Respiratory dysfunction in severe stroke, stemming from its unique and complex central-peripheral pathophysiological mechanisms, poses significant challenges to conventional respiratory support and rehabilitation strategies. In clinical practice, the need to maintain consciousness and intracranial homeostasis restricts conventional assessment of respiratory function and mechanical ventilation weaning potential, which restricts the implementation of early rehabilitation. To overcome these challenges, precision assessment techniques such as diaphragm ultrasound and esophagus pressure measurement, combined with ultra-early neurocritical care bundles and intelligent ventilation modes like neurally adjusted ventilatory assist, are driving a therapeutic paradigm shift in the management of severe stroke-related respiratory dysfunction from passive life support to active functional rehabilitation. This paper systematically reviews the core challenges and key technological breakthroughs in respiratory rehabilitation for severe stroke, aiming to provide a theoretical reference for the optimization of clinical practice and the identification of innovative research directions in this field.

Objective  To investigate the effects of different manually assisted coughing (MAC) techniques on cough function, sputum clearance ability, and pulmonary function in patients with brainstem stroke, and to provide clinical evidence for the rehabilitation intervention of post-stroke cough dysfunction.
Methods  A prospective randomized controlled design was used. Patients with brainstem stroke who were hospitalized at the Department of Sports Rehabilitation, Beijing Xiaotangshan Hospital from May 2023 to May 2024 were consecutively enrolled and randomly divided into the diaphragmatic stimulation group, the tracheal stimulation group, and the tongue root stimulation group. All three groups received routine rehabilitation therapy and respiratory training. On this basis, each group received diaphragm stimulation, tracheal stimulation, or tongue root stimulation, respectively. The intervention was performed twice daily, 30 minutes per session, 5 days per week for 2 consecutive weeks. Before and after intervention, peak cough flow (PCF) was measured using a pulmonary function tester, and 24-hour sputum volume was recorded. Static pulmonary function indices, including forced vital capacity (FVC), forced expiratory volume in one second (FEV1), and peak expiratory flow (PEF), were tested using a cardiopulmonary exercise testing system. Ultrasonic examination was used to evaluate diaphragmatic parameters, including diaphragm thickness at end of inspiration (TEI), diaphragm thickness at end of expiration (TEE), diaphragm thickness fraction (DTF), and diaphragmatic mobility (DM). The incidence of pneumonia after intervention was recorded.
Results  A total of 33 patients with brainstem stroke were enrolled, with 11 patients in each group. After intervention, PCF, FVC, FEV1, PEF, TEI, TEE, and DTF were significantly increased, and 24-hour sputum volume was significantly decreased in all three groups compared with those before intervention (P<0.05). Pairwise comparisons between groups after intervention showed that the tongue root stimulation group achieved greater improvement in 24-hour sputum volume than the diaphragmatic stimulation group [(35.64±10.71) mL vs. (46.09±9.52) mL, P=0.025]. FVC, FEV1, and PEF in the tongue root stimulation group and the tracheal stimulation group were better than those in the diaphragmatic stimulation group (all P<0.05). There were no statistically significant differences in PCF, TEI, TEE, and DTF among the three groups after intervention. The incidence of pneumonia after intervention was 36.4% in the diaphragmatic stimulation group, 18.2% in the tracheal stimulation group, and 9.1% in the tongue root stimulation group, with no statistically significant difference among groups (P=0.439).
Conclusions  Diaphragmatic stimulation, tracheal stimulation, and tongue root stimulation— three MAC techniques—all effectively improve cough function, pulmonary function, and sputum clearance in patients with brainstem stroke. Tongue root stimulation yields superior effects in eliciting cough reflex and promoting sputum excretion.

Objective  To compare the effects of respiratory training, external diaphragm pacing (EDP), and repetitive peripheral magnetic stimulation (rPMS) on diaphragmatic function in patients with cerebral infarction, evaluate the clinical efficacy of EDP and rPMS, and provide a reference for clinical application intervention of diaphragmatic dysfunction after cerebral infarction. 
Methods  Patients with cerebral infarction admitted to the Department of Sports Rehabilitation, Beijing Xiaotangshan Hospital from January to December 2022 were consecutively enrolled and randomly divided into a control group, an EDP group, and an rPMS group. The control group received respiratory training, the EDP group received EDP at the superficial position of the phrenic nerve, and the rPMS group received high-frequency magnetic stimulation at the spinous process of C7. The intervention protocol for all three groups was 20 minutes per session, once a day, five days a week for four consecutive weeks. The changes in diaphragm thickness, diaphragm mobility (DM), phrenic nerve motor conduction, static pulmonary function, thoracic mobility, and Sheikh trunk control test (TCT) of the three groups were observed and compared before and after intervention. The effects of the three intervention methods on various diaphragmatic indexes were analyzed.
Results  A total of 36 patients with cerebral infarction were enrolled, with 12 cases in each group. Four patients dropped out during the intervention, including one case in both the control group and the rPMS group, and two cases in the EDP group. Finally, 32 patients completed the entire research process. Intragroup comparison showed that after intervention, the diaphragm thickness at the end of inspiration (TEI), diaphragm thickening fraction (DTF), DM, compound muscle action potential (CMAP) amplitude, forced vital capacity (FVC), forced expiratory volume in first second (FEV1), FEV1/FVC, peak expiratory flow (PEF), thoracic mobility, and TCT were significantly higher in the three groups, while the phrenic nerve conduction time (PNCT) was significantly shorter in the three groups compared with those before intervention (all P<0.05). There was no significant change in  diaphragm thickness at the end of expiration (TEE). Intergroup comparison showed that after intervention, the DTF, DM, FVC, FEV1, PEF, thoracic mobility, TCT, and the improvement in PNCT in the EDP group and the rPMS group were significantly better than those in the control group (all P<0.05). The rPMS group had better effects in shortening PNCT (P=0.017) and increasing FEV1/FVC (P=0.008) than the EDP group. There were no significant differences in TEI and CMAP amplitude among the three groups after intervention by multiple comparison.
Conclusions  Respiratory training, EDP, and rPMS can all improve diaphragmatic function in patients with cerebral infarction. EDP and rPMS have better rehabilitation efficacy on diaphragmatic dysfunction after cerebral infarction than respiratory training, and rPMS has more advantages in promoting the recovery of phrenic nerve motor conduction function in patients with cerebral infarction.

Respiratory dysfunction after severe stroke often involves multiple complex and heterogeneous pathological mechanisms, including dysregulated central respiratory drive, respiratory pump failure, and airway protection failure. Conventional rehabilitation models are difficult to meet individualized clinical demands. This paper reviews an integrated respiratory rehabilitation framework guided by precision assessment. Based on multi-modal physiological assessments, this framework constructs a respiratory function profile for patients with severe stroke, thereby providing evidence for the multi-disciplinary team to formulate individualized intervention protocols. Although the field still faces challenges such as insufficient high-level evidence, precision rehabilitation models, which integrate artificial intelligence-assisted decision-making and targeted neuromodulation technology, may provide new insights into the development of respiratory rehabilitation in severe stroke.

Respiratory dysfunction is an important factor for poor respiratory prognosis in patients with severe stroke. Conventional single rehabilitation methods cannot accurately address multiple pathological mechanisms after stroke, including dysregulated respiratory central drive, peripheral respiratory pump failure, and “brain-lung interaction” disorder. This article systematically reviews the research progress of neuromodulation technologies in respiratory function reconstruction for severe stroke patients. Based on the theoretical framework of “brain-lung interaction”, it elaborates on the mechanisms of action and relevant clinical evidence of central and peripheral neuromodulation technologies, while emphasizing their potential application value in restoring respiratory central drive and regulating respiratory rhythm synchrony. Although several challenges remain in this field, such as inadequate elaboration of mechanisms of action and lack of standardized intervention parameters, the integration of multimodal assessment and individualized regulation protocols is expected to promote the development of neuro-respiratory rehabilitation toward precision and closed-loop management in the future.

Stroke-associated pneumonia (SAP) is a common complication after stroke and an important risk factor for mortality. Clinically, SAP is often complicated by multidimensional dysfunctions such as hemiplegia, dysphagia, speech disorders, and cognitive impairment, leading to poor prognosis and severely compromised quality of daily life in patients. This paper reports the process of early comprehensive rehabilitation management for a patient with cerebral infarction in the left middle cerebral artery territory complicated by SAP, including respiratory rehabilitation training, limb function training, swallowing and speech training, as well as occupational and physical therapy. By summarizing the clinical data of this patient and reviewing relevant literature, this paper explores the early comprehensive rehabilitation protocol for SAP patients with multidimensional dysfunctions, analyzes the key links and influencing factors of early rehabilitation for SAP, and clarifies the importance of rehabilitation team collaboration and the multi-disciplinary team model in the management of such patients.

Objective  To analyze the effects of serum glial fibrillary acidic protein (GFAP) and tenascin-X (TNX) levels on the risk of secondary brain injury in patients with intracerebral hemorrhage (ICH).
Methods  This observational study prospectively and consecutively enrolled patients with ICH who were admitted to the Affiliated Hospital of Jinggangshan University within 24 hours of onset between January 2023 and January 2025. Based on the occurrence of secondary brain injury on the 5th day post-treatment, patients were divided into the secondary brain injury group and the non-secondary brain injury group. The general data, clinical characteristics, and serum GFAP and TNX levels before and on the 5th day post-treatment were compared between the two groups. Multivariate logistic regression analysis was used to identify independent influencing factors associated with secondary brain injury in ICH patients and to construct a prediction model. The predictive value of the model for secondary brain injury in ICH patients was assessed using ROC curve analysis.
Results  A total of 116 ICH patients were finally included in this study, with an age range of 44-77 years and a median age of 61 (57-67) years. Among them, 69 were male. Secondary brain injury occurred in 37 patients, while 79 did not develop it. Before treatment, serum GFAP levels [(9.21±3.60) ng/mL vs. (8.05±1.67) ng/mL, P=0.019] and TNX levels [(11.52±4.71) ng/mL vs. (9.51±3.68) ng/mL, P=0.014] were significantly higher in the secondary brain injury group compared to the non-secondary brain injury group. After treatment, serum GFAP and TNX levels increased in both groups compared with baseline levels, with the secondary brain injury group showing significantly higher GFAP levels [(11.17±2.25) ng/mL vs. (8.96±1.71) ng/mL, P＜0.001] and TNX levels [(15.47±4.27) ng/mL vs. (13.08±4.29) ng/mL, P＜0.001] compared to the non-secondary brain injury group. Multivariate logistic regression analysis indicated that larger hematoma volume (OR 4.460, 95%CI 1.430-13.908, P=0.010), higher pre-treatment serum GFAP level (OR 1.128, 95%CI 1.020-1.247, P=0.019), and higher pre-treatment serum TNX level (OR 1.092, 95%CI 1.010-1.181, P=0.028) were independent risk factors for secondary brain injury in ICH patients, while a higher GCS (OR 0.574, 95%CI 0.385-0.857, P=0.007) was an independent protective factor. ROC curve analysis showed that the combined prediction model using these four indicators had an AUC of 0.819 (95%CI 0.711-0.887), with a sensitivity of 89.32% and a specificity of 80.15%.
Conclusions  Larger hematoma volume and higher pre-treatment serum GFAP and TNX levels are independent risk factors for secondary brain injury in ICH patients, while a higher GCS is an independent protective factor. The prediction model integrating these four indicators demonstrates high predictive value for secondary brain injury in ICH patients and can provide a reference for early clinical risk assessment.

Objective  To investigate the association between different obesity metabolic phenotypes and acute cerebral infarction (ACI) subtypes, including large-artery atherosclerosis (LAA) and small-artery occlusion (SAO). 
Methods  A retrospective analysis was conducted on patients with LAA-type and SAO-type ACI admitted to the First Ward, Department of Neurology, Hanzhong Central Hospital from January 2023 to March 2024. Participants were divided into four groups according to BMI and metabolic status: metabolically healthy non-obese, metabolically healthy obese, metabolically unhealthy non-obese, and metabolically unhealthy obese. The distribution differences of obesity metabolic phenotypes between the LAA group and the SAO group were compared. Multivariate logistic regression analysis was performed to explore the association between obesity metabolic phenotypes and ACI subtypes. 
Results  A total of 207 ACI patients were enrolled, including 105 patients in the LAA group and 102 patients in the SAO group. The metabolically unhealthy obese phenotype was predominant in the LAA group (66.67%), while the metabolically unhealthy non-obese phenotype was the most common in the SAO group (47.06%). The difference in the distribution of obesity metabolic phenotypes between the two groups was statistically significant (P<0.001). Multivariate logistic regression analysis showed that in Model 1 without adjusting for confounding factors, metabolically unhealthy obese patients had a higher probability of developing LAA-type ACI (OR 2.283, 95%CI 1.093-4.766, P=0.028), while metabolically unhealthy non-obese patients were less likely to be LAA-type ACI (OR 0.171, 95%CI 0.068-0.432 P<0.001). After adjusting for confounding factors, the negative association between metabolically unhealthy non-obese phenotype and LAA-type ACI remained statistically significant (OR 0.058, 95%CI 0.017-0.194, P<0.001), while no significant association was found between the metabolically unhealthy obese phenotype and either ACI subtype.
Conclusions  In ACI patients, the metabolically unhealthy obese phenotype is associated with LAA-type ACI, but this association may be affected by confounding factors. The metabolically unhealthy non-obese phenotype is strongly associated with SAO-type ACI (relative to LAA-type ACI), suggesting that specific associations exist between different obesity metabolic phenotypes and ACI subtypes.

Problems such as inconsistent definitions, non-uniform standards, and fragmentation of data elements in clinical research reduce research efficiency, and also fail to meet the ever-increasing demands for efficient data collection, integration, and standardization in medical research against the backdrop of the rapid development of big data and artificial intelligence technologies. The National Center for Healthcare Quality Management in Neurological Diseases, the China National Clinical Research Center for Neurological Diseases, the Beijing Office for Cerebrovascular Disease Prevention and Control, and the Medical Quality Management and Promotion Branch of the Chinese Stroke Association jointly established an expert working group. The group completed a comprehensive update of the 2020 Expert Consensus on Core Data Elements and Definitions for Diagnosis, Treatment, and Management of Stroke and formulated the Expert Consensus on Clinical Core Common Data Element Standards for Artificial Intelligence-Driven Stroke Research and Management. This revision and update further expanded the breadth and depth of the common data elements for stroke research and management, and systematically defined the standards and specifications for 13 data element modules, including demographic information, admission and transfer information, past medical history and risk factors, prior medication history, clinical symptoms and signs, auxiliary examinations, etiological classification, reperfusion therapy, acute-phase treatment and secondary prevention medications, in-hospital complications, discharge status, rehabilitation therapy, and follow-up information. Based on current practical needs and future research directions, this revised expert consensus provides unified and standardized data specifications for stroke clinical practice, disease management, and real-world research. It also supports the development and application of artificial intelligence models in clinical decision-making, disease risk prediction, and prognostic assessment for stroke diagnosis and treatment.

The ‌2026 Guideline for the Early Management of Patients with Acute Ischemic Stroke: A Guideline from the American Heart Association/American Stroke Association was released in January 2026. It contains substantial updates to the recommendations on intravenous thrombolysis, covering agent selection, screening of pediatric patients, extended therapeutic time windows, reduced contraindications, and that intravenous thrombolysis is not recommended for mild non-disabling stroke. This article summarizes and interprets these key updates, aiming to provide a reference for the clinical practice of intravenous thrombolysis among patients with acute ischemic stroke in China.

Acute ischemic stroke (AIS) is a major global cause of death and disability, with reperfusion therapy being crucial for improving outcomes in the acute phase. In January 2026, the 2026 Guideline for the Early Management of Patients with Acute Ischemic Stroke: A Guideline from the American Heart Association/American Stroke Association was officially released, updating relevant recommendations. This article provides an interpretation focused on the endovascular therapy (EVT) section. Building on existing evidence for patients with proximal anterior circulation large vessel occlusion, the 2026 guideline further clarifies imaging eligibility criteria of EVT within the extended time window and delineates more explicit selection criteria of EVT for AIS patients with large ischemic cores and pre-stroke functional impairment. In addition, for posterior circulation stroke, particularly basilar artery occlusion, the 2026 guideline stratifies decision-making basis according to clinical severity and the extent of imaging-defined injury. Notably, the 2026 guideline systematically includes EVT recommendations in pediatric AIS for the first time, providing a reference framework for clinical decision-making in pediatric patients across different age groups and time windows.

Cerebral venous sinus thrombosis (CVST) is a special type of cerebrovascular disease. Severe CVST is characterized by extensive venous sinus thrombosis, significant intracranial hypertension, and secondary brain injury. Anticoagulation therapy alone yields limited efficacy in severe cases, often necessitating surgical intervention. This article reports a case of severe CVST in which the patient developed an intracranial hypertension crisis due to an insufficient duration of anticoagulation therapy, and was subsequently treated with ventricular Ommaya reservoir placement as a bridging treatment. This intervention secured a therapeutic time window for anticoagulation therapy, ultimately achieving a successful reversal of the patient’s condition. This case provides a practical reference for the exploration of minimally invasive intervention strategies for severe CVST.

Cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL) is an autosomal dominant cerebral small vessel disease. Multiple sclerosis is an inflammatory demyelinating disease of the central nervous system. The two diseases share numerous similarities in clinical manifestations and neuroimaging features, which readily leads to misdiagnosis. This paper reports one patient with concurrent intracranial and spinal cord lesions who was initially diagnosed with multiple sclerosis, and was ultimately confirmed to have CADASIL caused by the homozygous missense mutation p.Arg544Cys (c.1630C>T) in the Notch3 (Notch receptor 3) via genetic testing. This case analysis is intended to reduce misdiagnosis and underdiagnosis of CADASIL, enrich the clinical phenotypes associated with this specific Notch3 mutation, and enhance the awareness, diagnosis, and treatment of the disease.

X-linked adrenoleukodystrophy (X-ALD) is a group of rare neurological disorders characterized by diverse mutation sites and clinical manifestations, exhibiting both genetic and clinical heterogeneity. The diagnosis of this disease requires a combination of very long-chain fatty acid concentration testing and genetic analysis of the ATP-binding cassette transporter D member 1 (ABCD1). This article reports the clinical diagnosis and treatment process of an atypical case of X-ALD that was initially misdiagnosed as cerebral infarction. The patient was a young male presenting with progressive gait disturbance and lower limb spasticity accompanied by weakness. During the diagnostic process, the patient was initially diagnosed with cerebral infarction and treated accordingly. Subsequently, genetic testing revealed a mutation in the ABCD1, leading to a definitive diagnosis of X-ALD. This case highlights the importance of considering X-ALD as a potential differential diagnosis in patients presenting with progressive spastic paraplegia.