EAN 2022: highlights and expert commentary

What were the key findings presented at the 8th Congress of the European Academy of Neurology (EAN), held in Vienna, Austria, on 25-28 June? Here, two leading specialists provide commentary on important studies, highlighting the implications for care.

Dr Ambra Stefani (Postdoctoral Research Fellow at the Massachusetts General Hospital, Boston, USA) and Professor Angelo Antonini (Professor of Neurology at the University of Padua, Italy) comment on advances in headache, stroke, multiple sclerosis, neuro-ophthalmology, dementia, sleep, pediatric neurology, Guillain-Barré syndrome, epilepsy, Parkinson’s disease, CANVAS, and myotonic dystrophy type 1.

For more details of the new findings presented at EAN 2022, see Neurodiem’s daily coverage of the event.



Attendees at EAN 2022 learned how a machine-learning based approach can predict anti-calcitonin gene-related peptide (CGRP) response in patients with migraine.

Machine-learning approaches were used for features selection to generate a prediction model of 30%, 50%, and 75% response to anti-CGRP therapies at 6 months. A 30%, 50%, and 75% response rate was achieved by 60%, 43%, and 17% of patients, respectively, at 6 months

Dr Stefani commented: “This study has relevant clinical implications, although they might be applicable only to female patients (as 93% of the migraine patients included were women). Through a machine-learning based approach, demographic and headache variables predict response to anti-CGRP, enabling a first step towards individualized treatment.

“Potentially, based on this approach, different phenotypes of migraine patients could be identified, allowing in the future more effective treatment approaches.”



Specialists at EAN 2022 reported on the direct association between susceptibility vessel sign (SVS) status and the presence of active malignancy in acute ischemic stroke (AIS) patients who underwent mechanical thrombectomy (MT).

The absence of SVS was strongly associated with:

  • Active malignancy: adjusted odds ratio [aOR]: 4.85; 95% confidence interval [CI]: 1.94–12.11
  • Occult malignancy alone: aOR: 11.42; 95% CI: 2.36–55.20

Dr Stefani commented: “Based on the results of this study, when MRI is performed in acute ischemic stroke, SVS status should be evaluated. In the absence of SVS, an active and occult malignancy should be suspected. Thus, considering SVS status may improve detection of malignancy in acute ischemic stroke.”

“Considering SVS status may improve detection of malignancy in acute ischemic stroke”

Dr Ambra Stefani

Also at EAN, new data indicated that the establishment of mechanical thrombectomy offers the possibility to enhance the etiological determination of stroke by histological and immunohistochemical analysis of retrieved thrombotic material (see box).

Analysis of retrieved thrombotic material


Clot histology

  • Mixed (n=123): 62%
  • Thrombocyte/fibrin-rich (n=45): 23%
  • Erythrocyte-rich (n=18): 9%


Clot etiology

  • Cardio-embolic (n=87): 44%
  • Arterio-embolic (n=37): 19%
  • Undetermined (n=26): 13%



Dr Stefani commented: “Histological analysis of thrombotic material obtained through mechanical thrombectomy allows distinction between erythrocyte-rich and thrombocyte/fibrin-rich thrombi, which are associated with arterio-embolic and cardio-embolic stroke, respectively.

“Performance of this analysis whenever possible would likely help in identifying stroke etiology, particularly in patients with embolic strokes of undetermined source. In these patients, the presence of thrombocyte/fibrin-rich clots should increase suspicion of cardio-embolic origin and may lead to consequent changes in diagnostic and treatment algorithms.”

“Performance of this analysis whenever possible would likely help in identifying stroke etiology”

Dr Ambra Stefani

As part of the Sleep Deficiency and Stroke Outcome Study, researchers reported that blood pressure variability (BPV) and, to a lesser extent, heart rate variability (HRV), predict future cerebro-cardiovascular events (CCVE) in acute stroke patients.

Both systolic blood pressure (SBP) and diastolic blood pressure (DBP) variability were associated with an increased risk of CCVE:

  • SBP-standard deviation (SD): odds ratio (OR) 1.70; 95% confidence interval (CI), 1.17–2.48; P=0.005
  • DBP-SD: OR, 1.41; 95% CI, 1.02–1.96; P=0.037

Professor Antonini commented: “There are limited data on BPV and HRV as potential predictors of future cerebro-cardiovascular events in acute stroke patients. This study analyzed 92 subjects who suffered a cerebro-cardiovascular event within 3 years, out of 437 subjects who suffered a stroke. Best predictors were systolic BP (OR 1.7) and non-linear HRV (OR 1.26), indicating that BP control is critical to prevent recurrence of acute vascular events following the first stroke.”

“BP control is critical to prevent recurrence of acute vascular events following the first stroke”

Professor Angelo Antonini

Multiple sclerosis

In multiple sclerosis (MS), serum glial fibrillary acidic protein (sGFAP) was highlighted as an important biomarker of disease progression.

Researchers analyzed levels of sGFAP and serum neurofilament light chain (sNfL) – quantifiers of neurodegeneration and astrocytic activation – in 259 patients within 6 months of a confirmed Expanded Disability Status Score of 3 or more. The patients were followed up for a median of 7.6 years.

After adjusting for age and sex, the researchers reported that:

  • sGFAP was positively correlated with a higher risk of 6-month confirmed disease progression (6mCDP; hazard ratio 1.64, P=0.006)
  • The correlation between sGFAP and 6mCDP was stronger in patients with low levels of sNfL (HR 2.30, P=0.006), although sNfL alone was not prognostic for 6mCDP
  • sNfL was increased in patients with disease activity after baseline (mean 12.5 pg/mL versus 11.7 pg/mL in non-active disease; P=0.036)

The findings suggest that, in this cohort of patients with progressive MS, higher levels of sGFAP were an indicator of progression, while higher sNfL levels reflected acute disease activity, the researchers concluded. Both could be used to more effectively stratify patients in research and clinical trials, they added.

Dr Stefani commented: “Markers of disease activity and progression in people with progressive MS are of utmost importance. This study assessed sGFAP, showing that it indicates disease progression in the next 6 months. This serum marker might be useful not only for its application in clinical trials, but potentially also to timely improve treatment strategies in people with progressive MS.”

Professor Antonini also commented on this study: “These results are important. A combination of both serum markers, that is relatively easy to test, could be used in the clinic to identify MS patients who have active disease and identify those who are at lower risk of progression.”

“This serum marker [sGFAP] might help improve treatment strategies in people with progressive MS”

Dr Ambra Stefani


Retinal layer thinning after optic neuritis (ON) may be useful as a predictor of future relapse remission in relapsing MS, according to a new study presented at EAN 2022.

In multivariable analyses, incomplete remission of non-ON relapse was found to be predicted by ganglion cell inner plexiform layer (GCIPL) thinning.

Dr Stefani commented: “Based on the results of this study, OCT can be a reliable instrument to predict future relapse remission in relapsing MS. After optic neuritis, optical layer thinning predicted future relapse remission.

“Future studies should assess if treatment strategy changes based on OCT results after optic neuritis improve long-term patient outcomes. If so, this fast exam may be implemented in treatment decision trees.”

“OCT can be a reliable instrument to predict future relapse remission in relapsing MS”

Dr Ambra Stefani


Attendees a EAN 2022 learned that Huntingtin intermediate alleles (IAs) in association with age and apolipoprotein E (APOE) ε4 increase the risk of progression to mild cognitive impairment (MCI) in patients with subjective cognitive decline (SCD):

  • 44 106 patients with SCD (41.51%) progressed to MCI and 62 patients (58.49%) did not
  • The rate of progression to MCI was associated with IAs, age at baseline, and APOE ε4

Dr Stefani commented: “Markers distinguishing people with subjective cognitive decline progressing to mild cognitive impairment (and eventually dementia) from those not converting are highly needed. Based on this study, Huntingtin intermediate alleles are promising as such markers.

“If their association with risk of progression to mild cognitive impairment is confirmed, they might be used in clinical practice, for example in memory clinics, to stratify patients according to risk of progression, and in the future might be included in diagnostic and treatment decision trees.”

“Huntingtin intermediate alleles might be useful in clinical practice to stratify patients according to risk of progression”

Dr Ambra Stefani

Meanwhile, plasma pTau 181 was highlighted as a favorable non-invasive and feasible diagnostic tool to detect patients early in the neuropathological cascade of Alzheimer’s disease (AD).

Researchers reported that the median concentration of plasma pTau 181 was 1.4, 2.0, 2.4, and 3.4 pg/ml in healthy controls, subjective cognitive decline (SCD), mild cognitive impairment (MCI), and AD, respectively.

Dr Stefani commented: “The suggested diagnostic tool is non-invasive and feasible. The presented results deriving from a large cohort suggest potential clinical applications of plasma pTau in the diagnosis of Alzheimer’s disease and its prodromal stages.

“This is particularly of interest in the light of the increasing focus on prevention of neurodegeneration in order to improve brain health.”



The specificity and positive predictive value of rapid eye movement (REM) sleep behavior disorder (RBD) screening questionnaires were found to be low to very low, in a prospective study presented at the EAN congress.

Patients in the study completed the RBD screening questionnaire (RBDSQ), RBD single question (RBD1Q), and Innsbruck RBD inventory. The sensitivity, specificity, and accuracy were:

  • 79.3%, 47.3%, and 50.3% for the RBDSQ
  • 75.9%, 66.1%, and 67% for the RBD1Q
  • 89.7%, 54.6%, and 57.9% for the Innsbruck RBD inventory
  • 96.6%, 33.3%, and 39.4% for the three questionnaires combined

Dr Stefani commented: “Previous smaller studies and this large video-polysomnography based one by the SINBAR (Sleep Innsbruck Barcelona) group showed that the performance of RBD screening questionnaires is low outside the context of validation studies. If such questionnaires are used in clinical practice, general neurologists should be aware of the high number of false positives and be careful if communicating questionnaire results to patients.

“If diagnosed with ‘probable’ RBD based on questionnaires, patients may get worried about developing alpha-synucleinopathy despite not having a definite diagnosis of RBD, thus careful counselling is of utmost importance in this context.”

“General neurologists should be aware of the high number of false positives when using RBD questionnaires, and carefully communicate questionnaire results to patients.”

Dr Ambra Stefani

In other research presented at EAN 2022, specialists reported that disturbed sleep is a risk factor for cognitive decline in the elderly. Sleep efficiency, wake after sleep onset, and fragmentation index (P<0.01) predicted follow-up memory performance, in both unadjusted and adjusted models.

Professor Antonini commented: “Several studies have indicated that there is an increased risk of Aβ deposition with reduced night-time sleep duration, and that this may occur early before patients manifest cognitive impairment or significant Aβ deposition. This study evaluated prospectively a group of elderly individuals followed for 18 months.

“The results indicate that sleep efficiency, wake after sleep onset, and fragmentation index predicted follow-up memory performance, supporting the concept of disturbed sleep as risk factor for cognitive decline.”


Pediatric neurology

In an innovative study, researchers developed a child-like robot Pepper for social skills training in children with social deficits. At the EAN congress they presented promising findings on children’s interactions with Pepper during the neurorehabilitation of social deficit.

Observational data showed that children became quickly engaged (within up to 3 minutes) with the robot. Survey results revealed a likability of 98%, while perceived safety was 83%, perceived intelligence was 81%, and anthropomorphism was 64%.

Dr Stefani commented: “New technologies and artificial intelligence are part of modern medicine. The approach described her