267 Development and Validation of a Prognostic Tool: Pulmonary Embolism Short-Term Clinical Outcomes Risk Estimation

Abstract

Develop and validate a prognostic prediction model for clinical deterioration or death within days of pulmonary embolism (PE) diagnosis using point-of-care criteria. We used data from the PE Short-term Clinical Outcomes Registry populated by emergency departments in six states. The primary composite outcome was death, cardiac arrest, respiratory failure, dysrhythmia, sustained hypotension treated with fluid boluses or adrenergic agents, or reperfusion intervention within 5 days after PE. Candidate predictors were demographics, vital signs, comorbidities, and immediate right ventricle (RV) assessments (biomarkers and imaging). The prediction rule was developed from 935 PE patients using univariate analyses of 107 candidate variables followed by penalized and standard logistic regression on retained variables and tested on a validation database (N = 801). Logistic regression yielded a nine-variable model, then simplified to a nine-point tool (PE-SCORE): one point each for abnormal RV by echocardiography, abnormal RV by computed tomography, systolic blood pressure < 100 mmHg, dysrhythmia, suspected sepsis, syncope, medico-social reason for admission, abnormal heart rate [< 50 or > 100 bpm], and two points for renal disease. In the development database, 22.4% had the primary outcome. Predictive accuracy of logistic regression versus PE-SCORE models: 0.83 vs. 0.78 using area under receiver operating curve (AUC) and 0.61 vs. 0.50 using precision recall curve (AUCpr). In the validation database, 26.6% had the primary outcome. PE-SCORE had AUC 0.77 and AUCpr 0.63. As points increased, outcome proportions increased: a score of zero had 2% outcome, whereas scores of six and above had > 69.6% outcomes. In the validation dataset, scores of six and above had 100% outcomes. PE-SCORE model identifies PE patients at low and high-risk for clinical deterioration and may guide decisions about early outpatient management versus need for hospital-based monitoring and considerations for escalated PE interventions.