Introduction: The severity of an acute pulmonary embolism (PE) can be assessed using hemodynamics, right ventricular (RV) strain on imaging, and cardiac biomarkers. The spatial ventricular gradient (SVG) is a vectorcardiographic measurement acquired from a 12-lead ECG that reflects cardiac loading conditions via electromechanical coupling and could aid in risk stratification in acute PE. Hypothesis: SVG is correlated with RV strain, and predicts adverse events in patients with acute PE. Methods: Retrospective cohort study of patients with acute PE. ECG signs of RV strain (S1Q3T3, precordial T-wave inversions, RBBB); imaging (CT and TTE RV/LV ratios); and outcome data (30-day mortality, need for vasopressor or advanced therapy) were obtained. The SVG vector was regressed on degree of RV dysfunction. Odds of adverse outcome after PE were regressed on the SVG and covariates (imaging data, age, gender, ECG signs) using a logit model. Results: ECGs from 318 patients (48% male, age 62.2 ± 16.4 y) with acute PE were analyzed; 44 adverse events (13. 8%) were reported. Worse RV hypokinesis and decreasing tricuspid annular plane systolic excursion (TAPSE) were both associated with less positive X and Y, and a more positive SVG Z vector components (p<0.001 for all). The Figure shows receiver operating characteristic (ROC) curves of models using conventional ECG, imaging, and SVG data. Compared to conventional ECG markers and imaging, the addition of SVG to imaging improved the area under the ROC curve from 0.62 vs 0.72 (p=0.013). Conclusion: The SVG vector is correlated with right heart dysfunction in acute PE. The SVG vector has a better prognostic value than traditional ECG markers, and may improve risk stratification in acute PE.