Temporal evolution and spatial distribution of quantitative T2 MRI following acute ischemia reperfusion injury. Issue 5 (July 2020)

• Record Type: Journal Article
• Authors: Sah, Rani Gupta; Nobakht, Sama; Rajashekar, Deepthi; Mouches, Pauline; Forkert, Nils D; Sitaram, Amith; Tsang, Adrian; Hill, Michael D; Demchuk, Andrew M; d'Esterre, Christopher D; Barber, Philip A
• Is Part Of: International journal of stroke. Volume 15:Issue 5(2020)
• Page Start: 495
• Page End: 506
• Publication Date: 2020-07
• Subjects: Quantitative T2 -- endovascular therapy -- diffusion-weighted magnetic resonance imaging -- stroke characterization -- infarct evolution; 616.8005
• DOI: 10.1177/1747493019895673
• Languages: English
• ISSNs: 1747-4930

Abstract

Background: Determining mechanisms of secondary stroke injury related to cerebral blood flow and the severity of microvascular injury contributing to edema and blood-brain barrier breakdown will be critical for the development of adjuvant therapies for revascularization treatment. Aim: To characterize the heterogeneity of the ischemic lesion using quantitative T2 imaging along with diffusion-weighted magnetic resonance imaging (DWI) within five hours of treatment. Methods: Quantitative T2 magnetic resonance imaging was acquired within 5 h (baseline) and at 24 h (follow-up) of stroke treatment in 29 patients. Dynamic contrast enhanced permeability imaging was performed at baseline in a subgroup of patients. Absolute volume change and lesion percent change was determined for the quantitative T2, DWI, and absolute volume change sequences. A Gaussian process with RRELIEFF feature selection algorithm was used for prediction of relative quantitative T2 and DWI lesion growth, baseline and follow-up quantitative T2/DWI lesion ratios, and also NIHSS at 24 h and change in NIHSS from admission to 24 h. Results: In n = 27 patients, median (interquartile range) lesion percent change was 114.8% (48.9%, 259.1%) for quantitative T2, 48.2% (−12.6%, 179.6%) for absolute volume change, and 62.7% (26.3%, 230.9%) for DWI, respectively. Our model, consisting of baseline NIHSS, CT ASPECTS, and systolic blood pressure, showed a strong correlation with quantitative T2 percent change (cross correlation R 2 = 0.80). There was a strong predictive ability for quantitative T2/DWI lesion ratio at 24 h using baseline NIHSS and last seen normal to 24 h magnetic resonance imaging time (cross correlation R 2 = 0.93). Baseline dynamic contrast enhanced permeability was moderately correlated to the baseline quantitative T2 values (rho = 0.38). Conclusion: Quantitative T2 imaging provides critical information for development of therapeutic approaches that could ameliorate microvascular damage during ischemia reperfusion.