Non-parametric deconvolution using Bézier curves for quantification of cerebral perfusion in dynamic susceptibility contrast MRI

Objective Deconvolution is an ill-posed inverse problem that tends to yield non-physiological residue functions R(t) in dynamic susceptibility contrast magnetic resonance imaging (DSC-MRI). In this study, the use of Bézier curves is proposed for obtaining physiologically reasonable residue functions in perfusion MRI. Materials and methods Cubic Bézier curves were employed, ensuring R(0) = 1, bounded-input, bounded-output stability and a non-negative monotonically decreasing solution, resulting in 5 parameters to be optimized. Bézier deconvolution (BzD), implemented in a Bayesian framework, was tested by simulation under realistic conditions, including effects of arterial delay and dispersion. BzD was also applied to DSC-MRI data from a healthy volunteer. Results Bézier deconvolution showed robustness to different underlying residue function shapes. Accurate perfusion estimates were observed, except for boxcar residue functions at low signal-to-noise ratio. BzD involving corrections for delay, dispersion, and delay with dispersion generally returned accurate results, except for some degree of cerebral blood flow (CBF) overestimation at low levels of each effect. Maps of mean transit time and delay were markedly different between BzD and block-circulant singular value decomposition (oSVD) deconvolution. Discussion A novel DSC-MRI deconvolution method based on Bézier curves was implemented and evaluated. BzD produced physiologically plausible impulse response, without spurious oscillations, with generally less CBF underestimation than oSVD.

Diagnostic Performance of Computed Tomography Angiography and Computed Tomography Perfusion Tissue Time‐to‐Maximum in Vasospasm Following Aneurysmal Subarachnoid Hemorrhage

Background Vasospasm is a treatable cause of deterioration following aneurysmal subarachnoid hemorrhage. Cerebral computed tomography perfusion mean transit times have been proposed as a predictor of vasospasm but suffer from well‐known technical limitations. We evaluated fully automated, thresholded time‐to‐maxima of the tissue residue function ( T max ) for determination of vasospasm following aneurysmal subarachnoid hemorrhage. Methods and Results Retrospective analysis of 540 arterial segments from 36 encounters in 31 consecutive patients with aneurysmal subarachnoid hemorrhage undergoing computed tomography angiography (CTA), computed tomography perfusion, and digital subtraction angiography (DSA) within 24 hours. T max at 4, 6, 8, and 10 s was generated using RAPID (iSchemaView Inc., Menlo Park, CA). Dual‐reader CTA and computed tomography perfusion interpretations were compared for patients with and without vasospasm on DSA (DSA+ and DSA−). Logistic regression models were developed using CTA and T max as input predictors and DSA vasospasm as outcome in adjusted and unadjusted models. Imaging studies from all 31 subjects (mean age 47.3±11.1, 77% female, 65% with single aneurysm with mean size of 6.0±2.9 mm) were included. Vasospasm was identified in 42 segments on DSA and 59 segments on CTA, with significant associations across individual vessel segments ( P <0.001). In adjusted analyses, DSA vasospasm was associated with CTA (odds ratio [OR], 2.43; 95% CI, 0.94–6.32; P =0.068) as well as territory‐specific T max >6 seconds delays (OR, 3.57; 95% CI, 1.36–9.35; P =0.009). Sensitivity/specificity for DSA vasospasm was 31%/91% for CTA, 26%/89% for T max >6 seconds, and 12%/99% for CTA+ T max >6 seconds. Conclusions CTA and T max offer high specificity for presence of vasospasm; their utility, even in combination, as screening tests is, however, limited by poor sensitivity.

Systematic CT perfusion acquisition in acute stroke increases vascular occlusion detection and thrombectomy rates

BackgroundIn patients with stroke, current guidelines recommend non-invasive vascular imaging to identify intracranial vessel occlusions (VO) that may benefit from endovascular treatment (EVT). However, VO can be missed in CT angiography (CTA) readings. We aim to evaluate the impact of consistently including CT perfusion (CTP) in admission stroke imaging protocols.MethodsFrom April to October 2020 all patients admitted with a suspected acute ischemic stroke underwent urgent non-contrast CT, CTA and CTP and were treated accordingly. Hypoperfusion areas defined by time-to-maximum of the tissue residue function (Tmax) >6 s, congruent with the clinical symptoms and a vascular territory, were considered VO (CTP-VO). In addition, two experienced neuroradiologists blinded to CTP but not to clinical symptoms retrospectively evaluated non-contrast CT and CTA to identify intracranial VO (CTA-VO).ResultsOf the 338 patients included in the analysis, 157 (46.5%) presented with CTP-VO (median Tmax >6s: 73 (29–127) mL). CTA-VO was identified in 83 (24.5%) of the cases. Overall CTA-VO sensitivity for the detection of CTP-VO was 50.3% and specificity was 97.8%. Higher hypoperfusion volume was associated with increased CTA-VO detection (OR 1.03; 95% CI 1.02 to 1.04). EVT was performed in 103 patients (30.5%; Tmax >6s: 102 (63–160) mL), representing 65.6% of all CTP-VO. Overall CTA-VO sensitivity for the detection of EVT-VO was 69.9% and specificity was 95.3%. Among patients who received EVT, the rate of false negative CTA-VO was 30.1% (Tmax >6s: 69 (46–99.5) mL).ConclusionSystematically including CTP in acute stroke admission imaging protocols may increase the diagnosis of VO and rate of EVT.

Power of Hypoperfusion in Predicting Recurrent Transient Ischemic Attacks: Protocol of a Prospective Cohort Study

Background: Transient ischemic attack (TIA) has a high incidence of recurrent vascular events. Hypoperfusion is one of the factors that are closely correlated with 7-day recurrence of TIA. This study aimed to evaluate the power of hypoperfusion shown on magnetic resonance (MR) perfusion imaging in predicting the incidence of 7-day recurrence of ischemic events after TIA.Methods/Design: REATTACK is a prospective multi-centered cohort study on the correlation between MR perfusion and TIA recurrence. Ninety patients aged ≥18 years with recent (&lt;7 days after onset) clinical TIA will be continuously included. All the patients will undergo diffusion-weighted imaging (DWI) and perfusion-weighted imaging (PWI) assessments within 24 h after the onset of TIA. The subjects will then be divided into a PWI positive group and a PWI negative group according to the time-to-maximum of the residue function (Tmax). PWI will be repeated after 7 days and in 3 months. The primary clinical outcome will be the recurrence of TIA within 7 days after the onset of TIA. Secondary outcomes will be the recurrence of TIA in 3 months and modified Rankin scale (mRS) score. A chi-square test will be performed to compare the difference in the incidence of recurrent TIA between the two groups, and rank sum test in the mRS score. Multivariate logistic regression will be simultaneously performed to analyze the risk factors for the recurrence of TIA.Discussion: The results of this study will confirm whether abnormal Tmax helps to identify the patients with TIA who have high risks of recurrent ischemic events. This would largely improve the prognosis of patients with TIA.Trial Registration:www.chictr.org.cn, registration number: ChiCTR2000031863, registered on 12 April 2020.

Reversible Edema in the Penumbra Correlates With Severity of Hypoperfusion

Background and Purpose: We aimed to investigate fluid-attenuated inversion recovery changes in the penumbra. Methods: We determined core and perfusion lesions in subjects from the WAKE-UP trial (Efficacy and Safety of MRI-Based Thrombolysis in Wake-Up Stroke) and AXIS 2 trial (Granulocyte Colony-Stimulating Factor in Patients With Acute Ischemic Stroke) with perfusion- and diffusion-weighted imaging at baseline. Only subjects with a mismatch volume >15 mL and ratio >1.2 were included. We created voxel-based relative fluid-attenuated inversion recovery signal intensity (rFLAIR SI) maps at baseline and follow-up. We studied rFLAIR SI in 2 regions of interest: baseline penumbra (baseline perfusion lesion−[core lesion+voxels with apparent diffusion coefficient <620 10 −6 mm 2 /s]) and noninfarcted penumbra (baseline perfusion lesion−follow-up fluid-attenuated inversion recovery lesion) at 24 hours (WAKE-UP) or 30 days (AXIS 2). We analyzed the association between rFLAIR SI and severity of hypoperfusion, defined as time to maximum of the residue function. Results: In the baseline penumbra, rFLAIR SI was elevated (ratio, 1.04; P =1.7×10 − 13 ; n=126) and correlated with severity of hypoperfusion (Pearson r, 0.03; P <1.0×10 − 4 ; n=126). In WAKE-UP, imaging at 24 hours revealed a further increase of rFLAIR SI in the noninfarcted penumbra (ratio, 1.05 at 24 hours versus 1.03 at baseline; P =7.1×10 −3 ; n=43). In AXIS 2, imaging at 30 days identified reversibility of the rFLAIR SI (ratio, 1.02 at 30 days versus 1.04 at baseline; P =1.5×10 −3 ; n=26) since it was no longer different from 1 (ratio, 1.01 at 30 days; P =0.099; n=26). Conclusions: Penumbral rFLAIR SI increases appear early after stroke onset, correlate with severity of hypoperfusion, further increase at 24 hours, and are reversible by 30 days. Registration: URL: https://clinicaltrials.gov ; Unique identifier: NCT01525290. URL: https://clinicaltrials.gov ; Unique identifier: NCT00927836.

Risk Factors of Hypoperfusion on MRI of Ischemic Stroke Patients Within 7 Days of Onset

Objective: Hypoperfusion is an important factor determining the prognosis of ischemic stroke patients. The present study aimed to investigate possible predictors of hypoperfusion on MRI of ischemic stroke patients within 7 days of stroke onset.Methods: Ischemic stroke patients, admitted to the comprehensive Stroke Center of Shanghai Fourth People's Hospital affiliated to Tongji University within 7 days of onset between January 2016 and June 2017, were recruited to the present study. Magnetic resonance imaging (MRI), including both diffusion-weighted imaging (DWI) and perfusion-weighted imaging (PWI), was performed within 7 days of the symptom onset. Time to maximum of the residue function (Tmax) maps were automatically evaluated using the RAPID software. The volume of hypoperfusion was measured outside the infarct area based on ADC &lt; 620 × 10−6 mm2/s. The 90 d mRS score was assessed through either clinic visits or telephone calls. Multivariate step-wise analysis was used to assess the correlation between MR findings and clinical variables, including the demographic information, cardio-metabolic characteristics, and functional outcomes.Results: Among 635 patients admitted due to acute ischemic stroke within 7 days of onset, 241 met the inclusion criteria. Hypoperfusion volume of 38 ml was the best cut-off value for predicting poor prognosis of patients with cerebral infarction (90 d-mRS score ≥ 2). The incidences of MR perfusion Tmax &gt; 4–6 s maps with a volume of 0–38 mL or &gt;38 mL were 51.9% (125/241) and 48.1% (116/241), respectively. Prior stroke and vascular stenosis (≥70%) were associated with MR hypoperfusion. Multivariate step-wise analysis showed that prior stroke and vascular stenosis (≥70%) were risk factors of Tmax &gt; 4–6 s maps, and the odds ratios (OR) were 3.418 (adjusted OR 95% CI: 1.537–7.600), and 2.265 (adjusted OR, 95% CI: 1.199–4.278), respectively.Conclusion: Our results suggest that prior stroke and vascular stenosis (≥70%) are strong predictors of hypoperfusion in patients with acute ischemic stroke within 7 days of stroke onset.

Ischemic Core Overestimation on Computed Tomography Perfusion

Background and Purpose: Different studies have pointed that CT perfusion (CTP) could overestimate ischemic core in early time window. We aim to evaluate the influence of time and collateral status on ischemic core overestimation. Methods: Retrospective single-center study including patients with anterior circulation large-vessel stroke that achieved reperfusion after endovascular treatment. Ischemic core and collateral status were automatically estimated on baseline CTP using commercially available software. CTP-derived core was considered as tissue with a relative reduction of cerebral blood flow <30%, as compared with contralateral hemisphere. Collateral status was assessed using the hypoperfusion intensity ratio (defined by the proportion of the time to maximum of tissue residue function >6 seconds with time to maximum of tissue residue function >10 seconds). Final infarct volume was measured on 24 to 48 hours noncontrast CT. Ischemic core overestimation was considered when CTP-derived core was larger than final infarct. Results: Four hundred and seven patients were included in the analysis. Median CTP-derived core and final infarct volume were 7 mL (interquartile range, 0–27) and 20 mL (interquartile range, 5–55), respectively. Median hypoperfusion intensity ratio was 0.46 (interquartile range, 0.23–0.59). Eighty-three patients (20%) presented ischemic core overestimation (median overestimation, 12 mL [interquartile range, 41–5]). Multivariable logistic regression analysis adjusted by CTP-derived core and confounding variables showed that poor collateral status (per 0.1 hypoperfusion intensity ratio increase; adjusted odds ratio, 1.41 [95% CI, 1.20–1.65]) and earlier onset to imaging time (per 60 minutes earlier; adjusted odds ratio, 1.14 [CI, 1.04–1.25]) were independently associated with core overestimation. No significant association was found with imaging to reperfusion time (per 30 minutes earlier; adjusted odds ratio, 1.17 [CI, 0.96–1.44]). Poor collateral status influence on core overestimation differed according to onset to imaging time, with a stronger size of effect on early imaging patients( P interaction <0.01). Conclusions: In patients with large-vessel stroke that achieve reperfusion after endovascular therapy, poor collateral status might induce higher rates of ischemic core overestimation on CTP, especially in patients in earlier window time. CTP reflects a hemodynamic state rather than tissue fate; collateral status and onset to imaging time are important factors to consider when estimating core on CTP.

Abstract P352: Reversible Relative FLAIR Signal Intensity Changes in the Penumbra Correlate With Severity of Hypoperfusion

In ischemic stroke, the study of edema, measurable as fluid attenuated inversion recovery (FLAIR) signal increase, has mainly focused on the ischemic core and less on the surrounding penumbra. To the naked eye, no FLAIR changes are present in the penumbra. However, changes in perfusion status could induce physiological changes resulting in subtle penumbral FLAIR signal alterations. To investigate penumbral FLAIR changes, we included subjects from the Efficacy and Safety of MRI-Based Thrombolysis in Wake-Up Stroke (WAKE-UP) and Granulocyte Colony-Stimulating Factor in Patients With Acute Ischemic Stroke (AXIS 2) trial with perfusion- and diffusion-weighted imaging (PWI, DWI) and FLAIR at baseline. We used RAPID software to calculate the core and perfusion lesion on DWI and PWI and selected subjects with a minimal mismatch volume (15 ml) and ratio (1.2). We created voxel-based relative FLAIR signal intensity (rFLAIR SI) maps at baseline and follow up (FU) by calculating the ratio of the FLAIR intensity in one voxel and the median FLAIR intensity in a sphere with 15 mm radius around a contralateral homologues voxel. We studied rFLAIR SI in two regions of interest: the baseline penumbra (baseline perfusion lesion - [core lesion + voxels with apparent diffusion coefficient <620 10 -6 mm 2 /s]) and the non-infarcted penumbra (baseline perfusion lesion - FU FLAIR lesion) at 24 hours (WAKE-UP) or 30 days (AXIS 2). Severity of hypoperfusion was defined as the time to maximum of the residue function. In the baseline penumbra, rFLAIR SI was elevated (ratio=1.04, p=1.7*10 -13 , n=126) and correlated with severity of hypoperfusion (Pearson’s r 0.03, p<1.0*10 -4 , n=126). At 24 hours in a subgroup from WAKE-UP, rFLAIR SI in the non-infarcted penumbra further increased (ratio=1.05 at 24h vs 1.03 at baseline, p=7.1*10 -3 , n=43). In a different subgroup from AXIS 2, this increase in rFLAIR SI was reversible (ratio=1.02 at 30d vs 1.04 at baseline, p=1.5*10 -3 n=26) since it was no longer different from 1 at 30 days (ratio=1.01 at 30 days, p=0.099, n=26). Increases in rFLAIR SI, likely representing edema, are not restricted to the ischemic core and correlate with severity of hypoperfusion in the penumbra. They appear early after stroke onset, further increase at 24 hours and are reversible by 30 days.

Clinical Outcomes and Identification of Patients With Persistent Penumbral Profiles Beyond 24 Hours From Last Known Well

Background and Purpose: DEFUSE 3 (Endovascular Therapy Following Imaging Evaluation for Ischemic Stroke 3) infarct volumes at 24 hours did not significantly differ in the endovascular thrombectomy (EVT) versus medical management (MM) only groups. We hypothesized that this was due to underestimation of the final infarct volume among patients with persistent penumbral tissue 24 hours after randomization that subsequently progressed to infarction. We sought to assess the clinical outcomes in patients with persistent penumbral profile >24 hours from last known well and identify them based on the Persistent Penumbra Index (PPI, time-to-maximum of the residue function >6 s perfusion lesion divided by diffusion-weighted magnetic resonance imaging lesion volume on 24-hour postrandomization imaging). Methods: Patients were stratified into those with a 24-hour postrandomization penumbral (PPI>1) versus a nonpenumbral (PPI≤1) profile. The primary outcome was 90-day–modified Rankin Scale. Results: One hundred eighty-two patients were randomized (EVT: 92, MM: 90). Twenty-four–hour postrandomization time-to-maximum of the residue function and infarct volumes were assessable for 144 (EVT: 75, MM: 69). Infarct volumes did not differ between EVT and MM (median [interquartile range] mL: 35.0 [17.6–81.6] versus 41.0 [25.4–106.2], P =0.185). Thirty-two patients had persistent penumbral profile (PPI>1), of these 29 (91%) received MM. PPI was 0 (0–0.07) for EVT, and 0.77 (0.23–1.79) for MM, P <0.001. Patients with clinical-imaging mismatch (more severe strokes and smaller infarct volumes) were more likely to have persistent penumbral profile (PPI>1; adjusted odds ratio, 1.20 [1.11–1.30] for every 1-point National Institutes of Health Stroke Scale-increment and adjusted odds ratio, 0.977 [0.964–0.990] for every 10 cc smaller infarct volume, P <0.001). Patients with nonpenumbral profile (PPI≤1) had higher odds of achieving functional independence (39% versus 9%; adjusted odds ratio, 9.9[95% CI, 2.3–42.8], P =0.002), a trend towards lower mortality (12% versus 34%, P =0.002; adjusted odds ratio, 0.34 [95% CI, 0.11–1.03], P =0.057) and early clinical improvement (24-hour National Institutes of Health Stroke Scale—decrease ≥8 points or 0–1): 29% vs 9%, P =0.034) which persisted at discharge and 90-day follow-up. For a given volume, patients with PPI≤1 had significantly higher likelihood of functional independence as compared to those with PPI>1. Conclusions: Patients with persistent penumbral profile who have salvageable tissue beyond 24 hours from last known well can be identified by PPI and clinical-imaging mismatch. They have a poor prognosis and may benefit from very late window reperfusion therapies. Clinical trials in these patients are warranted. Registration: URL: https://www.clinicaltrials.gov . Unique identifier: NCT02586415.