Use of CT Angiography in a Country with Low Pulmonary Embolism Prevalence: Correlation with Clinical Pretest Probability and D-dimer Values

Abstract

Departments of Radiology and Cardiovascular and Thoracic Surgery, Pusan National University Hospital, Pusan National University School of Medicine and Medical Research Institute Department of Radiology, Pusan National University Yangsan Hospital, Pusan National University School of Medicine and Medical Research Institute Department of Radiology, Haeundae Paik Hospital, Inje University School of Medicine All authors have no financial or other potential conflicts of interest to disclose. Received November 16, 2010 ; Accepted March 2, 2011 Address reprint requests to : Yeon Joo Jeong, M.D., Department of Radiology, Pusan National University Hospital, 1-10, Ami-dong, Seo-gu, Pusan 602-739, Korea. Tel. 82-51-240-7354 Fax. 82-51-244-7534 E-mail: jeongyj@pusan.ac.kr Purpose: To assess the use of CT angiography (CTA) in the diagnostic evaluation of pulmonary thromboembolism (PE) in a country with low PE prevalence and correlate the diagnostic performance of CTA with the clinical pretest probability and D-dimer values. Materials and Methods: The institutional review board approved this retrospective study. The observers reviewed all 660 CTAs and calculated the PE clot burden scores. The pretest probability of PE according to the Wells criteria and D-dimer values were calculated (clinical data were available for 371 of the 660 patients). We correlated the PE positivity rates of CTA and a PE clot burden score with the D-dimer values and pretest probability using Pearson’s correlation coefficient. Results: Of the 371 patients whose clinical data were available, 122 (32.8%) had PEs. None of the patients with both a normal D-dimer value and a low clinical probability had a PE. PE positivity rates of CTA were correlated with clinical pretest probability (r = 0.164, p = 0.002) and D-dimer values (r = 0.361, p < 0.001). PE clot burden scores were correlated with D-dimer values (r = 0.296, p < 0.001). Conclusion: Although PE positivity rates of CTA in a country with low prevalence were higher than those in a country with a higher prevalence, approximately 30% of the yield still represents an overuse of CTA. CTA should be performed after the pretest probability has been assigned and if the result of a D-dimer assay is abnormal.