Abstract Number ‐ 50: Multiplane Reconstruction Modifies The Diagnostic Performance Of CTA Imaging In Carotid Web Cases

Abstract

Introduction Carotid Web (CaW) represents an important and overlooked etiology for ischemic stroke and has been associated with high rates of stroke recurrence. Computed tomography angiography (CTA) has been shown to have comparable performance to digital subtraction angiography (DSA) and has been suggested to be the non‐invasive imaging of choice for CaW detection. However, misdiagnosis has been demonstrated to be common even in specialized centers. We evaluated the impact of adding CTA multiplane reconstruction (MPR) andthree‐dimensional maximum intensity projection (3D MIP) reformat on the diagnostic performance of CTA in CaW detection. Methods After exclusion of patients aged >65 years old and patients with no available/poor quality CTA,CaW cases (n = 31 consecutive patients leading to 31 ipsilateral carotids to the stroke derived from out prospective CaW database), as well as two other groups: 1)carotid atherosclerosis (n = 27consecutivepatients from out carotid stenting database leading to 27 carotids contralateral to the index lesion) and 2) consecutive normal carotid cases (n = 49 patients with normal carotids extracted from the electronic medical records for patients imaged due to suspected blunt cerebrovascular trauma) were included. All CTA images were deidentified and reviewed independently by three stroke neurologists to record the diagnosis and level of diagnostic certainty (in form of a scale (1[lowest]‐to‐5[highest]) after evaluating the CTA axial plane alone, then after sagittal and coronal planes (MPR) reconstructions, and then after evaluation of3D MIP reformatted images.The analyses were made for the total number of observations for all readers (93 CaWs, 81 atherosclerosis cases and 147 normal carotids). Results On reviewing CTA axial projection alone, raters correctly diagnosed 44.1% of CaW, 87.7% of carotid atherosclerosis and 83% of normal carotid images. Sagittal and coronal MPR significantly increased the rate of accurate CaW diagnosis (76.3%‐Table 1) The certainty level for CaW diagnosis was lower when compared to atherosclerosis as well as normal carotid using the CTA axial projection alone (3.0[3.0‐4.0] vs 4.0[3.0‐5.0];p< 0.001 and vs 4.0[3.0‐5.0];p< 0.001) as well as after adding sagittal/coronal MPR (4.0[3.0‐5.0] vs 5.0[4.0‐5.0],p = 0.01; and vs 4.0[4.0‐5.0],p< 0.001). The certainty level became similar between CaW and atherosclerosis as well as normal carotids with the addition of 3D MIP (5.0[5.0‐5.0] vs 5.0[4.5‐5.0], p = 0.61; and vs 5.0[5.0‐5.0],p = 0.15) respectively. Inter‐rater agreement in CaW detection increased from k = 0.46(0.35‐0.57);p< 0.05usingaxial section to k = 0.80(0.69‐0.91);p< 0.05 with MPR. Axial projection alone had lower sensitivity in CaW detection (AUC = 0.69(0.62‐0.76),sensitivity = 44%,specificity = 95%,p< 0.05) compared to MPR (AUC = 0.86(0.80‐0.91),sensitivity = 76%,specificity = 96%,p< 0.05). Misdiagnosed CaW cases, after using all planes with 3D MIP (n = 23/93), were older (56[46‐61] vs 52[46‐57] years,p = 0.04) and lower length/base ratio (0.51[0.49‐0.87] vs 0.92[0.74‐1.19],p< 0.001) compared to the correctly diagnosed CaW cases (n = 70/93). Conclusions CTA axial plane alone is unreliable to detect CaW and the addition of sagittal/coronal MPR and 3D MIPs are important to increase accurate diagnosis and perceived reader diagnostic certainty.