Study Design T-PODS follows a prospective cohort study design to achieve its objectives. A total of 2,000 patients will be recruited across 14 medical centers, with competitive enrollment. Data will be collected at baseline, 3 months, and 12 months post-enrollment through questionnaires, assessment scales, and imaging analyses to comprehensively determine the prevalence, risk factors, and long-term trends of various dimensions of post-stroke disability in the Taiwanese population. Eligible patients will be screened at each participating hospital. Enrollment will occur either during hospitalization for acute stroke or during follow-up outpatient visits after discharge. However, enrollment will not take place in the emergency department. While there is no strict minimum time required before enrollment after an acute stroke, patients should be recruited within one month of stroke onset, as soon as their general medical and neurological conditions have stabilized. The preferred baseline visit is between 7 and 30 days post-stroke, depending on the severity of the stroke. Follow-up evaluations will be conducted during outpatient clinic visits at 3 and 12 months post-stroke, with a permissible window of ±14 days for scheduling.
Assessment module The study assessment involves three primary module, questionnaires and scales assessment, brain MRI, and blood sample collection (Figure 1). Of them, questionnaires and scales assessment and blood sample collection will take place at baseline, 3 months, and 12 months visit, while brain MRI is arranged at baseline and 12 month visit. Questionnaires and Scales Assessment encompasses six domains, including motor, swallowing, cognition, depression, epilepsy, and quality of life. Each session of assessment takes approximately 40 minutes. Specific items include the following (Table 2).
If the participant has undergone a brain MRI at baseline, a follow-up brain MRI will be scheduled at one year post-enrollment (within a three-month window) to evaluate the impact of stroke on brain structure. Of note, all patients with diagnosis of ischemic stroke will require a baseline MRI before enrollment, while patients with diagnosis of ICH may only have CT at baseline. The MRI procedure will take approximately 40 minutes. The MRI scanning protocol is not strictly mandated and can be conducted according to the clinical routine of each hospital. However, it must include an iron-sensitive sequence (e.g., susceptibility-weighted imaging, T2\* or gradient echo). At baseline, a single 10 mL blood sample will be collected from a peripheral vein to conduct genetic and molecular biomarker studies related to stroke.
Besides the above disability assessment, brain MRI, and blood samples, the research team will complete a case report form for each participant at baseline, 3 months, and 12 months post-enrollment through medical record review to collect key relevant clinical information. The following paragraph will briefly introduce each assessment tool.
Motor function The National Institutes of Health Stroke Scale (NIHSS) is the globally recognized standard tool for quantifying stroke severity, ranged from 0 to 42, with higher scores indicating more severe stroke. It provides a comprehensive evaluation of neurological deficits and is widely adopted in clinical and research settings. The assessment will be conducted by trained personnel certified in NIHSS scoring to ensure consistency and reliability. The modified Rankin Scale (mRS) is universally utilized to classify post-stroke disability levels. It is also a key criterion for screening in Taiwan's Post-Acute Care (PAC) Stroke Program. This scale is effective for tracking functional outcomes after a stroke, with scores ranging from 0 (no symptoms) to 6 (death). The Modified Ashworth Scale (MAS) is the most commonly used tool in clinical practice for assessing muscle tone during passive movements. The evaluator passively moves the participant's relaxed limb and assigns a score based on the perceived resistance. This scale evaluates the resistance experienced during passive stretching and categorizes muscle tone severity on a scale of 0 to 4. Higher scores indicate increased muscle tone or spasticity. The scale is also a valuable tool for assessing treatment efficacy, with a one-grade reduction being considered clinically significant. The Functional Ambulation Categories (FAC) is a qualitative tool used to classify a patient's walking ability. It has been widely validated in retrospective studies for its reliability and validity and is frequently used in stroke rehabilitation research as a predictor of functional recovery. This tool does not require any equipment, making it quick, convenient, and easy to implement, and it also includes assessments for patients unable to walk. The scale categorizes patients into six levels, from 0 (unable to walk) to 5 (completely independent walking), based on observations and inquiries about their daily walking ability. The Barthel Index (BI) is primarily used to evaluate the ability of elderly or physically disabled individuals to perform daily activities and is a crucial tool for assessing individual self-care capabilities and determining the level of nursing care required. It includes 10 basic activities such as eating, dressing, bathing, transferring, and toileting. Each activity is scored based on the level of independence. The total score ranges from 0 to 100, with higher scores indicating greater ability. All of the motor disability evaluation will be conducted at baseline, 3 months, and 12 months post-stroke, except for the BI which will only be done at 12 months. Besides, only selected NIHSS items will be evaluated at 12 months, including consciousness (1a), aphasia (9), and dysarthria (10).
Swallowing function The Functional Oral Intake Scale (FOIS) assesses oral intake ability and categorizes patients into seven levels. Levels 1-3 indicate varying reliance on tube feeding, with limited or supplemental oral intake. Levels 4-6 reflect full oral intake, progressing from single textures to multiple textures with or without special modification or restrictions. Level 7 represents unrestricted oral intake. This scale provides a clear framework for evaluating and monitoring feeding abilities. The Repetitive Saliva Swallowing Test (RSST) evaluates swallowing function by number of well-performed swallowing behavior within 30 seconds. A score of three or more is considered normal, with fewer swallows indicating potential dysphagia. The test is simple, non-invasive, and widely used for screening and monitoring swallowing disorders.
Cognitive function The primary cognitive assessment tool used in this study is the Montreal Cognitive Assessment - Taiwan version (MoCA-T), a widely recognized screening tool for cognitive impairment and has been validated in Taiwan population. The MoCA-T evaluates nine key cognitive domains: attention, executive function, memory, language, visuospatial construction (e.g., clock drawing), abstract thinking, calculation, and orientation. The total score ranges from 0 to 30, with higher scores indicating better cognitive function. If the participant has received 12 years or less of formal education, 1 point will be added; however, the maximum total score should not exceed 30 points. A total score of 26 or above is considered normal. The MoCA-T will be administered at baseline, 3 months, and 12 months post-stroke to monitor cognitive changes over time and facilitate early detection and intervention for PSCID. At baseline, Informant Questionnaire on Cognitive Decline in the Elderly (IQCODE) is used to assess pre-stroke cognitive function.3 The IQCODE assesses cognitive decline over a 10-year period based on feedback from a caregiver or family member. It evaluates changes in areas like memory, problem-solving, and daily functioning, using a 5-point scale ranging from 1 (much improved) to 5 (much worse). Unlike self-reported tools, it relies on the informant's observations, providing a complementary perspective when the individual may lack insight into their condition. At 3 and 12 months post-stroke, participants with a MoCA-T score \< 26 (education level adjusted) will undergo a Clinical Dementia Rating (CDR) assessment to evaluate the presence and severity of dementia. The CDR will be administered by a qualified neuropsychologist to ensure accurate and reliable staging of cognitive and functional impairments. It evaluates six domains: memory, orientation, judgment and problem solving, community affairs, home and hobbies, and personal care, with scores ranging from 0 (no impairment) to 3 (severe impairment). A global score summarizes the overall severity, helping to classify patients as having no dementia (CDR = 0), very mild dementia (CDR = 0.5), or dementia of varying severity (CDR ≥ 1).
Depression Enrolled patients will undergo the Patient Health Questionnaire-9 (PHQ-9) assessment to evaluate the presence of depressive symptoms at baseline, 3-month follow-up, and 12-month follow-up. The PHQ-9 consists of 9 items, with a maximum score of 27. Scoring criteria are as follows: 0-9: no depression; 10-14: mild depression; 15-19: moderate depression; ≥20: severe depression. Since this is an observational study, data will be collected on the use of antidepressants and mood-stabilizing medications without influencing treatment decisions or prescriptions.
Post-Stroke Epilepsy A specialized questionnaire has been designed to collect the occurrence of epilepsy in stroke patients during the early hospitalization phase, at 3 months, and at 12 months post-stroke. The focus will include whether epilepsy events occurred, the timing of seizures, the nature of the episodes, diagnostic tests, treatments, and the impact on the patient.
Quality of Life T-PODS will evaluate the overall quality of life of stroke patients using the EuroQol 5-Dimension 5-Level Health Questionnaire (EQ-5D-5L). Patients will complete the questionnaire at baseline, 3 months, and 12 months post-stroke, with each session taking approximately five minutes. The EQ-5D-5L assesses five dimensions of self-reported health: mobility, self-care, usual activities, pain or discomfort, and anxiety or depression. Each dimension is rated on a scale from the worst to the best across five levels, generating a score for each dimension. Additionally, patients will rate their perceived health on the day of assessment using a visual analog scale (VAS), selecting a score between 0 (worst imaginable health) and 100 (best imaginable health). This dual approach provides a comprehensive evaluation of both specific aspects of quality of life and overall health status.
Brain MRI The suggested brain MRI protocol includes the following: structural scans with T1-weighted (3D MPRAGE on Siemens or SPGR on GE) imaging, T2-weighted FLAIR 3D SPACE (1.05x1.0x1.0mm³ voxels), susceptibility-weighted imaging (SWI; TE = 9.42 ms, 20 ms, iPAT = 2, 0.8x0.8x3.0mm³), MR angiography, diffusion weighted imaging (DWI; using 64 encoding directions with HARDI sequence), and arterial spin labeling (ASL). The DICOM files, after de-identification, will be sent to a designated imaging core laboratory and stored for analysis.
The location of stroke lesions themselves, as well as the presence of chronic brain conditions such as cerebral small vessel disease, significantly impact functional prognosis. Therefore, in this study, the enrolling physicians are asked to select the most recent brain imaging available at the time of enrollment-either CT or MRI, prioritizing MRI where possible. A brain imaging study with the highest quality will be selected and registered. The recorded data will include the location and evolution of brain infarctions, the location and size of brain hemorrhages, the ICH score, and markers of cerebral small vessel disease, such as white matter hyperintensities, lacunes, and cerebral microbleeds.
Apart from visual rating of brain MRI, advanced imaging biomarkers will be quantified using state-of-the-art analytical tools and artificial intelligence methodologies to derive insights into brain structure and function. Specifically, quantitative analysis will be performed to evaluate brain morphometry, such as cortical thickness and subcortical volume, using tools like FreeSurfer or FSL.45 These platforms enable detailed segmentation and volumetric measurements, providing precise data on structural changes in the brain. White matter tractography will be analyzed using DWI data processed by new tools which apply advanced fiber tracking algorithms to assess the integrity, directionality, and connectivity of white matter pathways. Recent advancements such as machine learning-enhanced tractography will be leveraged to improve the accuracy and reproducibility of fiber reconstructions, particularly in areas of complex crossing fibers. To explore emerging biomarkers, radiomics will be applied to extract high-dimensional features from imaging data, which can identify subtle texture, shape, and intensity patterns that correlate with functional recovery.48 Blood Biomarkers The genetic and molecular biomarker study in this project will collect one 10 mL EDTA tube (purple-top tube) of blood at three time points: baseline, 3-month follow-up, and 12-month follow-up, coordinated with routine blood draws. After centrifugation of blood samples at 3000 rpm for 20 minutes, the buffy coat and plasma will be collected and stored at -80°C. Processed blood specimens will then be sent to a central or genomic and proteomics research.
Purified DNA samples will be submitted to the National Center for Genomic Medicine according to specific submission standards. Genome-wide association study (GWAS) will be performed using the Axiom Genome-Wide TWB 2.0 Array Plate as the analytical template. By systematically reviewing GWAS studies related to post-stroke function published in PubMed or utilizing data from the GWAS Catalog and PRS Catalog, summary statistics from these studies will be used to obtain genetic association estimates for constructing polygenic risk scores. In the screening phase, 30 patients will be randomly selected for an untargeted screening platform to identify candidate molecular biomarkers with potential clinical relevance. In the validation phase, these candidate biomarkers will be further tested and analyzed
