Detection of deep vein thrombosis with Doppler sonography

Abstract

Sir, I read with interest the recent paper by Ozbudak et al. entitled ‘‘Doppler ultrasonography versus venography in the detection of deep vein thrombosis with pulmonary embolism’’ in the October 2006 issue of Journal of Thrombosis and Thrombolysis [1]. The authors stated that ‘‘The sensitivity and specificity of Doppler ultrasonography in the diagnosis of deep vein thrombosis (DVT) were 76% and 100%, respectively and the negative and positive predictive values were 81% and 100%, respectively. Although Doppler ultrasonography alone is considered sufficient for the diagnosis of DVT, venography still remains the gold standard in the diagnosis of DVT’’ [1]. I’m surprised this results and statement. Although contrast venography (CV) has been considered goldstandard technique for the diagnosis of DVT it is underused currently due to different technical problems (such as to cannulate an appropriate pedal vein is not possible in every patient; difficulty correctly interpretation of venograms; incomplete venous filling and inadequate studies) or its side effects [2]. Post-procedure DVT and skin necrosis may be observed in about 8% of examinations in spite of using non-ionic contrast media. About 5% of CV examinations are inadequate and non-diagnostic [2, 3]. Currently, ultrasound (US) using venous compression is considered the gold standard for confirmation of DVT and compression US has effectively replaced CV for the assessment of patients with suspected DVT [4, 5]. US examination is highly sensitive and specific for the diagnosis of DVT: several series have reported sensitivities and specificities approaching 95–100% for the diagnosis of DVT in the proximal lower extremity [6]. In a large, pooled series of DVT studies, the accuracy of compression US for DVT has been shown to reach 95% with 98% specificity [4]. Additional studies report a sensitivity of 95%, specificity of 99%, and accuracy of 98% for color Doppler flow imaging [7, 8]. Color flow Doppler alone without compression has shown excellent correlation with venography for evaluation of DVT with 95% sensitivity and 99% specificity; 95% of positive predictive value and 99% of negative predictive value [9]. In a recent Meta analysis related with diagnostic accuracy of sonography for DVT shown that while compression US only has 93.8% pooled sensitivity and 97.8% specificity; color Doppler only has 95.8% pooled sensitivity and 92.7% specificity for proximal DVT [10]. We also recently reviewed retrospectively 428 patients with 467 DVT examinations and 120 of 467 examinations were abnormal. About 49 of 120 patients had complete thrombus, 71 of 120 patients had partial thrombus. We observed the thrombus on gray scale ultrasound in all 120 positive patients. Color flow Doppler examinations of these patients were also abnormal [11]. In this published study, it is unclear that which diagnostic equipment and which diagnostic criteria were used for diagnosis of DVT. Who was performed DVT examination? (i.e., vascular surgeon, sonographer, or radiologist) Examination level (such as proximal or calf veins) and experience of examiner is also very important for reliable diagnosis. The American College of Radiology (ACR) Practice Guidelines for lower extremity duplex Doppler sonography includes compression, color, and spectral Doppler E. Kocakoc (&) Department of Radiology, Faculty of Medicine, Firat University, Elazig 23119, Turkey e-mail: ercankocakoc@yahoo.com