ABSTRACT
Objetivos. Establecer la importancia de la profilaxis para prevenir la hemorragia digestiva en pacientes críticos, determinar los predictores del desarrollo de hemorragia digestiva en estos enfermos y cuantificar la neumonía intrahospitalaria. Materiales y métodos. Diseño prospectivo, observacional, longitudinal. Se estudiaron 3438 pacientes durante dos años. Luego de aplicar los criterios de exclusión y de eliminación, se evaluó a 3213 pacientes. La muestra fue consecutiva, se formaron dos grupos: con profilaxis (sucralfato o ranitidina) y sin profilaxis. Se registraron los siguientes datos: puntaje Apache II, disfunciones, diagnósticos de riesgo, hemorragia digestiva, neumonía intrahospitalaria, mortalidad. Resultados. Incidencia de hemorragia digestiva: 4,57%; con profilaxis frente a sin profilaxis: 4,64% y 4,50% (p = 0,91). Predictores de hemorragia digestiva en análisis multivariado: trauma grave (p = 0,0004), insuficiencia respiratoria (p = 0,0021), pancreatitis (p = 0,0202), insuficiencia renal aguda (p = 0,0220) y ayuno (p <0,0001). En pacientes con puntaje Apache II ≥15: trauma grave (p = 0,0228) y ayuno (p = 0,0000). Incidencia de neumonía intrahospitalaria 10,05%; con profilaxis y sin profilaxis: 11,47% y 8,68% (p = 0,001)(AU)
Objectives. Set the importance of prophylaxis for gastrointestinal bleeding in criticallly ill patients; determine predictors of gastrointestinal bleeding and quantify hospital-acquired pneumonia. Materials and methods. Prospective, observational and longitudinal design. A total of 3438 patients were studied during two years. After applying exclusion and elimination criteria, 3213 patients were left. It was a consecutive sample, and there were two groups; with prophylaxis (sucralfate or ranitidine) and without prophylaxis. Apache II, organ failures, risk diagnoses, digestive tract bleeding, hospital-acquired pneumonia and mortality were registered. Results. Gastrointestinal bleeding incidence: 4.57%, with vs. without prophylaxis: 4.64% vs. 4.50% (p = 0.91). Gastrointestinal bleeding predictors in the multivariate analysis: serious trauma: (p = 0.0004), respiratory failure (p = 0.0021), pancreatitis (p = 0.0202), acute renal failure (p = 0.0220) and fasting (p <0.0001). In Apache II ≥15 patients: serious trauma (p = 0.0228) and fasting (p = 0.0000). Incidence of hospital-acquired pneumonia: 10.05%, with vs. without prophylaxis: 11.47% vs. 8.68% (p = 0.001). Conclusions. Prophylaxis did not reduce gastrointestinal bleeding, it was associated with serious trauma, respiratory failure, pancreatitis, acute renal failure and fasting, and considering only patients with the worst Apache II score, it was associated with serious trauma and fasting. Acquired pneumonia was higher in the prophylaxis group.(AU)