Premium
Detection of deep venous thrombosis: Prospective comparison of MR imaging and sonography
Author(s) -
Evans Avery J.,
Sostman H. Dirk,
Witty Lynn A.,
Paulson Erik K.,
Spritzer Charles E.,
Hertzberg Barbara S.,
Carroll Barbara A.,
Tapson Victor F.,
Saltzman Herbert A.,
Delong David M.
Publication year - 1996
Publication title -
journal of magnetic resonance imaging
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.563
H-Index - 160
eISSN - 1522-2586
pISSN - 1053-1807
DOI - 10.1002/jmri.1880060109
Subject(s) - medicine , pelvis , radiology , venous thrombosis , thrombus , confidence interval , thrombosis , prospective cohort study , nuclear medicine , surgery
Seventy‐five patients (41 women and 34 men, 20‐85 years old) with clinically suspected deep venous thrombosis (DVT) were examined with MR imaging and sonography. In 26 patients, the final diagnosis was acute femoropopliteal DVT. The sensitivity of MR imaging for detecting this disease was 100% with a 95% confidence interval (CI) of 87‐100%; the specificity was 100% with a CI of 92‐100%; and the accuracy was 96% with a CI of 89‐99%. The correspond‐ ing sensitivity of sonography was 77% with a CI of 53‐92%; the specificity was 98% with a CI of 89‐100%; and the accuracy was 83% with a CI of 72‐90%. In four of the 75 patients, MR images revealed thrombus of the pelvis ( n =1) or calf ( n =3) without femoropopliteal involvement. The estimated prevalence of isolated calf and/or pelvic DVT was 5% with a CI of 1‐13%. MR imaging is significantly more sensitive ( P =.02) and accurate ( P < .01) than sonography in the detection of lower extremity DVT, but there was no difference in the specificity of MR imaging and that of sonography ( P =.31).