Magenresektion wegen eines Gastroduodenalulcus (Billroth l, Billroth II) wurden die intragastralen Konzentrationen an Lysolecithin und Gallensäuren bestimmt und mit dem pH-Wert, der Schleimhauthistologie, dem subjektiven Beschwerdegrad und der bakteriellen Besiedlung korreliert. Hierbei fand sich ein maximaler Reflux bei retrocolischen B II-Anastomosen. Geringere, aber noch hochpathologische Werte liesßen sich im antecolischen B 11 mit Braunscher Fußpunktanastomose und im B 1-Magen unterschiedlicher Anastomosenform beobachten. 70 % der Resektionsmägen waren bakteriell besiedelt. Refluxwerte und bakterielle Kontamination zeigten eine enge Korrelation zum pH-Wert. Oberhalb pH 6 war kein Magensaft steril, die Lysolecithinkonzentration stieg überproportional an. Die Schleimhautveränderungen korrelierten am besten mit der intragastralen Lysolecithinkonzentration, aber auch gut mit der Keimbesiedlung. Keiner der Refluxparameter korrelierte mit der Symptomatik, eine Beschwerdearmut schloß eine Refluxschädigung nicht aus. Das Krankheitsbild der Refluxgastritis wird dargestellt, ihre Folgen bis hin zum Stumpfcarcinom aufgezeigt, auf chirurgisch-taktische Konsequenzen in der Ulcuschirurgie wird hingewiesen.

---

Reflux and reflux disease following gastric resection

Summary Intragastric concentrations of lysolecithin and bile acids were determined in 112 patients with different types of gastrectomy for peptic ulcer (Billroth I and Billroth II). These values were correlated with intragastric pH, bacterial contamination, gastric mucosal changes and Visick grading. A maximal amount of duodenogastric reflux was found in retrocolic gastrojejunostomy (Billroth II). Smaller, but still highly pathological values were observed in the antecolic gastrojejunostomy with Braun enteroanastomosis as well as in different types of gastroduodenostomy (Billroth 1). 70 % of the operated stomachs were bacterially comtaminated. There was a good correlation between amount of reflux, bacterial contamination and pH-value. Above pH 6 no stomach was sterile, the lysolecithin-concentration increased more than proportional. Gastric mucosal changes correlated closely with the intragastric lysolecithin concentration and the bacterial contamination. There was no correlation between amount of reflux and Visick-grading, no complaints were found in some cases with extreme reflux values as vice versa. The postoperative alkaline reflux gastritis is described, the consequences including the carcinoma of the gastric stump are mentioned. The principles of ulcer surgery are analysed under these aspects.

     

The extent of bile reflux and development of gastric cancer after resections in rat

Authors performed gastric resections (Billroth II, Billroth I, Billroth II+, Braun anastomosis and Roux Y reconstruction) and laparotomies in five groups, of 110 Wistar male rats. Thirty-eight weeks postoperatively the surviving 91 animals were sacrificed, and histological study was made of the frequency of gastric stump cancer in the individual groups and the extent of bile reflux characteristic of the individual GEA types was measured. Based on their results, the risk of stump cancer was higher in operation types associated with considerable bile reflux (a cancer incidence rate of 50% after Billroth II, 28.5% after Billroth I). Following gastric resections accompanied by insignificant bile reflux (Billroth II + Braun, Roux Y) the risk of gastric stump cancer was significantly lower.     

Biliary bile acid concentration is a simple and reliable indicator for liver function after hepatobiliary resection for biliary cancer

BACKGROUND: The functional recovery of the remnant liver after an extended hepatectomy is critical for the outcome of the patient. The aim of this prospective study was to examine whether biliary bile acids could be an indicator for postoperative liver function. METHODS: Externally drained bile samples were obtained from 51 patients with biliary or periampullary carcinomas before and after surgery. Patients were categorized into 3 groups: group A, 29 hepatectomized patients without liver failure; group B, 7 hepatectomized patients with liver failure (maximum serum bilirubin level, >10 mg/dL); and group C, 15 patients who underwent biliopancreatic resection without hepatectomy, with a good postoperative course. Bile samples were withdrawn 1 day before surgery and on postoperative days 1, 2, 3, 4, 6, and 7. Total bile acids were measured with a 3 alpha-hydroxysteroid dehydrogenase method. RESULTS: Before surgery, the concentration of bile acids was higher in groups A and C than in group B, and correlated significantly with the indocyamine green disappearance rate (KICG) values (R(2) = 0.557; P <.0001). After surgery, bile acid concentrations decreased in all 3 groups until postoperative day 2, which was followed by a gradual increase. The concentration recovered to the preoperative level in groups A and C but remained low in group B. Biliary bile acid concentrations on day 2 correlated significantly with remnant liver KICG values (R(2) = 0.257; P =.0019). Among several parameters studied, including KICG, remnant liver KICG, biliary bile acids, and biliary bilirubin, biliary bile acid concentration had the most predictive power for occurrence of postoperative liver failure. CONCLUSION: Biliary bile acid concentration could be a simple, real-time, reliable indicator of preoperative and postoperative liver function.     

Pylorus and antroseromuscular flap-preserving gastrectomy--a new type of reconstruction after subtotal gastrectomy for treatment of gastroduodenal ulcer: clinical and experimental study

To avoid motility disturbances after Billroth gastrectomy, the authors designed pylorus and antroseromuscular flap-preserving subtotal gastrectomy (PAFPG). Results showed that gastric motility and emptying time of dogs after PAFPG were close to normal. PAFPG was applied to 125 consecutive patients with gastroduodenal ulcers (gastric ulcer, 15 patients; duodenal ulcer, 94 patients; and combined ulcers, 16 patients) confirmed by barium examination and fibro-gastroendoscopy. All patients recovered smoothly, none of them had postoperative complication. Gastric acid output reduction rates were as follows: basal acid output, 85.05% +/- 8.13%; maximal acid output, 81.76% +/- 10.85%; peak acid output, 81.42% +/- 10.15%. The incidence of postoperative enterogastric reflux (endoscopically) and the concentration of cholic acids in gastric juice were significantly lower in patients after PAFPG than after Billroth I or II gastrectomy. Results suggest that PAFPG reduced gastric acid outputs definitely and overcame adverse motility consequences after Billroth I or II gastrectomy desirably.     

Bile reflux and degree of gastritis in patients with gastric ulcer: before and after operation

The incidence and degree of bile reflux and gastritis has been measured in normal subjects and in patients with gastric ulcer before operation and after treatment by highly selective vagotomy with ulcer excision, Billroth 1 partial gastrectomy, and truncal vagotomy and drainage. Before operation patients had significantly higher (P less than 0.001) bile acid concentrations in the stomach than normal subjects. Treatment by highly selective vagotomy resulted in significantly lower bile acid concentrations than those before operation and those found after Billroth 1 partial gastrectomy. Antral and body gastritis was significantly less in normal subjects than in the preoperative and all postoperative groups. There was no significant difference in antral or body gastritis between the preoperative gastric ulcer patients and the patients after any of the surgical procedures despite the significant differences in bile acids. Though highly selective vagotomy in the treatment of gastric ulcer results in a reduction in duodenogastric reflux of bile there is no improvement in the gastritis that is present.     

Effectiveness of Roux-en-Y Reconstruction After Distal Gastrectomy Based on an Assessment of Biliary Kinetics

Purpose: To evaluate the efficiency of Roux-en-Y reconstruction (RY) after distal gastrectomy we compared postoperative physiological functions and disorders among patients who underwent RY, conventional Billroth I reconstruction (BI), or Billroth II reconstruction (BII). Methods: The subjects were 91 patients who had undergone distal gastrectomy for gastric cancer more than 1 month earlier. To examine the severity of gastroesophageal reflux, acid reflux and alkali reflux were assessed, and to examine the severity of duodenal reflux into the remnant stomach, biliary scintigraphy was performed. The degree of inflammation in the esophagus and remnant stomach was examined by endoscopy. Questionnaires on postoperative complaints were sent out to the patients to determine how serious their reflux symptoms were. Results: Both acid and alkali reflux were mild in the RY group. Biliary reflux into the remnant stomach, as assessed by biliary scintigraphy, was significantly less severe in the RY group than in the BI and BII groups. Endoscopy showed that inflammation of the lower esophagus and remnant stomach was much less severe in the RY group than in the BI and BII groups. According to the questionnaire survey, none of the patients in the RY group reported any reflux symptoms. Conclusions: In this series, RY was found to be a superior reconstruction method after distal gastrectomy since it was rarely accompanied by the reflux of duodenal juice into the remnant stomach or gastric reflux into the lower esophagus. Received: May 9, 2001 / Accepted: January 8, 2002 Reprint requests to: K. Shinoto     

Effects of duodenogastric bile reflux on gastric mucosal hexosamine concentrations after selective proximal vagotomy or extended distal gastrectomy for duodenal ulcer

Effects of duodenogastric reflux (DGR) of bile on hexosamine concentrations in gastric mucosa were studied in 17 healthy controls and 133 patients with duodenal ulcer patients before and after surgery. Total bile acid concentration in gastric juice was measured using enzyme method to estimate DGR. Mucosal hexosamine concentration of the biopsy specimens taken from the gastric corpus and antrum was measured according to Boas's method. The operative procedures included selective proximal vagotomy (SPV) with or without pyloroplasty, and extended distal gastrectomy with Billroth I(BI) or II(BII) anastomosis. The rate of DGR were significantly higher in cases after gastrectomy, especially in BII cases than in cases after SPV. In the early postoperative period after SPV with or without pyloroplasty, DGR was increased significantly. However, the reflux rate was decreased gradually to the preoperative level thereafter, suggesting that normal function of gastric emptying might be recovered with time. The hexosamine concentration of the antral mucosa showed clearly an inverse relationship to the changes in DGR rate. These results suggested that SPV could be the more physiological procedure than gastrectomy from the point of DGR.