Background: Computed tomography pulmonary angiography (CTPA) plays an increasingly important role in the diagnosis of pulmonary embolism (PE). Although accurate in the detection of large PE, its accuracy in other patient groups is yet to be defined.
Aim: To compare CTPA with pulmonary angiography as a second-line investigation in patients with a ventilation/perfusion (VQ) scan indicating an intermediate probability of PE.
Methods: We recruited 25 patients over a 17-month period. Subjects were eligible if they: (i) had clinically suspected PE, (ii) had a VQ scan indicating an intermediate probability of PE and (iii) were referred for pulmonary angiography. Subjects underwent CTPA within 36 h of the VQ scan. CTPA was interpreted without knowledge of the results of the pulmonary angiogram by two of the authors.
Results: PE was prevalent (i.e. embolus detected at pulmonary angiography) in seven of 25 subjects (28%). The sensitivity of CTPA was 57% and the specificity was 94%.
Conclusions: In the setting of intermediate-probability VQ scanning, CTPA may be used to clarify the diagnosis of PE. However, a negative CTPA cannot definitely exclude PE. Conventional pulmonary angiography may be necessary to determine the presence of PE if CTPA is negative. (Intern Med J 2003; 33: 74−78)