Dose reduction in perfusion CT in stroke patients by lowering scan frequency does not affect automatically calculated infarct core volumes

Abstract Background and purpose CT Perfusion technique (CTP) is a quantitative, easily performed, accepted and reliable method for detection of ischemic brain changes. Based on calculated parameters, the size of ischemic penumbra and irreversibly damaged infarct core can be determined which helps guide treatment decisions. However, due to the dynamic nature of the CTP study, it is dose intensive. This study determines the consequences of retrospectively reducing the number of scans in the dynamic acquisition by half on the volume of the automatically calculated infarct core (non-viable tissue) and penumbra (tissue at risk) volumes. Our hypothesis was that equivalent volumetric information could be obtained at a substantial dose savings. Materials and methods Fifty one consecutive patients with occlusion of M1 and/or M2 segment of the middle cerebral artery and ischemic stroke proven by follow-up MRI were included. CTP scans were first analyzed in a standard fashion and automatically generated volumes measured in milliliters were recorded in a database. A second analysis was conducted after removing every second data acquisition from the sequential CTP scans. Automatic volume measurements were repeated, recorded and compared to the initial values obtained using the full dataset. Results The two CTP protocols were statistically equivalent pertaining to automatic infarct core volume calculation but a case-by-case analysis revealed substantial overestimation in some cases. Conclusion Reduction of radiation exposure in CTP without objective loss of accuracy of automatically calculated infarct core volume is feasible but might lead to clinically relevant infarct core overestimation in individual cases.