The aim of this study is to analyse possible clinical predictors of computed tomography perfusion (CTP) alterations in stroke mimics (SM) attended as stroke-code. 

SM account for up to 25% of admissions with stroke suspicion; moreover, it is believed that the number of admissions will increase over the next decades. Recent studies show CTP alterations in some SM, especially in seizures.

We conducted a retrospective study including a series of SM attended as stroke-code (up to 24 hours of neurological signs or symptoms) by neurologists at a stroke center, between January 2017 and March 2018. All of them underwent multimodal-CT study including baseline computed tomography (CT), CT-angiography and CTP as part of our stroke-code protocol. We registered clinical variables, diagnosis, CTP alteration patterns and type of map affected: time to peak (TTP), cerebral blood flow (CBF) and cerebral blood volume (CBV).

772 patients were attended as stroke-code, 117 (15,15%) were SM.  Mean age (64±17.7 years), women (59%), NIHSS median 2 (1-5 QR). CTP alterations were detected in 31 (26,5%) of SM, more than half of them had a final diagnose of seizure or status epilepticus (67,74%). The most commonly altered CTP map was TTP (96,8%) and the most frequent pattern of alteration was unilateral (90,3%), in diffuse hemispheric (38,7%) or focal non-vascular territory (41,93%). In a multivariate model two factors were significantly associated with the altered CTP, the first one was age (odds ratio (OR) at the age of 65 1,036; 95% CI:1,005-1,069); OR every 5 years 1,259 (IC95%:1,087-1,458) and presence of aphasia (any type) OR 3,800 (95% CI:1,404-10,287).

Our results show that older age and aphasia are risk factors associated to CTP changes in our series of SM. These findings may help decide which patients attended as stroke-code might have larger benefit from CTP for SM diagnosis.