Aim: Stroke is the third leading cause of mortality in the world and the stenosis degree is considered a fundamental parameter for the definition of the therapeutic approach. With the development of Multi-Detector-Row CT (MDCTA) scanner, computed tomography has become a widely used imaging technique for categorizing carotid artery stenosis degree. The purpose of this paper is to compare two CT post-processing procedures, maximum intensity projection (MIP) and multi planar reconstruction (MPR) in order to evaluate their sensitivity and inter-technique agreement.
Methods: For the purpose 45 patients (35 males and 10 females), that underwent MDCTA for carotid artery evaluation, have been retrospectively evaluated. Data set were processed with the study group's workstations, by using MPR and MIP algorithms. Each patient was assessed for stenosis degree by using North American Symptomatic Carotid Endarterectomy Trial method. Statistic analysis was performed to determine the sensitivity of the used procedure. The Cohen kappa test was applied to assess the level of intra-observer agreement. Image quality was also evaluated.
Results: MPR sensitivity was 87.8% (95% confidence interval [CI] 81-94.5%) and MIP sensitivity was 91.1% (95% CI 85.2-97%). Agreement in MPR was 0.792 with a standard error (SE) of 0.066, and agreement in MIP was 0.836, with a SE of 0.072.
Conclusion: Results of the study indicate the MIP algorithm is more sensitive than MPR. Best intra-observer agreement and image quality results were also observed in the MIP. Data suggest also that MIP should be the post-processing procedure to be utilized in the evaluation of carotid artery stenosis degree, when using MDCTA. Presence of big calcified plaque can determine difficulties in MIP evaluation of stenosis degree.