[In vitro validation of intravascular ultrasound, computerized and magnetic resonance tomography in diagnosis of atherosclerotic vascular segments in comparison with direct magnification radiography].

The depiction of atherosclerotic vessel abnormalities is a prerequisite for percutaneous interventional therapy and long-term observations of peripheral artery disease. The aim of this in-vitro study was to determine the potentials and limitations of 12.5 and 20 MHz intravascular ultrasound, computed tomography and magnetic resonance (MR) imaging in comparison to direct magnification radiography for the localization and quantification of peripheral vessel wall calcifications. Forty-three postmortem, human iliac segments were examined by intravascular ultrasound (12.5 and 20 MHz), computed tomography and magnetic resonance tomography (gradient echo-and spin echo-technique). For comparative analysis, each segment was divided into eight sectors of 45 degrees each; using all five methods, the presence of calcified wall areas was examined in each sector, and luminal area (42 segments) and plaque area (32 isolated plaques) were quantitatively estimated. In the sonograms, the circumferential extension of the boundary between intima and media was measured. 122 of 344 sectors showed regional vessel wall calcifications. Sensitivity of 20 MHz intravascular ultrasound was 73% versus 59% with the 12.5 probe, specificity was 97% with 20 MHz, 96% with 12.5 MHz. Sensitivity of both 12.5 and 20 MHz intravascular ultrasound was higher with increased thickness of the calcified structures. 20 MHz ultrasound identified the intima-media boundary averaging 146.8 degrees of the vessel circumference; the corresponding value of 131.8 degree with 12.5 MHz did not differ significantly. Computed tomography detected calcifications with a sensitivity of 88%, specificity was 88%. With MR imaging, sensitivity of the gradient echo-technique was 94% versus a sensitivity of 86% with spin echo-technique. Quantification of luminal and plaque areas showed that luminal area was precisely estimated only by 20 MHz ultrasound (no significant difference to direct magnification radiography), whereas all other techniques showed significant overestimation. Plaque areas were markedly overestimated by computed tomography and MR imaging, too. In an in vitro set-up, intravascular ultrasound, MR tomography and computed tomography do not allow an authentic depiction of peripheral vessel wall architecture. Limited resolution, subintimal shadowing and and distortion are the main limitations of these new techniques so that details of regional vessel wall calcifications cannot be presented thoroughly. Relevant over-estimation of luminal and plaque areas must be considered.