Computed tomography angiography in patients with suspected pulmonary embolism-too often considered?

Background: Pulmonary embolism (PE) is a major cause of morbidity and mortality associated with surgery and medical illnesses. In recent years, pulmonary computed tomography angiography (CTA) has become the diagnostic method of choice. However, it remains unclear when to perform CTA and how often a decision based on clinical judgment results in positive or negative findings. Methods: In a retrospective study, 261 patients admitted for suspected PE were evaluated with pulmonary CTA. Decisions to order CTA were based on clinical judgment and optionally quantitative d-dimer assays. Clinical, radiologic, and laboratory data were revisited and compared in patients with and without proven PE. Results: The patients' mean age was 63 ± 1 years; almost 30% of all participants had at least a moderately reduced renal function. Pulmonary CTA demonstrated PE in only 14.9%; both age and sex distribution was comparable in the PE and non-PE group. Proximal deep vein thrombosis or pathologic chest x-rays were significantly more likely in patients with PE (P <.001 and P <.05), whereas echocardiography results were comparable. d-dimer values were noticeably higher in the PE group (P <.001); however, C-reactive protein and troponin T levels were not helpful. Conclusions: Pulmonary CTA confirmed PE in only a minority of patients and may be overused. Clinical judgment in conjunction with d-dimer evaluation was of limited help to predict positive results but surprisingly comparable with previous results using pretest probability scoring systems. Using present and previous data, a simplified enhanced algorithm is proposed to reduce use of CTA.