Purpose To determine the negative predictive value of CT perfusion (CTP) imaging in excluding lesions amenable to endovascular reperfusion in patients who present with symptoms of acute ischemic stroke.

Materials and methods We retrospectively reviewed CTP studies performed between February 2013 and February 2015 who had follow up MRI imaging within 10 days. Patients who received intervenous thrombolysis were excluded. True negative was defined as a CTP study that had: no perfusion mismatch, a fixed matched defect or a small perfusion mismatch in a size (typically <1cm), location or distribution not amenable to endovascular reperfusion. On follow up imaging, all true negative lesions had: no diffusion abnormality or no lesions that, in retrospect, would have been amenable to endovascular therapy. True positive was defined as any prospectively identified perfusion mismatch where endovascular reperfusion was offered. False negative was defined as any CTP were a perfusion/ volume mismatch was not offered endovascular reperfusion subsequently had a diffusion abnormality which in retrospect would have been amenable to endovascular reperfusion. False positive was any CTP study with a perfusion mismatched deemed appropriate for endovascular intervention who at the time of the angiography had no identifiable occlusive lesion. Negative predictive value was calculated as true negative divided by true negative plus false negative.

Results 461 CTP studies performed in 450 patients (mean age 69.1 years, 57% male) were reviewed. 400 were true negative, 55 were true positive, 6 were false positive and 0 was false negative. Negative predictive value was 100%.

Conclusion CTP can confidently exclude lesions which are not amenable to endovascular reperfusion.

Disclosures M. Chehab: None. J. Wilseck: None.