Page 386

Karger_ESC London_2013

22. European Stroke Conference 195 Vascular imaging PROPOSAL OF A PROBABILITY MODEL FOR UNESTABLE CAROTID ATHEROMA PLAQUES IDENTIFICATION BASED ON GSM SYSTEM. AN INTERNAL VALIDATION G. Ruiz-Ares1, B. Fuentes2, P. Martinez-Sanchez3, E. Díez-Tejedor4 Neurology Department. Stroke Centre. IdiPAZ Health Research Institute. La Paz University Hospital, Madrid, SPAIN1, Neurology Department. Stroke Centre. IdiPAZ Health Research Insti-tute. La Paz University Hospital, Madrid, SPAIN2, Neurology Department. Stroke Centre. IdiPAZ Health Research Institute. La Paz University Hospital, Madrid, SPAIN3, Neurology Department. Stroke Centre. IdiPAZ Health Research Institute. La Paz University Hospital, Madrid, SPAIN4 Background and purpose: To date clinical and subjetive ultrasonografical criteria are used to classify unestable plaques but the standarized grey scale median (GSM) value is an useful objetive tool for this diagnosis. Our aim is to develop and validate a probability model for the identification of un-estable carotid atheroma plaques. Methods: Prospective study. Acute non-cardioembolic anterior cerebral circulation ischemic stroke patients with plaques in internal carotid artery ipsilateral to symptoms were included. Echogenic-ity of plaques was measured by a digital and standarized grey scale system in carotid ultrasound conducted within the first week after admission. Logistic regressions were performed in order to identify factors associated with unestable plaques. A probability model based on clinical and ultraso-nographic data independently associated with unstable plaques was developed and an internal vali-dation was conducted by the “leave one out cross validation” (LOOCV) method. Results: 52 patients included. Unestable plaques showed less echogenicity than stable ones (23 vs 37; p<0.0001). A Receiver Operation Curve (ROC) analysis point to GSM value of 29 as the point associated to higher sensitivity (76%) and specificity (82%) to identify a plaque as unestable (AUC=0,874; 95%CI=0,779-0,969; p<0,001). Male sex (OR=9.075), previous angiotensin convert-ing enzyme inhibitors (ACEI) treatment (OR=0.098) and GSM<29 (OR=25.712) were independent-ly associated with unestable plaques. According to the coefficients of these variables a probability model of unestable plaque was developed (Figure 1). LOOCV obtained AUC=0,834; IC=0,717- 0,951; p<0,001 as a very good internal validation data. Conclusions: We propose a probability model based on sex, previous treatment with ACEI and GSM for unestable carotid plaque prediction in acute stroke patients. 386 © 2013 S. Karger AG, Basel Scientific Programme 196 Vascular imaging The association of high resolution MRI-identified atherosclerotic plaque and the lesion loca-tion in patients with single subcortical infarctions Y. Yoon1, D.W. Kang2, S.U. Kwon3, J.S. Kim4 Asan Medical Center, Seoul, SOUTH KOREA1, Asan Medical Center, Seoul, SOUTH KOREA2, Asan Medical Center, Seoul, SOUTH KOREA3, Asan Medical Center, Seoul, SOUTH KOREA4 Background: Single subcortical infarctions (SSI) in the middle cerebral artery (MCA) territory may be categorized into two conditions by lesion location; (1) lesions which extend to the basal surface of the MCA (proximal SSI, pSSI) and (2) those which do not (distal SSI, dSSI). Previous studies have found that high-resolution MRI (HR-MRI) may detect atherosclerotic plaques even in patients without parental arterial lesion. We investigated whether there are any differences in the presence and location of plaque, infarct volume, and neurologic symptoms between pSSI and dSSI. Methods: We prospectively assessed patients with SSI in MCA territory with diffusion weighted MRI, MRA, and HR-MRI within 3 days of stroke onset. Lesions were classified as pSSI and dSSI. Patients with relevant MCA disease, internal carotid artery disease or potential embolic sources were excluded. Lesion patterns and microbleeds were examined. Neurologic function was assessed by NI-HSS on admission. The presence and the location (dorsal-superior vs. ventral-inferior side) of MCA plaque detected by HR-MRI in MCA were examined. Results: Thirty-nine patients were included; pSSI in 20 patients and dSSI in 19 patients. Although the prevalence of plaque did not differ between the groups (12, 63.2% vs. 8, 40%, p=0.205), that of plaque in dorsal-superior side was significantly higher in pSSI group than in dSSI group (6, 75% vs. 2, 16.7%, p=0.019). Initial lesion volume was bigger (1.96+/-1.18 vs. 1.11+/-1.11, p=0.025), NI-HSS score on admission was higher (4.90+/-2.34 vs. 3.05+/-2.17, p=0.015), and the number of mi-crobleeds was smaller in pSSI group than in dSSI group (1, 5% vs. 10, 52.6%, p=0.001). There were no significant differences in the aforementioned variables between patients with and without MCA plaques. Conclusion: Our results show that pSSI is more closely associated with severe clinical symptoms and dorsal-superiorly located MCA plaques, suggesting that SSI is not homogeneous entity.


Karger_ESC London_2013
To see the actual publication please follow the link above