Objective

Large pulmonary embolism (PE) may carry a mortality of 58%, with a 10% fatality rate in the first hour. Current radiological classifications do not identify patients most at risk of death. This lack of predictive capacity prevents effective utilisation of available treatment options such as thrombolysis ans surgery. This study has been undertaken to develop a new radiological scoring system for large PE, with a predictive capacity, related to clinical outcome which may then guide management.

Methods

The study was conducted using a radiology database to identify all patients who had a computer tomography pulmonary angiogram (CTPA) during 2014. 1923 scans were identified, of which 210 were reported to be large PE. Those with a segmental PE or poor quality imaging were excluded, leaving 154 patients, (91 females) in the study. The UHNM PE score assesses the degree of vascular occlusion and subsequent structural effects on the heart. The maximum score available is 40, and the minimum is 1.

Results

Average age was 69 years (range 20-98 years). 13 deaths occurred, 4 from large PE. Those who died of a PE had a median score of 24 (IQR 8.5-33.5) compared with survivors’ score of 6 (IQR 3-13.5), p= 0.0393. A major adverse event (MAE) included thrombolysis, surgery and death and the median score was 21 (IQR 15-30) compared with 5 (IQR 3-13) for those who did not have a MAE (p= <0.0001). Analysis also demonstrated a strong relationship between the score and clinical signs indicative of shock.

Conclusion

The results from this study demonstrate that the UHNM PE score is able to identify those patients with a large PE which are life-threatening. We recommend that the score be incorporated into a clinical algorithm which combines the score and clinical assessment. We suggest further evaluation through use of a prospective cohort study and randomised control trial.