Validation of an endoscopic anastomotic grading score as an intraoperative method for assessing stapled rectal anastomoses

PURPOSE

Anastomotic leak is a dreaded complication of colorectal surgery. An endoscopic grading score of the perianastomotic mucosa has been previously developed at our institution (UCI) to assess colorectal anastomotic integrity. The objective of this study is to validate the UCI anastomotic score and determine its impact in anastomotic failure.

METHODS

As a follow-up study of the UCI grading score implementation during 2011 to 2014, patients undergoing stapled colorectal anastomoses after sigmoidectomy or proctectomy at a single institution from 2015 to 2018 were retrospectively reviewed. Patients were grouped into three tiers based on endoscopic appearance (grade 1, circumferentially normal mucosa; grade 2, ischemia/congestion < 30% of circumference; grade 3, ischemia/congestion > 30% of circumference).

RESULTS

On the basis of endoscopic mucosal evaluation, grade 1 anastomosis was observed in 299 patients (94%), grade 2 anastomosis in 14 patients (4.4%), and grade 3 anastomosis in 5 patients (1.6%). All grade 3 classifications were immediately and successfully revised intraoperatively with reclassification as a grade 1 anastomosis. The anastomotic leak rate of the follow-up study period from 2015 to 2018 was 6.4% which was lower compared to the anastomotic leak rate of 12.2% in the original study period from 2011 to 2014 (p = 0.07). Anastomotic leak rate for the entire patient series was 8.5%. A grade 2 anastomosis was associated with higher anastomotic leak rate compared to a grade 1 anastomosis (35.7% vs. 7.4%, p < 0.05). None of the five grade 3 anastomoses resulted in an anastomotic leak upon revision.

CONCLUSION

This study further validates the anastomotic grading score and suggests that its systematic implementation can result in a reduction in anastomotic leaks.

The utility of intraoperative endoscopy in esophagogastric surgery

Flexible endoscopy (FE) plays a major role in the diagnosis and treatment of gastrointestinal disease. Although its intraoperative use has spread over the years, its use by surgeons is still limited in our setting. FE training opportunities are different among many institutions, specialties, and countries. Intraoperative endoscopy (IOE) presents peculiarities that increase its complexity compared to standard FE. IOE has a positive impact on surgical results, due to increased safety and quality, as well as a reduction in the complications. Due to its innumerable advantages, its intraoperative use by surgeons is currently a current project in many countries and is part of the near future in others because of the creation of better structured training projects. This manuscript reviews and updates the indications and uses of intraoperative upper gastrointestinal endoscopy in esophagogastric surgery.

Effectiveness of Intraoperative Endoscopy for Localization of Early Gastric Cancer during Laparoscopic Distal Gastrectomy

INTRODUCTION

Intraoperative localization of tumors has been considered crucial in determining adequate resection margins during laparoscopic gastrectomy for early gastric cancer (EGC). This study has evaluated the effectiveness of intraoperative endoscopy for localization of EGC during the totally laparoscopic distal gastrectomy.

METHODS

Patients with EGC who received totally laparoscopic distal gastrectomy from January 2018 to March 2020 were included in this study. Except the tumors located in the antrum, the patients were categorized into two groups: no localization procedure (n = 144) and intraoperative endoscopy (n = 65). To evaluate the effectiveness of the localization procedure, proximal resection margin (PRM) involvement by the tumor and approximation of optimal PRM were compared, including their postoperative outcomes.

RESULTS

There were 3 patients (2.1%) with tumor involvement of the PRM at the initial gastric resection in the no localization group. Distance from the tumor to the PRM was determined to be not significantly different between the no localization group and intraoperative endoscopy group. The PRM distribution pattern and reconstruction method were also not significantly different between the two groups.

DISCUSSION/CONCLUSION

Intraoperative endoscopy for localization of EGC is an effective method to avoid tumor involvement at the resection margin during the laparoscopic gastrectomy with intracorporeal gastric resection and reconstruction.

Recurrent massive bleeding from a small intestinal arteriovenous malformation after surgery for biliary atresia in an infant: a case report and literature review

BACKGROUND

Small intestinal arteriovenous malformation (AVM) can cause bleeding. Most small intestinal AVMs occur during adulthood, rarely in infancy. We report a case of an infant with hemorrhage due to small intestinal AVM early and recurrently after Kasai portoenterostomy (PE) for biliary atresia (BA).

CASE PRESENTATION

A 51-day-old male infant was admitted to our institution for obstructive jaundice. Laparotomic cholangiography revealed BA (IIIb1μ), and Kasai PE was performed at 60 days of age. On postoperative day 17, he developed massive melena and severe anemia. Contrast-enhanced computed tomography (CT) revealed that the jejunum around the PE site was strongly enhanced with enhancing nodules in the arterial phase, and a wide area of the Roux limb wall was slightly enhanced in the venous phase. As melena continued, emergency laparotomy was performed. There were no abnormal macroscopic findings at the PE site except for a clot in the Roux limb 5 cm away from the PE site, and the Roux limb was resected 5 cm. On further investigation, a red spot was detected on the jejunal serosa 30 cm away from the Roux-en-Y anastomosis site. PE and wedge resection for the red spot were performed. Histopathologically, both specimens indicated AVM. He was jaundice-free 65 days after the first surgery. However, at 7 months of age, he developed massive melena again. Contrast-enhanced CT and upper gastrointestinal endoscopy revealed no bleeding lesions. Hemorrhagic scintigraphy showed a slight accumulation at the hepatic hilum prompting an emergency surgery. Intraoperative endoscopy detected a bleeding lesion at the PE site, and the Roux limb was resected (approximately 6 cm). Intraoperative frozen section analysis of the stump of the resected jejunum revealed no abnormal vessels. PE was performed, and permanent section analysis revealed an AVM in the resected jejunum. The postoperative course was uneventful without re-bleeding.

CONCLUSIONS

We experienced a case of recurrent massive bleeding from small intestinal AVM in an infant after surgery for BA. Intraoperative endoscopy and frozen section analysis helped identify the bleeding lesion and perform a complete resection of the small intestinal AVM, even after surgery, in the infant.

Laparoscopic hand-sewn cardioplasty: an alternative procedure for end-stage achalasia

BACKGROUND

Primary achalasia is a rare oesophageal motor disorder characterized by the absence of swallow-induced relaxation of the lower oesophageal sphincter and diminished or absent oesophageal body peristalsis. Around 5% of these patients will develop end-stage achalasia, where oesophagectomy may be advocated. We present the laparoscopic hand-sewn cardioplasty as an alternative 'oesophagus-preserving' procedure in patients with end-stage achalasia.

METHODS

We present a retrospective review of four patients who underwent laparoscopic hand-sewn cardioplasty. Data collected included pre-operative demographic information and investigations; and post-operative outcomes. Patients were scored pre- and post-operatively using Reflux Symptom Index, Eating Assessment Tool-10 and Voice Handicap Index-10 questionnaires.

RESULTS

Four patients underwent laparoscopic hand-sewn cardioplasty during the study period. In one patient, it was performed as a rescue procedure during attempted myotomy following multiple perforations of friable mucosa. In the other three patients, laparoscopic hand-sewn cardioplasty was performed for end-stage achalasia. None of the patients had post-operative complications and all patients were discharged on the second post-operative day. All patients experienced improvement in swallowing symptoms (EAT-10; p = 0.03) but developed post-operative gastroesophageal reflux.

CONCLUSION

To our knowledge, this is the first published case series of laparoscopic hand-sewn cardioplasty for end-stage achalasia. It appears to be a safe and effective procedure for the treatment of end-stage achalasia, offering an alternative minimally invasive procedure to oesophagectomy. Laparoscopic hand-sewn cardioplasty can also be used as a 'rescue' procedure during myotomy in patients who have poor-quality mucosa which perforates intra-operatively or is at high risk of perforation/leaking post-operatively.

A case of submucosal abscess of the esophagus mimicking a mediastinal abscess

We report a case of a submucosal abscess of the esophagus that required differentiation from a mediastinal abscess. A 48-year-old man presented with a chief complaint of fever and sore throat. He did not remember swallowing a foreign body, and his oral cavity showed no signs of inflammation. Contrast-enhanced computed tomography showed a low density area with enhancement in the mediastinum, especially around the esophageal wall. We planned to perform surgical drainage with the intention of performing intraoperative endoscopy from the beginning. We performed surgical drainage through a left cervical oblique incision; however, there was no exudate obtained from the mediastinal space. Despite the reported absence of accidental ingestion of a foreign substance, intraoperative endoscopy was performed that revealed a laceration in the esophageal mucosa 24 cm from the incisors. We diagnosed it as a submucosal abscess of the esophagus and prescribed parenteral antibiotics. Submucosal abscess of the esophagus can occur even in the absence of awareness of a foreign body ingestion or oral infection. In case that the abscess was not localized clearly within the outer membrane of the esophagus, the coincidental mediastinal drainage via a cervical incision and intraoperative endoscopy seemed to be useful.

Clinical and economic burden of colorectal and bariatric anastomotic leaks

BACKGROUND

Anastomotic leaks cause a significant clinical and economic burden on patients undergoing bariatric and colorectal surgeries. Current literature shows a wide variation in incidence of anastomotic leaks and a significant gap in associated economic metrics. This analysis utilized claims data to quantify the full episode-of-care cost burden of leaks following colorectal and bariatric surgeries.

METHODS

Medicare Fee-for-Service and commercial claims data from a large U.S.-based health plan were queried for cost and utilization of members that underwent bariatric and colorectal surgical procedures between January 1, 2013 and August 31, 2015. Outcomes were collected for members with anastomotic leaks versus those without leaks during the initial hospital stay (index) and within 30 days of the procedure. These outcomes included leak frequency, payer reimbursement, and length of stay (LOS).

RESULTS

The colorectal Medicare analysis identified 239,350 patients undergoing colorectal surgery. For patients with a leak compared to those without, index admission costs were $30,670 greater ($48,982 vs. $18,312; p < 0.0001) and the index LOS was 12 days longer (19 vs. 7 days; p < 0.0001). This finding was similar for the bariatric patients (n = 62,292) where cost was $30,885 higher ($43,918 vs. $13,033; p < 0.0001) and LOS was 15 days longer (17 vs. 2 days; p < 0.0001). Furthermore, readmissions and associated costs were also substantially higher for those with an index leak. The commercial analysis of both the bariatric and colorectal populations trended similarly to the Medicare population in regards to all outcomes measured.

CONCLUSION

Patients experiencing anastomotic leaks during and after bariatric and colorectal surgery have significantly higher costs and longer LOS both at the initial stay and within 30 days of the procedure. It is important that providers and hospitals understand the economic consequences of these procedures and implement technologies and techniques to prevent/reduce anastomotic leaks.

Endoscopy Used as Provocative Testing in Bariatric Surgery: An Analysis of the Texas Public Use Data File

Background

Many bariatric surgeons test the anastomosis and staple lines with some sort of provocative test. This can take the form of an air leak test with a nasogastric tube with methylene blue dye or with an endoscopy. The State Department of Health Statistics in Texas tracks outcomes using the Texas Public Use Data File (PUDF).

Methods

We queried the Texas Inpatient and Outpatient PUDFs for 2013 to 2017 to examine the number of bariatric surgeries with endoscopy performed at the same time. We used the International Classification of Diseases Clinical Modification Version 9 (ICD-9-CM) and ICD-10 procedure codes and Current Procedural Terminology for Sleeve Gastrectomy (SG) and laparoscopic Roux-en-Y gastric bypass (LRYGB) and endoscopy, and the ICD-9-CM and ICD-10 diagnosis codes for morbid obesity.

Results

There were 74,075 SG reported in the Texas Inpatient and Outpatient PUDF for the years 2013-2017. Of the SG performed, 5,521 (7.4%) had an intraoperative endoscopy. For the 19,192 LRYGB reported, 1640 (8.6%) underwent LRYGB + endoscopy. This was broken down by SG only vs SG + endoscopy and LRYGB only vs LRYGB + endoscopy. Overall, SG + endoscopy had a significantly shorter length of stay (LOS) vs LRYGB + endoscopy at 1.74 d vs 2.34 d (P < .001) and a significantly less cost of $71,685 vs $91,093 (P < .001).

Conclusions

A small percentage of SG and LRYGB patients underwent endoscopy for provocative testing over the study period. Provocative testing with endoscopy costs more for SG and LRYGB and was associated with a shorter LOS.

Near-infrared fluorescent clip guided gastrectomy: Report of 2 cases (Case reports)

INTRODUCTION

This is the first report on near-infrared fluorescent (NIRF) clip-guided gastrectomy. The NIRF clip, ZEOCLIP FS, emits NIRF signals when excited. We hypothesized that preoperative placement of the ZEOCLIP FS near a gastric lesion would allow fluorescence laparoscopic localization of the clip, and hence, the lesion, during surgery. We report this technique in two cases.

CASE PRESENTATION

Case 1: An 81-year-old female was diagnosed with early gastric cancer and a pedunculated 4 cm large hyperplastic polyp that had prolapsed into the duodenum, and was scheduled for laparoscopy-assisted distal gastrectomy, due to the potential risk of dissection of the polyp with the duodenal wall. On the day before surgery, ZEOCLIP FS clips were endoscopically placed at the cancer site and the polyp. The locations of the fluorescent clips were confirmed intraoperatively using a full-color fluorescence laparoscope.

CASE 2

An 81-year-old male was scheduled for laparoscopy-assisted total gastrectomy for gastric cancer under fluorescent clip-guidance. Clip locations could not be confirmed during initial intraoperative observation. However, when the stomach wall was raised using forceps during a second observation attempt, the fluorescent clip locations were confirmed.

DISCUSSION

In case 1, it was easy to confirm clip location, facilitating complete resection of early gastric cancer without dissecting the polyp. In case 2, the fluorescent clip was located by raising the stomach and adjusting the angle between the stomach wall and the fluorescence laparoscope.

CONCLUSION

The positive results of these two cases warrant conducting feasibility studies for use of this method.

Effectiveness and safety of tumor site marking with near-infrared fluorescent clips in colorectal laparoscopic surgery: A case series study

BACKGROUND

In colorectal laparoscopic surgery, accuracy of tumor marking has been an important but not fully resolved issue. The tattoo marking technique or intraoperative endoscopy have been used but they either carry the risk of accidental intestinal puncture or require either longer operation times, a skilled endoscopist and/or intraoperative colon insufflation. We supposed that tumor site marking with the near-infrared fluorescent clips, ZEOCLIP FS clips (Zeon Medical Co., Ltd., Tokyo, Japan) might overcome disadvantages of both tattoo marking and intraoperative endoscopy-based tumor localization methods. This is the first report on the case series using near-infrared fluorescent marking clip. We summarize the early results in 30 patients, who underwent colorectal laparoscopic surgery; we focus particularly on effectiveness and safety of the method.

MATERIALS AND METHODS

Thirty consecutive patients, who underwent laparoscopic surgery for colorectal cancer after previous endoscopic ZEOCLIP FS placement were enrolled from May 2019 till October 2019. The primary endpoint was the rate of intraoperative clip detection and the secondary endpoints were: the rate of adverse effects, percentage of slipped clips and usefulness of plain abdominal radiography to preoperatively confirm the clip retention. Locations of fluorescent clips were identified with a full-color fluorescence laparoscope. All operations and clip placements were performed by the same senior surgeon with sufficient experience in both procedures.

RESULTS

Fluorescent clips could be detected in 94.1% of tumor lesions. Three (2.1%) clips dropped before surgery. Plain abdominal radiography was sufficient to assess clip retention in all cases. No adverse effects related to either clip placement or clip detection were observed.

CONCLUSION

The ZEOCLIP FS could be easily detected from the serosal side of the intestinal tract when placed 1-2 days before surgery. Fluorescent clip-guided laparoscopy may be considered a safe and effective method for localization of colorectal tumor sites. The Research Registry UIN: researchregistry5400.

Safety and efficacy of post-anastomotic intraoperative endoscopy to avoid early anastomotic complications during gastrectomy for gastric cancer. Surg Endosc. 2019;. [PMID: 31834512 DOI: 10.1007/s00464-019-07319-3]

BACKGROUND

Anastomotic complications such as leaks, bleeding, and stricture remain the most serious complications of surgery for gastric cancer. No perfect method exists for an accurate and reliable prevention of these complications. This study investigated the safety and efficacy of post-anastomotic intraoperative endoscopy (PAIOE) for avoidance of early anastomotic complications during gastrectomy in gastric cancer.

METHODS

This retrospective case-control study enrolled patients from a tertiary care, academic medical center. Routine PAIOE was performed on 319 patients undergoing gastrectomy for gastric cancer between 2015 and 2016. As controls, without PAIOE 270 patients from 2013 to 2014 were used for comparison. Early anastomotic complications and outcomes after PAIOE were determined.

RESULTS

Although there were no differences between the PAIOE and non-PAIOE group in terms of overall complication rates (20.1% vs 26.7%; P > 0.05), there were fewer complications related to anastomosis (3.4% vs 8.9%; P < 0.01) in the PAIOE group. The PAIOE group had rates of 2.5% for anastomotic leakage, 0.9% for intra-luminal bleeding, and 0% for anastomotic stenosis, while the non-PAIOE group exhibited rates of 5.6%, 2.6%, and 0.7%, respectively. Thirty-one abnormalities were detected in 26 PAIOE patients (9.71%) (20 venous bleeding, 7 mucosal tearing, 2 air leaks, 1 arterial bleeding, and 1 anastomotic stricture). All abnormalities were corrected by proper interventions (13 reinforced additional suture, 13 endoscopic hemostasis, and 2 re-anastomosis). There were no morbidities associated with PAIOE.

CONCLUSIONS

PAIOE appears to be a safe and reliable procedure to evaluate the stability of gastrointestinal anastomosis for gastric cancer patients. Further data collection and a well-designed prospective study are needed to confirm the validity of PAIOE.

Does intraoperative endoscopy decrease complications after bariatric surgery? Analysis of American College of Surgeons National Surgical Quality Improvement Program database

BACKGROUND

Intraoperative endoscopy (IOE) has been proposed to decrease serious complications following bariatric surgeries such as leaks, bleeding, and stenosis. Such complications can lead to sepsis and eventually can be fatal. We aim to compare major postoperative complications in patients with and without IOE.

METHODS

Data from the American College of Surgeons National Surgical Quality Improvement Program database years 2011 till 2016 were used to identify laparoscopic sleeve gastrectomy (LSG) and laparoscopic Roux-en-Y gastric bypass (LRYGB) patients. We compared outcomes of IOE and non-IOE using bivariate and multivariate analysis. Thirty-day outcomes included sepsis, organ space infection, unplanned reoperations, unplanned readmissions, prolonged hospital stay, bleeding, and mortality.

RESULTS

Out of 62,805 cases of LSG and 50,047 cases of LRYGB, 17.9%, and 19.7% had IOE, respectively. Endoscopy-assisted LSG was associated with a decrease in sepsis [0.37% vs. 0.21%, adjusted odds ratio (AOR) = 0.55 (0.36, 0.84)], unplanned reoperation [0.58% vs. 0.38%, AOR = 0.61 (0.44, 0.85)], prolonged hospital stay [14.9% vs. 14.0%, AOR = 0.87 (0.82, 0.92)], and composite complications [1.43% vs. 1.17%, AOR = 0.78 (0.65, 0.94)]. Outcomes after LRYGB were similar in both groups, except for decreased prolonged hospital stay with IOE [22.4% vs. 20.6%, AOR = 0.89 (0.84, 0.94)].

CONCLUSIONS

IOE is generally underutilized in baraitric procedures. IOE is associated with decreased risk of postoperative complications particularly sepsis, unplanned reoperations, prolonged hospital stay, and composite complications after LSG; and hospital stay after LRYGB. Large multicenter prospective studies are needed to explore the benefits of IOE in bariatric surgery, particularly the intermediate or long-term benefits.

Intraoperative endoscopy prevents technical defect related leaks in laparoscopic Roux-en-Y gastric bypass: A randomized control trial

BACKGROUND

Postoperative anastomotic leaks, bleeding and stenosis are major causes of morbidity after laparoscopic Roux-en-Y gastric bypass (LRYGB). Retrospective studies suggest that intraoperative endoscopy reduces the incidence of these complications.

METHODS

We conducted a prospective randomized controlled trial in a single institution between March 2013 and January 2016. Patients were assigned to one of two groups: LRYGB with Intraoperative Endoscopy (IOE) or LRYGB without IOE. Patient selection criteria were morbidly obese patients, 18 years or older who were candidates to LRYGB. The primary outcome was the frequency of technical defect related anastomotic leaks. Secondary outcomes were operative time, length of hospital stay, anastomotic related complications, reoperations and 30-day mortality.

RESULTS

50 patients were randomly assigned in the IOE group and 50 in the control group. The IOE group had statistically significant lower rate of anastomotic leak (0 vs. 8%, p = .0412), and lower need for reoperation (0 vs. 8%, p = .0412). The IOE group had longer operative time (194.10 vs. 159 min, p < .001), and shorter mean length of hospital stay (2.44 vs. 3.46 days, p = .025). No differences were found in the rate of bleeding of the anastomosis, narrow anastomosis and 30-day mortality.

CONCLUSION

This study specifically provides evidence that air leak test performed by intraoperative endoscopy is superior to simple visual inspection in preventing technical defect related leaks after laparoscopic Roux-en-Y gastric bypass.

Hemosuccus Pancreaticus following a Puestow Procedure in a Patient with Chronic Pancreatitis

Hemosuccus pancreaticus is an unusual cause of gastrointestinal bleeding that occurs as a complication of chronic or acute pancreatitis. We report a case of extremely acute-onset hemosuccus pancreaticus occurring in a patient with chronic pancreatitis over a long-term follow-up after a Puestow procedure (side-to-side pancreaticojejunostomy). The patient was admitted to our hospital due to severe anemia and tarry stools indicative of gastrointestinal bleeding. Emergent endoscopy, including gastrointestinal fiberscopy and colon fiberscopy, showed no abnormal findings. Abdominal contrast-enhanced computed tomography and hemorrhagic scintigraphy did not detect a hemorrhagic lesion. Although interventional radiology was considered for diagnosis and treatment, conservative therapy seemed sufficient to affect hemostasis. Two weeks later, however, acute intestinal bleeding with hemodynamic shock occurred, and exploration was performed without delay. Intraoperative endoscopy through an incision of the reconstructed jejunal loop in the close proximal end revealed a site of active bleeding from the side-to-side anastomotic pancreatic duct. Following a longitudinal incision of the jejunal loop, a bleeding point was sutured and ligated on direct inspection. The patient showed a good postoperative course.

Gastrointestinal bleeding 30 years after a complicated cholecystectomy

Gastrointestinal bleeding from small-bowel varices is a rare and difficult to treat complication of portal hypertension. We describe the case of a 79-year-old female patient with recurrent severe hemorrhage from small-bowel varices 30 years after a complicated cholecystectomy. When double balloon enteroscopy was unsuccessful to reach the site of bleeding, a rendezvous approach was favored with intraoperative endoscopy. Active bleeding from varices within a biliodigestive anastomosis was found and controlled by endoscopic injection of cyanoacrylate. Intraoperative endoscopy should be considered in the case of life-threatening gastrointestinal hemorrhage that is not accessible by conventional endoscopy.

Diagnostic value of intraoperative endoscopic examination in patients with intestinal lymphangiectasia: an analysis of 6 cases

AIM: To assess the diagnostic value of intraoperative endoscopic examination in patients with intestinal lymphangiectasia (IL).

METHODS: Six patients with IL and protein-losing enteropathy underwent intraoperative endoscopic examination during the study. Intraoperative endoscopic examination was performed via an intestinal incision to locate the site of lymph fluid leakage.

RESULTS: Endoscopic findings included white fluid in the intestinal lumen, multiple scattered white spots, and white and swollen villi in all the 6 patients, polyps in 3 patients, and mucosal erosion in 2 patients. After subserosal injection of methylene blue, the sites of lymph fluid leakage were successfully located in 5 patients (83.3%).

CONCLUSION: Endoscopic examination is a very important diagnostic test for patients with IL, especially for locating the site of lymph fluid leakage.

Diagnosis and treatment of arteriovenous malformation of gastrointestinal tract: an analysis of 8 cases

AIM: To summarize the clinical diagnosis and treatment of gastrointestinal hemorrhage induced by arteriovenous malformation of the gastrointestinal tract.

METHODS: From December 1999 to January 2005, 8 patients were diagnosed with arteriovenous malformation of the gastrointestinal tract in our hospital, and their clinical data were retrospectively analyzed.

RESULTS: Of the 8 patients, enteroscopy was totally performed 12 times on 5 patients due to the hemorrhage of unknown reasons, but no lesions were found. Arteriovenous malformations were confirmed in 3 cases by angiography and two by angiography integrated with intraoperative endoscopy. Blind surgeries were totally performed 5 times on 3 patients, and therapeutic embolization was successfully followed on 2 of them. One case died of severe hemorrhage. The 7 patients were in the follow up for 6-66 mo, and 1 died of cerebral apoplexy.

CONCLUSION: Angiography and intraoperative endoscopy are more effective in the diagnosis of arteriovenous malformation of the gastrointestinal tract.