RESUMO
AIMS OF THE STUDY: Long-term intensive care treatment confers a substantial physical, psychological and social burden on patients, their relatives and the treatment team. It is essential to know the outcome of patients with long-term treatment and to establish factors that possibly can predict mortality. Only few Swiss studies have previously addressed this issue. METHODS: This retrospective observational study at a Swiss tertiary academic medical care centre included patients who were treated for ≥7 consecutive days at the surgical intensive care unit (ICU) between 1 January 2011 and 31 December 2012. Follow-up ended on 30 September 2018. RESULTS: Two-hundred and fifty patients were included, and three were lost to follow-up. Fifty-two patients (21.1%) died in the ICU, 25 (10.1%) after transfer to the normal ward. Thirty-one patients (12.5%) died within one year after the beginning of intensive care treatment. Altogether, the one-year mortality was 43.7% (108 patients). At the end of follow-up, 99 patients (40.1%) were still alive. Polytrauma patients represent a special group with a survival of more than 90%. Median patient age was 66 years (interquartile range 56-75); two thirds were men. Patients who died within one year of beginning treatment in the ICU were significantly older (median 71 vs 63 years, p <0.001), had a higher Charlson comorbidity index (mean 2.3 vs 1.2, p <0.001), a longer intensive care stay (median 13.9 vs 10.6 days, p = 0.001), a higher SAPS-II score (mean 52.7 vs 45.6, p = 0.001), a higher NEMS score (mean 1772.4 vs 1230.4, p <0.001) and more complications (mean 2.9 vs 2.0, p <0.001) than patients who survived at least 1 year. Those who died within 1 year more often developed pneumonia (50.9% vs 29.5%, p = 0.001), pleural empyema (13.0% vs 2.9%, p = 0.005), septic shock (51.9% vs 20.1%, p <0.001) or critical illness polyneuropathy (16.7% vs 2.9%, p <0.001). Moreover, they more frequently (30.6% vs 15.1%, p = 0.006) required a renal replacement therapy. CONCLUSIONS: Long-term mortality of patients with prolonged intensive-care treatment is high. Scores combined with factors shown to be associated with an increased short- and long-term mortality can help to identify patients at risk for death within one year after ICU treatment.