Interrater reliability of plaque morphology classification in patients with severe carotid artery stenosis

OBJECTIVE: Ultrasonographic assessment of carotid artery plaque morphology is widely used to identify patients at high risk for stroke. However, the reliability of plaque analysis in high-grade stenosis is uncertain. We determined the interrater reliability of sonographic plaque morphology analysis in patients with severe carotid artery stenosis. MATERIALS AND METHODS: Duplex Doppler was performed on 114 patients with 80-99% stenosis of the internal carotid artery using a Siemens Quantum 2000 D with a handheld 7.5 MHz transducer. B-mode pictures with and without color coding were printed on a Sony color video printer UP-5000 W. Three raters independently evaluated plaque echolucency, heterogeneity, calcification, and surface structure. Interrater agreement was calculated by a jackknife procedure generating kappa values and two-sided 95% confidence intervals. RESULTS: Kappa values and 95% confidence intervals were 0.05 (-0.07 to 0.16) for plaque surface structure, 0.15 (0.02 to 0.28) for plaque heterogeneity, 0.18 (0.09 to 0.29) for plaque echogenicity, and 0.29 (0.19 to 0.39) for plaque calcification. The upper bounds of all of the confidence intervals were below the 0.40 level suggested for minimal reliability. CONCLUSION: The low interrater agreement indicated that unaided visual assessment of static B-mode pictures to assess plaque morphology in patients with severe carotid artery stenosis is not reliable. Other evaluation procedures and standardized criteria, as yet undeveloped, are needed to improve reliability.