Faculty Bibliography

GOALS:To report a case series of ischemic gastritis and discuss its etiology, management, and associated mortality according to our results and the published English literature. BACKGROUND:Ischemic gastritis is rare, given the rich blood supply of the stomach. It has been reported in isolated case reports and small case series. Most cases are vascular in origin and associated with a high mortality. STUDY:Pathology databases from 3 hospitals affiliated with the University of Minnesota Medical School were searched for cases of ischemic gastritis in the last 10 years. Patients' demographics, clinical course, and 1-month and 1-year mortalities were collected from electronic medical records. RESULTS:A total of 12 patients were identified (age range, 32.1 to 83.2), the largest series reported to date. The presenting symptom was gastrointestinal bleeding (8), abdominal pain (2), nausea (1), and symptomatic anemia (1). The etiology included postinterventional radiology embolization (2), hemodynamic changes in the setting of celiac axis stenosis (2), vasculitis (1), systemic hypotension (1), and unknown (6). Treatment included steroid therapy, revascularization by interventional radiology, surgery, or supportive treatment. Thirty-day and 1-year mortalities were 33% and 41%, respectively. CONCLUSIONS:Ischemic gastritis is rare, but associated with a high mortality. Evaluation for treatable etiologies should be sought and corrected if present.

BACKGROUND AND AIMS/OBJECTIVE:The biological environment varies across the colorectum and may therefore affect neoplastic growth differently in the proximal and distal colon. The aim of the study was to evaluate the risk for recurrent adenomas and their anatomic location based on adenoma location at baseline colonoscopy. METHODS:Data were extracted from 3 adenoma prevention trials (n = 2430). Participants had at least 1 adenoma at baseline colonoscopy and underwent subsequent surveillance colonoscopy, at which time metachronous adenomas could be detected. We calculated the risk ratio (RR) and the 95% confidence interval (CI) for metachronous adenomas by location of the baseline lesion and considered the impact of advanced neoplasia and multiplicity. RESULTS:At baseline, 522 patients (21.5%) had adenomas only in the proximal colon, 1266 patients (52.1%) had adenomas only in the distal colorectum, and 642 (26.4%) had adenomas in both regions. Overall, 877 patients (36.5%) had metachronous adenomas during the follow-up period. Those with only proximal adenomas at baseline had a higher risk of metachronous adenomas compared with patients with only distal adenomas (RR, 1.17; 95% CI, 1.01-1.35). A greater proximal risk was found after restricting the analysis to patients with multiple proximal adenomas versus multiple distal adenomas (RR, 1.35; 95% CI, 1.10-1.67). The risk of recurrent adenomas on the same side was 48% higher for patients with only proximal adenomas at baseline compared with those with only distal adenomas at baseline (RR, 1.48; 95% CI, 1.22-1.80). CONCLUSIONS:Patients with proximal adenomas only have a modestly greater risk of adenoma recurrence than patients with adenomas limited to the distal colon, and have a greater likelihood of adenoma recurrence on the same side compared with patients with distal adenomas. This observation suggests that biological factors may differentially affect neoplasia growth across the colon.

BACKGROUND & AIMS/OBJECTIVE:Cancelled and missed colonoscopy appointments waste resources, increase colonoscopy delays, and can adversely affect patient outcomes. We examined individual and organizational factors associated with missed and cancelled colonoscopy appointments in Veteran Health Administration facilities. METHODS:From 69 facilities meeting inclusion criteria, we identified 27,994 patients with colonoscopy appointments scheduled for follow-up, on the basis of positive fecal occult blood test results, between August 16, 2009 and September 30, 2011. We identified factors associated with colonoscopy appointment status (completed, cancelled, or missed) by using hierarchical multinomial regression. Individual factors examined included age, race, sex, marital status, residence, drive time to nearest specialty care facility, limited life expectancy, comorbidities, colonoscopy in the past decade, referring facility type, referral month, and appointment lead time. Organizational factors included facility region, complexity, appointment reminders, scheduling, and prep education practices. RESULTS:Missed appointments were associated with limited life expectancy (odds ratio [OR], 2.74; P = .0004), no personal history of polyps (OR, 2.74; P < .0001), high facility complexity (OR, 2.69; P = .007), dual diagnosis of psychiatric disorders and substance abuse (OR, 1.82; P < .0001), and opt-out scheduling (OR, 1.57; P = .02). Cancelled appointments were associated with age (OR, 1.61; P = .0005 for 85 years or older and OR, 1.44; P < .0001 for 65-84 years old), no history of polyps (OR, 1.51; P < .0001), and opt-out scheduling (OR, 1.26; P = .04). Additional predictors of both outcomes included race, marital status, and lead time. CONCLUSIONS:Several factors within Veterans Health Administration clinic control can be targeted to reduce missed and cancelled colonoscopy appointments. Specifically, developing systems to minimize referrals for patients with limited life expectancy could reduce missed appointments, and use of opt-in scheduling and reductions in appointment lead time could improve both outcomes.