Dados do Trabalho

Título

Decoding Prolonged ICU Stays: Key Insights from a Comprehensive Patient Analysis

Objetivo

To identify patients with prolonged ICU stays and the factors associated with extended hospitalization.

Métodos

A cohort study involving patients admitted to a private ICU in João Pessoa, Brazil, focusing on those with prolonged stays (above the 75th percentile of ICU stay duration).

Resultados

Among 2337 ICU patients, the median stay was 3 (1-5) days. Of these, 695 (29.7%) had stays longer than 5 days. These patients exhibited higher frailty scores (MFI 0.27 [0.09-0.36] vs. 0.09 [0-0.27], p<0.001), Charlson Comorbidity Index (CCI 2 [1-5] vs. 1 [0-2], p<0.001), SAPS3 (58 [48-68] vs. 45 [36-54], p<0.001), and EPM (15.84 [5.81-36.52] vs. 3.57 [1.38-11.22], p<0.001). In multivariate regression, SAPS3 had an odds ratio of 1.054 (95% CI 1.041-1.066, p<0.001), and EPM had an odds ratio of 1.037 (95% CI 1.032-1.042, p<0.001). Other factors associated with prolonged ICU stay included the use of vasoactive drugs at admission (OR 1.526, 95% CI 1.049-2.221), invasive ventilatory support (OR 1.693, 95% CI 1.13-2.537), and septic shock at admission (OR 1.255, 95% CI 1.046-1.506). The ROC curve area for SAPS3 in predicting prolonged stay was 0.737 (95% CI 0.715-0.758, p<0.001), and for EPM, it was 0.749 (95% CI 0.728-0.770, p<0.001).

Conclusão

Patients with prolonged ICU stays exhibited greater frailty, more comorbidities, and higher severity scores. Key factors at admission associated with prolonged stays included the use of vasoactive drugs, invasive ventilation, and septic shock. SAPS3 and EPM demonstrated acceptable discriminative power, with each point increase in SAPS3 raising the risk of prolonged stay by 5%, and EPM by 3.7%