S60 Utility of a novel radiological score in predicting clinical outcomes in large pulmonary embolism. a comparison with simplified PESI score

Introduction Pulmonary Embolism (PE) is associated with a notable risk of morbidity and mortality. CT pulmonary angiograms (CTPA) are used in the diagnosis of pulmonary embolism and also to determine the clot burden and right ventricular strain. University Hospital of North Midlands (UHNM) PE score is a radiological score created to optimize risk analysis (haemodynamic instability) of patients with large PE using radiological features. The score assesses location of clot (L) in major pulmonary arteries (scores 1, 2 and 4), the degree of occlusion (o) (score 1,2 and 3) and the impact on the right ventricle (RVR) (scores 1,2 and 3). Interventricular septum morphology (S) is also assessed (scores 1, 2, 3 and 4). Multipliers are used to obtain the total score, [(L × O) × RVR] + S. Maximum score is 64. This score is correlated with clinical outcome. The PE severity is classified into mild (1–9), moderate (10–14) and severe ≥15. (Figure 1) Methods A retrospective analysis of CTPAs (From 1/3/2014 till 1/07/2018) was performed. A total of 485 patients, whose CTPA showed Large PE/Saddle embolism, were included in study. All CTPAs showing segmental/sub-segmental PEs and CTPAs with poor quality imaging were excluded. S-PESI was calculated retrospectively using clinical notes. Results Out of 485 patients, 99 patients had S-PESI score of 0 (20.4%), these patients had average UHNM PE score of 9 and their average inpatient hospital stay was 5 days. 5 patients out of 99 ended up in having thrombolysis, 7 had inpatient cardiothoracic referral for surgical intervention. 3 patients out of 99 died in 30-day period. Conclusion UHNM PE score has comparable performance in identifying the risk of haemodynamic compromise in large PE. It improves the specificity by risk stratifying large PEs into 3 groups, hence can be utilized in clinical decision making regarding appropriate treatment and predicting clinical outcomes. Further studies are needed to validate its utility before its universal application.