CTP infarct core may predict poor outcome in stroke patients treated with IV t-PA

Abstract

BackgroundComputerized tomography perfusion (CTP) has been widely studied in assessing physiological brain tissue parameters in patients with acute ischemic stroke (AIS). The utility of CTP to predict clinical outcome in patients with AIS treated with intravenous tissue plasminogen activator (IV t-PA) is controversial. We reviewed CTP data in AIS patients treated with IV t-PA to uncover potential predictors of clinical outcome. MethodsWe retrospectively identified AIS patients from our stroke registry (7/07 to 2/10) who underwent CTP on arrival and then received IV t-PA. A neuroradiologist blinded to outcome performed all CTP parameter measurements on a commercially available Siemens Neuro PCT workstation. Tissue at risk (TAR) was defined as the area of infarct territory with a relative time to peak (rTTP) greater than 4s. Non-viable tissue (NVT) was defined as the area of infarct territory with absolute cerebral blood volume (CBV) less than 2ml/100g and cerebral blood flow (CBF) less than 12.7ml/100g/min. Penumbra was defined as the area of (TAR) minus the area of (NVT). Excellent clinical outcome was defined as mRS (0–1), good clinical outcome was defined as mRS (0–2), and poor clinical outcome was defined as mRS (4–6), all measured at hospital discharge and 90days if available. Recanalization data was obtained when available by comparing pre-thrombolytic CTA data and post-treatment MRA/CTA images by a single blinded radiologist. ResultsWe identified 61 patients that met our inclusion criteria with a mean age of 68 (29–94), median NIHSS on admission of 13 (1–40), and median discharge mRS of 4 (0–6). Using multivariate logistic regression and ordinal logistic regression controlling for age and admission NIHSS, none of the CTP parameters were statistically associated with excellent or good clinical outcome (mRS<2). Using multivariate analysis controlling for age and admission NIHSS, NVT area>30cm2 (OR=5.12, CI: 0.95–27, p=0.05) was statistically associated with poor clinical outcome at discharge. NVT area≥30cm2 was a potential predictor of poor outcome at discharge even when controlling for age and NIHSS. ConclusionCTP parameters derived from commercially available software and published thresholds yield little predictive value for good clinical outcomes for AIS patients treated with IV t-PA but may be useful in predicting poor clinical outcome especially if the area of non-viable tissue is greater than 30cm2.