Management of lower gastrointestinal bleeding in the emergency department short-stay unit

Fernández Alonso C, García Lamberechts EJ, Fuentes Ferrer M, Chaparro D, Cuervo Pinto R, Villarroel Elipe P, López Farré AJ, Esteban JM, González Del Castillo J, González Armengol JJ, Martín-Sánchez FJ

Objectives: To profile the patient with lower gastrointestinal bleeding admitted to an

emergency department short-stay unit and to determine predictors of admission to a

conventional ward.

Material and methods: Retrospective study of a series of patients with lower

gastrointestinal bleeding in the short-stay unit of Hospital Clínico San Carlos between

January 1 and December 31, 2008. Information gathered from the discharge reports

included patient characteristics and clinical data, diagnostic and therapeutic procedures,

the definitive diagnosis, hospital mortality, length of stay, and final destination. Two

groups defined according to risk of admission (high-risk: colon cancer, colitis, or unknown

cause of bleeding; and low-risk: anal-rectal disease, diverticulitis, or angiodysplasia). A

logistic regression model was constructed to identify variables that were independently

associated with transfer to the ward.

Results: A total of 177 patients (50.8% women) with a mean (SD) age of 70.5 (17.0)

years were treated. Sixty-six patients (37%) had high-risk of admission. The most

frequent definitive diagnosis was internal hemorrhoids (35.0%). The median stay was

2.0 (interquartile range, 1.0-3.0) days. Most patients were discharged home (71.2%).

Factors related to admission to a ward in the univariate analysis (high-risk vs low-risk

groups) were bloody stools (24.6% vs 9.1%, P=.05), abdominal pain (49.2% vs 21.8%,

P<.0001), tachycardia (22.2% vs 9.1%, P=.022), computed tomography (14.5% vs 3.7%, P=.015), white blood cell count (9871.4/mL vs 8280.0/mL, P=.006), scheduling of prompt surgery (14.8% vs 0.0%, P<.0001), and antibiotic treatment (37.7% vs 3.7%, P<.0001). Multivariate analysis of the likelihood of admission identified the following risk factors: abdominal pain (odds ratio [OR], 2.37, 95% confidence interval [CI], 1.4-5.0; P=.025) and fever (OR, 3.68; 95% CI, 1-13.7; P=.042). Conclusions: The short-stay unit seems to have a high capacity for diagnosing and resolving cases of low gastrointestinal bleeding. It complements hospitalization by offering an alternative to admission to a conventional ward.

Chaparro D, Cuervo Pinto R, Esteban JM, Fernández-Alonso C, Fuentes Ferrer M, García-Lamberechts EJ, González Armengol JJ, González Del Castillo J, López Farré AJ, Martín-Sánchez FJ, Villarroel Elipe P