Keywords:
Ischaemia / Infarction, Acute, Computer Applications-Detection, diagnosis, CT-Quantitative, CNS
Authors:
K. Endo, K. Suzuki, E. Okaniwa, H. Hashimoto, K. Abe, S. Sakai; Tokyo/JP
DOI:
10.26044/ecr2019/C-1722
Conclusion
Much discussion has been held regarding the quantitative thresholds for penumbra and ischemic core.
CBF is gradually reduced throughout all stages of ischemia; consequently,
it is occasionally difficult to define a clearthreshold between infarct and penumbra based on CBF.
On the other hand,
CBV increases during ischemic penumbra,
whereas CBV decreases during irreversible tissue damage.
Therefore,decreased CBV was used to define an infarcted core[11]. Although the quantitative stability of CBF is worse than that of CBV,
decreased CBF is has been considered an accurate definition of infarcted core in recent clinical trials,
such as the DAWN and DEFUSE 3 trials.
In addition,
decreased CBF (<30% of normal baseline),
particularly using the Bayesian algorithm versus the standard deconvolution algorithm,
has recently been reported as the accurate definition of an infarcted core and indicated in diffusion-weighted magnetic resonance imaging (MR DWI)[12].
Our results showed increased CBV and prolonged MTT and TTP in penumbra.
Although we identified decreased CBF,
no significant difference was observed between penumbraand normal tissue.
CTP parameters for AISwere different than those in our previous study for the chronic cerebral ischemic phase in ECR2018[13]. The Bayesian algorithm indicated chronic ischemic lesions with longer MTT and increased CBV.
Conversely,
the rSVD algorithm indicated those with longer TTP and increased CBV.
Limitations of this study include the small number of patients and potential for bias. It is possible that ischemic lesions are not calculated accurately as CTP values owing to auto-segmentation.
In conclusion,
non-infarcted acute ischemic lesions could be diagnosed with prolonged MTT and TTP without decreased CBF and CBV using either algorithm.
No clinical difference may exist between the Bayesian and rSVD algorithms in establishing a diagnosis of non-infarcted AIS at low radiation doses.