{
  "abstract": "Introduction/Purpose Acute ischemic stroke due to a large vessel occlusion (LVO) is often diagnosed with a computerized tomographic angiography (CTA) and treated with endovascular mechanical thrombectomy (EMT). Stroke core volume can be measured by computerized tomographic perfusion (CTP) and confirmed with a follow-up 24-hour MRI. However, it is uncertain whether CTP and MRI core measurements correlate with each other. The aim of this study is to evaluate the role of CTP in core volume prediction and its correlation with the stroke volume as measured in a follow-up 24-hour MRI scan.Materials and Methods We conducted a single-center retrospective observational study of adult patients with acute ischemic stroke due to LVO treated with EMT between 2018 and 2022. Patients underwent a CTP at first work up and a follow-up 24-hour MRI. Demographic, clinical and imaging data were extracted from the medical chart review. Subjects were categorized according to onset of symptoms as early window (<6 hours) and late window (>6 hours). Core volume in MRI DWI sequence was obtained by two independent observers using the ABC2 formula. Level of agreement and correlation coefficient between observers was calculated. A subgroup analysis was conducted for the early and late window groups stratified by location.Results Ninety patients were included, 45 with early window and 45 with late window onset. The most common comorbidities were hypertension (74.11%) and diabetes (25.56%). 80% had a pre-stroke modified Rankin Scale of 0-2. Mean NIHSS total score was 16.44 (SD 7.77). TICI grade was 2a and 3 in 45.98% and 42.53% of patients, respectively. Less than 5% reported severe complications from thrombolytic therapy. The most frequent vessel occluded was the middle cerebral artery (MCA) in 92.94% of cases. Median CTP CBF < 30% correlated moderately with median MRI core volume (rs = 0.561, p<.0001), both in the early (rs = 0.529, p<.0002) and late window (rs = 0.646, p<.0001). The linear relationship between MRI and CTP measurements was weak in all patients (R 2 = 0.294), as well as in both the early (R2=0.273) and late (R2=0.365) windows. There was a strong linear relationship between CTP and MRI core measurements in patients with occlusion of the anterior cerebral artery (ACA) within the early window (R2 = 0.783, p = 0.1154), whereas a moderate linear relationship (R2=0.351, p<.0001) in patients with MCA stroke in late window. A Bland Altman analysis showed a strong correlation in the ABC2 estimated scores between the two observers (r=0.988, p <.0001).Conclusion There is a weak correlation between CTP predicted core and MRI-defined infarct volume for subjects with both early and late presentation window. Final infarct volume at follow-up MRI is variable and sometimes discordant to initial perfusion-based imaging during first work up, questioning the previously assumed linear relationship between both imaging modalities. Other clinical and imaging variables should be considered to improve estimation of ischemic penumbra and better guide treatment selection.Disclosures M. Aguilera-Pena: None. M. Michel: None. S. Shahrestani: None. S. Ramirez-Guerrero: None. O. Jallow: None. P. Eboli: None. N. Gonzalez: None.",
  "authors": [
    {
      "affiliations": [
        "Neurosurgery, Cedars Sinai Medical Center, Los Angeles, CA",
        "Neurology, Tulane University School of Medicine, New Orleans, LA"
      ],
      "name": "M Aguilera-Pena"
    },
    {
      "affiliations": [
        "Neurosurgery, Cedars Sinai Medical Center, Los Angeles, CA",
        "Neurosurgery, University of Pennsylvania, Philadelphia, PA"
      ],
      "name": "M Michel"
    },
    {
      "affiliations": [
        "Neurosurgery, Cedars Sinai Medical Center, Los Angeles, CA"
      ],
      "name": "S Shahrestani"
    },
    {
      "affiliations": [
        "Neurosurgery, Cedars Sinai Medical Center, Los Angeles, CA"
      ],
      "name": "S Ramirez-Guerrero"
    },
    {
      "affiliations": [
        "Neurosurgery, Cedars Sinai Medical Center, Los Angeles, CA",
        "Neurosurgery, Virginia Tech Carilion School of Medicine, Roanoke, VA"
      ],
      "name": "O Jallow"
    },
    {
      "affiliations": [
        "Neurosurgery, Cedars Sinai Medical Center, Los Angeles, CA"
      ],
      "name": "P Eboli"
    },
    {
      "affiliations": [
        "Neurosurgery, Cedars Sinai Medical Center, Los Angeles, CA"
      ],
      "name": "N Gonzalez"
    }
  ],
  "title": "E-358 Evaluating the accuracy of CTP in predicting stroke core volume: correlation with 24-hours MRI and implications for endovascular mechanical thrombectomy eligibility",
  "uid": "17f9794d-94d3-58da-ab28-6ef7f276c7bb"
}