S 11 - My scientific paper in the field of neuroimaging

Purpose:

The DAWN trial has shown that patients with acute stroke who had a mismatch between clinical deficit and infarct volume could benefit from endovascular thrombectomy even 24h after the onset of stroke. As the assessment of the patients’ neurological deficits is by definition a subjective assessment, it raises the need for an objective marker of salvageable brain tissue. Perfused brain slices prove an ex-vivo model of the brain in which the perfusion can be controlled. In this study, we were interested in monitoring instantaneous changes in lactate dehydrogenase (LDH) and pyruvate dehydrogenase (PDH) activities in perfused rat brain slices by hyperpolarised [1-¹³C] pyruvate MRS. The slices were exposed to an ischaemic insult which was achieved by arrested-perfusion. In this model, [1-¹³C] pyruvate delivery is not hindered by the blood-brain-barrier and therefore its metabolism by brain cells can be determined directly.

Methods and materials:

Brain slices were produced from Sprague-Dawley rats (n=11) and placed in an NMR tube perfused with oxygenated aCSF. In the control group (n=6), well-oxygenated [1-¹³C] pyruvate was administered continuously. In the arrested-perfusion group (n=5), the perfusion was stopped 30 seconds before the injection of [1-¹³C] pyruvate. LDH and PDH activities were quantified by measuring the production of [1-¹³C] lactate and [¹³C] bicarbonate with product-selective-saturating-excitations that were designed to quantify enzymatic rates.

Results:

During 94 seconds of ischaemic insult, LDH activity increased by 56±25% and PDH activity decreased by 51±16% (n=5). No significant changes in LDH nor PDH activities were observed in the control group (n=6).

Conclusion:

Hyperpolarised [1-¹³C] pyruvate MRS shows promise in identifying the dynamically changing ischaemic penumbra as it could detect rapidly and non-invasively temporal changes in cerebral metabolic activity following ischaemia.

Limitations:

Further research is needed to test this approach in-vivo.

Ethics committee approval

The study was approved by the Hebrew-University Institutional-Animal-Care

Funding:

Horizon-2020.

Purpose:

Cerebral vasospasm is the major cause of morbidity and mortality in patients with ruptured cerebral aneurysms who survived an initial haemorrhagic event. A recent randomised clinical trial validated the significant benefit of intra-arterial treatment over conservative treatment. However, risk-benefit-profiles of the drugs used for pharmaceutical angioplasty have not yet been analysed. Our study compares the efficacy and side-effects of intra-arterial nimodipine versus milrinone in patients treated for symptomatic vasospasm.

Methods and materials:

35 patients were treated intra-arterially for severe cerebral vasospasm. 12 received nimodipine only, 6 received milrinone only, and 3 were treated with a combination of both. Totally, 144 procedures were reviewed, 82 finally included. Spasm-relief, dosage and duration of application, mean-arterial-pressure (MAP), the requirement for systemic vasopressor therapy, and functional outcomes after 3-6 months were reviewed.

Results:

Both drugs strongly increased the calibre of MCA- and ACA-segments. Less markedly, they reduced the vasotonus of the intradural ICA-segments (192% vs 120%). The time required to achieve significant spasm-relief was much higher for nimodipine compared to milrinone (6mg:55min vs 20mg/20min). The effect on the vasotonus after milrinone application was dose-dependent (maximum increase in diameter: 187% vs 166%, 20mg vs10mg). Milrinone showed the maximum effect after 10 minutes, whereas nimodipine efficacy appeared in a more delayed manner. Comparing the compensatorily required vasopressor dosages to avoid critical drops in MAP and nimodipine demanded significantly more systemic vasopressor than milrinone (0.08+/-0.13μg/kg/min vs 0.07+/-0.07μg/kg/min). All except for one patient (nimodipine) developed DCI but unfavourable outcomes were lower in the nimodipine cohort (mRS mean 3 vs 4.2).

Conclusion:

The immediate vaso-relaxing-effect of nimodipine is inferior to milrinone, but nimodipine may have additional neuroprotective effects. However, cardiovascular comorbidities, an initial Hunt and Hess scale and individual susceptibility to pharmaceutical agents, are most considerable confounders.

Limitations:

A retrospective, unicentric study with a small sample size.

Ethics committee approval

Local IRB granted.

Funding:

No funding was received for this work.

Purpose:

To summarise our experience from the last decade with stenting of symptomatic intracranial stenosis (SIS) causing acute, progressive stroke.

Methods and materials:

Our retrospective study included 71 interventions (66 patients, mean age: 68.5 years) with average stenosis of 76% (modified NASCET-calculation). Balloon-expandable-stents (BES: Liberté, Pharos, Rebel, Coroflex) and self-expandable-stents (SES: Enterprise, Solitaire, Wingspan, Leo+Baby, p48MW_HPC, Silk Vista Baby) were applied. In total, 86 stents were used, but only 69 were implanted with adequate success. 77.3% of patients presented in the context of acute progressive stroke. Lysis (rtPA) was initiated in 12 patients.

Results:

77.3% were successfully treated. In 60.6%, BES, in 22.7%, SES+PTA ,and in 7.6%, other combinations (BES+SES+balloon, low-profile flow diverter) were used. In 9.1%, proper vessel-reconstruction failed.

77.3% regained at least partial functional independence (90days-mRS: ≤3), however, 22.7% died despite successful intervention. Except for one patient, all remaining presented with initial mRS-scores>3. 66.7% of deaths were related to poorly-collateralised basilar stenosis and 33.3% had high-grade stenosis at the C7/M1 segment.

Conclusion:

SIS, especially in the posterior circulation, bears an extraordinarily high-risk for mortality and permanent disability despite early causative treatment. In this condition, collateralisation and comorbidities are most significant for the individual outcome. Off-label use of BES yields a good success-to-risk ratio, however, considerable stiffness and restricted manoeuvrability of BES empirically renders implantation impossible in demanding cases. Application of SES+balloon-PTA represents a promising bail-out-technique (especially with low-profile-flow-diverters) in these situations. Hence, considering the devastating natural course of SIS and accounting for the substantial evolution of neuro-interventional techniques which revolutionised stroke-treatment in the post-SAMMPRIS era, endovascular treatment of SIS has become obligatory especially, but not exclusively, during acute interventions as thrombectomy. Yet asymptomatic high-grade intracranial stenosis should be evaluated critically before deciding for conservative treatment only.

Limitations:

Only single-centre experiences were presented.

Ethics committee approval

n/a

Funding:

No funding was received for this work.

Purpose:

This audit assessed the local consenting practice in interventional radiology using a patient feedback form provided by the RCR. The questionnaire contained 14 questions looking at different aspects of consenting and recorded the patient’s satisfaction in terms of yes, no, and not sure. The target was for 95% of patients answering ‘yes’ to all questions. In the first cycle, 30% of patients answered ‘Yes’ to all questions. Areas needing improvement included providing information leaflets and providing information about pain and procedure benefits. The action plan after the first cycle included providing feedback to the radiologists to improve awareness and providing leaflets in advance.

Methods and materials:

In the second cycle, 33 patients filled out a questionnaire form within 48 hours of their procedure. The anonymised data was quantitatively analysed.

Results:

36% of patients answered ‘Yes’ to all questions, showing a 6% increase. 90% answered yes to 10 out of 14 questions when analysed separately, confirming good progress. 100% of patients understood their procedure and had enough information to give consent. 55% answered yes to having sufficient information regarding alternative treatment.

Conclusion:

A separate analysis of questions showed improvements in most areas, although the overall improvement was small. The worst outcome was seen in the question regarding discussing alternative treatments. Arguably, the day of the procedure is unsuitable for these conversations due to time limitations and a lack of complete history. It is assumed these conversations have taken place by the referring clinical team. Actions to take forward include revision of the questionnaire; we feel the RCR set questionnaire is not well suited to our practice. We intend to analyse these results alongside the relevant consent form to look for discrepancies and a repeat audit is prudent.

Limitations:

N/A

Ethics committee approval

N/A

Funding:

No funding was received for this work.

Purpose:

Diabetic polyneuropathy (DPN), being one of the most common and most burdening complications in patients with diabetes mellitus, is still poorly understood. Magnetic resonance neurography (MRN) allows for the detection of pathological changes in the tibial compartment of the sciatic nerve. The aim of this study is to combine MRN with clinical, serological, and electrophysiological patient assessments in order to achieve a better understanding of the pathophysiology and possible treatment options in DPN.

Methods and materials:

MRN of the sciatic nerve at 3 Tesla (Siemens Magnetom TIM TRIO) was performed in 130 patients with diabetes mellitus using a T2-weighted, fat-suppressed sequence with the following parameters: Tr: 5970ms, TE: 55ms, FOV: 160x150 mm^2, matrix size: 512x512, slice thickness: 4mm, interslice gap: 0.35 mm, voxel size: 0.5x0.3x4.0 mm³)

Segmentation of the tibial compartment of the sciatic nerve was carried out manually using the software Image J (version 1.52o, 2019) while nerve lesions were quantified semi-automatically with MATLAB (version 7.14.0.0739 (R2012a)).

In addition, every patient underwent clinical, electrophysiological, and serological assessment. MRN and clinical data were subsequently correlated using GraphPad Prism (Version 8).

Results:

At the time of abstract submission, preliminary results of 42 patients were available. The cross-sectional surface area (CSA) correlated negatively with tibial nerve conduction velocity (NCV) (r=-0.37, p<0.01) and negatively with tibial compound motor action potential (Amp) (r=-0.37,p<0.019).

Conclusion:

The preliminary results described above indicate that the thickening of nerve fascicles seen in DPN patients results in poorer nerve conduction values (NCV, Amp) implying advanced nerve damage. With all 130 patients being analysed, we expect more correlations of nerve pathologies with clinical and serological data.

Limitations:

A cross-sectional study.

Ethics committee approval

Ethics number: S-383/2016, clinicaltrials.gov identifier: NCT03022721.

Funding:

No funding was received for this work.

Purpose:

To assess differences in visual cortex BOLD signal between myopic and emmetropic subjects.

Methods and materials:

fMRI EPI was performed using a block-design stimulation in 6 emmetropic (3 men) and 6 myopic (3 men) healthy young volunteers. Two medical students manually mined the following data in a commercial workstation: highest BOLD magnitude and cross-correlation coefficient (CCC) using a 9-voxel ROI, whole cluster BOLD magnitude/CCC, total activated pixels, and number of activated slices. Variables were compared between groups (Mann-Whitney U test) and correlated (Spearman’s correlation) with age, myopia duration, and dioptres in myopic subjects. One of the students was further trained in imaging analysis within the Erasmus Program. He performed a two-sample t-test group comparison and multiple regression modelling dioptres and myopia duration as regressors within SPM software.

Results:

Total activated pixels (P=0.046) were significantly higher in myopic subjects. Significant correlations were observed between the highest BOLD magnitude and number of activated slices (ρ -0.928; P=0.008), the highest CCC and total activated pixels (ρ 0.943; P=0.005), and the total activated pixels and number of activated slices (ρ 0.943; P=0.005) only in myopic subjects. In addition, the highest BOLD magnitude (ρ 0.986; P<0.001) and whole cluster CCC (ρ -0.899; P=0.015) were correlated with dioptres. Imaging analysis showed four significant voxels in the occipital lobe in myopic>emmetropic and none in myopic<emmetropic. Some voxels correlating positively to myopia duration and negatively to dioptres were also demonstrated.

Conclusion:

Myopic may have more extensive activation than emmetropic subjects, which might be a first adaptive mechanism. Increasing peak activation might be a subsequent compensation.

Limitations:

The sample size, short block length, and lack of refraction correction.

Ethics committee approval:

Ethics committee approval waived.

Funding:

No funding was received for this work.