Abstract
Background: Hospital admissions for COPD exacerbations account for 70% of total costs of COPD treatment whereas the length of hospital stay (LOS) is directly related to this cost.
Objectives: The aim of the present study was to investigate possible associations of demographic, clinical, laboratory and functional parameters with LOS of patients admitted for an acute exacerbation of COPD (AECOPD) and to provide a score for the prediction of prolonged LOS
Methods: We included 164 consecutive patients admitted to two tertiary hospitals for AECOPD and we evaluated parameters possibly related to the duration of hospital stay.
Results: Seven parameters evaluated on patient' admission (Antonisen type of exacerbation, Number of Exacerbations in the previous year, Charlson index of comorbidities, Oxygenation, Partial Pressure of PaCO2 in arterial blood gases, Dyspnea according to the Borg dyspnea scale and history of Chronic Respiratory Failure, were able to predict LOS and were included in a simple score named AECOPD-F Score. The AUC of the score for the prediction of prolonged hospital stay is 0.960, and a cut-off point ≥3 predicts prolonged length of stay with a sensitivity 84.5% and specificity 92.5%, 95% CI 0.917-0.984. The AECOPD-F score was validated in a second group of 88 patients admitted for AECOPD. In the validation group patients with AECOPD-F score ≥3 required prolonged LOS compared to those with a score<3, [8.0 (6.0, 10.0) vs 6.5 (4.0, 9.0), p=0.007], respectively.
Conclusion: The AECOPD-F Score could accurately predict LOS in hospitalized COPD patients. The implementation of this score in clinical practice could be useful in the discharge planning of such patients.
- © 2014 ERS