Colonic Diverticular Bleeding Research Articles

THERAPEUTIC ENDOSCOPY FOR NONVARICEAL GASTROINTESTINAL BLEEDING

This article reviews the role of therapeutic endoscopy in the diagnosis and treatment of nonvariceal upper and lower gastrointestinal (GI) hemorrhage. The initial approach to patients with GI bleeding is reviewed. Endoscopic treatment of various stigmata of recent peptic ulcer hemorrhage is discussed in detail. Management of less common causes of nonvariceal bleeding, such as Dieulafoy's lesions, Mallory-Weiss tears, angiomas, and bleeding colonic diverticula is described. Recommendations for endoscopic techniques are based on the results of UCLA-CURE hemostasis studies.

4473 Factors predictive of persistent or recurrent colonic diverticular bleeding.

Background:The majority of colonic diverticular bleeding ceases spontaneously. However, approximately 20% of patients with diverticular bleeding require endoscopic, angiographic, or surgical intervention to stop the bleeding.We aim to identify the admission clinical features that predict the need for intervention to stop bleeding. Methods:Using our previously described diverticular bleeding database, we identified clinical features felt to be associated with the risk of persistent or recurrent diverticular bleeding, and hence the need for therapeutic intervention. These included gender, age, hypotension, admission hematocrit, lowest hematocrit, prothrombin time, creatinine, ASA, NSAIDs or Coumadin usage, comorbidities (incl CHF,HTN, COPD), and history of previous lower gastrointestinal bleeding. The outcome was defined as the need to intervene to stop colonic diverticular bleeding. Statistical analyses were performed using SAS. Individual predictors were analyzed using Chi-Square and t-Test analysis. Multiple Discriminant Analysis was performed to perdict the outcome based on five factors: gender, chronic renal failure, congestive heart failure, ASA usage and lowest recorded hematocrit. Finally, an exploratory Stepwise Discriminant Analysis was then performed using all possible individual predictive factors. Results: The only individual predictor which was significantly (p<0.05) related to outcome was gender. Intervention was required more often in women than in men. The hypothesized model was not significant. The Stepwise analysis identified five possible perdictors: in order of entry, gender, chronic renal failure, congestive heart failure, ASA usage and lowest recorded hematocrit were associated with requiring an intervention to stop bleeding. Other factors did not appear to be predictive of the risk of persistent or recurrent bleeding. Conclusions: Identification of factors predicting recurrent or persistent bleeding would be useful clinically, and help to triage patients on admission. Other than gender, there are no single strong predictors of the risk of persistent or recurrent diverticular bleeding, based on hospital admission data. Interestingly, anticoagulation status and history of previous lower gastrointestinal bleed did not appear to be predictive of persistent or recurrent diverticular bleeding. We intend to apply the Stepwise model to a future 60 patients with diverticular bleeding to validate or invalidate their predictive qualities. Background:The majority of colonic diverticular bleeding ceases spontaneously. However, approximately 20% of patients with diverticular bleeding require endoscopic, angiographic, or surgical intervention to stop the bleeding.We aim to identify the admission clinical features that predict the need for intervention to stop bleeding. Methods:Using our previously described diverticular bleeding database, we identified clinical features felt to be associated with the risk of persistent or recurrent diverticular bleeding, and hence the need for therapeutic intervention. These included gender, age, hypotension, admission hematocrit, lowest hematocrit, prothrombin time, creatinine, ASA, NSAIDs or Coumadin usage, comorbidities (incl CHF,HTN, COPD), and history of previous lower gastrointestinal bleeding. The outcome was defined as the need to intervene to stop colonic diverticular bleeding. Statistical analyses were performed using SAS. Individual predictors were analyzed using Chi-Square and t-Test analysis. Multiple Discriminant Analysis was performed to perdict the outcome based on five factors: gender, chronic renal failure, congestive heart failure, ASA usage and lowest recorded hematocrit. Finally, an exploratory Stepwise Discriminant Analysis was then performed using all possible individual predictive factors. Results: The only individual predictor which was significantly (p<0.05) related to outcome was gender. Intervention was required more often in women than in men. The hypothesized model was not significant. The Stepwise analysis identified five possible perdictors: in order of entry, gender, chronic renal failure, congestive heart failure, ASA usage and lowest recorded hematocrit were associated with requiring an intervention to stop bleeding. Other factors did not appear to be predictive of the risk of persistent or recurrent bleeding. Conclusions: Identification of factors predicting recurrent or persistent bleeding would be useful clinically, and help to triage patients on admission. Other than gender, there are no single strong predictors of the risk of persistent or recurrent diverticular bleeding, based on hospital admission data. Interestingly, anticoagulation status and history of previous lower gastrointestinal bleed did not appear to be predictive of persistent or recurrent diverticular bleeding. We intend to apply the Stepwise model to a future 60 patients with diverticular bleeding to validate or invalidate their predictive qualities.

The Role of Endoscopy in Managing Acute Lower Gastrointestinal Bleeding

In this issue of the Journal, Jensen et al. report their experience with the use of colonoscopy performed on an urgent basis for the diagnosis and treatment of acute hemorrhage from colonic diverticulosis.1 The study has two key messages. The first is that the endoscopic findings that characterize recent hemorrhage, routinely used in the evaluation of acute upper gastrointestinal bleeding, are also useful in establishing the diagnosis in a subgroup of patients with colonic diverticular hemorrhage. This group of findings, referred to as stigmata of recent hemorrhage, includes active bleeding (e.g., arterial spurting), nonbleeding visible vessels, appearing as discrete pigmented . . .

Bleeding colonic diverticula. A reappraisal of natural history and management.

The study was undertaken to correct or reaffirm current recommendations based on old observations of doubtful validity because of their lack of routine colonoscopy, scintigraphy, or angiography. Patterns of bleeding were derived from transfusion records of 78 patients admitted 106 times for lower gastrointestinal bleeding with no detectable cause other than colon diverticula. Bleeding stopped spontaneously in 82 of 108 episodes and in 66 of 67 patients requiring less than four units of transfusion on any day. When four or more units were required in a day, 25 of 42 patients required emergency surgery. When a bleeding site was identified and removed, only 1 of 25 patients bled again from another diverticulum. After discharge without surgery, 28 of 73 began to bled again. After "blind" colectomy and ileoproctostomy, four of seven patients developed leaks or abscesses, and two died. Bleeding stopped spontaneously in 75% of episodes and in 99% of patients requiring less than four units of transfusion per day. Bleeding continued in 25% of episodes and in most patients who required four or more units per day. Bleeding sites of those patients who continued to bleed were shown by scintigraphy or angiography. When a bleeding diverticulum is removed, rebleeding is rare. "Blind" resection is unsafe.

"Lower" gastrointestinal tract hemorrhage: new concepts based on arteriography.

1. Arteriography is the diagnostic method of choice in acute rectal hemorrhage and results in precise preoperative diagnosis and rational surgery. The potential for selective therapy through the catheter is also evident.2. Colonic diverticulosis is, by far, the most common cause of acute rectal bleeding in the middle-aged or elderly adult.3. Most bleeding colonic diverticula are located to the right of the splenic flexure.4. Carcinoma is a rare cause of rectal hemorrhage.5. The bleeding point is occasionally located outside of the colon.6. A history of trauma or previous surgery should suggest the possibility of arteriovenous fistula.7. Arteriovenous malformations of the right colon should be diligently searched for in elderly patients with aortic stenosis and recurrent hemorrhage.8. Hemorrhage from duodenal ulcer may present as rectal hemorrhage and lead to diagnostic confusion.

Arteriographic Diagnosis and Management of Suspected Colonic Diverticular Hemorrhage

Abstract Selective arteriography was performed in 12 patients with rectal hemorrhage who were suspected of having diverticulosis as the cause of bleeding. Four patients had negative arteriography and have had no recurrence of hemorrhage. In 4 patients, the hemorrhage was due to a lesion other than diverticulosis. These included 1 patient each with malignant lymphoma, intestinal ischemia, cecal ulcer, and arteriovenous malformation of the small bowel. Four patients were bleeding from diverticular disease of the colon, including 3 patients with diverticulosis and 1 patient with diverticulitis. Three of these patients were treated with intraarterial infusion of vasopressin. In each case bleeding stopped for at least 24 hr resulting in significant improvement in the preoperative condition of these patients. Selective arteriography in patients in whom hemorrhage is thought to be due to colonic diverticular disease is indicated to (1) exclude other causes of rectal hemorrhage, (2) accurately localize the site of the bleeding diverticulum, (3) allow a trial of nonoperative control of hemorrhage by infusion of vasopressin or other vasoconstrictor agents, and (4) allow more specific and limited colonic resection when bleeding is uncontrolled or recurrent.

Arterial ligation for control of colonic diverticular hemorrhage

Arterial ligation for control of colonic diverticular hemorrhage

Arterial Ligation for Control of Colonic Diverticular Hemorrhage

Massive diverticular hemorrhage in an elderly poor-risk woman was controlled by simple ligation of the feeding artery which had been clearly demonstrated by emergency mesenteric arteriography. Since the basic lesion is erosion of a single submucosal artery, mesenteric arteriography is advocated as a primary diagnostic tool in this clinical setting because it may provide the necessary localization which will permit simple ligation. Thus major or blind resection and multiple operations with inherent high risk may be avoided.