Increased Risk of Progression to Gastric Adenocarcinoma in Patients with Non-dysplastic Gastric Intestinal Metaplasia Versus a Control Population.

AIM: In previous studies, the 5-year progression rate of gastric intestinal metaplasia to gastric adenocarcinoma has varied substantially. We investigated the incidence rate of dysplasia and gastric adenocarcinoma and the rate of progression among a cohort of patients with non-dysplastic gastric intestinal metaplasia. METHODS: This is a single-center, single-cohort retrospective study. Patients who had undergone an EGD with biopsies from 01/01/1993 to 12/31/2013 were included. The primary outcome of interest was the composite of low-grade dysplasia, high-grade dysplasia, or adenocarcinoma. Time to progression and risk factor subgroup analyses were performed. RESULTS: A total of 1628 subjects were screened, of whom 358 met the inclusion criteria. A total of 21 first-time events were recorded. The annual incidence rate of low-grade dysplasia was 2.1 (95% CI 1.3-3.5) cases per 1000 person-years, 0.5 (95% 0.2-1.3) per 1000 person-years for high-grade dysplasia, and 0.8 (95% CI 0.3-1.6) cases per 1000 person-years for gastric adenocarcinoma. The historical control group had an annual adenocarcinoma incidence rate of 0.07 per 1000 person-years. The event rate in Asians was also noted to be significantly higher between years 0-8 as compared with patients of non-Asian race, and extensive intestinal metaplasia was an independent risk factor (HR = 4.06 (95% CI 1.45-11.34), p = 0.007). CONCLUSIONS: Patients with non-dysplastic gastric intestinal metaplasia may progress to dysplasia and gastric adenocarcinoma. The incidence rate of gastric adenocarcinoma is higher than that of the historical control population (0.07 per 1000 person-years). The presence of extensive intestinal metaplasia was a risk factor for progression of disease. Triennial EGD may be warranted in patients with non-dysplastic gastric intestinal metaplasia.

Home Hospital Outcomes for Acute Decompensated Heart Failure and Factors Associated With Escalation of Care.

BACKGROUND: Overall outcomes and the escalation rate for home hospital admissions for heart failure (HF) are not known. We report overall outcomes, predict escalation, and describe care provided after escalation among patients admitted to home hospital for HF. METHODS: Our retrospective analysis included all patients admitted for HF to 2 home hospital programs in Massachusetts between February 2020 and October 2022. Escalation of care was defined as transfer to an inpatient hospital setting (emergency department, inpatient medical unit) for at least 1 overnight stay. Unexpected mortality was defined as mortality excluding those who desired to pass away at home on admission or transitioned to hospice. We performed the least absolute shrinkage and selection operator logistic regression to predict escalation. RESULTS: We included 437 hospitalizations; patients had a median age of 80 (interquartile range, 69-89) years, 58.1% were women, and 64.8% were White. Of the cohort, 29.2% had reduced ejection fraction, 50.9% had chronic kidney disease, and 60.6% had atrial fibrillation. Median admission Get With The Guidelines HF score was 39 (interquartile range, 35-45; 1%-5% predicted inpatient mortality). Escalation occurred in 10.3% of hospitalizations. Thirty-day readmission occurred in 15.1%, 90-day readmission occurred in 33.8%, and 6-month mortality occurred in 11.5%. There was no unexpected mortality during home hospitalization. Patients who experienced escalation had significantly longer median length of stays (19 versus 7.5 days, P<0.001). The most common reason for escalation was progressive renal dysfunction (36.2%). A low mean arterial pressure at the time of admission to home hospital was the most significant predictor of escalation in the least absolute shrinkage and selection operator regression. CONCLUSIONS: About 1 in 10 home hospital patients with HF required escalation; none had unexpected mortality. Patients requiring escalation had longer length of stays. A low mean arterial pressure at the time of admission to home hospital was the most important predictor of escalation of care in the least absolute shrinkage and selection operator logistic regression model.