CT pulmonary arteriography (CTPA) is evolving as the predominant noninvasive modality for the diagnosis of pulmonary embolism. CTPA is more often definitive than pulmonary scintigraphy, and CTPA may also disclose causes of hypoxemia other than pulmonary embolism. However, as there has been increased usage of CTPA, there has not been a concomitant increase in educative efforts in its interpretation. Examine this normal CTPA below.

Recently, the Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED) II study, a multicenter, prospective trial, evaluated the utility of CT in the evaluation of suspected PE. Included in this study were CTPA and pelvic-lower extremity CT venography (CTV). Using a composite reference standard and 4-, 8-, and 16-MDCT, PIOPED II demonstrated sensitivity and specificity of 83% and 96% (PPV 86%, NPV 95%) for CTPA alone and 90% and 95% (PPV 85%, NPV 97%) for CTPA-CTV. The accuracy of these tests were improved when used in conjunction with clinical evaluation, with a positive predictive value of 96% and 92% in patients with high and intermediate clinical probability, and a negative predictive value of 97% in patients with low clinical probability. Other recent large-scale outcome studies have also demonstrated high negative predictive values for CTPA, with a very low rate of subsequent thromboembolism in untreated patients following negative CTPA. In PIOPED II, discordant finding, however, showed much lower accuracy, with a 42% false positive rate in low clinical probability cases and an 18-40% false negative rate in high clinical probability cases.