Utility of CT perfusion with 64-row multi-detector CT for acute ischemic brain stroke

Emerg Radiol. 2011 Apr;18(2):95-101. doi: 10.1007/s10140-010-0905-8. Epub 2010 Sep 15.

Abstract

We investigated the utility of computed tomographic (CT) perfusion (CTP) with 64-row multi-detector row CT (MDCT) to diagnose acute infarction and ischemic penumbra. We reviewed 58 clinical cases with acute ischemic stroke with CTP, compared the size of the area with long mean transit time (MTT) to that with abnormal intensity in magnetic resonance (MR) diffusion-weighted imaging (DWI) to diagnose penumbra, and compared the size of the area with reduced cerebral blood volume (CBV) in CTP to that in MR DWI to evaluate sensitivity for infarction. The total sensitivity of MTT to acute ischemic lesions was 81% (47/58). Sensitivity of MTT to segmental lesions was 100% (42/42) and for spot and focal lesions, 31% (5/16). In 13 patients, penumbra was diagnosed as lesions mismatched between MTT in CTP and MR DWI. When we regarded a lesion with decreased CBV as infarction, the sensitivity of CBV to segmental lesions was 85% (11/13), and the sensitivity to small infarction was 14% (4/28). Use of 64-row MDCT improves coverage and radiation exposure in head CTP. The combination of plain CT, CT angiography, and CTP with MDCT can demonstrate all segmental ischemic lesions and most large segmental infarctions, and their combined application is useful in considering indication and contraindication for thrombolysis. The problem of low sensitivity for small lesions remains, and MR DWI may be required to assess small infarctions when findings from combined plain CT, CT angiography, and CTP are negative in patients with suspected acute brain stroke.

Publication types

  • Evaluation Study

MeSH terms

  • Aged
  • Aged, 80 and over
  • Brain Ischemia / diagnosis*
  • Brain Ischemia / diagnostic imaging
  • Female
  • Humans
  • Male
  • Middle Aged
  • Retrospective Studies
  • Stroke / diagnosis*
  • Stroke / diagnostic imaging
  • Tomography, X-Ray Computed*