- Email: [email protected]
Gastric outlet obstruction with benign endoscopic biopsy should be further explored for malignancy
Gastric outlet obstruction with benign endoscopic biopsy should be further explored for malignancy Amjad Awan, MD, David E. Johnston, MD, M. Mazen Jamal, MD Albuquerque, New Mexico
Background: Gastric outlet obstruction is commonly considered a complication of peptic ulcer disease. Malignancy accounts for up to 39% of gastric outlet obstruction. The object of this study was to evaluate the reliability of endoscopic biopsies in excluding malignancy as the cause of gastric outlet obstruction. Methods: A retrospective study of 40 consecutive patients admitted with gastric outlet obstruction was conducted. Patient demographics, their use of H2-receptor antagonists or nonsteroidal anti-inflammatory drugs, and history of peptic ulcer disease were recorded. Histopathologic results of the endoscopic biopsy and surgical specimen were reviewed. The diagnosis based on the surgical specimen was considered the gold standard. Results: Sixteen patients (40%) had malignant gastric outlet obstruction. Seven patients had gastric adenocarcinoma and nine had extragastric tumors. The patients with malignant obstruction were significantly older (> 55 years) (p = 0.03; odds ratio: 95% CI: 5.21 [1.05-23.49]). Gastric cancer patients had less frequently a history of peptic ulcer disease when compared with patients with benign gastric outlet obstruction (p = 0.04; odds ratio: 95% CI: 5 [1.04-38.13]). Endoscopic biopsy to detect malignant obstruction had poor sensitivity (i.e., 37%) when compared with biopsies of the surgical specimen. In three of seven patients with gastric cancer (40%), repeated jumbo biopsies were negative for malignancy. Conclusion: Patients with gastric outlet obstruction who had endoscopic biopsies negative for cancer should be explored surgically before embarking on medical therapy. The surgical exploration is especially important in gastric outlet obstruction patients who are considered at high risk for malignancy, that is, those who are older and have no history of peptic ulcer disease. (Gastrointest Endosc 1998;48:497-500.)
Gastric outlet obstruction (GOO) is commonly considered a complication of peptic ulcer disease (PUD).1 However, malignant tumors account for up to 39% of cases of GOO.2-4 Some small studies have suggested that endoscopic balloon dilatation of the obstructed gastric outlet is an acceptable alternative to surgery.5-8 A conclusion that GOO is of benign etiology can be misleading, if determined Received July 10, 1997. For revision November 17, 1997. Accepted June 5, 1998. From the Department of Gastroenterology and Hepatology, University of New Mexico Health Sciences Center, Albuquerque, New Mexico. Reprint requests: M. Mazen Jamal, MD, Department of Gastroenterology and Hepatology, University of New Mexico Health Sciences Center, Albuquerque, NM 87131-5271. Copyright © 1998 by the American Society for Gastrointestinal Endoscopy 0016-5107/98/$5.00 + 0 37/1/92250 VOLUME 48, NO. 5, 1998
solely by endoscopic biopsy. This observation has been made repeatedly in patients in our experience in whom biopsies of the obstructing lesion, taken at endoscopy, were reported as benign, and yet patients were subsequently found to have malignant tumors at surgery. The objective of this study was to evaluate the reliability of negative endoscopic biopsies in excluding malignant disease as a cause of gastric outlet obstruction. PATIENTS AND METHODS The records of 40 consecutive patients admitted during 1993 to 1996 with GOO diagnosed by upper endoscopy or upper gastrointestinal barium x-ray study were reviewed. GOO was defined as the inability to pass a 9.5 mm diameter endoscope (GIF 130; Olympus America, Inc., Melville, N.Y.) through the obstruction or a barium study that GASTROINTESTINAL ENDOSCOPY
497
A Awan, D Johnston, M Jamal, et al.
Gastric outlet obstruction with benign endoscopic biopsy and malignancy
Table 1. Clinical characteristics of patients
Mean age (yr) Race White Non-white Gender Male Female Endoscopic biopsy HP+ H2RA PPI NSAID Hx of PUD
Total (n = 40)
Non-cancer (n = 24)
Cancer (surgical path, n = 16)
p Value
53 ± 16
48 ± 15
62 ± 14
0.03
15 25
11 13
4 12
NS
20 20 38 19 20 6 4 12
14 10 24 Benign, 0 malignant 13 15 4 3 10
6 10 10 Benign, 6 malignant 6 5 2 1 2
NS 0.02 NS NS NS NS 0.04
PUD, Peptic ulcer disease; EGD, esophagogastroduodenoscopy, HP, Helicobacter pylori; H2RA, H2 receptor antagonist; PPI, proton pump inhibitor.
Table 2. Multivariate analysis of factors predictive of malignant gastric outlet obstruction Factor
p Value
Odds ratio (95% CI)
Age > 55 No history of PUD
0.03 0.04
5.21 [1.05-23.49] 5 [1.04-38.13]
PUD, Peptic ulcer disease.
demonstrated narrowing of the gastric outlet. All patients had symptoms suggestive of GOO (i.e., intractable vomiting). Patient demographic data, use of H2-receptor antagonists (H2RAs), nonsteroidal anti-inflammatory drugs (NSAIDs), and ulcer history were recorded. Histopathologic results of biopsies taken during upper endoscopy where available were recorded, and the presence of Helicobacter pylori or gastritis was noted. The biopsies were taken from the obstructive site with regular forceps on initial examination and jumbo forceps on repeat examinations using a therapeutic upper endoscope. Operative findings and results of surgical pathology were reviewed, the latter being the gold standard for the diagnosis of malignancy. The charts of the patients diagnosed with malignant disease at the time of surgery were further reviewed for clinical presentation, endoscopic findings, type of biopsy (jumbo or regular), and results of abdominal CT performed before surgery. Statistical analysis The statistical analysis was performed using SAS/STAT software.15 Any p values < 0.05 were taken to be indicative of statistical significance. Fisher’s exact test was used for comparisons involving binary variables and the t test for continuous variables. Multivariate logistic regression was used to examine potential predictors of malignant GOO (both gastric and extragastric). Results of this analysis were expressed in terms of odds ratios and their 95% confidence intervals. 498
GASTROINTESTINAL ENDOSCOPY
RESULTS The clinical characteristics of the 40 patients who presented with GOO are shown in Table 1. Fifteen were white, 22 Hispanic, 2 Native Americans, and 1 Asian; age ranged from 16 to 82 years. Thirty-eight patients underwent upper endoscopy and biopsy. H. pylori status was determined in 34 patients, of whom 19 were positive by either the CLO test (Tri Med Specialties, Inc., Lenexa, Kan.) or histology. Thirty-five patients underwent surgery. The remaining five patients treated medically were asymptomatic at 6 months and were judged to be free of cancer. Patients were further divided into non-cancer and cancer groups based on the surgical pathology report (Table 1). Of the 16 patients (40%) with malignant GOO, 7 had gastric adenocarcinoma, 4 had pancreatic carcinoma, 3 had gallbladder cancer, and 2 had cancer of undetermined origin. The patients with malignant disease were significantly older when compared with those with benign disease (p = 0.03; odds ratio: 95% CI: 5.21 [1.05-23.49]) (Table 2). Here, age is treated as a binary variable, whereas in Table 1 it is a continuous variable. Fewer patients in the cancer group had a history of peptic ulcer disease (PUD) when compared with those with nonmalignant obstruction (p = 0.04; odds ratio: 95% CI: 5 [1.0438.13]) (Table 2). When compared with the surgical specimen, endoscopic biopsy had poor sensitivity (37%) in detecting malignant disease. Six patients in the cancer group, all with gastric cancer, and 13 in the non-cancer group were H. pylori positive. Aspirin and NSAID use was not remarkable in either group. Seven patients had gastric cancer causing GOO (Table 3). In four, additional endoscopic biopsies obtained from the obstructing lesion with jumbo forVOLUME 48, NO. 5, 1998
Gastric outlet obstruction with benign endoscopic biopsy and malignancy
A Awan, D Johnston, M Jamal, et al.
Table 3. Gastric and extragastric cancers causing GOO (n = 16) No. of patients
CT of abdomen positive for mass
CT of abdomen negative for mass
CT of abdomen not done
4
1
1
2
3
0
3
0
2
2
0
0
7
6
1
0
Endoscopic biopsy positive for malignancy Endoscopic biopsy negative for malignancy Endoscopic biopsy positive for malignancy Endoscopic biopsy negative for malignancy GOO, Gastric outlet obstruction.
ceps were positive for malignancy. Three of these four patients had an obstructive mass seen on upper endoscopy and one had no visible mass, although a biopsy was taken from the site of obstruction. Three patients with gastric cancer had negative endoscopic biopsies which remained negative despite obtaining repeated biopsies with a jumbo forceps. None of the three patients had an ulcer or mass seen on upper endoscopy. CT of the abdomen also failed to reveal any malignant lesion in these patients (Table 3). Nine patients had extragastric cancer causing GOO (Table 3). In two, endoscopic biopsies of the obstructing lesion revealed the presence of a tumor. Seven patients had a negative biopsy. CT of the abdomen revealed a mass in eight patients. One extragastric cancer was not detected by CT. Multivariant analysis by logistic regression showed that age and negative history of PUD were associated with malignant GOO (Table 2). DISCUSSION In this study, the importance of careful evaluation of patients with GOO has been emphasized. Because the majority of patients underwent upper gastrointestinal endoscopy and subsequently surgery, there was an opportunity to objectively confirm or disprove the endoscopic biopsy findings. In this era of “less invasive” therapy,9,10 it is tempting to undertake endoscopic management of GOO. However, as malignant tumors are becoming increasingly important as a cause of GOO and because the incidence of complications of long-term PUD is decreasing, it is essential to carefully consider malignancy as a cause of GOO.11,12 In a predominantly Hispanic population, malignant tumors occurred in 40% of patients presenting with GOO. Although not statistically significant, a considerable number of young, Hispanic women presented with GOO and had repeated endoscopies with jumbo biopsies which failed to reveal the presence of tumor. Surgical exploration revealed malignancy. VOLUME 48, NO. 5, 1998
In view of major changes over the last 2 decades in the use of H2RAs,11,12 aspirin, NSAIDs,13,14 and eradication of H. pylori, as well as in the availability of new diagnostic methods, the approach to GOO may need to change. We suggest the following algorithm for investigating patients with GOO. When there is intractable vomiting of more than 2 weeks duration or an upper gastrointestinal barium study shows GOO, the patient should undergo upper endoscopy. If an obstructing lesion is found, biopsy specimens should be obtained. If the biopsy shows malignancy, a CT of the abdomen should be obtained to stage the tumor and the patient should be referred for surgery. If the initial biopsy of the obstructing lesion is negative for malignancy, endoscopy should be repeated and several jumbo biopsies obtained. If these are positive for malignancy, a CT should be obtained and the patient should be referred for surgery. If the second set of larger biopsies of the obstructing lesion is negative but the patient is at high risk for malignancy (age greater than 55 years, no history of PUD), a CT of the abdomen should be performed. Where available, endosonography with fine-needle aspiration of any suspicious lesion may be performed and the patient should then be referred for surgical exploration. When there is no clinical suspicion of malignancy and endoscopic biopsies are negative, endoscopic balloon dilation and H2RA therapy may be appropriate, provided the patient is monitored for at least 1 year. ACKNOWLEDGMENTS We thank Denis M. McCarthy, MD, MSc, FACP, Professor of Medicine and Chief, Division of Gastroenterology for valuable comments and suggestions. REFERENCES 1. Kreel L, Ellis H. Pylorus stenosis in adults: a clinical and radiological study of 100 consecutive patients. Gut 1965;6:253. GASTROINTESTINAL ENDOSCOPY
499
A Awan, D Johnston, M Jamal, et al.
Gastric outlet obstruction with benign endoscopic biopsy and malignancy
2. Grahm DY. Ulcer complications and their nonoperative treatment. In: Sleisinger MH, Fordtran JS, editors. Gastrointestinal disease: pathophysiology, diagnosis, management. 5th ed. Philadelphia: WB Saunders; 1993. p. 707. 3. Jaffin BW, Kaye MD. The prognosis of gastric outlet obstruction. Ann Surg 1985;2:176-9. 4. Goldstein H, Janin M, Schapiro M, Boyle JD, Gastric retention associated with gastroduodenal disease. Am J Dig Dis 1966;11:887-97. 5. Perng CL, Lin HJ, Lo WC, Lai CR, Guo WS, Lee SD. Characteristics of patients with benign gastric outlet obstruction requiring surgery after endoscopic balloon dilation. Am J Gastroenterol 1996;5:987-90. 6. Kozarek RA, Botoman VA, Patterson DJ. Long-term follow-up in patients who have undergone balloon dilation for gastric outlet obstruction. Gastrointest Endosc 1990;17:2-4. 7. Kozarek RA, Dilation therapy for gastric outlet obstruction. Are balloons a bust [editorial]? J Clin Gastro 1993;17:2-4. 8. Sommer A, Bethge N. Relief of malignant external gastric obstruction by endoscopic implantation of a self-expanding metal stent. Endoscopy 1995;27:210-1. 9. Feritis C, Benakis P, Dimopoulos C, Georgopoulos K, Milas F,
10.
11.
12. 13.
14.
15.
Manouras A, et al. Palliation of malignant gastric outlet obstruction with self-expanding metal stents. Endoscopy 1996;28:225-8. Kuwada SK, Alexander GL. Long-term outcome of endoscopic dilation of non-malignant pyloric stenosis. Gastrointest Endosc 1995;41:15-7. Shone DN, Nikoomanesh P, Snuth-Meek MM, Bender JS. Malignancy is the most common cause of gastric outlet obstruction in the era of H2 blockers. Am J Gastroenterol 1995;90:1769-70. Johnson CD, Ellis H. Gastric outlet obstruction now predicts malignancy. Br J Surg 1990;77:1023-4. Weaver GA, Merrell NB, Harper RL, Storey JA, Jenkins PL. Nonsteroidal anti-inflammatory drugs are associated with gastric outlet obstruction. J Clin Gastroenterol 1995;20:196-8. Griffin MR, Piper JM, Daugherty JR, Snowden M, Ray W. Nonsteroidal anti-inflammatory drugs and increased risk for peptic ulcer disease in elderly persons. Ann Intern Med 1991;114:257-63. SAS/STAT Software: changes and enhancements through release 6.12. Cary (NC): SAS Institute; 1997.
Moving? To ensure continued service please notify us of a change of address at least 6 weeks before your move. Phone Subscription Services at 800-453-4351 (outside the U.S. call 314-453-4351) or fax your information to 314-432-1158.
500
GASTROINTESTINAL ENDOSCOPY
VOLUME 48, NO. 5, 1998
Background: Gastric outlet obstruction is commonly considered a complication of peptic ulcer disease. Malignancy accounts for up to 39% of gastric outlet obstruction. The object of this study was to evaluate the reliability of endoscopic biopsies in excluding malignancy as the cause of gastric outlet obstruction. Methods: A retrospective study of 40 consecutive patients admitted with gastric outlet obstruction was conducted. Patient demographics, their use of H2-receptor antagonists or nonsteroidal anti-inflammatory drugs, and history of peptic ulcer disease were recorded. Histopathologic results of the endoscopic biopsy and surgical specimen were reviewed. The diagnosis based on the surgical specimen was considered the gold standard. Results: Sixteen patients (40%) had malignant gastric outlet obstruction. Seven patients had gastric adenocarcinoma and nine had extragastric tumors. The patients with malignant obstruction were significantly older (> 55 years) (p = 0.03; odds ratio: 95% CI: 5.21 [1.05-23.49]). Gastric cancer patients had less frequently a history of peptic ulcer disease when compared with patients with benign gastric outlet obstruction (p = 0.04; odds ratio: 95% CI: 5 [1.04-38.13]). Endoscopic biopsy to detect malignant obstruction had poor sensitivity (i.e., 37%) when compared with biopsies of the surgical specimen. In three of seven patients with gastric cancer (40%), repeated jumbo biopsies were negative for malignancy. Conclusion: Patients with gastric outlet obstruction who had endoscopic biopsies negative for cancer should be explored surgically before embarking on medical therapy. The surgical exploration is especially important in gastric outlet obstruction patients who are considered at high risk for malignancy, that is, those who are older and have no history of peptic ulcer disease. (Gastrointest Endosc 1998;48:497-500.)
Gastric outlet obstruction (GOO) is commonly considered a complication of peptic ulcer disease (PUD).1 However, malignant tumors account for up to 39% of cases of GOO.2-4 Some small studies have suggested that endoscopic balloon dilatation of the obstructed gastric outlet is an acceptable alternative to surgery.5-8 A conclusion that GOO is of benign etiology can be misleading, if determined Received July 10, 1997. For revision November 17, 1997. Accepted June 5, 1998. From the Department of Gastroenterology and Hepatology, University of New Mexico Health Sciences Center, Albuquerque, New Mexico. Reprint requests: M. Mazen Jamal, MD, Department of Gastroenterology and Hepatology, University of New Mexico Health Sciences Center, Albuquerque, NM 87131-5271. Copyright © 1998 by the American Society for Gastrointestinal Endoscopy 0016-5107/98/$5.00 + 0 37/1/92250 VOLUME 48, NO. 5, 1998
solely by endoscopic biopsy. This observation has been made repeatedly in patients in our experience in whom biopsies of the obstructing lesion, taken at endoscopy, were reported as benign, and yet patients were subsequently found to have malignant tumors at surgery. The objective of this study was to evaluate the reliability of negative endoscopic biopsies in excluding malignant disease as a cause of gastric outlet obstruction. PATIENTS AND METHODS The records of 40 consecutive patients admitted during 1993 to 1996 with GOO diagnosed by upper endoscopy or upper gastrointestinal barium x-ray study were reviewed. GOO was defined as the inability to pass a 9.5 mm diameter endoscope (GIF 130; Olympus America, Inc., Melville, N.Y.) through the obstruction or a barium study that GASTROINTESTINAL ENDOSCOPY
497
A Awan, D Johnston, M Jamal, et al.
Gastric outlet obstruction with benign endoscopic biopsy and malignancy
Table 1. Clinical characteristics of patients
Mean age (yr) Race White Non-white Gender Male Female Endoscopic biopsy HP+ H2RA PPI NSAID Hx of PUD
Total (n = 40)
Non-cancer (n = 24)
Cancer (surgical path, n = 16)
p Value
53 ± 16
48 ± 15
62 ± 14
0.03
15 25
11 13
4 12
NS
20 20 38 19 20 6 4 12
14 10 24 Benign, 0 malignant 13 15 4 3 10
6 10 10 Benign, 6 malignant 6 5 2 1 2
NS 0.02 NS NS NS NS 0.04
PUD, Peptic ulcer disease; EGD, esophagogastroduodenoscopy, HP, Helicobacter pylori; H2RA, H2 receptor antagonist; PPI, proton pump inhibitor.
Table 2. Multivariate analysis of factors predictive of malignant gastric outlet obstruction Factor
p Value
Odds ratio (95% CI)
Age > 55 No history of PUD
0.03 0.04
5.21 [1.05-23.49] 5 [1.04-38.13]
PUD, Peptic ulcer disease.
demonstrated narrowing of the gastric outlet. All patients had symptoms suggestive of GOO (i.e., intractable vomiting). Patient demographic data, use of H2-receptor antagonists (H2RAs), nonsteroidal anti-inflammatory drugs (NSAIDs), and ulcer history were recorded. Histopathologic results of biopsies taken during upper endoscopy where available were recorded, and the presence of Helicobacter pylori or gastritis was noted. The biopsies were taken from the obstructive site with regular forceps on initial examination and jumbo forceps on repeat examinations using a therapeutic upper endoscope. Operative findings and results of surgical pathology were reviewed, the latter being the gold standard for the diagnosis of malignancy. The charts of the patients diagnosed with malignant disease at the time of surgery were further reviewed for clinical presentation, endoscopic findings, type of biopsy (jumbo or regular), and results of abdominal CT performed before surgery. Statistical analysis The statistical analysis was performed using SAS/STAT software.15 Any p values < 0.05 were taken to be indicative of statistical significance. Fisher’s exact test was used for comparisons involving binary variables and the t test for continuous variables. Multivariate logistic regression was used to examine potential predictors of malignant GOO (both gastric and extragastric). Results of this analysis were expressed in terms of odds ratios and their 95% confidence intervals. 498
GASTROINTESTINAL ENDOSCOPY
RESULTS The clinical characteristics of the 40 patients who presented with GOO are shown in Table 1. Fifteen were white, 22 Hispanic, 2 Native Americans, and 1 Asian; age ranged from 16 to 82 years. Thirty-eight patients underwent upper endoscopy and biopsy. H. pylori status was determined in 34 patients, of whom 19 were positive by either the CLO test (Tri Med Specialties, Inc., Lenexa, Kan.) or histology. Thirty-five patients underwent surgery. The remaining five patients treated medically were asymptomatic at 6 months and were judged to be free of cancer. Patients were further divided into non-cancer and cancer groups based on the surgical pathology report (Table 1). Of the 16 patients (40%) with malignant GOO, 7 had gastric adenocarcinoma, 4 had pancreatic carcinoma, 3 had gallbladder cancer, and 2 had cancer of undetermined origin. The patients with malignant disease were significantly older when compared with those with benign disease (p = 0.03; odds ratio: 95% CI: 5.21 [1.05-23.49]) (Table 2). Here, age is treated as a binary variable, whereas in Table 1 it is a continuous variable. Fewer patients in the cancer group had a history of peptic ulcer disease (PUD) when compared with those with nonmalignant obstruction (p = 0.04; odds ratio: 95% CI: 5 [1.0438.13]) (Table 2). When compared with the surgical specimen, endoscopic biopsy had poor sensitivity (37%) in detecting malignant disease. Six patients in the cancer group, all with gastric cancer, and 13 in the non-cancer group were H. pylori positive. Aspirin and NSAID use was not remarkable in either group. Seven patients had gastric cancer causing GOO (Table 3). In four, additional endoscopic biopsies obtained from the obstructing lesion with jumbo forVOLUME 48, NO. 5, 1998
Gastric outlet obstruction with benign endoscopic biopsy and malignancy
A Awan, D Johnston, M Jamal, et al.
Table 3. Gastric and extragastric cancers causing GOO (n = 16) No. of patients
CT of abdomen positive for mass
CT of abdomen negative for mass
CT of abdomen not done
4
1
1
2
3
0
3
0
2
2
0
0
7
6
1
0
Endoscopic biopsy positive for malignancy Endoscopic biopsy negative for malignancy Endoscopic biopsy positive for malignancy Endoscopic biopsy negative for malignancy GOO, Gastric outlet obstruction.
ceps were positive for malignancy. Three of these four patients had an obstructive mass seen on upper endoscopy and one had no visible mass, although a biopsy was taken from the site of obstruction. Three patients with gastric cancer had negative endoscopic biopsies which remained negative despite obtaining repeated biopsies with a jumbo forceps. None of the three patients had an ulcer or mass seen on upper endoscopy. CT of the abdomen also failed to reveal any malignant lesion in these patients (Table 3). Nine patients had extragastric cancer causing GOO (Table 3). In two, endoscopic biopsies of the obstructing lesion revealed the presence of a tumor. Seven patients had a negative biopsy. CT of the abdomen revealed a mass in eight patients. One extragastric cancer was not detected by CT. Multivariant analysis by logistic regression showed that age and negative history of PUD were associated with malignant GOO (Table 2). DISCUSSION In this study, the importance of careful evaluation of patients with GOO has been emphasized. Because the majority of patients underwent upper gastrointestinal endoscopy and subsequently surgery, there was an opportunity to objectively confirm or disprove the endoscopic biopsy findings. In this era of “less invasive” therapy,9,10 it is tempting to undertake endoscopic management of GOO. However, as malignant tumors are becoming increasingly important as a cause of GOO and because the incidence of complications of long-term PUD is decreasing, it is essential to carefully consider malignancy as a cause of GOO.11,12 In a predominantly Hispanic population, malignant tumors occurred in 40% of patients presenting with GOO. Although not statistically significant, a considerable number of young, Hispanic women presented with GOO and had repeated endoscopies with jumbo biopsies which failed to reveal the presence of tumor. Surgical exploration revealed malignancy. VOLUME 48, NO. 5, 1998
In view of major changes over the last 2 decades in the use of H2RAs,11,12 aspirin, NSAIDs,13,14 and eradication of H. pylori, as well as in the availability of new diagnostic methods, the approach to GOO may need to change. We suggest the following algorithm for investigating patients with GOO. When there is intractable vomiting of more than 2 weeks duration or an upper gastrointestinal barium study shows GOO, the patient should undergo upper endoscopy. If an obstructing lesion is found, biopsy specimens should be obtained. If the biopsy shows malignancy, a CT of the abdomen should be obtained to stage the tumor and the patient should be referred for surgery. If the initial biopsy of the obstructing lesion is negative for malignancy, endoscopy should be repeated and several jumbo biopsies obtained. If these are positive for malignancy, a CT should be obtained and the patient should be referred for surgery. If the second set of larger biopsies of the obstructing lesion is negative but the patient is at high risk for malignancy (age greater than 55 years, no history of PUD), a CT of the abdomen should be performed. Where available, endosonography with fine-needle aspiration of any suspicious lesion may be performed and the patient should then be referred for surgical exploration. When there is no clinical suspicion of malignancy and endoscopic biopsies are negative, endoscopic balloon dilation and H2RA therapy may be appropriate, provided the patient is monitored for at least 1 year. ACKNOWLEDGMENTS We thank Denis M. McCarthy, MD, MSc, FACP, Professor of Medicine and Chief, Division of Gastroenterology for valuable comments and suggestions. REFERENCES 1. Kreel L, Ellis H. Pylorus stenosis in adults: a clinical and radiological study of 100 consecutive patients. Gut 1965;6:253. GASTROINTESTINAL ENDOSCOPY
499
A Awan, D Johnston, M Jamal, et al.
Gastric outlet obstruction with benign endoscopic biopsy and malignancy
2. Grahm DY. Ulcer complications and their nonoperative treatment. In: Sleisinger MH, Fordtran JS, editors. Gastrointestinal disease: pathophysiology, diagnosis, management. 5th ed. Philadelphia: WB Saunders; 1993. p. 707. 3. Jaffin BW, Kaye MD. The prognosis of gastric outlet obstruction. Ann Surg 1985;2:176-9. 4. Goldstein H, Janin M, Schapiro M, Boyle JD, Gastric retention associated with gastroduodenal disease. Am J Dig Dis 1966;11:887-97. 5. Perng CL, Lin HJ, Lo WC, Lai CR, Guo WS, Lee SD. Characteristics of patients with benign gastric outlet obstruction requiring surgery after endoscopic balloon dilation. Am J Gastroenterol 1996;5:987-90. 6. Kozarek RA, Botoman VA, Patterson DJ. Long-term follow-up in patients who have undergone balloon dilation for gastric outlet obstruction. Gastrointest Endosc 1990;17:2-4. 7. Kozarek RA, Dilation therapy for gastric outlet obstruction. Are balloons a bust [editorial]? J Clin Gastro 1993;17:2-4. 8. Sommer A, Bethge N. Relief of malignant external gastric obstruction by endoscopic implantation of a self-expanding metal stent. Endoscopy 1995;27:210-1. 9. Feritis C, Benakis P, Dimopoulos C, Georgopoulos K, Milas F,
10.
11.
12. 13.
14.
15.
Manouras A, et al. Palliation of malignant gastric outlet obstruction with self-expanding metal stents. Endoscopy 1996;28:225-8. Kuwada SK, Alexander GL. Long-term outcome of endoscopic dilation of non-malignant pyloric stenosis. Gastrointest Endosc 1995;41:15-7. Shone DN, Nikoomanesh P, Snuth-Meek MM, Bender JS. Malignancy is the most common cause of gastric outlet obstruction in the era of H2 blockers. Am J Gastroenterol 1995;90:1769-70. Johnson CD, Ellis H. Gastric outlet obstruction now predicts malignancy. Br J Surg 1990;77:1023-4. Weaver GA, Merrell NB, Harper RL, Storey JA, Jenkins PL. Nonsteroidal anti-inflammatory drugs are associated with gastric outlet obstruction. J Clin Gastroenterol 1995;20:196-8. Griffin MR, Piper JM, Daugherty JR, Snowden M, Ray W. Nonsteroidal anti-inflammatory drugs and increased risk for peptic ulcer disease in elderly persons. Ann Intern Med 1991;114:257-63. SAS/STAT Software: changes and enhancements through release 6.12. Cary (NC): SAS Institute; 1997.
Moving? To ensure continued service please notify us of a change of address at least 6 weeks before your move. Phone Subscription Services at 800-453-4351 (outside the U.S. call 314-453-4351) or fax your information to 314-432-1158.
500
GASTROINTESTINAL ENDOSCOPY
VOLUME 48, NO. 5, 1998