Angiodysplasia Research Articles

Gastrointestinal bleeding in von Willebrand disease

Gastrointestinal bleeding due to colonic angiodysplasia is a well-recognised complication of von Willebrand disease (vWD), occurring almost exclusively in subtypes of the disease associated with a reduction in high-molecular-weight (HMW) multimers of von Willebrand factor (vWF). A deficiency of vWF HMW multimers also provides the link between aortic stenosis and gastrointestinal (GI) bleeding in Heyde's syndrome. The diagnosis and treatment of the angiodysplastic bleeding in vWD can be very difficult and the role of vWF containing concentrates in treatment and prophylaxis remains to be established.

Crohn's disease with life-threatening hemorrhage from terminal ileum: successful control by superselective arterial embolization

A case of life-threatening lower gastrointestinal hemorrhage from Crohn's disease is reported. Several promising studies have recently been published that describe superselective embolization for the treatment of massive lower gastrointestinal hemorrhage in patients with bleeding colonic diverticular disease and angiodysplasia, and success rates of 74%-93% have been reported. But in patients with Crohn's disease, successful superselective embolization has rarely been reported. This is a report of successful superselective embolization in a patient with Crohn's disease; this should be the initial treatment of choice in Crohn's disease in an attempt to avoid surgical resection, because repeated resections predispose patients to the development of short-bowel syndrome.

Gastrointestinal haemorrhage from a second source in patients with colonic angiodysplasia.

Angiodysplasia of the colon is increasingly recognized as a cause of obscure gastrointestinal haemorrhage. On review of 71 patients with angiodysplasia, 40 came to surgery. Of these, nine (22.5 per cent) were also found to have small bowel lesions, either at the original laparotomy (7) or later when further investigations were performed for recurrent bleeding. Angiodysplasia can only be confidently diagnosed as the source of blood loss if seen to be actively bleeding. If a synchronous small bowel lesion is seen, we recommend that it should be removed at the same time as the colectomy.

Peritonitis following Argon Plasma Coagulation of Colonic Angiodysplasia in a CAPD Patient — an Avoidable Complication?

Peritonitis following Argon Plasma Coagulation of Colonic Angiodysplasia in a CAPD Patient — an Avoidable Complication?

Multiple large angiodysplasia of the right colon and aortic stenosis (Heyde-syndrome) in a patient with gastrointestinal bleeding – a case report

Multiple large angiodysplasia of the right colon and aortic stenosis (Heyde-syndrome) in a patient with gastrointestinal bleeding – a case report

Colonic angiodysplasia and true diverticula: is there an association?

To investigate the pathology of colectomy specimens, from patients presenting with lower gastrointestinal haemorrhage, who had undergone preoperative mesenteric angiography. The angiography diagnoses ranged from active bleeding of unknown aetiology to angiodysplasia. The macroscopic and microscopic pathology of 11 colectomy specimens was studied. All the specimens contained blood on receipt with no obvious macroscopic source of haemorrhage identified. In all cases the large bowel demonstrated diverticula with an unusual topography. These were wide-mouthed, up to 20 mm across and exhibited a partly scattered distribution across the bowel; in two cases the diverticula contained blood clot. Histology demonstrated true diverticula invested by all muscle layers, exhibiting thick-walled vessels, within the submucosa and within the muscularis propria fibres, resembling the vascular elements identified in congenital vascular malformations. The two cases that contained blood clot revealed the thick-walled vessels to have ruptured into the diverticular lumen. Histology from the background colon between the diverticula demonstrated vascular lesions fulfilling the histological criteria for angiodysplasia; these features were also represented in the true-type diverticula. Most cases showed right-sided angiodysplasia and right-sided true diverticula. However, left-sided angiodysplastic lesions were also seen in association with left-sided true-type diverticula. We have identified the dual pathologies of angiodysplasia and multiple true-type diverticula, and suggest a topographical relationship between these two lesions. Furthermore, we propose that the thick-walled vessels within the true diverticula are congenitally abnormal, and raise the possibility that these vessels exert local haemodynamic effects, that effectively predispose the colon to acquired angiodysplastic phenomena.

Nonsteroidal anti-inflammatory drug associated lesions and mortality among African American and Hispanic patients with lower gastrointestinal bleeding

Nonsteroidal anti-inflammatory drugs (NSAIDs) are toxic to the entire gastrointestinal (GI) tract. Although there is a plethora of data about upper GI toxicity of NSAIDs, the data are scant about lower GI toxicity, particularly in African Americans and Hispanics, our main patient population. We studied the association between NSAIDs use and lower GI lesions in patients with lower GI bleeding (LGIB). Records of 988 patients, 19 to 103 years old, admitted with a diagnosis of LGIB over 12 years (from 1991 to 2002), were reviewed retrospectively. We excluded 230 patients (44 white, 45 Asian, 98 with upper GI source of LGIB, and 43 with incomplete data). The information abstracted included admission diagnoses, NSAIDs and/or aspirin (NSAIDs/AS) use, colonoscopic findings, and outcome measures (duration of hospitalization; surgery; mortality). Among the 758 patients (411 African American and 347 Hispanics), 47% were taking NSAIDs/AS regularly and 53% were not taking NSAIDs/AS. Colonoscopic findings revealed that the common lesions were colonic diverticulosis (36%), angiodysplasia (29%), colonic polyps (25%), and hemorrhoids (24%). Colonic diverticulosis, colonic ulcers, and inflammatory bowel diseases were significantly higher among NSAIDs/AS users than among nonusers ( P < 0.05). On average, NSAIDs/AS users had longer hospital stays (5 days) than nonusers (3 days). Of the 758 patients, 13% had surgery because of severe bleeding. Surgical intervention was more likely among NSAIDs/AS users (16%) than among nonusers (11%) (odds ratio OR = 1.6, 95% confidence interval CI = 1.0–2.4, P = 0.04). Overall, 160 deaths occurred (21%). Mortality was significantly higher among NSAIDs/AS users (29%) than among nonusers (14%) (OR = 2.6, 95% CI = 1.8–3.7, P = 0.0001). Morbidity, mortality, and need for surgery were higher among NSAIDs/AS users than nonusers. Future prospective studies in this population may help in better management and possible risk reduction.

Argon Plasma Coagulation in the Prevention of Recurrent Bleeding from Colonic Angiodysplasias

BACKGROUND: Angiodysplasias have also been found in 3 to 6% of subjects undergoing colonoscopy. Despite its relatively low prevalence, this disorder is a frequent cause of recurrent lower gastrointestinal tract bleeding. Argon plasma coagulation (APC) has proven useful to arrest bleeding from angiodysplasia, but its effectiveness in the prevention of rebleeding has been poorly evaluated. AIM: To asses the effectiveness and safety of the endoscopic argon plasma coagulation in the prevention of recurrent bleeding due to colonic angiodysplasias. PATIENTS AND METHODS: A total of 41 patients (20 females and 21 males; mean age, 65 years; range 10-85) with endoscopic diagnosis of gastrointestinal bleeding due to angiodysplasia of the colon were included in the study. Lesions were treated with an argon plasma coagulator APC 300 (ERBE Elektromedizin GmbH. Germany). APC was performed using a 2.3 mm probe with a 2.5 L/m gas flow and different power ranging from 60 watts for left colon and rectum to 40 watts for right colon, transverse and cecum. The endoscopic intervention was considered successful if there was no further overt bleeding and hemoglobin concentration level was stabilized without transfusion or iron therapy, except for the initial iron storage replacement. Rebleeding was defined as any detectable bleeding (melena or hematoquezia) or a drop in hemoglobin level >1 g/dl after angiodysplasia eradication. RESULTS: Follow-up data were obtained from all 41 patients. Overt bleeding resolved and hemoglobin levels were stabilized without transfusions or iron therapy in 36 of 41 patients (88%) with a mean follow-up of 19 months (range, 6 - 39). In the subgroup of patients with anemia, mean hemoglobin levels increased from 8.8 g/dl (range, 6.1-12.2) before the treatment to 12.3 g/dl (range, 8.1-15.2) after it (paired t-test p<0.01). The actuarial probability of remaining free of rebleeding at 2 years of follow-up was 88% (CI, 72% - 95%). Six patients required more than one therapeutic session. On 48 procedures, only two complications were observed (4.1%): one patient had bacteremia and another pneumoperitoneum without laparoscopic evidence of perforation. There were no treatment-related deaths. CONCLUSIONS: Endoscopic argon plasma coagulation in the prevention of recurrent bleeding from colonic angiodysplasia is both effective and safe.

Usefulness of Angiography in Patients with Undiagnosed Lower Gastrointestinal Bleeding

Colonoscopy is considered the most effective first test in the evaluation of patients with significant lower gastrointestinal bleeding (LGB). However, if colonoscopy can not identify a definite bleeding source, the optimal approach remains uncertain. Aims of the study: 1) To assess the effectiveness of colonoscopy as initial diagnostic test for acute LGB, and 2) To determine the diagnostic yield of selective mesenteric angiography in patients with undiagnosed LGB. Patients and methods: patients hospitalized for acute LGB over a 50 month period in two teaching hospitals, who underwent urgent colonoscopy and eventual gastroscopy as initial diagnostic procedures were retrospectively analyzed. By using predetermined criteria, diagnosis were categorized either as definite, when a lesion showed stigmata of recent hemorrhage, or undiagnosed, in the absence of lesions or signs of recent bleeding. Results:162 patients (50% male) with a mean age of 69+/−1.6 yrs (75% equal or > 65 yrs) were included. Presentation of hemorrhage was severe with shock in 19%, 67% requiered blood transfusions (mean: 3.4+/−0.4 UU) and 60% had comorbidity. Colonoscopy found potential bleeding lesions in 109/162 patients (38% had diverticulosis), but definite diagnosis could be established in only 42/162 (26%) patients. Diverticulosis (7%), cancer and polyps (38%), ischemic colitis (12%) and angiodysplasia (14%) were the main sources of bleeding. Angiography, performed in 69 out of 120 patients with undiagnosed hemorrhage (58%), established a definite diagnosis in 48/69 (70%) patients. Angiodysplasia was the bleeding source in 15 out of 17 (88%) patients with further bleeding during hospitalization, whereas in the 52 patients with self-limited hemorrhage angiodysplasia was found in 31 (60%) patients. Angiography also established the diagnosis in 2 additional patients (1 small bowel tumor and 1 actively bleeding diverticuli). Conclusions: Acute LGB often remains undiagnosed after initial colonoscopy. In this condition, colonic angiodysplasia is the most common cause and mesenteric angiography is a useful procedure, particularly in patients with further bleeding.

Are Colonic Angiodysplasia a Frequent Cause of Brisk Lower Intestinal Bleeding?

Background/Aims: Estimates of colonic angiodysplasia (AVMs) as a source of acute lower intestinal bleeding range from 5-40%. The aims of this study were to gather population-based data on the prevalence and treatment of bleeding colonic AVMs, and to identify predictors of active bleeding. Methods: Colonoscopy reports with documented AVMs generated between January 2000 and December 2002 were retrieved from the Clinical Outcomes Research Initiative (CORI) national endoscopic database. Data were collected on patient and hospital demographics, as well as exam findings. Multiple logistic regression was used to identify risk factors for active bleeding. Results: Of the 229,727 unique patients undergoing colonoscopy, a total of 1804 (0.79%) were found to have AVMs. The mean age of patients with AVMs was 68 years (±12.2); 53% of patients were male, 84% were Caucasian, and 15% had an American Association of Anaesthetists (ASA) score of ≥ III. AVMs were more common in the proximal colon: cecum 42%, ascending 21%, transverse 7%, descending and sigmoid 20%, rectum 8%. Among patients with AVMs, common exam indications included hematochezia (25%), occult blood in stool (19%) and anemia (19%). Only 47 patients with AVMs (3%) underwent inpatient colonoscopy for evaluation of hematochezia. Active bleeding was noted at the time of endoscopy in 114 patients with AVMs (6.3%). 48 of the bleeding AVMs (42%) and 164 of the total AVMs (9%) were treated endoscopically. Bipolar coagulation (69%) and APC (27%) were the most commonly utilized treatment modalities. Independent risk factors for active bleeding were [OR, 95% CI]: hematochezia (2.8, 1.9-4.3), increasing age (1.03, 1.02-1.05), academic practice setting (1.8, 1.2-2.9) and occult blood in stool (1.8, 1.1-2.8). AVM location was not a predictor of bleeding. Conclusions: Contrary to frequently cited estimates, colonic AVMs do not appear to play a major role in brisk lower intestinal bleeding. Fewer than half of bleeding AVMs receive endoscopic treatment. Older age and hematochezia were the most significant predictors of active bleeding at colonoscopy.

Functional outcome of Horton's vaginoplasty.

To describe the functional results of Horton's vaginoplasty. A descriptive study. Department of Plastic Surgery, Hayatabad Medical Complex, Peshawar. The duration of study was three years from May 1998 to May 2001. During the three years period, 17 patients were operated with congenital vaginal agenesis. Horton's vaginoplasty with full thickness skin grafts was performed in all these patients. Patients with a follow-up of less than 6 months were excluded from the study. Sexual assessment was done in married patients after Horton's vaginoplasty. These patients were categorized into four categories according to their sexual experience. Functional assessment revealed excellent sexual relations in 36.36%, good in 45.45% and satisfactory in 18.18% patients. Vaginal stenosis and loss of graft take was observed in 5.8% patients. The average depth of vagina was 8.5 cm. The width was two fingers in 64.70% and three fingers in 29.41% patients. Horton's vaginoplasty is a versatile, single-stage procedure, with minimal donor site morbidity. The reconstructed vagina is excellent both in function and cosmesis.

Pedunculated angiodysplasia of the colon successfully treated with endoscopy: a new type of angiodysplasia

Pedunculated angiodysplasia of the colon successfully treated with endoscopy: a new type of angiodysplasia

Atypical angiodysplasia with varicose appearance

Angiodysplasia (AD) is a degenerative disorder in which a vascular malformation extends to the submucosa, the mucosa, or both. In 1956, Rutter1 described the histopathologic diagnosis of AD in the GI tract. In 1960, a case of AD associated with GI bleeding was reported by Margulis et al.2 The number of reports on AD in the stomach and lower intestinal tract has increased because of the use of advanced endoscopic and angiographic techniques.

Argon plasma coagulation (APC)for the treatment of bleeding angiodysplasia (AD)

Argon plasma coagulation (APC)for the treatment of bleeding angiodysplasia (AD)

Bleeding colonic angiomas and radiation telangiectasias: Endoscopic diagnosis and treatment

Colonic angiodysplasia and angiomas are common sources of lower gastrointestinal (Gl) bleeding in elderly patients. Approximately one third of our patients with lower Gl bleeding whose diagnosis was colon angiomas presented with recurrent hematochezia. Many of these patients have severe comorbid conditions such as heart, renal, or liver disease. Additionally, some are on medications that are associated with Gl hemorrhage, such as aspirin, NSAIDs, (nonsteroidal anti-inflammatory agents) or anticoagulants. Urgent colonoscopy after thorough cleansing of the colon of stool, blood, and clots with colonic purge is a very effective way to make the diagnosis. A more common presentation of colon angioma hemorrhage is iron deficiency anemia and stools with occult blood. Colonic cleansing is also important in this subgroup, but can be done with standard oral preparations. The goal of the colonoscopy in patients with colonic angiomas is to perform 1 complete examination for both diagnosis and treatment. Endoscopic coagulation with contact probes such as bipolar or heater probes should be with a low power setting, gentle apposition pressure, and short coagulation pulses directly on the angiomas. Often multiple angiomas are found, and the endpoints are control of active bleeding and good coagulation (whitening) of all angiomas in the colon segment. Overdistension, repeated coagulation at the same site, and high power settings should be avoided. Radiation telangiectasias are a common cause of recurrent, often severe hematochezia in patients who were treated for pelvic cancers with radiotherapy. Multiple oozing sites or moderately severe bleeding from radiation telangiectasias is diagnosed on colonoscopy or flexible sigmoidoscopy. Both diagnosis and endoscopic treatment are facilitated by target irrigation and contact probes. Control of bleeding depends on coagulation of multiple telangiectasias, usually in the rectum. Outcomes of patients with severe hematochezia from either colonic angiomas or radiation telangiectasia on endoscopic therapy are significantly better than those of patients who undergo medical therapies. Details on the diagnosis, treatment techniques, results, and complications are presented in this report.

Is Helical CT Angiography Useful for Diagnosing Colonic Angiodysplasia

Angiodysplasia of the colon typically is diagnosed by colonoscopy, or in a few cases by angiography. The authors of this prospective study evaluated 28

Accuracy of helical computed tomographic angiography for the diagnosis of colonic angiodysplasia

Background & Aims: The diagnosis of colonic angiodysplasia is often challenging and relies on endoscopy or catheter angiography. We investigated whether computed tomographic angiography (CTA) contributes to the diagnosis of colonic angiodysplasia. Methods: Twenty-eight patients with suspected bleeding from colonic angiodysplasia were prospectively evaluated. Gastrointestinal bleeding was investigated by colonoscopy plus visceral angiography and by CTA. The level of agreement between CTA and the former procedures was determined. Results: CTA images of diagnostic quality were obtained in 26 patients. Eighteen patients were diagnosed with colonic angiodysplasia by colonoscopy plus visceral angiography, and 14 by CTA (κ = 0.68; P < 0.001). Sensitivity, specificity, and positive predictive values of CTA for detection of colonic angiodysplasia were 70%, 100%, and 100%, respectively. CTA signs including accumulation of vessels in the colonic wall, early filling vein, and supplying enlarged artery were present in 55%, 50%, and 22% of cases, respectively. None of these signs were present in the 8 patients with obscure gastrointestinal bleeding and negative diagnostic investigation of the digestive tract. Conclusions: CTA is a sensitive, specific, well-tolerated, and minimally invasive tool for the diagnosis of colonic angiodysplasia.GASTROENTEROLOGY 2000;119:293-299

3387 Yield of panendoscopy and colonoscopy for iron deficiency in post-gastrectomy patients.

Intro: Iron (Fe) deficiency is a well documented late complication of partial gastrectomy (PG). The yield of endoscopy in the evaluation of Fe deficient PG patients is unknown. Aim: Determine the utility of panendoscopy (EGD) and colonoscopy (CSP) in Fe deficient patients with prior Billroth I (B-I) or Billroth II (B-II) gastrectomy compared to non-PG patients. Methods: Patients with prior PG who had CSP and/or EGD for evaluation of Fe deficiency were identified and compared to a control (cont) group of Fe deficient patients without history of gastric surgery. The two groups were retrospectively identified using the CORI (Clinical Outcomes Research Initiative) database, between periods of 6/1/96 and 7/31/99 at Oregon Health Sciences University and the Portland VAMC. Patients with a ferritin ≤ 50 μg/L or a transferrin saturation (%sat) ≤ 20% were included and their records reviewed. Patients were excluded for prior gastric surgeries other than a B-I or B-II. Laboratory data included hemoglobin, hematocrit, mean corpuscular volume, ferritin, and %sat. Colonoscopic findings of tumor, polyps ≥1 cm, angiodysplasia >5 in number, active colitis, or colonic ulceration were recorded. The presence of any adenomatous polyp (regardless of size) was also noted. EGD findings evaluated were ulcers, esophagitis, or esophageal/gastric varices (Vxs). Fisher's exact test and Student's T-test were used for statistical analysis. Results: 52 PG patients (35 CSP, 52 EGD) and 53 controls (44 CSP, 53 EGD) were studied. There were 13 BI and 39 BII patients. Mean interval between PG and endoscopy was 22 yr (range 4-46 yr). Demographics and laboratory values were not significantly different between PG and cont groups. Mean ferritin levels were similar (PG=20.6, cont=23.4 μg/L, p=0.3). There were no significant differences (p>0.05) in the proportion of patients with polyps ≥1 cm (PG=9%, cont=7%) or any adenomatous polyps (PG=29% vs 18%). No patient had malignant colonic tumors, active colitis, or significant colonic angiodysplasia. Two patients with PG had colonic ulcers, versus no control patients. Comparing EGD findings, there were no significant differences between groups for the presence of ulcers (PG=14% vs 23%), esophagitis (PG=12% vs 13%), Vxs (PG=4% vs 6%), or presence of any of these three lesions (PG=25% vs 38%). There were no upper GI malignancies. Conclusions: The yield of EGD and CSP is similar in PG and non-PG Fe deficient patients. The endoscopic evaluation of Fe deficiency should not differ in patients with PG.

A Case of Colonic Angiodysplasia

A Case of Colonic Angiodysplasia

Increased expression of angiogenic factors in human colonic angiodysplasia

OBJECTIVE: Angiodysplasia of the colon is a distinct vascular abnormality characterized by focal accumulation of ectatic vessels in the mucosa and submucosa. To investigate whether angiogenesis contributes to the pathogenesis of human colonic angiodysplasia, we examined the expression of basic fibroblast growth factor (bFGF) and vascular endothelial growth factor (VEGF), and its endothelial cell receptors flt-1 and KDR. METHODS: Immunohistochemistry was performed in sections of specimens obtained from 18 patients with colonic angiodysplasia and from eight patients with colon cancer and its adjacent, histologically normal margins of resection. We used affinity-purified rabbit polyclonal antibodies and a streptoavidin–biotin peroxidase method. RESULTS: We detected strong immunoreactivity for vascular endothelial growth factor, homogeneously distributed in the endothelial lining of blood vessels of all sizes in 16 (89%) specimens of colonic angiodysplasia and in seven (88%) patients with colon cancer. In contrast, very limited immunoreactivity was found in normal colon. Vascular staining for flt-1 was observed in eight (44%) and one (12.5%) of the colonic angiodysplasia or colon cancer specimens, respectively, but not in normal colon. Vascular immunoreactivity for basic fibroblast growth factor was observed in seven (39%) specimens from patients with colonic angiodysplasia, whereas either very limited or no immunostaining was found in sections from specimens of patients with colon cancer and its normal margins. CONCLUSIONS: In human colonic angiodysplasia, increased expression of angiogenic factors is likely to play a pathogenic role.