Setting: Reported predictors of the adverse evolution of patients with chronic obstructive pulmonary disease exacerbations (eCOPD) are various and inconsistent in the bibliography.
Objective: To develop clinical prediction rules for short-term outcomes in eCOPD patients attending an emergency department (ED).
Design: Prospective cohort study of patients with an eCOPD. Short-term outcomes were admission to an intensive care unit (ICU), admission to an intermediate respiratory care unit (IRCU) and death in these groups. Multivariate logistic regression models were developed for each of the outcomes.
Results: Predictors of ICU or IRCU admission were use of long-term home oxygen therapy (LT-HOT) or non-invasive mechanical ventilation (NIMV), elevated PCO2 and decreased pH upon ED arrival (area under the curve [AUC] 0.87 in the derivation sample; 0.89 in the validation sample). Among those admitted to an ICU or IRCU, predictors of death were increased age, use at home of LT-HOT or NIMV, use of inspiratory accessory muscles upon ED arrival and altered Glasgow Coma Scale (<15 points) (AUC 0.78).
Conclusions: Three clinical predictors available in the ED can be used to create a simple score to predict the need for intensive treatment among eCOPD patients. Such a score can be a tool for clinical practice.