Evolution and prognosis of long intensive care unit stay patients suffering a deterioration: A multicenter study.

Autor: Hernández-Tejedor A; Unidad de Cuidados Críticos, Hospital Universitario Fundación Alcorcón, 28922 Alcorcón, Madrid, Spain. Electronic address: albertohmed@hotmail.com., Cabré-Pericas L; Unidad de Cuidados Intensivos, Hospital de Barcelona SCIAS, 08034 Barcelona, Spain. Electronic address: 10654lcp@comb.cat., Martín-Delgado MC; Unidad de Cuidados Intensivos, Hospital Universitario de Torrejón, 28850 Torrejón de Ardoz, Madrid, Spain. Electronic address: mmartin@torrejonsalud.com., Leal-Micharet AM; Unidad de Cuidados Críticos, Hospital Universitario Fundación Alcorcón, 28922 Alcorcón, Madrid, Spain. Electronic address: alealmicharet@gmail.com., Algora-Weber A; Unidad de Cuidados Críticos, Hospital Universitario Fundación Alcorcón, 28922 Alcorcón, Madrid, Spain. Electronic address: aalgora@fhalcorcon.es.
Jazyk: angličtina
Zdroj: Journal of critical care [J Crit Care] 2015 Jun; Vol. 30 (3), pp. 654.e1-7. Date of Electronic Publication: 2015 Jan 14.
DOI: 10.1016/j.jcrc.2015.01.011
Abstrakt: Purpose: The prognosis of a patient who deteriorates during a prolonged intensive care unit (ICU) stay is difficult to predict. We analyze the prognostic value of the serialized Sequential Organ Failure Assessment (SOFA) score and other variables in the early days after a complication and to build a new predictive score.
Materials and Methods: EPIPUSE (Evolución y pronóstico de los pacientes con ingreso prolongado en UCI que sufren un empeoramiento, Evolution and prognosis of long intensive care unit stay patients suffering a deterioration) study is a prospective, observational study during a 3-month recruitment period in 75 Spanish ICUs. We focused on patients admitted in the ICU for 7 days or more with complications of adverse events that involve organ dysfunction impairment. Demographics, clinical variables, and serialized SOFA after a supervening clinical deterioration were recorded. Univariate and multivariate analyses were performed, and a predictive model was created with the most discriminating variables.
Results: We included 589 patients who experienced 777 cases of severe complication or adverse event. The entire sample was randomly divided into 2 subsamples, one for development purposes (528 cases) and the other for validation (249 cases). The predictive model maximizing specificity is calculated by minimum SOFA + 2 * cardiovascular risk factors + 2 * history of any oncologic disease or immunosuppressive treatment + 3 * dependence for basic activities of daily living. The area under the receiver operating characteristic curve is 0.82. A 14-point cutoff has a positive predictive value of 100% (92.7%-100%) and negative predictive value of 51% (46.4%-55.5%) for death.
Conclusions: EPIPUSE model can predict mortality with a specificity and positive predictive value of 99% in some groups of patients.
(Copyright © 2015 Elsevier Inc. All rights reserved.)
Databáze: MEDLINE
Externí odkaz: