Welcome to the ESOC 2021 Virtual Conference Calendar

Group Name

The International Cerebral Venous Thrombosis Consortium

Background And Aims

After cerebral venous thrombosis (CVT), 10-15% of patients die or become dependent in activities of daily living. To improve individualized patient education and treatment decisions, we developed models to predict individual risk of death and dependency after CVT.

Methods

We included patients from the International CVT Consortium. We used logistic regression to predict poor outcome at 6 months (CVT-PO-6, modified Rankin Scale 3-6) and Cox regression to predict 1-year mortality (CVT-MORT). Potential predictors derived from previous studies were selected with backward stepwise selection. Internal validation was assessed in terms of discrimination (c-statistic, 1000-bootstrap samples) and goodness-of-fit (calibration plots).

Results

Of 834 CVT patients, 125 (15.0%) were dependent or dead at 6 months and 62 (7.4%) died within 1 year. Odds ratios for CVT-PO-6 and hazard ratios for CVT-MORT are listed in the Table. C-statistic was 0.74 (95%CI 0.68-0.80) for CVT-PO-6 and 0.83 (95%CI 0.76-0.89) for CVT-MORT. Calibration plots showed good fit between predicted and observed values.

Conclusions

CVT-PO-6 and CVT-MORT show adequate performance for estimating individual risk of poor outcome and mortality after CVT. External validation is required before implementation in clinical practice by use of practical tools which predict individual risks.

Trial Registration Number

Not applicable

Background And Aims

Determining stroke etiology is crucial for optimal secondary prevention in young stroke patients. Unfortunately, cause of stroke remains unknown in many young patients according to TOAST classification, whereas the child- and adolescent stroke classification (International Pediatric Stroke Study) may close this diagnostic gap. Therefore we investigated the cause and prevalence of cryptogenic young stroke with IPSS criteria.

Methods

The ODYSSEY study prospectively included patients aged 18-50 years old with a first-ever, imaging-confirmed ischemic stroke or TIA between 2013 and 2021 in the Netherlands. All underwent complete work-up and etiology was classified according to modified TOAST classification and IPSS-criteria.

Results

We included 1322 patients (median age 44.2 years; 52.7% men, median NIHSS at admission 3.0), 105 had a TIA and 1217 had ischemic stroke. Stroke was due to likely or definite large vessel atherosclerosis in 17.4%, small vessel disease in 12.6%, cardio-embolism in 21.0%, rare causes in 21.0%, multiple causes in 5.9% and unknown cause (cryptogenic) in 25.2% of patients. According to IPSS criteria, 90.1% of patients with cryptogenic stroke had ≥1 traditional vascular risk-factors, 44.7% had a prothrombotic state, 29.1% had a chronic or acute head-neck condition, 15.6% a chronic or acute systemic condition and 4.8% had an arteriopathy. Only 3% of all TOAST-classified cryptogenic strokes, had no identifiable risk-factors when using IPSS-criteria.

Conclusions

We found that true cryptogenic stroke may be very rare or even non-existent. Using a dedicated classification significantly improved the detection of risk-factors and etiology of stroke in the young. This may allow for better individualized treatment and prognosis.

Trial Registration Number

Not applicable

Group Name

for the CADISP consortium

Background And Aims

The association between vascular risk factors and cervical artery dissections (CeAD), a leading cause of ischemic stroke in the young, remains controversial.

Methods

We applied a suite of genetic instrumental variable analysis (Mendelian randomization) in a two-sample setting, to study the association between vascular risk factors and CeAD using the summary statistics from the latest published genome-wide association studies of CeAD (N=1,393/14,416), blood pressure traits (N=757,601), lipids (N=60,812), type 2 diabetes (N=62,892/596,424), and Body Mass Index (BMI) (N=795,640).

Results

We found that genetically determined higher systolic blood pressure (SBP) and diastolic blood pressure (DBP) were significantly associated with an increased risk of CeAD of all, carotid and vertebral arteries (p=5.12x10^-5 to p=1.16x10^-16), the strongest effect being observed for DBP and all CeAD, with no indication of horizontal pleiotropy. Using GSMR and leveraging additional instruments we also found a significant association of genetically determined higher BMI with a lower risk of all CeAD (p=6.35x10^-3) and carotid CeAD (p=3.93x10^-3), while no significant association was observed between lipid levels or type 2 diabetes and CeAD. In a case-only analysis we observed a significant association of a weighted genetic risk score for DBP and LDL-cholesterol with increased risk of multiple vs. single CeAD.

Conclusions

Our results strongly suggest a causal association of higher SBP and DBP with increased CeAD risk (all locations) and multiple dissections. We also show weaker evidence for a causal relation between high BMI and lower all/carotid CeAD risk, and between high LDL and multiple vs. single CeAD risk.

Group Name

Y-ESUS investigators

Background And Aims

Cryptogenic strokes constitute ~ 30% of ischemic strokes in young adults (< 50 years), and most are considered to be embolic and of undetermined source. Two large RCTs, NAVIGATE-ESUS and RESPECT-ESUS, showed a high rate of recurrence (annualized rate ~5%) in older adults with ESUS. Although these trials provided us with valuable information about older ESUS population there is a gap in the knowledge regarding younger individuals with ESUS.

Methods

We prospectivwly enrolled 535 ESUS patients age < 50 years presenting to 41 sites in 13 countries between October 2017-October 2019. Stroke recurrence and detection of AF was ascertained at follow-up (median 16 months)

Results

Mean age was 40.4 +/- 7.3 years; 56% were male. Median NIHSS score on admission was 2 (IQR1-6). Current/former tobacco use (45%), followed by hypertension (22%) and dyslipidemia (20%) were the most common vascular risk factors. The majority of patients were discharged on antiplatelet therapy (84%) and less commonly anticoagulation (16%). Recurrence rate of stroke was 1.92 per 100-patient years and rate of death was 0.4 per 100-patient years. Detection of AF was low at 2.6% during follow up. On multivariate analysis, history of stroke/TIA [HR 5.27(1.82-15.29)] and presence of diabetes [HR 4.44(1.52-12.93)] were predictive of recurrent stroke.

Conclusions

In this large cohort of Young ESUS patients, the rate of recurrent ischemic stroke was comparably low than in older patients. Prior stroke or TIA, and diabetes mellitus were independently associated with risk of ischemic stroke recurrence.

Trial Registration Number

NCT03185520

Group Name

The KIDCLOT group of investigators:

Manoelle KOSSOROTOFF, Celine BELLESME, Christian DENIER, Peggy REINER, Catherine LAMY, Sandrine DELTOUR, Michele LEVASSEUR, François LUN, Hassan HOSSEINI, Adrien VILLAIN, Joanna BELLEVILLE GOFFENEY, Alessandra BIONDI, Gaultier MARNAT, Pauline RENOU, Marie THIBAUD, Nathalie BACH, Ganaelle REMERAND, Yannick BEJOT, Benoit DELPONT, Sylvie JORIOT, Cecile LAROCHE, Maryline CARNEIRO, Omer EKER, Frederique AUDIC, Pierre MEYER, Caroline ARQUIZAN, Claire BILBAULT, Sebastien RICHARD, Emmanuelle GONDON, Pascal AUZOU, Guillaume CAMI, Canan OZSANCAK, Jean-Philippe NEAU, Nicolas RAYNAUD, Lena DAMAJ, Francois EUGENE, Aude TRIQUENOT-BAGAN, Alexandra PEREZ, Valerie WOLFF, Remy BEAUJEUX, Emmanuel CHEURET, Jean DARCOURT, Maximilien PERIVIER

Background And Aims

Background: Safety and efficacy of recanalization treatments, i.e. intravenous thrombolysis (IVT) and/or mechanical thrombectomy (MT), have to be refined in pediatric stroke.

Methods

Methods: The French KIDCLOT multisourcing nationwide retrospective study (NCT03887143) collected data of consecutive children aged >28 days and <18 years old with acute arterial ischemic stroke who had a recanalization treatment, during a 41 month-period.

Results

Results: 63 patients, 41 boys, median age at stroke onset 11.3 years old, were included. Pre-stroke mRS was 0 for 90.5%. Main stroke etiologies were cardio-embolic (31.7%), focal cerebral arteriopathy (FCA, 23.8%), cervical (carotid/vertebral) arteriopathy (17.4%), and thrombotic (12.7%). 64 pediatric procedures were recorded: IVT alone (n=28), MT alone (n=21), combined IVT and MT (n=13), and DSA without recanalization treatment (n=2). Median stroke onset to first recanalization treatment delay was shorter in adult stroke units than in pediatric wards (209 vs. 252 minutes respectively, p=0.03). An early complication was reported for 4 procedures, including two significant intracranial hemorrhages. After MT, cardio-embolic stroke patients achieved good arterial recanalization more frequently than FCA patients (88.9% vs 33.3%). Although high initial severity markers (median NIHSScore=13, decreased consciousness in 52.4%, decompressive craniectomy in 17.2%), outcome was relatively good in the 60/63 survivors (median mRS=1 and PSOM=1.5 at 12 month-follow-up).

Conclusions

Conclusions: This study provides encouraging safety data despite severe stroke presentations. It suggests potential ways to improve recanalization treatments efficacy in children: reducing management delays in pediatric wards and optimizing etiological orientation since the hyperacute phase, as cardio-embolic strokes seem to display a better recanalization rate profile.

Trial Registration Number

NCT03887143

Group Name

the DECOMPRESS2 study group

Background And Aims

Decompressive neurosurgery is recommended in patients with cerebral venous thrombosis (CVT) and impending brain herniation. We aimed to evaluate predictors of mortality in a large prospective cohort of CVT patients treated with decompressive craniectomy.

Methods

DECOMPRESS2 was a multinational prospective register of consecutive patients with CVT (2011-2020) treated by decompressive neurosurgery in which 118 patients were included (80 women, median age 38 years, 115 craniectomies, 37 hematoma evacuations). Outcomes were evaluated at discharge, 6 and 12 months.Twelve-month follow up data was available for 113 patients (96%). Causes of death were adjudicated and Firth penalized maximum likelihood logistic regression was performed to identify predictors of mortality.

Results

Forty patients (34%) died after decompressive neurosurgery, of whom 25 (21%) during the first 30 days. The median time between diagnosis and surgery was 1 day (IQR 0-2). Median time between CVT diagnosis and death was 14 days, and between surgery and death, 9 days. The most common causes were brain herniation (n=17, 43%) and infection (n=12, 30%). Independent predictors of death were age (OR [10 years] 1.52; 95%CI:1.06-2.18), posterior fossa lesion (OR 9.28; 95%CI:1.33-64.52), bilateral dilated fixed pupils before surgery (OR 6.47; 95%CI:1.47-28.34) and bilateral decompressive surgery (OR 7.51; 95%CI:1.40-40.37).

Conclusions

Most deaths occurred within the first 30 days after surgery. Older age, posterior fossa lesion, bilateral fixed pupils before surgery and bilateral decompressive surgery were independent predictors of death. Improving patient selection and optimizing the timing of surgery may further improve the outcome of these severely affected patients with CVT.

Trial Registration Number

Not applicable

Background And Aims

Anecdotal reports suggest a relation between dural arteriovenous fistulas (dAVF) and cerebral venous thrombosis (CVT). We aimed to investigate the anatomical and temporal relationship between the two conditions.

Methods

We used data from a consecutive cohort of adult patients with dAVF from Amsterdam University Medical Centers between 2007 and 2020. An experienced interventional neuroradiologist reevaluated the presence and characteristics of dAVF and CVT on all available imaging. Among patients who had both diagnoses, we assessed the temporal (previous/concurrent/subsequent) and anatomical (same/adjacent/unrelated venous sinus or vein) relation between CVT and dAVF.

Results

Of 178 included patients with dAVF, 85 (48%) were female and median age was 61 years (IQR 51-68). Fifty-nine (33%) patients had CVT. Four patients (7%) were diagnosed with CVT before dAVF diagnosis, 33 (56%) concomitant with dAVF diagnosis, and 22 (37%) developed CVT after dAVF diagnosis (median follow-up time 9 months, IQR 4-23. The incidence rate of CVT after dAVF diagnosis was 97 per 1000 person-years (95%-CI 62-145). In 47 (80%) of the patients with both dAVF and CVT, thrombosis was located in the same venous sinus as the dAVF, and in an additional 9 (15%) CVT was located in an adjacent venous sinus.

Conclusions

CVT occurred in one-third of all patients with dAVF and could both precede or develop after dAVF diagnosis. In 95% of cases there was an anatomical relation between dAVF and CVT. These data support the hypothesis of a bidirectional relationship between the CVT and dAVF.

Trial Registration Number

Not applicable