Enhancement Characteristics of the Computed Tomography Pulmonary Angiography Test Bolus Curve and Its Use in Predicting Right Ventricular Dysfunction and Mortality in Patients with Acute Pulmonary Embolism

PURPOSE: The purpose of the study was to evaluate the relationship between computed tomography pulmonary angiography (CTPA) test bolus curve data and mortality in patients with pulmonary embolism (PE) in comparison with conventional methods of right ventricular (RV) dysfunction. MATERIALS AND METHODS: The study was approved by our institutional review board and is HIPAA-compliant. We retrospectively evaluated consecutive CTPA studies performed with a test bolus technique in a 2-year period. A time-density curve was derived from each test bolus. For comparison, left ventricular (LV) and RV dimensions (area, diameter) and PE load score (Qanadli method) were measured using CT data. A cardiologist blinded to the clinical and other imaging data reviewed a subset of the corresponding echocardiographic images to assess for RV dysfunction. Demographic data, mode of treatment, and patient outcome information were gathered using electronic medical records. Test bolus and anatomic data were correlated with PE-related mortality. RESULTS: A total of 71 patients (34 men and 37 women, average age 54.4 y) who had a CTPA performed using a test bolus technique were diagnosed with acute PE. Factors that significantly correlated with PE-related mortality on univariate analysis were: age above 60 years (odds ratio 19.1, P = 0.05), RV/LV diameter >1.5 (odds ratio 48.8, P 1 (odds ratio 8.6, P = 0.06), bolus curve upslope time >6 seconds (odds ratio 23.3, P = 0.04), 50% downslope time >6 seconds (odds ratio 20, P = 0.01), and embolus load score >15 (odds ratio 25, P = 0.03). The predictive value of upslope time (Exp(B) 1.65, P = 0.05), RV/LV diameter (Exp(B) 43.8, P = 0.01), and RV/LV area (Exp(B) 16.7, P = 0.01) were confirmed to be statistically significant in multivariate analyses. CONCLUSIONS: Data from the CTPA timing bolus curve provide prognostic value similar to the best conventional methods for predicting PE-related mortality.