Predictores de mortalidad en hemorragia digestiva alta en pacientes ingresados por otra condición: estudio prospectivo.

Resum

Introduction and aims: Patients with in-hospital upper gastrointestinal bleeding have a higher mortality rate, however, the reasons for this increase have not been consistently identified. The aim of our study is to compare outpatients presenting with upper gastrointestinal bleeding (UGIB) to those who started GI bleeding while they were hospitalized for another condition; as well as to identify predictors of in-hospital mortality in this subgroup of patients. Material and methods: Prospective study on consecutive UGIB patients (variceal and non-variceal) treated in "Virgen de las Nieves" University Hospital (2013-2017). Information regarding clinical, biochemical data and procedures was collected. Documented clinical outcomes were in-hospital mortality, rebleeding and delayed (6-months) mortality. Results: 540 patients with a diagnosis of UGIB were included, 59 inpatients (IPs) and 481 outpatients (OPs). IPs differed from OPs in in-hospital mortality rate (42.1% vs 7.8%; P<0.001), rebleeding (29.3% vs. 15.3%; P=0.014), AIMS65≥2 (25.6% vs. 13.7%; P=0,043), ASA score (3-4 86,4% vs. 65,5; P<0,001), active bleeding in endoscopy (55.9% vs. 26.4%; P<0.001), endoscopic treatment (71.2% vs. 35.8%, P<0.001), blood units transfusions (5.42 vs. 2.48; P<0.001); hospital stay (23.69 vs. 6.91 days; P<0.001). Independent predictors for mortality in IPs were ASA ≥3 (OR 7.8; 95% CI 2.089-29.143, P=0.002) and need for endoscopic intervention (OR 2.00; 95% CI 1.063-3.746, P=0.032). Conclusions: Inpatients with UGIB have higher morbidity and mortality rates, and the ASA class is clearly a predictor of mortality in patients with UGIB both variceal and non-variceal.