Decline in functional status after intensive care unit discharge is associated with ICU readmission: a prospective cohort study

Abstract

ObjectivesTo compare the functional status at intensive care unit (ICU) discharge of patients who were later readmitted to the ICU and patients discharged home and to verify whether a decline in functional status is associated with ICU readmission. DesignProspective cohort study. SettingICU at a tertiary teaching hospital. ParticipantsPatients admitted to the ICU, ≥18 years old, submitted to invasive mechanical ventilation (IMV), and discharged to the ward. InterventionsFunctional assessment at ICU discharge. Discharge Group (DG) (patients discharged home) and Readmission Group (RG) (patients who returned to the ICU) were compared with Mann–Whitney and Chi-square or Exact Fisher tests. Multiple logistic regression verified association. Main outcome measuresBarthel Index, key pinch strength, clinical and demographic data. ResultsPatients in the readmission group presented lower Barthel Index [Median 40 (IQR 20–75) vs 60 (33–83), P=0.033], greater relative variation (pre and post ICU) of the Barthel Index (P=0.04), lower key pinch strength [3.4 (1.8–4.5) vs 4.5 (2.7–6.8)kg·f, P=0.006] and higher APACHE II [18 (12–22) vs 15 (11–20), P=0.027]. Multiple regression found that the relative variation of the Barthel Index was independently associated with ICU readmission (P<0.001), as well as higher APACHE II (P=0.020), shorter IMV duration (P<0.001) and ICU admission without clear diagnosis (P=0.020). The Hosmer–Lemeshow test indicated good adjustment of the model (P=0.99). ConclusionReadmitted patients presented poorer functional status and lower pinch strength. Relative variation of the Barthel Index was associated with ICU readmission despite other factors, as was higher APACHE II, shorter IMV duration and admission without clear diagnosis.Trial registration number: Not applicable.