Degree of concordance between double-balloon enteroscopy and capsule endoscopy in obscure gastrointestinal bleeding: a multicenter study.

BACKGROUND AND STUDY AIMS: Capsule endoscopy is considered the diagnostic procedure of choice in patients with obscure gastrointestinal bleeding (OGIB). Double-balloon endoscopy (DBE) offers both diagnostic and therapeutic potential, but is invasive, complex, and time-consuming. The aim was to evaluate diagnostic agreement between capsule endoscopy and DBE in patients with OGIB, and secondarily the diagnostic gain of DBE when capsule endoscopy detected only blood or clots in the small-bowel lumen. METHODS: Multicenter prospective study carried out at six institutions in Italy. RESULTS: 193 patients (119 men, mean age 61.6 +/- 16.2) first underwent capsule endoscopy and then DBE. The most frequent positive findings at capsule endoscopy were vascular lesions (74 patients, 38.3 %), blood or clot in the lumen (34, 17.6 %), and tumor mass (20, 10.4 %). The most frequent findings at DBE were vascular lesions (72 patients, 37.3 %), neoplasia (30, 15.5 %) and ulcers/inflammatory lesions (12, 6.2 %). Overall kappa coefficient was 0.46 (95 %CI 0.38 - 0.54), with maximum concordance for vascular (0.72 [95 %CI 0.59 - 0.84]) and inflammatory (0.78 [0.58 - 0.99]) lesions and minimum for polyps (0.46 [0.16 - 0.80]). Blood in the lumen was the only positive finding at capsule endoscopy in 34 cases; of these, 12 had negative DBE findings whereas 10 had vascular lesions, 6 neoplasia, 1 ulcer, and 5 diverticula. CONCLUSION: Capsule endoscopy and DBE have good agreement for vascular and inflammatory lesions but not for polyps or neoplasia. DBE provides valuable adjunctive information, particularly in patients with neoplasia or polyp at capsule endoscopy. DBE clarified the origin of bleeding in two-thirds of patients with capsule endoscopy showing only blood in the lumen.

Occult intestinal haemorrhage due to lipoma of the small bowel detected with the combined use of the new endoscopic techniques. A report of two cases.

We report two unusual cases of occult intestinal bleeding due to lipoma of the small bowel. Both the patients underwent several inconclusive endoscopic and radiological procedures before the diagnosis of these lesions that was possible using video capsule endoscopy and double balloon enteroscopy. In the first case, the finding of capsule endoscopy was confirmed using enteroscopy, allowing us to diagnose the ileal mass. In the second and more recent case, we used double balloon enteroscopy for the further characterisation and management of the ileal lipoma. Lipoma is a very rare cause of intestinal bleeding and we describe the management and the outcome of our patients.

Clinical identification and long-term surveillance of 22 hereditary non-polyposis colon cancer Italian families.

OBJECTIVE: To assess the efficacy of a hereditary non-polyposis colon cancer (HNPCC) identification and surveillance policy. METHODS: Familial clustering of colorectal cancer (CRC) and extracolonic cancers (ECs) was investigated in 1520 consecutive CRC patients and relatives. HNPCC was identified by Amsterdam criteria, and individuals at risk were offered biennial colonoscopy and other examinations, starting from age 25 years. RESULTS: Twenty-two HNPCC families were identified. The CRC prevalence was 27.8% (121/435), decreasing from 59.4% in the first generation to 24.4% and 8% in the second and third generation, respectively. Twenty-nine patients had multiple CRC and 34 patients (in 12 families) had ECs.A total of 199/331 at-risk individuals accepted surveillance. The mean follow-up was 48+/-32 months. CRCs were detected at first surveillance in four out of 199 surveilled individuals (2%); in two surveilled individuals (1%), three CRCs developed during follow-up. The overall CRC incidence was 7/199 (3.5%) in surveilled individuals and 5/132 (3.7%) in unsurveilled individuals. CRCs were less advanced in surveilled than in unsurveilled patients. Eleven individuals had 22 adenomas (one with high-grade dysplasia). Three individuals had adenomas at first surveillance; two of them and eight more individuals during surveillance. Seven surveilled individuals and six unsurveilled individuals, all belonging to families with a history of EC, had EC during the study period. All patients with CRC detected by surveillance are alive. One of the unsurveilled patients who had CRC died 18 months after the diagnosis. CONCLUSIONS: Data confirm the importance of the family history collected in each patient with CRC for identification of HNPCC and support the efficacy of repeated colonoscopies for early diagnosis and prevention of CRC in at-risk members. Reasons for surveillance failure could be an accelerated progression of small adenomas and a lesion missing at colonoscopy. Longer follow-up is required to assess the efficacy of surveillance for EC.