Stroke
Conditions
Keywords
acute ischemic stroke, artificial intelligence, clinical decision support system, cluster randomized controlled trial, evidence-based performance
Brief summary
This study will conduct a cluster randomized controlled trial to evaluate the impact of an artificial intelligence-based clinical decision support system for the integrated management of patients with acute ischemic stroke on the adherence to guideline-based therapies and the incidence of new clinical vascular events.
Detailed description
Artificial intelligence (AI) and clinical decision support system (CDSS) have demonstrated great progress in the diagnosis and treatment of cerebrovascular diseases.CDSS can use computer technology to simulate and extend expert knowledge and empirical evidence in a timely and efficient manner. The combination of AI and CDSS is a potential solution to the shortage of medical resources and can improve the quality of and promote the standardization of medical services. At intervention sites, neurologists will receive support on use of the AI-based CDSS. Aim: To evaluate the effectiveness of an AI-based CDSS for stroke management in patients with acute ischemic stroke within 7 days of symptom onset. Intervention: An AI-based CDSS for an integrated management of patients with acute ischemic stroke. The strategy includes automatically identifying acute stroke lesions and lesion patterns, automated classification of stroke subtypes and mechanisms, evidence-based alerts, and guideline-recommended secondary stroke prevention strategies. Eighty eligible hospitals in china, stratified by hospital capacity (secondary grade or tertiary) and economic-geographical regions (eastern, central, and western), will be randomized into either the CDSS group or the usual care group.
Interventions
DEVICEAI-based CDSS
1. Automatically identifying acute ischemic stroke lesions on DWI. 2. Classification of stroke subtypes and mechanisms. 3. Evidence-based alerts and guidelines for early stroke management. 4. Guideline-recommended secondary stroke prevention strategies.
Sponsors
Beijing Tiantan Hospital
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
PREVENTION
Masking
SINGLE (Outcomes Assessor)
Eligibility
Sex/Gender
ALL
Age
18 Years to 90 Years
Healthy volunteers
No
Inclusion criteria
* Cluster Inclusion Criteria: * Secondary or tertiary hospitals with an emergency department and neurologic wards that receive patients with AIS. * Hospitals with available brain MRI scans (1.5T or 3.0T). * Patient Inclusion Criteria: * Patients of 18 years or older. * Patients admitted with neurological deficit consistent with ischemic stroke within 7 days of of symptoms onset. (\*Symptom onset is defined by the last seen normal principle) * Confirmation of new ischemic stroke by objective modality of CT scan and MRI (had relevant lesions on DWI). * Informed consent from patient or legally authorised representative (primarily spouse, parents, adult children, otherwise indicated).
Exclusion criteria
* Cluster
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Incidence rate of new clinical vascular events (ischemic stroke, hemorrhagic stroke, myocardial infarction, or vascular death) | 3 months | To evaluate the efficacy of AI-based CDSS in reducing the risk of new clinical vascular events (ischemic stroke, hemorrhagic stroke, myocardial infarction, or vascular death) at 3 months after initial symptom onset. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Incidence rate of new clinical vascular events (ischemic stroke, hemorrhagic stroke, myocardial infarction, or vascular death) | 6, 12 months | Incidence rate of new clinical vascular events (ischemic stroke, hemorrhagic stroke, myocardial infarction, or vascular death) at discharge, 6,12-months after initial symptom onset. |
| Disability | 3, 6, 12 months | Degree of disability (measured by the Modified Rankin Scale) at discharge, 3, 6, and 12 months |
| All-or-none measure of evidence-based performance measures | Participants will be followed for the duration of hospital stay, an expected average of 2 weeks | Proportion of prescription of evidence-based performance measures. All or none measures including the evidence-based therapies: antithrombotic medication within 48 hours after symptom onset, dual antiplatelet therapy (aspirin and clopidogrel) started within 24 hours after symptom onset (NIHSS score ≤3), intensive statin therapy, antihypertensive, hypoglycemic medications, dysphagia screening, and deep vein thrombosis prophylaxis within 48 hours of admission. Discharge therapies include: antithrombotics, lipid lowering agents, anticoagulants for atrial fibrilation or flutter, antihypertensive, and hypoglycemic medications. |
| A composite measure score of performance measures | Participants will be followed for the duration of hospital stay, an expected average of 2 weeks | The composite measure score was defined as the total number of eligible performance measures performed divided by the total number of performance measures for which a given patient was eligible. |
| All-cause mortality | 3, 6, 12 months | All-cause mortality |
Other
| Measure | Time frame | Description |
|---|---|---|
| All-cause readmission | 3, 6, 12 months | All-cause readmission |
| Ischemic stroke readmission | 3, 6, 12 months | Ischemic stroke readmission |
| Hemorrhagic stroke readmission | 3, 6, 12 months | Hemorrhagic stroke readmission |
| Moderate and severe bleeding events according to the GUSTO criteria | 3, 6, 12 months | Severe bleeding incidence (GUSTO definition), including fatal bleeding and symptomatic intracranial hemorrhage. |
| Adverse events | 3, 6, 12 months | Adverse events |
| Stratified analysis | 3, 6, 12 months | Efficacy endpoint will also be analyzed stratified by Hospital level(secondary hospitals/tertiary hospitals), economic-geographical regions (eastern, central, and western), stroke severity (NIHSS≤3/NIHSS\>3) and stroke subtypes. |
| Cardiovascular readmission | 3, 6, 12 months | Cardiovascular readmission |
| All bleeding events | 3, 6, 12 months | All bleeding events (severe/moderate bleeding and intracranial hemorrhage) |
| Intracranial hemorrhagic events | 3, 6, 12 months | Intracranial hemorrhagic events |
| Total costs of care | Participants will be followed for the duration of hospital stay, an expected average of 2 weeks | Total costs of care |
| Average length of stay | Participants will be followed for the duration of hospital stay, an expected average of 2 weeks | Average length of stay |
| Home time | 3, 6, 12 months | We calculated home time as total days alive and not in a hospital or skilled nursing facility. |
Countries
China
Contacts
Primary ContactZixiao Li, MD
Backup ContactLingling Ding, MD
Outcome results
None listed