Introduction and objectives A number of clinical prediction rules have been described to predict adverse outcomes in patients admitted to hospital with an acute exacerbation of COPD (AECOPD).¹ None are used routinely, perhaps because of limitations including setting (confined to intensive care), use of subjectively defined or difficult to access clinical measurements and lack of external validation. None have undergone impact assessment.

The National Early Warning Score (NEWS) in unselected medical admissions accurately predicts risk of in-patient mortality. The NEWS is less discriminating in patients with COPD. We hypothesised that patients admitted with an AECOPD could be more accurately risk stratified based on a combination of the NEWS and other parameters.

Methods This was a twin site observational cohort study, over a two-year period (March 2012 – February 2014). 2361 admissions with COPD were identified (J40–44).

Results 123 died during admission (5.2%) and a further 36 (1.5%) were escalated to Intensive Care (ICU) and survived to discharge. We analysed these 159 patients against a control group (n = 159) matched only for month of admission (to address seasonal fluctuations in disease severity).

Major results of the study are summarised in Table 1. Those who died or had care escalated were older, had a higher NEWS and respiratory rate. Neutrophils, lymphocyte count, neutrophil-lymphocyte ratio, urea, albumin and CRP were significantly different between the two groups studied. On multivariable analysis lymphocyte count, urea, NEWS and age were independent predictors of adverse outcome.

• Download figure
• Open in new tab
• Download powerpoint

Abstract S31 Table 1

Admission parameters for patients with AECOPD - comparison between those who died or had care escalated versus those who remained on the ward and survived to discharge. Results given as Mean (Standard deviation). NLR = neutrophil-lymphocyte ratio, NEWS = National Early Warning Score

Conclusions Age, admission NEWS and blood parameters differed significantly between those who were managed on the ward with AECOPD and those who either died or whose care was escalated to ICU. This could form the basis for a prediction score, automatically calculable on admission to hospital using available technology to highlight those patients judged at greatest risk of deterioration.

Reference

1. Steer J, Gibson GJ, Bourke SC. Predicting outcomes following hospitalization for acute exacerbations of COPD. QJM 2010;103(11):817-29