Interobserver and intraobserver agreement in the evaluation of CT perfusion in ischemic stroke

Knowledge of interrater reliability in the evaluation of perfusion computed tomography (CTP) studies is very limited even though the method is widely used in the workup of acute stroke. The aims of this study were to estimate the inter- and intraolbserver agreement in the evaluation of CTP data and to evaluate the feasibility of the method. The CTP data of 20 consecutive patients (50% were females) aged 68 +/- 11 years with different categories of acute ischemic stroke were included in this retrospective analysis. Perfusion studies were evaluated independently by six radiologists on two different occasions. The overall inter- and intraolbserver agreement was substantial, showing a K value of 0.65 (95% confidence interval 0.39-0.91). The time required for the post-processing and interpretation ranged from 37 to 460 seconds. Evaluation of manually post-processed CTP data according to the maximum slope model appears to be reliable. Experience and also a short training period increase the reliability of the method and reduce the time needed for delivery of the results to the treating clinician.