MRI biomarkers in acute ischemic stroke: a conceptual framework and historical analysis

Magnetic resonance imaging-based deep learning imaging biomarker for predicting functional outcomes after acute ischemic stroke

PURPOSE

Clinical risk scores are essential for predicting outcomes in stroke patients. The advancements in deep learning (DL) techniques provide opportunities to develop prediction applications using magnetic resonance (MR) images. We aimed to develop an MR-based DL imaging biomarker for predicting outcomes in acute ischemic stroke (AIS) and evaluate its additional benefit to current risk scores.

METHOD

This study included 3338 AIS patients. We trained a DL model using deep neural network architectures on MR images and radiomics to predict poor functional outcomes at three months post-stroke. The DL model generated a DL score, which served as the DL imaging biomarker. We compared the predictive performance of this biomarker to five risk scores on a holdout test set. Additionally, we assessed whether incorporating the imaging biomarker into the risk scores improved the predictive performance.

RESULTS

The DL imaging biomarker achieved an area under the receiver operating characteristic curve (AUC) of 0.788. The AUCs of the five studied risk scores were 0.789, 0.793, 0.804, 0.810, and 0.826, respectively. The imaging biomarker's predictive performance was comparable to four of the risk scores but inferior to one (p = 0.038). Adding the imaging biomarker to the risk scores improved the AUCs (p-values) to 0.831 (0.003), 0.825 (0.001), 0.834 (0.003), 0.836 (0.003), and 0.839 (0.177), respectively. The net reclassification improvement and integrated discrimination improvement indices also showed significant improvements (all p < 0.001).

CONCLUSIONS

Using DL techniques to create an MR-based imaging biomarker is feasible and enhances the predictive ability of current risk scores.

Hyperoxia-Induced ΔR1: MRI Biomarker of Histological Infarction in Acute Cerebral Stroke

OBJECTIVE

To evaluate whether hyperoxia-induced ΔR₁ (hyperO₂ΔR₁) can accurately identify histological infarction in an acute cerebral stroke model.

MATERIALS AND METHODS

In 18 rats, MRI parameters, including hyperO₂ΔR₁, apparent diffusion coefficient (ADC), cerebral blood flow and volume, and ¹⁸F-fluorodeoxyglucose uptake on PET were measured 2.5, 4.5, and 6.5 hours after a 60-minutes occlusion of the right middle cerebral artery. Histological examination of the brain was performed immediately following the imaging studies. MRI and PET images were co-registered with digitized histological images. The ipsilateral hemisphere was divided into histological infarct (histological cell death), non-infarct ischemic (no cell death but ADC decrease), and non-ischemic (no cell death or ADC decrease) areas for comparisons of imaging parameters. The levels of hyperO₂ΔR₁ and ADC were measured voxel-wise from the infarct core to the non-ischemic region. The correlation between areas of hyperO₂ΔR₁-derived infarction and histological cell death was evaluated.

RESULTS

HyperO₂ΔR₁ increased only in the infarct area (p ≤ 0.046) compared to the other areas. ADC decreased stepwise from non-ischemic to infarct areas (p = 0.002 at all time points). The other parameters did not show consistent differences among the three areas across the three time points. HyperO₂ΔR₁ sharply declined from the core to the border of the infarct areas, whereas there was no change within the non-infarct areas. A hyperO₂ΔR₁ value of 0.04 s^-1 was considered the criterion to identify histological infarction. ADC increased gradually from the infarct core to the periphery, without a pronounced difference at the border between the infarct and non-infarct areas. Areas of hyperO₂ΔR₁ higher than 0.04 s^-1 on MRI were strongly positively correlated with histological cell death (r = 0.862; p < 0.001).

CONCLUSION

HyperO₂ΔR₁ may be used as an accurate and early (2.5 hours after onset) indicator of histological infarction in acute stroke.

Impact of aging and comorbidities on ischemic stroke outcomes in preclinical animal models: A translational perspective

Ischemic stroke is a highly complex and devastating neurological disease. The sudden loss of blood flow to a brain region due to an ischemic insult leads to severe damage to that area resulting in the formation of an infarcted tissue, also known as the ischemic core. This is surrounded by the peri-infarct region or penumbra that denotes the functionally impaired but potentially salvageable tissue. Thus, the penumbral tissue is the main target for the development of neuroprotective strategies to minimize the extent of ischemic brain damage by timely therapeutic intervention. Given the limitations of reperfusion therapies with recombinant tissue plasminogen activator or mechanical thrombectomy, there is high enthusiasm to combine reperfusion therapy with neuroprotective strategies to further reduce the progression of ischemic brain injury. Till date, a large number of candidate neuroprotective drugs have been identified as potential therapies based on highly promising results from studies in rodent ischemic stroke models. However, none of these interventions have shown therapeutic benefits in stroke patients in clinical trials. In this review article, we discussed the urgent need to utilize preclinical models of ischemic stroke that more accurately mimic the clinical conditions in stroke patients by incorporating aged animals and animal stroke models with comorbidities. We also outlined the recent findings that highlight the significant differences in stroke outcome between young and aged animals, and how major comorbid conditions such as hypertension, diabetes, obesity and hyperlipidemia dramatically increase the vulnerability of the brain to ischemic damage that eventually results in worse functional outcomes. It is evident from these earlier studies that including animal models of aging and comorbidities during the early stages of drug development could facilitate the identification of neuroprotective strategies with high likelihood of success in stroke clinical trials.

Amide Proton Transfer MRI Signal as a Surrogate Biomarker of Ischemic Stroke Recovery in Patients With Supportive Treatment

Background: Amide proton transfer (APT) MR imaging has shown great potential in the evaluation of stroke severity because of its sensitivity to acid environments. However, this promising MRI technique has not been used to assess treatment efficacy with regard to stroke recovery.

Purpose: To assess the therapeutic effect of supportive treatment in ischemic stroke patients using the pH-sensitive APT MRI technique.

Material and Methods: Forty-three ischemic stroke patients at an early stage were recruited and scanned with conventional and APT MRI sequences at 3T before treatment. After treatment, 26 patients underwent a follow-up MRI scan (one to three times on different days). The magnetization-transfer-ratio asymmetry at 3.5 ppm, usually called the APT-weighted (APTW) signal, was measured. The APTW signal changes following treatment were analyzed.

Results: Baseline APTW signal intensities in the infarcted lesions inversely correlated with baseline stroke severity. Lesion APTW values gradually increased with time in 24 cases (92.3%) with a follow-up MRI scan, showing clinical symptom improvements. Two cases (7.7%) showed further decreased APTW signal in the follow-up scan, accompanied by clinical symptom aggravation. Compared to the baseline, significant APTW signal increases were found for all post-treatment patients (efficacious), whether based on post-treatment or on stroke onset times. The increase in APTW signal in the ischemic stroke lesion after treatment was associated with an improvement in clinical symptoms.

Conclusion: The APTW signal would be a useful imaging biomarker by which to assess the therapeutic efficacy of ischemic stroke treatment.

The different clinical value of susceptibility vessel sign in acute ischemic stroke patients under different interventional therapy: A systematic review and meta-analysis

Acute ischemic stroke (AIS) subtype, one of the most important factors for selecting therapeutic strategies, is difficult to be accurately diagnosed at admission sometimes. The magnetic susceptibility effect of deoxygenated hemoglobin in red thrombi appeared as hypointense signals in magnetic resonance imaging (MRI) scans. The prognostic value of susceptibility vessel sign (SVS) for stroke subtype, recanalization and outcomes in AIS patients will be comprehensively determined in the present study. A comprehensive search of databases was conducted including the PubMed, Embase, and Cochrane Library from inception up to August 2017. Statistical tests were performed to check for heterogeneity and publication bias. Subgroup and sensitivity analysis were also conducted to evaluate the robustness of the conclusions. Overall, 21 studies including 1832 patients were identified. The presence of SVS was significantly associated with cardioembolic stroke than absence of SVS (RR = 1.53, 95% CI = 1.30-1.81, p < 0.001). The patients with SVS were less likely to achieve recanalization (RR = 0.70, 95% CI = 0.56-0.88, p = 0.002) and poor functional outcome (RR = 1.68, 95% CI = 1.44-1.97, p < 0.001) after intravenous thrombolysis (IVT), whereas it was similar between two group after endovascular treatment (EVT) (p = 0.990 and p = 0.335). The SVS length was smaller in recanalization group than that in non-recanalization group (RR = -0.49, 95% CI = -0.72 to -0.27, p < 0.001), however, no significant difference between SVS width and recanalization rate was found. The presence of SVS appears to be a stronger predictor of cardioembolic stroke. Furthermore, the SVS was associated with a decreasing recanalization rate and poor outcome in AIS patients after IVT but not after EVT. Which offered a practical information to select optimal therapeutic strategies for stroke patients with SVS though the level of evidence seems to be quite shaky.

An Increase of Sigma-1 Receptor in the Penumbra Neuron after Acute Ischemic Stroke

BACKGROUND

Penumbra salvage from infarction by early reperfusion within the time window is the target of acute ischemic stroke therapies. Although the penumbral imaging is potently usable in clinic trial, additional work needs to be performed to advancing the field with better-defined, evaluated, and validated imaging measures.

METHODS

Mice were subjected to permanent stroke by right middle cerebral artery (MCA) occlusion. Multimodel magnetic resonance imaging (MRI) method was assessed to define the penumbra as that brain region in which the perfusion and diffusion-weighted MR images are mismatched (perfusion-weighted imaging [PWI]-diffusion-weighted imaging [DWI] mismatch). MRI measurements were performed at 1 hour after MCA occlusion (MCAO). Sigma-1 receptor expression was assessed by immunoblotting and immunostaining in PWI-DWI-defined penumbra and core compared with sham or contralateral slice. Penumbral sigma-1 receptor identified the correlation with the neuron, astrocyte, and microglia by immuno-colocalization.

RESULTS

Sigma-1 receptor was significantly upregulated in penumbra or peri-infarct compared with sham and core tissue at 1 hour and 24 hours after MCAO. There was a colocalization of sigma-1 receptor and neuron in penumbra at 1 hour after stroke. Sigma-1 receptor is specifically increased in ischemic penumbral neuron at 1 hour after MCAO.

CONCLUSIONS

Sigma-1 receptor may act as an endogenous marker of penumbra after acute ischemic stroke.

Brain Vascular Imaging Techniques

Recent major improvements in a number of imaging techniques now allow for the study of the brain in ways that could not be considered previously. Researchers today have well-developed tools to specifically examine the dynamic nature of the blood vessels in the brain during development and adulthood; as well as to observe the vascular responses in disease situations in vivo. This review offers a concise summary and brief historical reference of different imaging techniques and how these tools can be applied to study the brain vasculature and the blood-brain barrier integrity in both healthy and disease states. Moreover, it offers an overview on available transgenic animal models to study vascular biology and a description of useful online brain atlases.

Can MRI quantification help evaluate stroke age?

BACKGROUND

Diffusion-weighted imaging (DWI) fluid-attenuated inversion recovery (FLAIR) mismatch has a proven ability to estimate stroke-to-magnetic resonance imaging (MRI) delay. We evaluated the possibility of enhancing this estimation by quantifying MRI (DWI and FLAIR) signals, and compared this approach to the visual evaluation of DWI-FLAIR mismatch.

MATERIALS AND METHODS

This retrospective study included 194 patients presenting an ischemic stroke in the middle cerebral artery territory that had been explored with 3T MRI within 12h. According to the study design, written informed consent was waived and patient information was anonymized and de-identified prior to analysis. DWI-FLAIR mismatch was visually estimated by two radiologists and a quantification of MRI signals based on a manual segmentation of stroke lesion volume was performed. Using their receiver operating curve and area under the curve (AUC), we identified the variables of MRI quantification that were predictive of stroke-to-MRI delay, then compared their performance against visual classification.

RESULTS

The quantitative variables identified as predictive of stroke-to-MRI delay were: 1st quartile, 3rd quartile and median values of B0; 1st quartile, 3rd quartile, median and relative values of B1000; 1st quartile and relative values of the apparent diffusion coefficient. FLAIR was not found to be predictive. The AUC values of these variables ranged between 0618±0.053 and 0.683±0.048. The relative value of B1000 appeared to be the best predictive quantitative variable, with predictive values comparable to visual classification.

CONCLUSIONS

The quantification of MRI signal may be a helpful tool for stroke dating but cannot outperform the visual estimation of stroke lesion age.

Cyclosporine in acute ischemic stroke

OBJECTIVES

We examined whether IV administration of cyclosporine in combination with thrombolysis might reduce cerebral infarct size.

METHODS

Patients aged 18 to 85 years, presenting with an anterior-circulation stroke and eligible for thrombolytic therapy, were enrolled in this multicenter, single-blinded, controlled trial. Fifteen minutes after randomization, patients received either an IV bolus injection of 2.0 mg/kg cyclosporine (Sandimmune, Novartis) or placebo. The primary endpoint was infarct volume on MRI at 30 days. Secondary endpoints included infarct volume according to the site (proximal/distal) of arterial occlusion and recanalization after thrombolysis.

RESULTS

From October 2009 to July 2013, 127 patients were enrolled. The primary endpoint was assessed in 110 of 127 patients. The reduction of infarct volume in the cyclosporine compared with the control group was overall not significant (21.8 mL [interquartile range, IQR 5.1, 69.2 mL] vs 28.8 mL [IQR 7.7, 95.0 mL], respectively; p = 0.18). However, in patients with proximal occlusion and effective recanalization, infarct volume was significantly reduced in the cyclosporine compared with the control group (14.9 mL [IQR 1.3, 23.2 mL] vs 48.3 mL [IQR 34.5, 118.2 mL], respectively; p = 0.009).

CONCLUSIONS

Cyclosporine was generally not effective in reducing infarct size. However, a smaller infarct size was observed in patients with proximal cerebral artery occlusion and efficient recanalization.

CLASSIFICATION OF EVIDENCE

This study provides Class I evidence that in patients with an acute anterior-circulation stroke, thrombolysis plus IV cyclosporine does not significantly decrease 30-day MRI infarct volume compared with thrombolysis alone.

Endovascular treatment of acute ischemic stroke

Endovascular therapy for acute stroke has evolved with the use of intra-arterial thrombolytics, intravenous/intra-arterial bridging strategies, and mechanical thrombectomy/aspiration devices. Despite widespread use in clinical practice, randomized trials of first-generation devices failed to demonstrate improved outcomes compared with standard care. New-generation stent retriever devices demonstrate higher rates of revascularization and clinical outcomes compared with first-generation devices. Additional randomized trials are underway and have the potential to confirm clinical efficacy of new-generation devices compared with standard care. The role of additional advanced imaging for patient selection remains unclear, and further trials are needed to demonstrate the role of these techniques for patient selection.

Identifying the ischaemic penumbra using pH-weighted magnetic resonance imaging

The original concept of the ischaemic penumbra suggested imaging of regional cerebral blood flow and metabolism would be required to identify tissue that may benefit from intervention. Amide proton transfer magnetic resonance imaging, a chemical exchange saturation transfer technique, has been used to derive cerebral intracellular pH in preclinical stroke models and has been proposed as a metabolic marker of ischaemic penumbra. In this proof of principle clinical study, we explored the potential of this pH-weighted magnetic resonance imaging technique at tissue-level. Detailed voxel-wise analysis was performed on data from a prospective cohort of 12 patients with acute ischaemic stroke. Voxels within ischaemic core had a more severe intracellular acidosis than hypoperfused tissue recruited to the final infarct (P < 0.0001), which in turn was more acidotic than hypoperfused tissue that survived (P < 0.0001). In addition, when confined to the grey matter perfusion deficit, intracellular pH (P < 0.0001), but not cerebral blood flow (P = 0.31), differed between tissue that infarcted and tissue that survived. Within the presenting apparent diffusion coefficient lesion, intracellular pH differed between tissue with early apparent diffusion lesion pseudonormalization and tissue with true radiographic recovery. These findings support the need for further investigation of pH-weighted imaging in patients with acute ischaemic stroke.

Editorial: Workforce needs for endovascular acute ischemic stroke therapy: myth or reality?

The acute ischemic strokes amenable to intraarterial therapy probably number no more than 20,000 per year in the United States. The future demand for intraarterial reperfusion techniques may change, but the fraction of patients who require intraarterial thrombolysis is currently rather low, and the number of neurointerventionists is adequate. Each hospital caring for patients with acute stroke will need to determine its own demand for intraarterial therapy and employ an adequate supply of qualified neurointerventionists available to meet demand. Comprehensive stroke centers are now being designated and hopefully will foster a rational, regionalized approach to the delivery of endovascular therapies for stroke.

Ascertainment of any and all neurologic and myocardial damage in carotid revascularization: the key to optimization?

The majority of carotid revascularization procedures performed at present are in asymptomatic patients. Since such procedures convey no immediate benefit, but rather protect from future hazard, optimization of procedural safety is mandatory. The authors focus their discussion on the methodologies that assess periprocedural myocardial damage and brain injury, as used in past clinical trials, from the fields of perioperative medicine and neurovascular imaging, and discuss methodologies to reduce both events in carotid revascularization.

A trial of imaging selection and endovascular treatment for ischemic stroke

BACKGROUND

Whether brain imaging can identify patients who are most likely to benefit from therapies for acute ischemic stroke and whether endovascular thrombectomy improves clinical outcomes in such patients remains unclear.

METHODS

In this study, we randomly assigned patients within 8 hours after the onset of large-vessel, anterior-circulation strokes to undergo mechanical embolectomy (Merci Retriever or Penumbra System) or receive standard care. All patients underwent pretreatment computed tomography or magnetic resonance imaging of the brain. Randomization was stratified according to whether the patient had a favorable penumbral pattern (substantial salvageable tissue and small infarct core) or a nonpenumbral pattern (large core or small or absent penumbra). We assessed outcomes using the 90-day modified Rankin scale, ranging from 0 (no symptoms) to 6 (dead).

RESULTS

Among 118 eligible patients, the mean age was 65.5 years, the mean time to enrollment was 5.5 hours, and 58% had a favorable penumbral pattern. Revascularization in the embolectomy group was achieved in 67% of the patients. Ninety-day mortality was 21%, and the rate of symptomatic intracranial hemorrhage was 4%; neither rate differed across groups. Among all patients, mean scores on the modified Rankin scale did not differ between embolectomy and standard care (3.9 vs. 3.9, P=0.99). Embolectomy was not superior to standard care in patients with either a favorable penumbral pattern (mean score, 3.9 vs. 3.4; P=0.23) or a nonpenumbral pattern (mean score, 4.0 vs. 4.4; P=0.32). In the primary analysis of scores on the 90-day modified Rankin scale, there was no interaction between the pretreatment imaging pattern and treatment assignment (P=0.14).

CONCLUSIONS

A favorable penumbral pattern on neuroimaging did not identify patients who would differentially benefit from endovascular therapy for acute ischemic stroke, nor was embolectomy shown to be superior to standard care. (Funded by the National Institute of Neurological Disorders and Stroke; MR RESCUE ClinicalTrials.gov number, NCT00389467.).