Purpose: Automatic and quantitative measures of penumbra and infarct core volumes have been used in CT brain perfusion for identifying viable brain tissue that could be saved following a stroke and determining appropriate patient management. The aim of this study was to evaluate the impact of radiation dose on the quantification accuracy of penumbra and infarct core volume.
Methods: Twenty-five CT brain perfusion patient data acquired using 4D adaptive spiral technique on a 128-slice scanner (Flash, Siemens) were retrospectively analyzed. A validated low-dose simulation tool was used to simulate three lower dose-levels (75%, 50% and 25%) from the original full dose data (CTDIvol=225mGy). A total of 100 perfusion datasets were analysed (25 patients x 4 dose levels). For each dose-level, six perfusion maps (CBF, CBV, Tmax, MTT, TTD, TTP) and penumbra and infarct volumes were derived for each patient dataset using commercial analysis software (Neuro-VPCT, Syngo.Via-VB30, Siemens). The infarct and penumbra volumes for the reduced dose datasets were compared with the volumes calculated from the original full-dose patient dataset using a Bland-Altman plot. The treatment criteria from the DEFUSE-3 Trial (infarct volume<70mL, penumbra volume>15mL and ischemic and infarct volume ratioâ‰¥1.8) were used to decide if the clinical treatment would have been changed due to the change of the stroke volumes at lower dose-levels.
Results: Excellent agreement between the full dose and both 75% and 50% dose-levels in infarct and penumbra volumes was demonstrated (95% Confidence Interval-Limits of agreement) for the patients evaluated. No patient treatment decision would have been altered using both of these lower dose levels. The overall percentage differences for 75%, 50% and 25% dose-levels were 14Â±10%, 13Â±9%, 29Â±24%, respectively.
Conclusion: Dose reductions of up to 50% for CT head perfusion procedures could be used without significantly changing stroke and penumbra volumes or treatment decision.