Intracranial Hemorrhage Brain Image Non-rigid Registration from Real-world Dataset to Reference Space

AbstractMachine learning (ML) holds great promise in transforming healthcare. While published studies have shown the utility of ML models in interpreting medical imaging examinations, these are often evaluated under laboratory settings. The importance of real world evaluation is best illustrated by case studies that have documented successes and failures in the translation of these models into clinical environments. A key prerequisite for the clinical adoption of these technologies is demonstrating generalizable ML model performance under real world circumstances. The purpose of this study was to demonstrate that ML model generalizability is achievable in medical imaging with the detection of intracranial hemorrhage (ICH) on non-contrast computed tomography (CT) scans serving as the use case. An ML model was trained using 21,784 scans from the RSNA Intracranial Hemorrhage CT dataset while generalizability was evaluated using an external validation dataset obtained from our busy trauma and neurosurgical center. This real world external validation dataset consisted of every unenhanced head CT scan (n = 5965) performed in our emergency department in 2019 without exclusion. The model demonstrated an AUC of 98.4%, sensitivity of 98.8%, and specificity of 98.0%, on the test dataset. On external validation, the model demonstrated an AUC of 95.4%, sensitivity of 91.3%, and specificity of 94.1%. Evaluating the ML model using a real world external validation dataset that is temporally and geographically distinct from the training dataset indicates that ML generalizability is achievable in medical imaging applications.

Download Full-text

Abstract 1122‐000068: Basilar Artery Occlusion Thrombectomy Between Evidence Based Medicine and the Real‐World Practice, Single Centre Experience

Stroke: Vascular and Interventional Neurology ◽

10.1161/svin.01.suppl_1.000068 ◽

2021 ◽

Vol 1 (S1) ◽

Author(s):

Islam El Malky ◽

Ali Hendi ◽

Hazem Abdelkhalek

Keyword(s):

Clinical Outcome ◽

Intracranial Hemorrhage ◽

Basilar Artery ◽

Real World ◽

Artery Occlusion ◽

Basilar Artery Occlusion ◽

Early Reperfusion ◽

Good Outcome ◽

Symptomatic Intracranial Hemorrhage ◽

Multiple Variables

Introduction : BAO (basilar artery occlusion) is well known by catastrophic outcomes whether death or disability in approximately 70 %. 1 Thrombectomy as an intervention in large vessel occlusion of anterior proximal circulation was approved after multiple RCTs and meta‐analyses. 2 In spite of two RCTs that appeared lately, there is still uncertainty about the effect of thrombectomy in BAO. 9, 10 Our study aims to report the outcome of BAO, as a further clue of MT effectiveness in BAO and variables affecting good outcome and mortality rate. Methods : We retrospectively collected the clinical and radiological data of 30 BAO patients treated in our center between 2016 and 2020. There is no limitation as regard age or presenting NIHHS. Twenty‐two patients who came to the emergency within 4.5 hours had I.V. thrombolytic therapy (73.3%). A favorable clinical outcome was considered if mRS ≤ 2. Angioplasty, stenting, or I.A thrombolysis were used as a rescue treatment. Symptomatic intracranial hemorrhage within two days after the initiation of treatment and mortality at 90 days were reported. The radiological outcome was assessed by modified Thrombolysis in Cerebral Infarction (mTICI) score where mTICI ≥ 2b or 3 at the end of the intervention was considered a favorable radiological result. Multiple variables were tested for their effect on favorable clinical outcomes and mortality (Table 1). Results : Among 30 patients, the mean age was 61.23 ± 16.81 years; 20/30 (66.7%) male. A favorable functional outcome was achieved in (40.7%). Successful revascularization was achieved in 26 patients (86.7 %). Four patients had procedural complications (13.3%). Symptomatic intracranial hemorrhage occurred in three cases (11%) and mortality at 90 days was 11 patients (36.7 %). The presenting NIHSS is the only predictor of mortality and the optimal cut‐off value for death was 15 with AUC = 0.758 (sensitivity 91 % and specificity 59%) and p‐value = 0.02. TOR (time of onset to recanalization) had no effect on the clinical outcome which is controversy with the paradigm of early reperfusion leading to a good outcome Conclusions : In spite of two RCSs approved no statistical difference between medical treatment and thrombectomy, thrombectomy is still an effective procedure in real‐world practice in selected cases. The presenting NIHSS is the only predictor of mortality in our studies. More studies are warranted to discover other predictors of BAO thrombectomy outcome to improve case selection and avoid futile recanalization.

Download Full-text

Abstract 1122‐000199: Real‐World Bridging with Intra‐Arterial Therapy is Safe After Tenectplase in Patients with Acute Ischemic Stroke

Stroke: Vascular and Interventional Neurology ◽

10.1161/svin.01.suppl_1.000199 ◽

2021 ◽

Vol 1 (S1) ◽

Author(s):

Sindhu Sahito ◽

Hemal Patel ◽

Nasar Ali ◽

Teye Umanah ◽

Siddhart Mehta ◽

...

Keyword(s):

Ischemic Stroke ◽

Acute Ischemic Stroke ◽

Intracranial Hemorrhage ◽

Real World ◽

Subgroup Analysis ◽

Hemorrhagic Transformation ◽

Small Subset ◽

Significant Difference ◽

Ct Head ◽

Ich Score

Introduction : The safety profile of IV tenecteplase (TNK) as a bridging therapy to Intra‐arterial therapy (IAT) is not well‐established in patients receiving acute ischemic stroke therapy. Our objective was to evaluate the incidence of hemorrhagic transformation in subjects who received TNK followed by IAT outside of clinical trial setting. Methods : Electronic medical records of subjects with stroke secondary to LVO who received TNK and IAT within 4.5 hours of last known normal were reviewed. CT head within 24 hours post‐TNK was evaluated for hemorrhagic transformation (HT). Severity was determined by ECASS III criteria Symptomatic intracranial hemorrhage was defined as an increase in NIHSS greater than or equal to 4. Clinical outcomes were assessed with NIHSS at admission, discharge and mRS scores at one month. Z score population proportions were used for subgroup analysis. Social Science Statistics was used for data analysis. Results : From October 2020 to April 2021, 20 subjects received IV tenecteplase. Four subjects did not have LVO or undergo IAT and were excluded from study. Four subjects (25%) developed hemorrhagic transformation. Of this subset, 2 subjects (12.5%) had asymptomatic HI‐1, 1 subject had symptomatic HI‐2, and 1 subject had asymptomatic PH‐1. One subject developed intracranial hemorrhage (ICH score 5) outside of stroke region (intraventricular, subarachnoid, infratentorial parenchymal) without evidence of hemorrhagic transformation of ischemic stroke. In subgroup analysis between subjects with HT and without HT, there was no statistically significant difference in intra‐arterial non‐thrombolytics (z = 0.1393, p = 0.44433); there was a trend towards significance in number of passes (z = 1.2534, p = .10565) and periprocedural IV heparin use for intracranial stenting and/or angioplasty (z = 0.9342, p = 0.17619). There was a statistically significant increase of HT when periprocedural IV integrilin (z = 1.6727, p = 0.04746) was used. Conclusions : Our small subset of early real‐world experience demonstrates a higher rate of symptomatic transformation in bridging with TNK when compared with alteplase. Larger prospective studies are needed to validate our findings.

Download Full-text

Abstract 12465: Dabigatran in Real World of Atrial Fibrillation: Systematic Review and Meta-analysis of Comparison Studies With Vitamin K Antagonists

Circulation ◽

10.1161/circ.132.suppl_3.12465 ◽

2015 ◽

Vol 132 (suppl_3) ◽

Author(s):

João Carmo ◽

Francisco M Costa ◽

Jorge Ferreira ◽

Miguel Mendes

Keyword(s):

Myocardial Infarction ◽

Atrial Fibrillation ◽

Systematic Review ◽

Intracranial Hemorrhage ◽

Vitamin K ◽

Major Bleeding ◽

Real World ◽

Observational Studies ◽

Meta Analysis ◽

Vitamin K Antagonists

Background: In the clinical trial RE-LY, dabigatran showed a better efficacy/safety profile in comparison with warfarin, but clinical trials are few representative of the real world. We aim to access if dabigatran in real-world patients with atrial fibrillation (AF) showed a better profile in comparison with warfarin, through a systematic review and meta-analysis of observational studies comparing with vitamin K antagonists. Methods: PubMed, Embase and Scopus databases were searched through December 2014. We include observational studies comparing dabigatran to warfarin for non-valvular AF that reported clinical events during a follow-up for dabigatran 75mg, 110 mg or 150 mg, and warfarin. We proceeded to the extraction and analysis of data for clinical thromboembolic events, bleeding and mortality. Data were pooled by meta-analysis using a random-effects model. Results: We selected 9 studies involving a total of 291,703 patients, 85,399 treated with dabigatran and the remaining 206,304 with warfarin. The incidence of stroke was 1.71 / 100 patient-years for dabigatran and 2.44 / 100 patient years for warfarin (relative risk [RR] 0.91, 95% CI 0.66 to 1.27, p=0.58). The major bleeding rate was 3.90 / 100 patient-years for dabigatran and 3.92 / 100 patient years for warfarin (RR 0.90; 0.78 to 1.03, p=0.11). The all-cause mortality (RR 0.81, 0.75-0.88, p<0.001) and intracranial hemorrhage (RR 0.45, from 0.27 to 0.76, p=0.002) were significantly lower in patients treated with dabigatran in comparison to those treated with warfarin. There were no significant differences in risk of myocardial infarction (RR 0.55; 0.29 to 1.07, p=0.08), total hemorrhage (RR 1.00; 0.57 to 1.77, p=0.99), and gastro-intestinal bleeding (RR 1.14; 0.78 to 1.69, p=0.50). Conclusions: In this combined analysis of observational studies of real world, dabigatran compared to warfarin was associated with a similar risk of stroke, myocardial infarction, major bleeding, total bleeding and gastrointestinal bleeding, and a lower risk of intracranial hemorrhage and mortality.

Download Full-text

Thrombectomy Versus Combined Thrombolysis and Thrombectomy in Patients With Acute Stroke

Stroke ◽

10.1161/strokeaha.120.031599 ◽

2021 ◽

Author(s):

Xu Tong ◽

Yilong Wang ◽

Jens Fiehler ◽

Clayton T. Bauer ◽

Baixue Jia ◽

...

Keyword(s):

Endovascular Treatment ◽

Intracranial Hemorrhage ◽

Real World ◽

Odds Ratio ◽

Large Vessel ◽

Practice Setting ◽

Modified Rankin Scale ◽

Large Vessel Occlusion ◽

Vessel Occlusion ◽

Symptomatic Intracranial Hemorrhage

Background and Purpose: A recent randomized controlled trial DIRECT-MT (Direct Intra-Arterial Thrombectomy to Revascularize AIS Patients With Large Vessel Occlusion Efficiently in Chinese Tertiary Hospitals) compared the safety and efficacy of mechanical thrombectomy (MT) versus combined intravenous thrombolysis (IVT) and MT for acute large vessel occlusion. The current study utilized a prospective, nationwide registry to validate the results of the DIRECT-MT trial in a real-world practice setting. Methods: Subjects were selected from a prospective cohort of acute large vessel occlusion patients undergoing endovascular treatment at 111 hospitals from 26 provinces in China (ANGEL-ACT registry [Endovascular Treatment Key Technique and Emergency Work Flow Improvement of Acute Ischemic Stroke]) between November 2017 and March 2019. All patients eligible for IVT and receiving MT were reviewed and then grouped according to whether prior IVT or not (MT and combined IVT+MT). After a 1:1 propensity score matching, the outcome measures including the 90-day modified Rankin Scale, successful recanalization, door-to-puncture time, symptomatic intracranial hemorrhage, and intraprocedural embolization were compared. Results: A total of 1026 patients, 600 in the MT group and 426 in the combined group, were included. Among 788 patients identified after matching, there were no significant differences in the 90-day modified Rankin Scale (median, 3 versus 3 points; P =0.82) and successful recanalization (86.6% versus 89.3%; P =0.23) between the two groups; however, patients of the MT group had a shorter door-to-puncture time (median, 112 versus 136 minutes; β=−45.02 [95% CI, −68.31 to −21.74]), lower rates of symptomatic intracranial hemorrhage (5.5% versus 10.1%; odds ratio, 0.52 [95% CI, 0.30–0.91]), and embolization (4.6% versus 8.1%; odds ratio, 0.54 [95% CI, 0.30–0.98]) than those of the combined group. Conclusions: This matched-control study largely confirmed the findings of the DIRECT-MT trial in a real-world practice setting, suggesting that MT may carry similar effectiveness to combined IVT+MT for acute large vessel occlusion patients, despite MT alone seems to be associated with a shorter in-hospital delay until procedure, lower risks of symptomatic intracranial hemorrhage, and embolization. Registration: URL: https://www.clinicaltrials.gov ; Unique identifier: NCT03370939.

Download Full-text

Abstract 261: Applying Clinical Trial Data to Real-World: Apixaban, Dabigatran, and Rivaroxaban

Circulation Cardiovascular Quality and Outcomes ◽

10.1161/circoutcomes.7.suppl_1.261 ◽

2014 ◽

Vol 7 (suppl_1) ◽

Author(s):

Alpesh Amin ◽

Michael Stokes ◽

Ning Wu ◽

Elyse Gatt ◽

Dinara Makenbaeva ◽

...

Keyword(s):

Atrial Fibrillation ◽

Clinical Trials ◽

Relative Risk ◽

Clinical Outcome ◽

Intracranial Hemorrhage ◽

Major Bleeding ◽

Real World ◽

Randomized Clinical Trials ◽

Absolute Risk ◽

The Real

BACKGROUND: Data from randomized controlled trials and a real-world sample of non-valvular atrial fibrillation patients were combined to estimate the absolute effect of each new oral anticoagulant (NOAC, apixaban, dabigatran, and rivaroxaban) versus warfarin on stroke and major bleeding rates in real-world clinical practice. METHODS: Non-valvular atrial fibrillation patients were selected from Medco healthplans during 2007-2010. Reference rates for stroke and major bleeding excluding intracranial hemorrhage (to avoid double counting) were calculated for real-world Medco patients during warfarin use. Real-world event rates for NOACs were estimated by multiplying the corresponding relative risk from the randomized clinical trials by each reference rate. Absolute risk reductions and numbers needed to treat (NNT) or numbers needed to harm (NNH) for each NOAC vs. warfarin were then estimated. Reduction in net clinical outcome was calculated by summing the absolute risk reductions for stroke and major bleeding excluding intracranial hemorrhage for each NOAC versus warfarin. RESULTS: Each NOAC resulted in a reduction in stroke events compared with warfarin in the real-world (TABLE). Apixaban was the only NOAC to reduce the rate of major bleeding excluding intracranial hemorrhage compared with warfarin. The NNT to avoid one net clinical outcome (stroke plus major bleeding excluding intracranial hemorrhage) per year was 32 and 84 for apixaban and dabigatran, respectively. Rivaroxaban resulted in an increase in net clinical outcome (NNH=166). CONCLUSIONS: If relative risk reductions from randomized clinical trials persist in the real-world, apixaban would result in the greatest clinical benefit versus warfarin of all NOACs in terms of stroke and major bleeding excluding intracranial hemorrhage events avoided.

Download Full-text