PREDICTING OUTCOMES FROM ACUTE EXACERBATIONS OF INTERSTITIAL LUNG DISEASE: A MULTICENTRE OBSERVATIONAL STUDY

Abstract

Introduction Acute exacerbations of interstitial lung disease (AE-ILD) often lead to hospitalisation but have limited evidence-based treatment options. Patient-specific outcomes from AE-ILD are difficult to predict, making decisions regarding prioritisation for specialist palliative care input challenging. The PCR (PaO­2/FiO2 ratio, C-reactive protein (CRP), and CT chest pattern) score has been shown to predict mortality risk in AE-idiopathic pulmonary fibrosis (AE-IPF),1 however this has not been assessed in AEs of non-IPF ILDs. Aims Describe a real-world patient population with AE-ILD and establish the relationship between PCR score and mortality. Methods Clinical records of ILD patients admitted to six NHS trusts over one year were reviewed. Patients with a deterioration in respiratory symptoms not explained by heart failure or pulmonary embolism were included. Demographic, treatment, investigation, and mortality data were collected. Participating hospitals provided information on local services. The PCR score, where one point is gained for each of CRP >55 mg/l, PaO2/FiO2 ratio <250, and diffuse ground glass changes on CT chest, was calculated. Results 443 patients with 602 admissions were included. IPF was the commonest ILD (29.8%), and other ILD diagnoses were represented (table 1). Antibiotics and steroids were prescribed in 82.9% and 66.9% of admissions, respectively. Just one participating hospital had a protocol for the management of AE-ILD, which was specific to AE-IPF. Mortality after AE-ILD was high, with 14% in-hospital, 21.7% 30-day, 39.8% 6-month, and 53.9% 12-month mortality. Higher PCR scores were associated with increased in-hospital (p=0.03), 30-day (p=0.01), and 6-month mortality (p=0.003), with a trend towards increased 12-month mortality (p=0.07) (88 index admissions, p for trend). No specialist palliative care input was recorded in 29.3% of admissions with in-hospital mortality, and 49.2% of admissions where death occurred within 30 days. This could be explained by hetereogeneity in specialist palliative care availability, particularly out-of-hours (table 1). VIEW INLINE VIEW POPUP Abstract S7 Table 1 Patient and participating hospital characteristics Conclusions AE-ILDs are associated with significant mortality, limited standardised treatment and heterogeneous palliative care provision. Higher PCR scores were associated with increased mortality in AE-ILD and may have utility when prioritising patients for palliative care input and advanced care planning.