Predictors for early readmission after COPD exacerbation

Aim: Identify predictors for early readmission in patients with COPD and severe exacerbation. Material and methods: Prospective study (July 2017-June 2018) that included all patients hospitalized for an acute exacerbations of COPD. Demographic date, clinical and spirometric parameters, arterial blood gases, length of stay, and evolutive parameters were collected. We used univariate and multivariate statistical techniques to identify risks for readmission. Results: 278 consecutive patients were enrolled. During the follow-up 31 (11%) patients were admitted within 30 days of discharge (early readmission). Univariate analysis showed that FEV1 (p=0.02), FEV1 (%) (p=0.01), total dose of steroids (p=0.03), close follow-up (p=0.04), mMRC scale (p=0.03) and number of visits to emergency room (p=0.004) increased the risk of readmission for exacerbation. Multivariate analysis showed that use of non invasive ventilation (NIV) (p=0.01) was independently related to risk of readmission. Stepwise logistic regression analysis (table 1) showed that infiltrates at chest radiograph (CXR), non invasive ventilation (NIV) and FEV1 (ml) were associated with the risk of readmission. The area under receiver-operating curve (ROC) of this model was 0.69. Conclusions: Close attention to these predictors may identify hospitalized COPD patients at elevated risk for early readmission.