Evaluation of Surgical Treatment of Duodenal Ulcer: Short- and Long-term Effects

Current state of gastric stump carcinoma in Japan: based on the results of a nationwide survey

Background

Carcinoma of the gastric remnant after partial gastrectomy for benign disease or cancer is unusual but an important cancer model. The Japanese Society for the Study of Postoperative Morbidity after Gastrectomy (JSSPMG) performed a nationwide questionnaire survey to understand the current state of gastric stump carcinoma in Japan.

Methods

In the questionnaire survey of November 2008, gastric stump carcinoma was defined as an adenocarcinoma of the stomach occurring 10 years or more after Billroth I or Billroth II gastrectomy for benign condition or cancer disease. The survey was conducted at the request of reports on five or more patients with gastric stump carcinoma for each institution. Items for the survey included gender, age, methods of reconstruction in an original gastrectomy, original diseases, time interval between original gastrectomy and first detection of stump carcinomas, locations of stump carcinomas, tumor histology, tumor depth, and extent of lymph node metastasis. The questionnaire was sent to 163 surgical institutions in the JSSPMG.

Results

Ninety-five institutions (58.3%) responded to the survey, and the data of 887 patients satisfied the required conditions for the survey. A total of 887 patients were composed of 368 patients who received Billroth I distal gastrectomy and 519 who received Billroth II. The Billroth II group has a significantly higher number of original benign lesions than the Billroth I group (P < 0.001). This study confirmed the following issues: (1) The remnant stomach after gastrectomy for cancer disease had a higher prevalence to develop stump carcinomas occurring in a shorter time interval since original gastrectomy; (2) Patients with Billroth II gastrectomy had stump carcinomas most frequently in the anastomotic area, but not in the non-stump area as in Billroth I gastrectomy; (3) Tumor histology of 72.4% of 304 stump carcinomas at an early stage was intestinal type adenocarcinoma, i.e., well or moderately differentiated adenocarcinoma, whereas it decreased to 42.2% at the locally advanced stage of 521 stump carcinomas (P = 0.0015), suggesting that stump carcinoma mostly may develop from intestinal type and change to diffuse type during the evolution to advanced stage cancers.

Conclusions

This large series of surveys suggest that there are two distinct biological plausibilities in the development of gastric stump carcinoma: (1) it develops in a shorter time interval of 10 years or less since the original gastrectomy, may come from a higher risk of gastric mucosa after gastrectomy for cancer diseases that highly predisposes to cancer, and (2) it develops during a longer time interval of 20 years or more, may come from gastrectomy-relating mechanisms after gastrectomy for original benign diseases.

Trends and predictors for vagotomy when performing oversew of acute bleeding duodenal ulcer in the United States

BACKGROUND

In the era of Helicobacter pylori treatment, the role of vagotomy in bleeding duodenal ulcers is debatable. National outcomes were evaluated to determine the current surgical treatment and use of vagotomy for bleeding duodenal ulcers.

METHODS

Data from the Nationwide Inpatient Sample (NIS) were used from years 1999 to 2003. Patients were selected using diagnostic codes for acute duodenal ulcer bleed and procedure codes for simple oversew of a bleeding ulcer and vagotomy. Data were analyzed using multiple linear and logistic regression.

RESULTS

Between 1999 and 2003, 100,931 patients with an acute bleeding duodenal ulcer were identified. Over time, there was a decrease in the number of acute bleeding ulcers (p = 0.027) and a decrease in the number of vagotomies (p = 0.027). A high co-morbidity index [odds ratio (OR), 0.60, p = 0.017], operation in the Midwest (OR 0.50, p < 0.001) and operation in the West (OR 0.68, p = 0.034) were predictive of no vagotomy during surgery for a bleeding duodenal ulcer.

CONCLUSIONS

A vagotomy is not commonly performed during surgical treatment of an acute bleeding duodenal ulcer. This variation in practice was not fully explained by patient characteristics. We must seek new evidence to determine the safety of combined medical and surgical management of this clinical problem.

Variations reported in surgical practice for bleeding duodenal ulcers

BACKGROUND

To determine the current surgical management of bleeding duodenal ulcers in our program, faculty (FAC) and residents (RES) were surveyed.

METHODS

FAC (n = 33) and RES (n = 42) were surveyed regarding their surgery of choice between oversew (OS) or acid-reducing procedures (ARPs) in 4 scenarios. FAC who had recertified in general surgery (RECERT) were compared with young FAC who had not and RES (RES/young FAC). Two-group comparisons were performed.

RESULTS

Seventy-three percent of FAC and 62% of RES responded. RES perform more ARPs on hemodynamic (HD), unstable, elderly patients than FAC (P = .013). On the elderly patient, RES/young FAC perform more ARPs in a HD stable (P = .07) and unstable condition (P = .18). HD unstable patients would undergo OS more frequently than stable patients (P = .016).

CONCLUSIONS

In this survey, the choice of optimal surgical procedure for an acute bleeding ulcer varies among surgeons based on years of surgical experience and individual patient factors.

Comparative Analysis of Vagotomy and Drainage Versus Vagotomy and Resection Procedures for Bleeding Peptic Ulcer Disease: Results of 907 Patients from the Department of Veterans Affairs National Surgical Quality Improvement Program Database

BACKGROUND

The purpose of this study was to determine postoperative outcomes and risk factors for morbidity and mortality in patients requiring surgery for bleeding peptic ulcer disease (PUD). Vagotomy and drainage procedures are technically simpler but are usually associated with higher ulcer recurrence rates. In contrast, vagotomy and resection approaches offer lower ulcer recurrences but represent much more challenging operations and are associated with considerable morbidity and mortality.

STUDY DESIGN

Data collected through the Department of Veterans Affairs National Surgical Quality Improvement Program database from 1991 to 2001 were submitted for stepwise logistic regression analysis for prediction of 30-day postoperative morbidity and mortality, rebleeding, and postoperative length of stay. The study population included all patients operated on for bleeding PUD within an 11-year period.

RESULTS

The 30-day morbidity, mortality, and rebleeding rates were comparable between surgical groups. Age, American Society of Anesthesiologists class, presence of ascites, coma, diabetes, functional status, hemiplegia, and history of steroid use were predictors of postoperative death. Risk factors for rebleeding included dependent functional status, history of congestive heart failure, smoking, steroid use, and preoperative transfusions. Having a resective procedure, American Society of Anesthesiologists class, hemiplegia, history of COPD, and requiring ventilator-assisted respirations before surgery were positively associated with increased length of hospital stay.

CONCLUSIONS

No differences were observed in 30-day mortality, morbidity, or rebleeding rates between surgical groups. Having a resective procedure was a predictor of prolonged postoperative stay. Dependent status and chronic use of steroids were predictors of both rebleeding and postoperative mortality.

[Late follow-up of gastrectomized patients for peptic ulcer: clinical, endoscopic and histopathological aspects]

BACKGROUND

The gastrectomy is an uncommon procedure because the proton bomb inhibitors associated to the antibiotic outlines used to eradicate the Helicobacter pylori changed the focus of the peptic ulcer treatment.

AIMS

Later evaluation on those patients who underwent partial gastrectomy as a treatment for peptic ulcer, at that time when any drug to eradicate the Helicobacter pylori was not used. The clinical evaluation included the late postoperative symptoms and postgastrectomy syndromes like dumping, diarrhea, alkaline gastritis and nutritional aspects. The upper digestive endoscopy analysed the surgery reconstruction and the gastric stump, the duodenum and the jejunum mucosa aspects. The histopathological evaluation included looking for Helicobacter pylori by using two different methods: histology and urease test.

CASUISTIC AND METHODS

Fifty-nine patients, 44 (74.6%) male, median age 55.5 years old (range from 31 to 77 years old), who underwent a clinical interview and an upper digestive endoscopy. Paraffin blocks from the surgical specimen were reviewed in order to find out if the patients did have or did not have Helicobacter pylori before surgery.

RESULTS

The final results show that most of the patients had very good and good clinical evolution (Visick I e II) in 96%. The most common symptoms on late postoperative are mild dyspepsia with or without Helicobacter pylori, and diarrhea, anemia and dumping occurred in, respectively, 11 (18.6%), 2 (3.4%) and 2 (3.4%) cases. The Billroth I reconstruction had the best clinical results on statistical rate. The endoscopic finding showed normal results in the most number of cases, and reflux alkaline gastritis or erosive gastritis in a few cases. Ulcer recurrences were diagnosed in two patients (3.4%), and both had positive Helicobacter pylori. Most of the patients had Helicobacter pylori (86%) before surgery and also in the postoperative time (89.9%).

CONCLUSIONS

The patients had a very good clinical evolution after the gastrectomy. The Billroth I reconstruction had the best clinical results. The Helicobacter pylori is still present on gastric stump in late postoperative time, and we believe that it does not bring any negative influence to surgical results.

Clinical analysis of pericardial devascularization by preserving vagus trunks in 42 patients with portal hypertension

Pericardial devascularization (PD) is less of a burden and provides better bleeding control to patients compared to shunt procedures, and so has been widely used in portal hypertension in China. However, because the vagus trunks are interrupted during surgery, patients easily suffer from postoperative stomach adynamia. Based on our understanding from autopsy of the path of vagus trunks along the distal oesophagus, we designed the operative procedure of PD by preserving vagus trunks (PDPVT) to treat portal hypertension. Between May 1991 and January 2003, patients with portal hypertension were treated surgically using PDPVT (n = 42), single PD (n = 32), or PD with pyloroplasty (PD+PP; n = 16). Operative mortality was 2.4% in the PDPVT group and 6.3% in both the PD and PD+PP groups (p > 0.05). The postoperative rebleeding rate was 9.5% in the PDPVT group and 12.5% in both the PD and PD+PP groups (p > 0.05). There were no differences in operative time and estimated blood loss between the three groups (p > 0.05). The recovery time for gastroenteric function was shorter with PDPVT (mean, 3.5 days) than with PD (mean, 5.7 days) and PD+PP (mean, 4.2 days; p < 0.02). Incidences of early satiety and enterogastric reflux were significantly lower in the PDPVT group (both 4.8%) than in the PD group (46.9% and 18.8%) and PD+PP group (12.5% and 100%; p < 0.005). Incidences of retention of gastric juice, diarrhoea and late gallstones were 12.5%, 15.6% and 7.1%, respectively, in the PD group, and 12.5%, 18.8% and 6.3%, respectively, in the PD+PP group, but none of these were observed in the PDPVT group. Since it preserves vagus trunks, PDPVT can maintain normal stomach dynamics and physiological function of hepatobiliary and gut systems better than PD and PD+PP, thus reducing incidences of postoperative complications. Thus, PDPVT is superior to PD and PD+PP in the treatment of portal hypertension.

Laparoscopic management of peptic ulcer disease

Laparoscopic surgery has heralded a new era for the operative management of peptic ulcer disease. With a mean hospital stay of 3.5 days,22 a recurrence rate of 4% to 11%,1,3 and a morbidity from dumping and diarrhea of 1% to 2%,21 laparoscopic proximal gastric vagotomy can truly provide a good alternative to medical therapy. Despite the high cost of medical care and surgical equipment, a laparoscopic vagotomy should be cost effective compared with life-long pharmacologic management of peptic ulcer disease. Several different operative procedures have been discussed, with similar outcomes. The surgeon has a choice of several approaches, depending on his or her training and level of skill. As surgeons gain experience with laparoscopic surgery, we are able to offer consistently good results with low recurrence rates and negligible morbidity and mortality. Minimally invasive surgery has rekindled the operative treatment of peptic ulcer disease.

Gastric motor physiology and pathophysiology

The stomach has two distinct physiologic motor areas: the proximal stomach and the distal stomach. The proximal stomach, with its slow, sustained contractions, has a key role in regulating intragastric pressure and gastric emptying of liquids, while the distal stomach, with its peristaltic contractions, has a major role in mixing, trituration, and emptying of solids. Diseases and operations that disturb the motility of these two areas can result in unique adverse motor sequelae. For example, operations that impair proximal gastric motility, such as proximal gastric resection, may cause rapid gastric emptying of liquids and subsequent dumping and diarrhea. In contrast, operations that impair distal gastric contractions, such as truncal vagotomy, may cause slow gastric emptying of solids and chronic gastric atony. Knowledge of the physiology of the stomach in health and of the pathophysiology with disease and after operation provides a basis for the successful treatment and prevention of these disorders.

Laparoscopic repair of duodenal ulcer and gastroesophageal reflux

The low morbidity and early recovery associated with laparoscopic procedures have heralded a new era for many types of surgery. In addition to the initial promising reports for duodenal ulcer disease and gastroesophageal reflux discussed above, there is a growing body of reports of gastric procedures performed laparoscopically, including omentopexy for perforated duodenal ulcer and laparoscopic repair of full-thickness stomach injury. Laws et al recently described the use of transthoracic vagotomy in recurrent peptic ulcer disease for four patients who had previously undergone a gastric drainage procedure. As with any new procedure, laparoscopic techniques for duodenal ulcer and Nissen fundoplication reviewed in this section need to be evaluated further for long-term effectiveness and comparability to existing therapy. At least one controlled multicenter trial is ongoing to compare the long-term results and cost-effectiveness of laparoscopic surgery for duodenal ulcer with those of standard medical therapy, and as surgeons gain more experience with these laparoscopic procedures, it is likely that other similar trials will be initiated.

Postgastrectomy syndromes

Anatomic and physiological changes introduced by gastric surgery result in postgastrectomy syndromes in approximately 20% of patients. Most of these disorders are caused by operation-induced abnormalities in the motor functions of the stomach, including disturbances in the gastric reservoir function, the mechanical-digestive function, and the transporting function. Division of the vagal innervation to the stomach and ablation or bypass of the pylorus are the most significant factors contributing to postgastrectomy syndromes. Either rapid or slow emptying may result, depending on the relative importance of lack of a compliant gastric reservoir, loss of an effective contractile force, and loss of controlling factors that slow or speed gastric emptying and result in duodenal-gastric reflux. Clearly defining which syndrome is present in a given patient is critical to developing a rational treatment plan. In syndromes with slow gastric emptying, bilious vomiting, or alkaline reflux gastritis, the use of endoscopy is essential to rule out mechanical causes of the syndrome. Contrast radiography and scintigraphic gastric emptying studies are useful to document rapid or delayed gastric emptying. Postgastrectomy syndromes often abate with time. Conservative measures, including medical, dietary, and behavioral therapy, should be given at least a 1-year trial. If these nonoperative measures fail, surgical therapy is recommended. The Roux-en-Y gastrojejunostomy is useful for patients with dumping, because it slows gastric emptying and the transit of chyme through the Roux limb. The same operation helps patients with alkaline reflux gastritis, because it diverts pancreaticobiliary secretions away from the gastric remnant. Near-total gastrectomy, which reduces the size of a flaccid gastric reservoir, can be used to treat delayed gastric emptying. This operation should be combined with the Roux procedure to prevent postoperative reflux gastritis and esophagitis. Newer techniques, such as gastrointestinal pacing and the uncut Roux operation, may improve the treatment of the postgastrectomy syndromes in the future.

Review article: maintenance treatment with H2-receptor antagonists for peptic ulcer disease

In recent years a number of different strategies for managing patients with peptic ulcer disease have become available. The present review discusses the relative merits of each form of treatment. Intermittent treatment (whether given in response to symptoms or as a prophylactic regimen prescribed seasonally or at weekends) fails to prevent ulcer recurrence and leaves patients at risk of haemorrhage and perforation. Anti-Helicobacter pylori therapy, although useful in certain circumstances, cannot be recommended for all patients with ulcer disease because of side effects and, in any case, requires further assessment of efficacy. Gastric surgery reduces ulcer recurrence and complications, but operations which have a low incidence of side effects are associated with higher rates of ulcer recurrence, particularly when patients are followed up for more than 10 years. Long-term continuous maintenance treatment with H2-receptor antagonists for 5 or more years effectively prevents ulcer recurrence in the majority of patients and significantly reduces the risk of ulcer complications. In addition, maintenance treatment has proved to be safe and is well tolerated by patients. Maintenance treatment with H2-receptor antagonists is the preferred option for the management of patients with peptic ulcer disease.

Complications associated with ulcer recurrence following gastric surgery for ulcer disease

The present study is an attempt to assess the risks of the complications associated with recurrent ulcers in patients who have undergone gastric surgery and to determine whether these risks differ from those observed in patients receiving long term maintenance treatment with H2-receptor antagonists for ulcer disease. One hundred and thirty studies reported in the literature during the past three decades have been analysed to determine both the approximate rate of ulcer recurrence and the proportion of patients with recurrent ulcers who have presented with either haemorrhage or perforation following the various types of gastric surgery for ulcer disease. From these data, estimates of the risks of haemorrhage and of perforation during the years following gastric surgery have been calculated. Vagotomy and antrectomy is associated with a low risk of ulcer recurrence (less than 1%) and the risk of complications in later years is accordingly very small (less than 0.5%). Partial gastrectomy, although associated with low recurrence rates, has a higher risk of complications (1.3% for haemorrhage, 0.3% for perforation) because the proportion of recurrent ulcers that present with haemorrhage or perforation is high (33% and 8%, respectively). Truncal vagotomy plus drainage (TV + D) and highly selective vagotomy (HSV) are associated with recurrence rates of 9% and 12%, respectively, but ulcer recurrences following these operations are less frequently accompanied by complications then recurrences after gastric resection and, as a result, the risks of haemorrhage (1.7% for TV + D; 1.3% for HSV) are similar to the risks after gastric resection. During long term (five years or more) maintenance treatment with H2-receptor antagonists, the risks of haemorrhage and perforation are less than 2% and less than 0.5%, respectively. It appears, therefore, that the likelihood of developing haemorrhage or perforation following gastric surgery is of the same order as that during maintenance treatment with H2-receptor antagonists, at least during the first decade of follow-up.

Effect of partial gastrectomy on serum anti-Helicobacter pylori immunoglobulins in peptic ulcer patients

Since biliary enterogastric reflux is suggested to eradicate gastric infection with Helicobacter pylori (HP), we have investigated in a prospective randomized study the effect of partial gastrectomy with either Billroth II or Roux-en-Y anastomosis on infection with HP as assessed by the titers of IgG and IgA antibodies against HP in serum. These antibodies were measured by ELISA in serum of 22 patients before and at 10 days and 6, 15, and 24 months after either Billroth II (N = 11) or Roux-en-Y (N = 11) gastrectomy for peptic ulcer. All patients had HP demonstrated in their preoperative endoscopic gastric biopsies. The preoperative serum IgA antibodies against HP (anti-HP IgA) were increased in 20 of the 22 patients (range 0.21-1.69) while the IgG antibodies (anti-HP IgG) were increased in all 22 patients (range 0.38-1.31). Four of the Billroth II patients had clearance of HP from gastric biopsies accompanied by rapid and pronounced decrease of anti-HP IgA. In contrast, the patients with Roux-en-Y gastrectomy and the Billroth II patients with persistent HP infection had no change in anti-HP IgA after surgery. Anti-HP IgG showed variable results in the four patients without gastric HP infection and was not affected by gastrectomy in the patients with persistent HP infection. We concluded that serum anti-HP IgA, but not anti-HP IgG, is helpful in identifying those patients in whom HP is no longer demonstrable after Billroth II gastrectomy. Gastrectomy with Roux-en-Y anastomosis had no effect on gastric HP infection.

Cancer incidence following subtotal gastrectomy

Hawaiian Japanese men (n = 432) who had undergone subtotal gastrectomy for peptic ulcers before 1971-1975 were followed up for detection of cancer development. They showed a significant increase in colon cancer risk (P = 0.008) and lung cancer risk (P = 0.002) compared with 6161 nongastrectomized men. The association with lung cancer persisted after adjustment for cigarette use (P = 0.03). Alcohol consumption was associated with colonic cancer in this cohort, and gastrectomized men consumed more alcohol than nongastrectomized men; however, the association of gastrectomy with colon cancer persisted after adjustment for alcohol use (P = 0.02). Gastrectomized men were lighter and had lower serum lipid levels than controls, suggesting that undernutrition might favor the development of some cancers. The type of gastroenteric anastomosis did not influence the cancer risk level in the colon or lung.

The Roux operation for postgastrectomy syndromes

The aim of this paper is to describe the technique, indications, and results of the Roux operation as used in the treatment of postgastrectomy syndromes. A Roux gastrojejunostomy with a 40-cm Roux limb is the procedure of choice for alkaline reflux gastritis, because it virtually eliminates reflux of bile and pancreatic juice into the stomach. The slow transit through a Roux limb can also be used to good advantage to slow gastric emptying in patients with dumping. Patients with delayed gastric emptying respond to the combination of near-total gastric resection, which removes the atonic gastric remnant and speeds emptying, and Roux-Y gastrojejunostomy, which prevents reflux esophagitis and provides a reservoir for ingesta in the upper gut. After all Roux operations, however, the Roux limb may slow emptying so much that pain, fullness, nausea, and food vomiting result, the so-called Roux stasis syndrome. Prevention of the Roux stasis syndrome with an "uncut" Roux limb and the treatment of the syndrome by using electrical pacing to suppress the ectopic pacemakers that emerge in the limb offer possible new solutions to this vexing problem.

Manometric evaluation of jejunal limb after total gastrectomy and Roux-Orr anastomosis for gastric cancer

Total gastrectomy with Roux-Orr anastomosis is frequently performed for gastric cancer. Since intestinal motility of the Roux limb has never been evaluated after this operation, pressure activity was investigated in the Roux limb of ten patients (aged 51-77 years) who had undergone total gastrectomy and Roux-Orr reconstruction. Investigations were carried out during a 6-h fast and 3 h after a 605 kcal mixed meal. During fasting only two patients had activity fronts and these were abnormal. All ten patients displayed non-propagating bursts of contractions and three had discrete clustered contractions and high amplitude jejunal contractions. The fed state was characterized by a severely reduced motor activity pattern and other abnormalities. Total gastrectomy with Roux-Orr anastomoses provokes a relatively severe disturbance in intestinal activity.

Effect of gastrectomy with either Roux-en-Y or Billroth II anastomosis on small-intestinal function

This prospective study was undertaken in patients scheduled for gastrectomy for peptic ulcer disease to determine the effect of partial gastrectomy with either Roux-en-Y (n = 11) or Billroth II anastomosis (n = 11) on the function of the small intestine. Patients were studied before and at 6 months (blood and small-intestinal function tests) and at 24 months (blood tests) postoperatively. Median postoperative body weights at 6 months (70.5 kg; p less than 0.01) and 12 months (70.3 kg; NS) were lower than preoperatively (73.0 kg). Haemoglobin concentrations at 6 months (8.9 mM; p less than 0.01) and at 24 months (9.1 mM; p less than 0.05) were also significantly reduced compared with the preoperative value (9.5 mM). However, neither at 6 nor at 24 months postoperatively were there significant changes for serum iron, iron saturation, folic acid, vitamin B12, protein, albumin, alkaline phosphatase, and calcium concentrations. Whereas no significant deterioration of the absorption of D-xylose and vitamin B12 or of faecal fat excretion was observed, the orocoecal transit time was significantly shortened from 98 to 50 min (p less than 0.01), the expiratory hydrogen excretion after a 50-g oral glucose load was significantly increased from 8 to 54 ppm (p less than 0.01), as was indicanuria from 257 to 368 mumol/24 h (p less than 0.01). Apart from a lower serum iron concentration and iron saturation index in the Roux-en-Y patients 6 months postoperatively (p less than 0.05), no significant differences between the two types of anastomosis were observed. It is therefore concluded that both in patients with Roux-en-Y and in those with Billroth II anastomosis most abnormalities observed after gastrectomy are secondary to an accelerated small-intestinal transit.

Costs of medical and surgical treatment of duodenal ulcer

Proximal gastric vagotomy and intermittent and maintenance therapy with H2-antagonists have all been claimed to be effective in long-term management of duodenal ulcer disease. The model of a Markov chain was used to compare their costs by a medical decision analysis. The high price of the initial procedure made proximal gastric vagotomy the most expensive therapy, its costs rising from +10,600 after 1 yr to +12,200 after 15 yr. The average costs of intermittent therapy per patient rose from +500 to +7500. Maintenance therapy cost as much as intermittent therapy but provided 8% and 4% more time spent free of ulcer relapse and pain, respectively. In a sensitivity analysis, the order of the therapeutic options regarding their cost-effectiveness remained robust to changes in the assumptions underlying the model. In a European health care system, the initial surgical procedure cost only one-seventh of the average annual income compared with two-thirds in the United States, and proximal gastric vagotomy turned out to be the cheapest therapy after 6 yr. These results suggest that maintenance therapy provides the best long-term management. Gastric surgery may represent a cost-effective measure of ulcer prevention in Europe but not in the United States.

24-H study of gastric acidity and bile acid concentration after parietal cell vagotomy

Intragastric pH and bile acid concentration (BAC) were measured over 24 h in 15 sham-feeding-negative patients, 7 with and 8 without recurrence after parietal cell vagotomy (PCV). Nocturnal acidity was significantly higher (p less than 0.02), and BAC at night significantly lower (p less than 0.02), in the group with recurrent ulcers after PCV. There were no significant differences postoperatively in acidity or BAC between seven patients who had duodenal ulcers preoperatively and eight patients with prepyloric ulcers. High nocturnal acidity may be a predisposing factor for the recurrence of ulcers after PCV.