S166 The role of clinical, metabolic and cardiac biomarkers in predicting outcome from COPD exacerbations requiring hospital admission: A prospective observational study

Introduction Baseline hyperglycaemia along with Respiratory Rate has been shown to represent a marker for mortality and failure of Non Invasive Ventilation in COPD exacerbations complicated by decompensated respiratory failure. This study aims to determine whether the same association holds true for those COPD exacerbations admitted to hospital in the absence of respiratory acidosis and if other markers may predict outcomes in such a population. Methodology COPD patients admitted to University Hospital Aintree with an acute exacerbation were recruited within 48 h of admission with the primary end point being 3 month mortality. Patients presenting with respiratory acidosis were excluded. Subjects underwent clinical assessment at recruitment; blood samples were drawn for Random Blood glucose (RBG), Brain Natriuretic Peptide (BNP) level (Siemens Healthcare©). Admission ECGs were analysed in order to calculate Cardiac Infarction Injury Score (CIIS). Results 116 patients (mean age 70 years; 55% female; FEV1 0.98(predicted 2.49) litres; admission ABGS pH 7.40; PCO2 5.93kPA; PO2 9.04kPA) were recruited; 18 (16%) patients had died by 3 months. Hyperglycaemia (defined as RBG ≥7 mmol/l) was not associated with increased 3 month mortality (observed in 6/18 (33%) of deaths at 3 months vs 35/98(36%) of survivors; p=0.85).The Respiratory Rate (RR) measured during clinical assessment appeared to significantly higher in those who had died by 3 months (27 (SD 5) vs 24 (SD 5); p=0.048). No association was observed between those who had died by 3 months in terms of BNP levels (18.71 vs 24.22 pg/ml; p=0.48), CIIS (7.65 vs 6.25; p=0.37), age (71 vs 70 years; p=0.63), PaCO2 (6.29 vs 5.87 kPA) or serum Bicarbonate (25 vs 26 mmol/l; p=0.64). An inverse correlation was noted between BNP values and admission PaCO2 levels (Correlation coefficient −0.25; p=0.018) and Bicarbonate levels (Correlation coefficient −0.35; p=0.001); a positive correlation was observed between BNP levels and patient age (Correlation coefficient −0.35; p=0.002). Discussion In patients admitted to hospital with COPD exacerbations, hyperglycaemia, BNP level and CIIS were not found to be predictors of mortality in the absence of acute respiratory acidosis. Interestingly, Respiratory Rate measured whilst in hospital appears to predict outcome at 3 months.