Welcome to the WSC 2022 Interactive Program

Background and Aims

Susceptibility-weighted imaging (SWI) is not only sensitive in detecting haemorrhage but it is used to complement other MRI sequences. In the era of revascularization therapy during hyperacute cerebral ischemia, SWI can provide useful information in decision making. Our objective is to determine whether SWI correlates well with magnetic resonance angiogram (MRA) and FLAIR, intravascular thrombus and illustration of distal slow flow possibly depicting penumbra.

Methods

Retrospective case-notes of patients admitted to Brunei Neuroscience Stroke and Rehabilitation Centre with hyperacute cerebral ischemia (within 6 hours of symptoms onset) were selected from January to March 2021. All imaging were reviewed by a neuroradiologist to see if SWI correlated with MRA, FLAIR, thrombus and distal slow flow.

Results

25 patients presented with hyperacute cerebral ischemia. 24 (96%) showed SWI findings that correlated with MRA in demonstration of stenosis or occlusion in branch or territorial infarct but not in perforator infarcts. One case with a cortical infarct, MRA did not cover the area of high cortex. In 11 patients intravascular thrombus were evidenced in SWI and all had slow flow distal to the thrombus in FLAIR. SWI correlates with 10 (91%)of these patients for distal slow flow.

Conclusions

Aside from advances in stroke treatment, imaging-based selection of eligible patients for revascularization therapy in hyperacute cerebral ischemia is crucial to success. MRI imaging is increasingly being used and in our audit, not only was SWI as good as MRA in determining vessel status, it was capable to depict intravascular thrombus and distal blood flow of a vessel.

Background and Aims

Cerebral venous drainage is a potential indicator to stroke prognosis. However, existing clinical CT venogram (cCTV) presents limited collaterals. Thus we developed the cerebral CT venography techinques to help evaluate venous drainage after stroke.

Methods

Multimodal CT images of 12 stroke patients visited to Beijing Tiantan Hospital from May to July 2021 were retrospectively collected. Firstly, marked by patient’s skull, CT images at different time points were registed to reduce image artifacts. Next, cerebrovascular pixel points were segmentated based on cerebrovascular temporal response differing from other tissues. Then, time for contrast to arrive the pixel points was calculated according to cerebral arterial and venous temporal response, which was used to distinguish arteries and veins, to obtain the proposed CT venogram (pCTV). To assess qualities of pCTV, scores were assigned to overall image quality (1-5) and venous visibility scores, including dural sinus (0-16), superficial vein (0-11) and deep vein (0-9). Two blinded experienced neuroradiologists scored shuffled images.

Results

The example and scores of pCTV (M1) and cCTV (M2) were compared in Figures 1&2. The median [25th, 75th percentiles] total scores were 34.25 [33.50, 37.00] vs 20.50 [18.00, 29.00]. Overall image quality; dural sinus, superficial vein and deep vein scores were 3.50 [3.50, 4.00] vs 2.50 [2.00, 3.00], 14.00 [14.00, 15.50] vs 7.50 [7.00, 10.50], 9.00 [9.00, 10.00] vs 3.25 [2.00, 8.00] and 8.00 [7.50, 8.00] vs 6.75 [6.00, 7.00], respectively.

Conclusions

This study proposed the advanced cerebral CT venography techinques to support analyzing venous drainage after stroke and their correlations.

Background and Aims

Intracranial vessel wall magnetic resonance (MR) imaging (VW-MRI) has been utilized more in recent years. It is useful in further assessing the cause of ischemic stroke, especially in the young. Intracranial VW imaging has advantages over DSA, CT angiography, and MR angiography, as it can depict nonstenotic lesions and it can further characterize stenotic lesions that have already been detected with common angiographic methods.

Methods

This case series illustrates 3 different conditions demonstrating Moyamoya disease, Takayasu arteritis and atherosclerotic plaque using VW-MRI.

Results

Case 1: 34-year-old female without known cardiovascular risk factors with CT angiography showing smooth mural thickening along the entire length of the left internal carotid artery and occlusion of left middle cerebral artery (from M1 to proximal M2). WW-MRI demonstrated circumferential wall thickening and intense enhancement involving the left common carotid artery. She is treated with immunotherapy for Takayasu arteritis.

Case 2: 50-year-old, female without known cardiovascular risk factors, suspected to have Moyamoya disease. VW-MRI demonstrated circumferential enhancement with prominent collateral vessel formation. She is evaluated for eligibility for surgery.

Case 3: 51-year-old, female with hypertension. VW-MRI demonstrated juxta luminal T2 hyperintensity with eccentric wall enhancement at the stenotic area suggesting underlying atherosclerotic plaque. She is advised to optimize her risk factors.

Conclusions

Intracranial vessel wall magnetic resonance (VW-MRI) imaging is a non-invasive technique that is extremely useful in evaluating intracranial vasculopathies, especially for stroke in younger adults. This imaging modality is complementary to luminal imaging techniques and has a potential for guiding treatment and even longitudinal follow-up.

Background and Aims

Brain reserve relates to the ability of the unaffected brain to compensate for acquired brain dysfunction and may explain variable outcomes in acute ischemic stroke (AIS) patients. Subcortical atrophy, one of the imaging markers for impeded brain reserve, can be measured using the hemi-intercaudate distance (h-ICD) in AIS since brain edema on the affected side can be observed in some patients. Although the h-ICD is routinely assessed on MRI, CT is most widely used for AIS patients. We aim to compare CT and MRI in measuring the h-ICD for subcortical brain atrophy in AIS.

Methods

Data are from the ESCAPE-NA1 trial, a multicenter randomized study that assessed the efficacy of intravenous nerinetide in subjects with acute ischemic stroke who underwent endovascular thrombectomy within 12 hours from onset. Patients with an interpretable baseline CT and available follow-up MRI were included. The h-ICD and inner table width (ITW) were measured at the same level, and the intercaudate distance – inner table width (ICD-ITW) ratio was calculated. Scatter plots and Pearson correlations were analyzed.

Results

Among 1105 patients, 558 (50%) had interpretable baseline CT and follow-up MRI available for h-ICD measurement. The medians of h-ICD on CT and MRI were 7 mm (IQR 5-8 mm) and 7 mm (IQR 5-9 mm). The h-ICD and the ICD-ITW ratio had a strong correlation between CT and MRI (Pearson correlation r=0.760, R²=0.578, p<0.0001; r=0.745, R²=0.555, p<0.0001 respectively).

Conclusions

The h-ICD and the ICD-ITW ratio on CT correlate well with MRI in acute stroke patients.

Background and Aims

White matter hyperintensities (WMH) and lacunes or lacunar infarctions are among important imaging markers for brain reserve, known as the ability of the normal brain to compensate for an ischemic event such as in acute ischemic stroke (AIS) patients. They are routinely assessed on MRI. However, since CT is more widely used in the acute setting, we aim to compare CT and MRI in measuring WMH and lacunes in AIS patients.

Methods

Patients with an interpretable baseline CT and available follow-up MRI were included from the ESCAPE-NA1 trial, a randomized trial in AIS patients. Periventricular and deep WMH were measured using the Fazekas scale and summed to generate a total WMH score. Lacunes were scored using the definition of a round or oval subcortical infarction between 3 mm and 15 mm in diameter. Spearman rho analysis was conducted.

Results

Among 1105 patients, 564 (51%) had both interpretable CT and MR images. Periventricular WMH and deep WMH showed moderate correlations (Spearman rho 0.547, p<0.0001; Spearman rho 0.517, p<0.0001 respectively) between CT and MRI. Total WMH scores showed a strong correlation between CT and MRI (Spearman rho 0.621, p<0.0001). Lacunes showed a weak correlation between the two modalities (Spearman rho 0.364, p<0.0001).

Conclusions

Total WMH score can be calculated from the measurement on CT scan, which is available in the acute setting of stroke and correlates well with MRI, while the assessment of lacunes has a poor correlation between the two modalities.

Background and Aims

Brain frailty may diminish the ability of acute stroke patients to cope with ischemic injury irrespective of their chronological age, leading to impaired functional recovery. We aim to investigate the impact of brain atrophy, an important surrogate for brain frailty, on functional outcome after EVT.

Methods

We analyzed CT imaging data from the ESCAPE-NA1 trial. Cortical atrophy was visually assessed using the GCA-scale; subcortical atrophy was captured by measuring the intercaudate-distance (ICD). To account for any swelling, we measured the hemi-ICD, on the non-affected-side and multiplied this by 2. The primary outcome was 90-day-mRS (ordinal-shift-analysis). The secondary outcome was repeated-mRS-measures at different time points (pre-stroke, day 5, 30 and 90). Adjustments were made for age, sex, final infarct volume (FIV), laterality, white matter hyperintensity burden, Nerinetide-Alteplase interaction, and study site.

Results

Out of 1102 included patients, 818 (74%) had no-atrophy(GCA-0), 220 (20%) mild(GCA-1) and 64 (6%) moderate-severe(GCA-2/3). The median ICD was 14mm (IQR10-18). Both, GCA and ICD were significantly associated with 90-day-mRS (GCA-1 vs GCA-0: cOR 1.67, 95%CI:1.22-2.29; GCA-2/3 vs GCA-0: cOR 2.25, 95%CI:1.34-3.79; ICD: cOR[per mm increase] 1.05, 95%CI:1.02-1.08). Furthermore, mRS outcomes over the course of follow-up varied by the degree of atrophy (figure1). For example, the mRS for patients with GCA-1 was on average 0.42 points higher at 30-days (95%CI:0.18-0.66) and 0.72 points higher at 90-days (95%CI:0.49-0.95), compared to GCA-0.

Figure 1.

Conclusions

Brain atrophy is associated with less favorable outcome over time, in AIS patients treated with EVT within 12-hours of symptom onset.

Background and Aims

Stroke limited to the bilateral middle cerebellar peduncles (MCPs) are extremely rare. This study aimed to investigate the neuroimaging characteristics of ischemic stroke with a proclivity for the bilateral MCPs and to explore the underlying mechanism.

Methods

This retrospective study enrolled consecutive patients with ischemic stroke of bilateral MCPs in the Department of Neurology at the First Affiliated Hospital of China Medical University between January 2016 and March 2019. The demographic, clinical, and neuroimaging characteristics were evaluated.

Results

Nine patients with ischemic stroke were included. Their ages were from 54 to 86 (69±10.11) years, and five were men. All patients had isolated bilateral symmetrical MCP hypointensities on T1-weighted imaging and hyperintensities on fluid-attenuated inversion recovery (FLAIR) imaging and diffusion-weighted imaging (DWI) (Figure 1). Eight underwent angiography of the involved vessels. Among these, four patients demonstrated severe stenosis of the basilar artery (BA) near the origin of the anterior inferior cerebellar artery (AICA), which caused bilateral AICAs occlusion (Figure 2). Two patients revealed severe stenosis of the right vertebral artery and BA. Complete occlusion of bilateral vertebral arteries and the left vertebral artery occlusion with right vertebral artery dysplasia could be seen, respectively.

Figure 1 A 75-year-old female patient, with a history of hypertension, presented with vertigo and hearing impairment for three days.

Figure 2 Eight patients underwent angiography of the involved vessels.

Conclusions

Significant anatomical variations exist in the vasculature supplying the MCPs. Severe stenosis of the BA causing bilateral AICAs occlusion might be the mostly underlying cause of isolated bilateral MCP infarctions.

Background and Aims

In acute ischaemic stroke, perfusion imaging increases diagnostic accuracy and aids in treatment decisions, helping identify patients who may benefit from reperfusion therapy. particularly those presenting outside the conventional time window or in whom the time of symptom onset is unknown. Thrombectomy is typically offered to patients within 4.5 hours of symptom onset, or with evidence of salvageable tissue on perfusion imaging, and produces demonstrable improvements in independent function. The aim was to demonstrate the role of perfusion imaging in determining those patients in whom reperfusion therapy would derive the greatest benefit.

Methods

Data was analysed of stroke patients presenting to Watford General Hospital between 1st September 2020 and 31st August 2021. These patients were assessed for undergoing perfusion imaging and their subsequent treatment.

Results

Fifty-eight CT Perfusion scans were performed, leading to 46 stroke diagnoses. Fourteen patients underwent thrombectomy (alone or in combination with thrombolysis). The average pre-stroke mRS is 0.67 and the average post-stroke mRS on discharge is 4.3. The thrombectomy conversion rate for the centre is 4.8% which is approximately twice the average (2.5%) for most stroke centres in the UK.

Conclusions

Our conversion rate to thrombectomy has arisen from successful use of perfusion imaging. This data corroborates that perfusion imaging plays a significant role in the investigation of stroke patients, particularly in identifying those patients who will benefit from reperfusion therapy. CT perfusion imaging helps to define additional numbers who would otherwise have not had thrombolysis and/or thrombectomy.

Background and Aims

We investigated the impact of BBB disruption on length of hospital stay (LOS) for patients admitted with ischemic stroke.

Methods

This was a retrospective analysis of a de-identified dataset of stroke patients admitted to a single center over 10 years. Patients were included if they had an MRI with perfusion weighted imaging (PWI) with an area of hypoperfusion and had data on LOS. BBB disruption was measured from the PWI source images, detected as gadolinium leakage, and averaged within the perfusion deficit. BBB disruption was compared with other variables using linear and logistic regression.

Results

181 patients were included; the median age was 62 and 40% were women. More severe BBB disruption was associated with larger stroke volume (p<0.001), higher NIHSS (p<0.001), and longer LOS (p<0.001), but not with age (p=0.710) or sex (p=0.105). In the multivariate analysis, stroke volume (p=0.006) and LOS (p=0.045) but not NIHSS (p=0.271) were independently associated with BBB disruption. 55 patients (30%) were admitted to the hospital for more than a week with a mean BBB disruption of 3.91% compared with those less than week who had a mean BBB disruption of 2.97%. For every 1% increase in BBB disruption the likelihood of being in the hospital for greater than a week increased by almost 20% (OR 1.18, CI 1.03:1.35, p=0.015).

Conclusions

More severe BBB disruption is associated with longer LOS independent of the size or severity of the stroke. BBB disruption may represent a marker, or potentially a target, for treatments aimed at improving stroke outcomes.

Background and Aims

Angiographic "slow flow" in the middle cerebral artery (MCA), caused by carotid stenosis, may be associated with high oxygen extraction fraction (OEF). Furthermore, if the MCA slow flow is related to an MCA's reduced relative signal intensity (rSI) on MR angiography, the reduced rSI may be related to a high OEF. Therefore, we investigated whether the MCA slow flow ipsilateral to carotid stenosis was associated with a high OEF and aimed to create a practical index to estimate the high OEF.

Methods

We included patients who underwent digital subtraction angiography (DSA) and MRA between 2015 and 2019 to evaluate carotid stenosis. We evaluated MCA slow flow by image count using DSA, MCA rSI, minimal luminal diameter (MLD) of the carotid artery, carotid artery stenosis rate (CASr), and whole-brain OEF (wb-OEF). When MCA slow flow was related to a high wb-OEF, and we identified the determinants of MCA slow flow, we evaluated their association with high wb-OEF.

Results

One hundred and twenty-seven patients met our inclusion criteria. Angiographic MCA slow flow was associated with high wb-OEF. We identified MCA rSI and MLD as determinants of angiographic MCA slow flow. The upper limits of MCA rSI and MLD for angiographic MCA slow flow were 0.89 and 1.06 mm, respectively. The wb-OEF was higher in patients with an MCA rSI ≤0.89 and ipsilateral MLD ≤1.06 mm than patients without this combination.

Conclusions

The combination of reduced MCA rSI and ipsilateral narrow MLD is a feasible index of high wb-OEF.

Background and Aims

Animal experiments have demonstrated the dependency of cerebrospinal fluid clearance function on age and sleep, which partially underlay the cognitive decline in the elderly. However, evidence is lacking in humans mainly due to limited method to assess cerebrospinal fluid clearance function.

Methods

Consecutive T1-weighted and T2-fluid attenuated inversion recovery imaging were performed before and at multiple time points including 4.5 hours, 15 hours and 39 hours after intrathecal injection of contrast agent to visualize the putative meningeal lymphatic pathway, peri-olfactory nerve pathway and peri-optic nerve pathway. We defined the clearance function as the percentage change in signal unit ratio of critical locations in these pathways from baseline to 39 hours after intrathecal injection, and further analyzed their relationships with age, sleep and cognitive function.

Results

Cerebrospinal fluid clearance through the putative meningeal lymphatic and perineural pathways were clearly visualized in 85 patients. The clearance function of putative meningeal lymphatic and perineural pathways were reduced with aging (all p < 0.05). The clearance function through peri-olfactory nerve pathway was positive correlated with sleep quality and cognitive function (both p < 0.05). Moreover, the clearance function through peri-olfactory nerve pathway mediated the association of sleep quality with cognitive function (percent change in β [bootstrap 95% CI]: 35% [-0.220, -0.003]).

Conclusions

The impaired clearance through putative peri-olfactory nerve pathway may explain the cognitive decline in patients with dyssomnia. The study shows a promising method to assess cerebrospinal fluid clearance function of putative peri-neural pathways by dynamic magnetic resonance imaging with intrathecal injection of contrast agent.

Background and Aims

A 77-year-old man was electively admitted for laparoscopic adhesiolysis and wedge resection of recurrent hepatocellular carcinoma. Intraoperatively, he developed transient hypotension secondary to carbon dioxide embolism that lasted 30 minutes. Post-operatively, patient developed altered mental status. On examination, his Glasgow Coma Scale score was 6 and he had sluggish pupil reaction. There was no motor response in all limbs even with pain. Arterial blood gas was unremarkable.

Methods

Computed tomography (CT) of the brain showed air embolism in the right hemisphere. Magnetic resonance imaging (MRI) of the brain showed multiple haemorrhagic recent infarcts within both cerebral hemispheres. Transthoracic echocardiogram (TTE) revealed no right-to-left shunt. During the stay, he developed a severe diffuse encephalopathy and unfortunately he succumbed to his illness.

Results

Even in the absence of intracardiac abnormalities, paradoxical embolism can occur in hepatic patients undergoing liver resection because of abnormal arteriovenous communications in the pulmonary circulation. In addition, medical gases like carbon dioxide can also cause or aggravate this condition.

Conclusions

Symptoms of air embolism are hypotension, arrhythmia, cardiovascular collapse, and asystole . Intraoperatively, end-tidal CO₂ (EtCO₂) has been widely used to monitor for air embolism and significant decrease in EtCO₂ is sensitive in detecting air embolism. Transoesophageal echocardiogram (TEE), is the most sensitive in detecting air embolism. However, our patient was not well enough after the operation to undergo TEE . Despite the low overall incidence of air embolism during laparoscopic surgery, the consequences of this complication can potentially be fatal. Early diagnosis and treatment are warranted.

Background and Aims

To determine whether dual-layer detector cone-beam CT is non-inferior to CT to detect acute ischemic stroke.

Methods

A prospective single center clinical trial (NCT04571099) enrolled consecutive patients, 50 years or older, with ischemic or hemorrhagic stroke on initial CT. Patients were imaged with both CT and dual-layer detector cone-beam CT one day after admission. Three neuroradiologists blinded to clinical data evaluated clinical routine CT images and cone-beam CT 75 keV monoenergetic images in a single-sequence two-period crossover design. An ASPECTS-region overlay was fitted to both CT and cone-beam CT images to standardize scoring of infarct localization and extent. Non-inferiority was determined with the exact binomial test, with a one-sided lower performance boundary for diagnostic accuracy prospectively set to 80% (98.75% CI). Fourteen subjects were necessary to show non-inferiority compared to CT.

Results

28 patients were enrolled, and two patients were imaged twice (n= 30 matched image sets). Two cone-beam CT scans had severe motion artifacts, thus 28 matched image sets were used for analysis. The diagnostic accuracy to determine whether an ASPECTS region was ischemic or not was 96%, 95% and 91% (CI lower performance boundary of 93%, 92%, 88%) for each reader, respectively. The consensus analysis resulted in 95% accuracy (CI 93%).

Conclusions

Prospectively defined performance criteria were fulfilled. Dual-layer detector cone-beam CT in the angiography suite could be an alternative to CT in the setting of acute ischemic stroke to shorten door to thrombectomy time. However, there were many true negative ASPECTS regions and patients were imaged one day after admission.

Background and Aims

Cerebral digital subtraction angiography (DSA) remains the gold standard method for evaluating many cerebrovascular abnormalities .As per the ALARA principle (which stands for "as low as reasonably achievable"), reducing the dose delivered to both patients and staff must be a priority for endovascular therapists, who should monitor their practice. Only fewer studies are available, especially from rural centres from developing countries, and there is still a lack of proper research on the same.

Methods

A retrospective observational study on 113 patients who did diagnostic DSA in a period of 2 years. Angiograms analysed with radiation dose parameters: total, fluoroscopy time, air kerma, and dose-area product. Data entry was done using Excel and analysed using the SPSS 16 version software.

Results

Of the 125 patients studied, 113 had completed data. Stroke 63%(71) was the most common indication for DSA. Vascular abnormalities were detected in 97.3% of the procedures, which were mostly stenosis 56% (64) and aneurysm 21% (24), followed by AV malformations 5%(6), vasculitis 3% (3).

The average estimated radiation dose was 542.26 mGy. The average total, fluoro and acquisition times for the procedures were 32.5 , 18.2 , and 2.2 minutes. The mean total dose area product(DAP), fluoro and acquisition DAP were 50,21 and 26 Gycm2, respectively. The average frame used was 328.

Conclusions

Increased awareness of radiation risks and the establishment of strategies to reduce radiation dose help lower exposure. It is expected that each institute should take the initiative to establish the nominal radiation dose by doing studies on the same.

Background and Aims

Using multi-modal MRI evaluation and follow-up, we intended to identify the clinical and imaging factors associated with recanalization after cerebral venous thrombosis (CVT).

Methods

We retrospectively analyzed the information of a group of prospectively included CVT patients who were hospitalized between May 2020 to May 2022 in Beijing Tiantan Hospital, Capital Medical University. We included CVT patients who underwent high-resolution MRI within one month after diagnosis and had at least one follow-up MRI. The protocol of multi-modal high-resolution MRI included: DWI, MRV, 3D T1-SPACE-DANTE and SWI. Two independent raters who were blinded to clinical information analyzed thrombosis location, thrombosis burden, venous collateral status and recanalization based on high-resolution MRI. We investigated the factors associated with recanalization after CVT using multivariate logistic regression methods.

Results

A total of 33 patients were included. The mean age was 40.2 years (72 years as the oldest and 15 as the youngest), and 90.9% of the patients belonged to the Han Chinese ethnic group. There were 22 (66.7%) male patients. Thirty-two patients (97%) presented with headaches, 14 patients (42.4%) had blurred vision, 23 patients (69.7%) had focal neurological deficits, and 5 patients (15.2%) had seizures. Among the 32 patients (97%) who were prescribed anticoagulants during hospitalization, 14 patients received direct contact thrombolysis via endovascular approach. plus anticoagulation treatment In the multivariate logistic regression analysis, contact thrombolysis plus anticoagulation treatment was independently associated with recanalization.

Conclusions

Contact thrombolysis via endovascular approach based on anticoagulation treatment was associated with imaging recanalization after CVT.

Background and Aims

Identification of computed tomography (CT) thrombus imaging characteristics can predict the degree of recanalization and outcome after endovascular thrombectomy (EVT) in patients with acute ischaemic stroke and large vessel occlusion.

Aim: We analyzed the thrombus imaging characteristics and procedural factors and correlated with the degree of recanalization and functional outcome after EVT.

Methods

We evaluated the thrombus imaging characteristics (hyperdense MCA sign, thrombus location, length and permeability) from thin slice CT and CT angiogram. In addition, groin to recanalization time, number of passes, and EVT technique were documented. The primary outcome was degree of recanalization (mTICI score) and secondary outcome was modified Rankin scale (mRS) at 3 months.

Results

The mean age of 102 patients was 60.5±11.8 years. Patients with hyperdense MCA sign (90 % vs 75%, p=0.07) and permeable thrombus (86 % vs 70 %, p=0.09) had good recanalization (mTICI grade 2b,2c or 3). The requirement of <3 passes (90 % vs 62 %, p= 0.001) was associated with good recanalization. Multiple logistic regression analysis showed thrombus permeability (OR 5.9; 95% CI 1.3-26.6, p=0.02), use of stent retreiver alone (without aspiration) (OR 5.4; 95% CI 1.3-22.5, p=0.02) and a puncture to recanalization ≤60 minutes (OR 7.9; 95% CI 1.7-36.8; p=0.008) were associated with good recanalization. The requirement of ≥3 passes was associated with poor functional outcome (OR 3.4 ;95% CI 1.2-9.8; p=0.02).

Conclusions

Thrombus permeability was a predictor of successful recanalization after EVT. The requirement of three or more passes during EVT was associated with poor recanalization and poor functional outcome.

Background and Aims

History and examination:

A 42 years old lady presented with few days’ history of left arm weakness, right facial droop, ptosis and right eye abduction. She had miscarriages in the past and smoked seven cigarettes a day. No family history of cancers.

Methods

CT head showed low density areas in left basal ganglia, thalamus and midbrain. MRI head with contrast indicated bilateral thalamic lesions with partial ring enhancement.
She was discussed at the regional neuro-surgical MDT and a likely diagnosis of Glioblastoma Multiforme was given. Further investigations were planned at the tertiary centre and a grave prognosis was given.
Meanwhile, the patient was noted to have severe oral thrush by the ward registrar and HIV screening was ordered as her neutrophil count was 2.9. It was positive. CSF analysis showed elevated protein and lymphocytosis. Further CSF analysis was found to show DNA of Toxoplasma.

Results

A follow up MRI brain with contrast showed further increasing in the size of the previously noted lesion with extension to the upper midbrain and more vivid enhancement.

Treatment: After consultation with infectious disease and neuroradiologist, she was given a diagnosis of Toxoplasmosis and was commenced on Sulphadiazine, Pyrimethamine and retroviral agents. Follow up MRI contrast study after four weeks of treatment, showed reduction in size and enhancement of the intracranial lesions.

Conclusions

Summary:

Only 350 people develop neuro-toxoplasmosis in UK annually and hence it is important to consider this treatable disease as a differential in patients with ring enhancing lesions in basal ganglia, thalami and midbrain structures.

Background and Aims

To determine whether the delayed filling of the superficial middle cerebral vein (SMCV) could be reversed after reperfusion therapy in patients with acute anterior large vessel occlusion (LVO) and the effect of the reversal on the neurological prognosis.

Methods

Consecutive acute LVO (middle cerebral artery M1 ± intracranial internal carotid artery) patients between March 2019 and May 2020 were included. The delayed filling of SMCV is defined as the delayed image in the affected side compared with the normal side in the venous phase of computed tomography (CT) angiography based on CT perfusion reconstruction in the ischemic hemisphere. The modified Rankin scale (mRS) was used to evaluate the prognoses of these patients at 3 months after surgery.

Results

In the total of 54 patients, 47 (87.0%) patients presented with baseline SMCV delayed and 36 (76.6%) of these patients achieved SMCV reversal after reperfusion therapy. Successful reperfusion (TICI≥2b) was independently associated with SMCV reversal (OR=47.097, 95%CI=4.764-465.606, P=0.001). Multivariate regression analysis showed a significant association between baseline SMCV delay and 3-month poor outcome (OR=10.518, 95%CI=1.081-102.354, P=0.043). Compared with patients with sustained delayed SMCV, SMCV reversal did not show significant difference in the risk of 3-month poor outcome (OR=0.722, 95%CI=0.136-3.819, P=0.701).

Conclusions

In acute LVO patients, successful reperfusion can reverse the delayed filling of SMCV to normal. However, the SMCV reversal can not reverse the 3-month stroke prognosis.

Background and Aims

Aortic arch calcification (AoAC) is linked with plaque development as well as associated with cardiovascular events. We aimed to estimate the predictive value of AoAC in stroke recurrence in Embolic stroke of undetermined source (ESUS) patients.

Methods

Consecutive ESUS patients who admitted to our center between October 2019 and October 2020 and who had one-year follow-up of stroke recurrence were retrospectively reviewed. According to our AoAC grading scale (AGS), AoAC was classified into 4 grades based on chest CT: no visible calcification (grade 0), spotty calcification (grade 1), lamellar calcification (grade 2) and circular calcification (grade 3).

Results

Of 158 ESUS patients (age, 62.1±14.5 years; 120 males) enrolled, 24 patients (15.2%) had recurrent stroke within one-year follow-up. Cox regression analysis showed that, stroke history (hazard ratio, 4.625, 95% confidence interval, 1.828-11.700, P=0.001) and AoAC (hazard ratio, 2.672; 95% confidence interval, 1.129-6.319; P=0.025) predicted recurrent stroke. AGS grade 1 associated with a significant higher risk of stroke recurrence in comparison of AGS grade 0 (hazard ratio, 5.033, 95% confidence interval, 1.858-13.635, P=0.001), and AGS grade 2 plus 3 (hazard ratio, 3.388, 95% confidence interval, 1.124-10.206, P=0.030), respectively. In AoAC patients, ROC analysis showed that AGS had a 2good value for predicting stroke recurrence in ESUS patient with an AUC of 0.735 (95%CI=0.601-0.869, P=0.005).

Conclusions

AoAC, especially the spotty calcification, had good values for predicting stroke recurrence in ESUS patients.

Background and Aims

The association between collateral status and imaging and clinical outcome is unclear in posterior circulation ischemic stroke (PCIS). We investigated the effect of collateral status on clinical and imaging outcome in patients with acute PCIS.

Methods

Consecutive PCIS patients who were admitted within 24 hours since stroke onset were retrospectively reviewed from the database of INSPIRE registry. The collateral index (CI)=100% * hypoperfusion volume of Delay Time>6 seconds / hypoperfusion volume of Delay Time>2 seconds, which was then divided into good and poor collaterals based on the median CI. TICI≥2b or reperfusion rate ≥80% on follow-up CT perfusion were defined as successful reperfusion, otherwise as unsuccessful reperfusion.

Results

A total of 79 patients were finally analyzed, including 39 patients with poor collaterals at baseline. In follow-up, 21(26.6%) patients had hemorrhagic transformation (HT) and 58 (73.4%) patients had 3-months poor outcome (mRS scored 3-6). Compared with patients with unsuccessful reperfusion and good collaterals, patients with successful reperfusion and poor collaterals had a significantly higher risk of HT (P = 0.012), and a larger infarct growth (P = 0.002). However, poor collaterals had no association with 3-months poor outcome.

Conclusions

Poor collaterals was associated with HT and infarct growth in patients achieved successful reperfusion in acute PCIS.

Background and Aims

Patients with structural cardiac abnormalities had considerable delays in the contrast bolus reaching and transiting the brain vessels on computed tomography perfusion (CTP). We hypothesized that impaired left ventricular ejection fraction (LVEF) resulted in ischemic core overestimation.

Methods

Patients with large vessel occlusion in anterior circulation presenting within 16 hours after ischemic symptom onset who received initial CTP evaluation and achieved successful reperfusion were retrospectively analyzed. To determine the cardiac function, LVEF was dichotomized as either normal (≥50%) or decreased (<50%). Admission infarct core was considered as tissue with a relative reduction of cerebral blood flow <30%, as compared with contralateral hemisphere. Final infarct volume was measured on 24 to 48 hours non-contrast CT. Ischemic core overestimation was considered when CTP-derived core was larger than final infarct.

Results

133 patients were included into the analysis. Median CTP-derived core and final infarct volume were 16 mL (interquartile range, 8-33) and 24 mL (interquartile range, 16-43), respectively. Thirty-four patients (25.6%) presented ischemic core overestimation (median overestimation, 8 mL, interquartile range, 5-13). Multivariable logistic regression analysis showed that decreased LVEF (odds ratio, 7.48; 95% CI, 2.10-26.6, P=0.006) and earlier onset to imaging time (per 30 minutes earlier, odds ratio, 1.13; 95% CI, 1.02-1.24, P=0.012) were independently associated with core overestimation. A multivariable predictive model including all significant parameters yielded the highest diagnostic ability with an area under the curve of 0.81 (sensitivity 82.4%, specificity 71.7%).

Conclusions

Impaired left ventricular function induced higher rates of ischemic core overestimation on CTP, especially in patients in earlier window time.

Background and Aims

Subjects with acute ischemic stroke(AIS) require fast treatment to avoid major complications. CT Angiography(CTA) allows assessment of brain lesion extent to guide treatment decision-making. However, techniques such as CT Perfusion(CTP) enable brain hemodynamics visualization for improved treatment effectiveness. We evaluate StrokeSENS Multiphase CTA Perfusion(mCTAp), a post-processing software which outputs CTP-like maps, only requiring two additional low-dose scans to the ubiquituous CTA workflow. We compare mCTAp against CTP in its ability to predict abnormal hemodynamics extent.

Methods

528 subjects with baseline CTP/multiphase-CTA(mCTA)/non-contrast-CT(NCCT)/24h-36h-follow-up NCCT/DWI were included. 378 were used for development, remaining 150 for this evaluation. StrokeSENS mCTA Perfusion(mCTAp) and GE Perfusion-4D CTP were used to generate hemodynamic maps. We compared mCTAp versus CTP in terms of agreement of abnormal hemodynamic extent (lesion) at 10 ASPECTS regions via clustered AUC-ROC. Region-lesion overlap volume ratio was used as a score for the ROC curve, and the reference was positive if CTP region-lesion overlap>20%.

Results

Results yielded mCTAp-CTP agreement ranging from AUC=83%-AUC=99%. At lesion thresholds of TMAX>16.2s:AUC=92%; rCBF<20%:AUC=99%. Additional measurements are shown in Table 1 and image examples in Figure 1.

Figure 1. Visual comparison of mCTAp and CTP

Conclusions

In this analysis, we show significant spatial agreement of acute lesion (abnormal hemodynamics) thresholds between GE CT Perfusion 4D and StrokeSENS mCTA Perfusion.