BACKGROUND AND OBJECTIVES: Currently, the reformulation of intensive care goals, often shifting from the search for a cure to offering comfort, has become more and more necessary. The intensivist is frequently confronted with the decision to suspend or not offer a specific therapy, despite its availability. The objective of this study was to estimate the developing risk of probability of death for individual ICU patients with respiratory failure, identify which life-sustaining therapies were administered, time of internment and outcome. Compare the death outcome in relation to UNICAMP II and APACHE II models, as well as verify if the life-sustaining therapies may be limited or suspended. METHODS: It is the observational, prospective cohort study of 150 patients with respiratory failure confined to the intensive care unit. Statistical analysis was carried out using Generalized Linear Models. RESULTS: Age, sex, race or morbidity did not reveal statistical significance in predicting outcome. This prediction was confirmed more accurately by means of changes in the individual prognostic index of death probability during the first seven days of ICU internment. A 10% worsening prognosis in patients who presented initial death risk of 70% to 80%, utilizing the UNICAMP II Model, showed a specificity of 97.4% - 98.6%. CONCLUSIONS: Prognostic changes in patients during the first seven days of ICU internment are of great aid, from an objective point of view, for ethical decision-making in relation to not-offering new life-sustaining therapies.
ICU; Medical Futility; Respiratory Failure
