The effect of helical computed tomography on diagnostic and treatment strategies in patients with suspected pulmonary embolism

Background Helical computed tomography (CT) has been proposed as a first-line test for the diagnosis of pulmonary embolism. How the test affects the diagnostic evaluation of patients with suspected pulmonary embolism is unknown. Methods We examined a cohort of 360 patients evaluated for pulmonary embolism at a teaching hospital in the 4 years following the introduction of the helical CT scan. We collected patient demographic and clinical data to calculate the pretest likelihood of pulmonary embolism; we then read the test results and determined rates of further testing and treatment for pulmonary embolism. Results After the helical CT scan became available, the number of patients referred for pulmonary embolism testing increased markedly from 170 to 624 total evaluations during 1997 to 2000 (P<0.01). This rise was due to increased use of the helical CT scan (9% to 83% of evaluations, P<0.01) as the use of ventilation-perfusion scanning (79% to 17%, P = 0.03) and pulmonary angiography (12% to <1%, P <0.01) fell. There was no change in the pre-test likelihood of disease over time, but the percentage of scans that were positive for pulmonary embolism rose (14% to 32%, P =0.02). Clinicians treated all patients who had a positive CT scan, but became less likely over time to order further testing for patients who had a negative scan (30% to 12%, P = 0.02). Conclusion At this academic medical center, introduction of the helical CT scan had a profound effect on the evaluation of pulmonary embolism, resulting in more frequent use of the CT scan, and more frequent diagnosis and treatment of pulmonary embolism, despite no change in the pretest probability of disease. Future studies should confirm our findings and determine whether increased detection of pulmonary emboli results in improved outcomes.