Determinants of mismatch in acute ischaemic stroke.

Background: Multimodal CT or MR imaging may be helpful in guiding reperfusion therapy for stroke. However, access to multimodal imaging may frequently be limited. We hypothesised that certain clinical and non-enhanced CT (NECT) findings at initial assessment can potentially predict mismatch on CT perfusion (CTP) in patients with acute ischaemic stroke.

Methods: We undertook an analysis of prospectively collected clinical and imaging data of consecutive patients with anterior circulation ischaemic stroke who underwent CTP during their initial assessment. NECT was read for early ischaemic change as measured by the Alberta Stroke Program Early CT Score (ASPECTS), and for hyperdense middle cerebral artery sign (HMCAS). CTP images were evaluated for mismatch. Independent clinical and imaging predictors of a CTP mismatch were identified using stepwise logistic regression.

Results: Of the 202 patients, 92 (46%) demonstrated a mismatch, 23 (11%) a matched deficit, and 87 (43%) no perfusion deficit. HMCAS on NECT (OR 13.65, 95% CI 6.04-30.81, p<0.001), female gender (OR 2.37, 95% CI 1.19-4.72, p = 0.015), atrial fibrillation (OR 2.05, 95% CI 1.02-4.11, p = 0.044), and absence of a history of hypertension (OR 0.46, 95% CI 0.22-0.96, p = 0.037) were independent predictors of a CTP mismatch. HMCAS had 58% sensitivity, 91% specificity, 84% positive predictive value and 72% negative predictive value.

Conclusions: A HMCAS on the initial NECT is associated with a high probability of mismatch in acute ischaemic stroke, and may identify patients most likely to benefit from recanalisation treatments when access to multimodal CT or MR facilities is limited.