Spiral computed tomography (CT) has shown promising results in the detection of acute pulmonary embolism. The aim of this study was to investigate whether the severity of acute pulmonary embolism could be quantitatively assessed with spiral CT examinations and to test the potential clinical impact of this information. In a consecutive series of 123 patients screened with spiral CT for suspected acute pulmonary embolism, 31 patients (25%) had evidence of emboli. The severity of pulmonary arterial obstruction in those 31 spiral CT examinations was evaluated by two independent observers using angiographic scores previously described by Walsh (29) and Miller (30), adapted to the needs of spiral CT. Clinical patient subgroups were defined according to oxygen saturation, heart rate, and echocardiographic signs of right ventricular strain. CT severity scores were then correlated to each other and to clinical parameters using the Spearman rank test. Interobserver agreement was calculated using the analysis of variance. Both modified Walsh and Miller scores were readily reproducible and showed interobserver agreements of 0.85 and 0.96, respectively (p = 0.001). Patients with mild and marked clinical abnormalities showed statistically significant differences between CT severity scores. Differences between severity scores of patients with moderate and marked clinical abnormalities were somewhat significant. No significant mean severity score differences were seen between patients with mild and moderate clinical abnormalities. Although correlations of severity scores and detailed clinical parameters within the defined subgroups were moderate to poor, threshold scores greater than 10 (Miller) and greater than 11 (Walsh) always indicated marked clinical abnormalities. The modified scores presented in this study constitute a readily reproducible method for the quantitative assessment of acute pulmonary embolism severity on spiral CT examinations.