Development of a Novel Prognostic Model to Predict 6-Month Swallowing Recovery After Ischemic Stroke

Artificial Intelligence in Speech-Language Pathology and Dysphagia: A Review From Latin American Perspective and Pilot Test of LLMs for Rehabilitation Planning

Artificial Intelligence (AI) is transforming speech-language pathology (SLP) and dysphagia management, offering innovative solutions for assessment, diagnosis, and rehabilitation. This narrative review examines AI applications in these fields from 2014 to 2024, with particular focus on implementation challenges in Latin America. We analyze key AI technologies-including deep learning, machine learning algorithms, and natural language processing-that have demonstrated high accuracy in detecting voice disorders, analyzing swallowing function, and supporting personalized rehabilitation. The review identifies three primary domains of AI application: diagnostic tools with improved sensitivity for speech-language disorders, rehabilitation technologies that enable customized therapy, and telehealth platforms that expand access to specialized care in underserved regions. However, significant barriers persist, particularly in Latin America, where limited infrastructure, insufficient linguistic adaptation, and scarce regional datasets hamper widespread implementation. Our pilot study evaluating commercially available large language models for rehabilitation planning demonstrates their potential utility in generating structured therapy activities, especially in resource-constrained settings. While AI shows promise in enhancing clinical workflows and expanding service delivery, the evidence suggests that current applications remain predominantly focused on diagnosis rather than integrated rehabilitation. This review highlights the need for culturally and linguistically adapted AI models, expanded regional research collaborations, and regulatory frameworks that ensure ethical AI integration into SLP and dysphagia care, positioning these technologies as complementary tools that enhance rather than replace clinical expertise.

Association of type 2 diabetes mellitus and cognitive function with the resumption of oral intake during stroke rehabilitation

BACKGROUND

Physical factors including motor and balance functions are associated with resumption of oral intake of the stroke patients receiving nasogastric tube feeding. However, association of non-physical factors remains unclear. The aims of this study are to explore non-physical factors associated with resumption of oral intake after stroke and to clarify how they affect the time course of the resumption.

METHODS

A total of 116 patients receiving nasogastric tube feeding after stroke were retrospectively analyzed from a single center observational cohort database. Associations of physical and non-physical factors with resumption of oral intake at discharge from the stroke rehabilitation facility were investigated. We also evaluated how the specific combination of non-physical prognostic factors affected the time course of resumption using survival analyses.

RESULTS

Among 116 patients, 59 (50.4 %) resumed oral intake. History of diabetes mellitus (DM) (P < 0.001) and higher cognitive functional independence scores (C-FIM) (P = 0.03) were significantly associated with the resumption of oral intake at discharge. These non-physical factors had an additive value on conventional risk factors to predict resumption of oral intake. Patients with a history of DM were significantly associated with improved resumption of oral intake than with non-DM patients with lower C-FIM scores (< 18) by survival analysis (P = 0.001), and this difference became prominent after 90 days post-stroke.

CONCLUSION

A history of DM and better cognitive function as non-physical factors were significantly associated with the resumption of oral intake in patients receiving nasogastric tube feeding after stroke.

Post-stroke dysphagia: Neurological regulation and recovery strategies

Swallowing is a complex process requiring precise coordination of numerous muscles in the head and neck to smoothly guide ingested material from the mouth to the stomach. Animal and human studies have revealed a complex network of neurons in the brainstem, cortex, and cerebellum that coordinate normal swallowing. The interactions between these regions ensure smooth and efficient swallowing. However, the current understanding of the neurophysiological mechanisms involved in post-stroke dysphagia (PSD) is incomplete, and complete functional connectivity for swallowing recovery remains understudied and requires further exploration. In this review, we discussed the neuroanatomy of swallowing and the pathogenesis of PSD and summarized the factors affecting PSD recovery. We also described the plasticity of neural networks affecting PSD, including enhancing activation of neural pathways, cortical reorganization, regulation of extracellular matrix dynamics and its components, modulation of neurotransmitter delivery, and identification of potential therapeutic targets for functional recovery in PSD. Finally, we discussed the therapeutic strategies based on functional compensation and motor learning. This review aimed to provide a reference for clinicians and researchers to promote the optimization of PSD treatments and explore future research directions.

Predictors of recovery from dysphagia after stroke: A systematic review and meta-analysis

Objective

This systematic review aimed to identify the predictors of recovery from dysphagia after stroke in the last ten years, thereby providing an evidence-based basis for nurses to identify high-risk patients and develop individualized rehabilitation plans to improve patient prognosis.

Methods

Databases including the China National Knowledge Infrastructure (CNKI), China Biology Medicine disc (CBMdisc), China Science and Technology Journal (VIP), WanFang, PubMed, Embase, CINAHL, Web of Science, the Cochrane Library, and Scopus were retrieved to search for literature on the predictors of recovery from dysphagia after stroke. The retrieval period was from January 2013 to December 2023. The quality of studies was assessed using the Newcastle-Ottawa Scale (NOS) and the Prediction model Risk of Bias Assessment Tool (PROBAST). Meta-analysis was performed using Revman5.3 and Stata15.1 software. The review protocol has been registered with PROSPERO (CRD42024605570).

Results

A total of 1,216 results were obtained, including 599 in English and 617 in Chinese. A total of 34 studies were included, involving 156,309 patients with post-stroke dysphagia, and the rate of dysphagia recovery increased from 13.53% at 1 week to 95% at 6 months after stroke. Meta-analysis results showed that older age [OR = 1.06, 95%CI (1.04, 1.08), P < 0.001], lower BMI [OR = 1.28, 95%CI (1.17, 1.40), P < 0.001], bilateral stroke [OR = 3.10, 95%CI (2.04, 4.72), P < 0.001], higher National Institutes of Health Stroke Scale (NIHSS) score [OR = 1.19, 95%CI (1.01, 1.39), P = 0.030], tracheal intubation [OR = 5.08, 95%CI (1.57, 16.39), P = 0.007] and aspiration [OR = 4.70, 95%CI (3.06, 7.20), P < 0.001] were unfavorable factors for the recovery of swallowing function in patients with post-stroke dysphagia.

Conclusions

The lack of standardized criteria for rehabilitation assessment of post-stroke dysphagia has resulted in reported recovery rates of swallowing function exhibiting wide variability. Nurses should take targeted preventive measures for patients aged ≥70 years, low BMI, bilateral stroke, high NIHSS score, tracheal intubation, and aspiration to promote the recovery of swallowing function in patients with post-stroke dysphagia.

Post-stroke dysphagia: identifying the evidence missing

Dysphagia is a high-profile dysfunction that often occurs after a stroke, with a prevalence of 50%-80%. Post-stroke dysphagia (PSD) often leads to serious complications such as pneumonia and malnutrition, reducing the quality of life and leading to poor prognosis or even death. PSD causes these adverse physical and psychological impairments to patients, which becomes a challenge for both patients and physicians. This review intends to contribute to the international debate on evidence-based options on Stroke Rehabilitation and to better understand the need for further research on PSD and summarizing evidence on some of the most relevant topics and clarifying its clinical practice value for Neurology, stroke rehabilitation experts, rehabilitation and nursing staff, as well as patients. The article identifies and discusses the gaps in knowledge on PSD and elaborates on current evidence concerning the selection of subjects, examination methods, patient data extraction and analysis, classification of stroke lesions, details of dysphagia, significance of results, and neuromodulation of dysphagia, from the perspective of rehabilitation physicians. The review identified a set of 10 points and parameters for the international debate on PSD, namely: stroke onset, cognitive impairment, feeding method, contrast medium, swallowing reflex delay, swallowing evaluation form, division of brainstem, multiple stroke sites, basal ganglia lesions and neuromodulation techniques. The article explores available evidence on factors associated with dysphagia and stroke site. Although there is plenty of evidence exploring the correlation between stroke site and swallowing disorders, the pathophysiological mechanisms between the two are complex, and expert interpretations of the evidence and clinical opinions vary on which swallowing abnormalities occur. The study generates evidence on current evidence-based options on Stroke Rehabilitation and a better understanding of the need for further research on Post-Stroke Dysphagia. Taking a patient-centric approach, the ultimate goal is to generate on how can available evidence influence policy or practice or research or clinical education. The article provides a structured discussion clarifying key points on the relationship between stroke lesions and swallowing dysfunctions and contributes to clarifying the gaps in evidence to further improve the quality of life of the patients suffering from Post-Stroke Dysphagia.

Factors associated with oropharyngeal dysphagia and unsuccessful nasogastric tube removal after endovascular thrombectomy for anterior circulation stroke

PURPOSE

To identify the predictive variables for post-stroke dysphagia (PSD) among anterior circulation large vessel occlusion (LVO) stroke patients who underwent endovascular thrombectomy (EVT).

METHODS

This retrospective cohort study enrolled hospitalized patients with anterior LVO stroke who underwent EVT between January 1, 2018 and October 31, 2022. PSD was defined as the unsuccessful removal of the nasogastric (NG) tube. Factors, such as premorbid characteristics, laboratory results, EVT, rehabilitation-related parameters, and neuro-imaging, were analyzed for correlations to PSD at 4 and 12 weeks.

RESULTS

The study enrolled 136 patients, with a mean age of 72.9 ± 13.0 years, and 59 patients (43.4%) were male. At 4 weeks, 47.1% of the patients needed an NG tube, and at 12 weeks, 16.2% still required an NG tube. We found that lower albumin, lower body mass index (BMI), higher initial and 24-h post-EVT National Institute of Health Stroke Scale (NIHSS) scores, stroke-associated pneumonia, poor initial sitting balance and ability to sit up, insula or frontal operculum lesions, and bilateral hemisphere involvement were all associated with PSD at both 4 and 12 weeks in the univariate logistic regression. Multivariate analysis revealed that significant predictors of unsuccessful NG tube removal at 4 weeks included lower BMI (adjusted OR [aOR] 0.73, p = 0.005), hemorrhagic transformation (aOR 4.01, p = 0.0335), higher NIHSS scores at 24 h post-EVT (aOR 1.13, p = 0.0288), poor initial sitting ability (aOR 0.52, p = 0.0231), insular cortex ischemia (aOR 7.26, p = 0.0056), and bilateral hemisphere involvement (aOR 41.19, p < 0.0001). At 12 weeks, lower BMI (aOR 0.78, p = 0.0098), poor initial sitting balance (aOR 0.57, p = 0.0287), insular cortex lesions (aOR 4.83, p = 0.0092), and bilateral hemisphere involvement (aOR 4.07, p = 0.0139) remained significant predictors.

CONCLUSIONS

In patients with anterior LVO following EVT, PSD was associated with lower BMI, higher NIHSS scores, poor initial sitting balance and sitting ability, insular lesions, and bilateral hemisphere involvement.

Prediction of Oral Intake at Discharge with Early Assessment of Swallowing Function within 24 h after Admission: A Retrospective Cohort Study

Dysphagia is one of the most common symptoms of stroke and can lead to other complications such as pneumonia, dehydration, and malnutrition. This retrospective cohort study evaluated the predictive value of a comprehensive swallowing assessment tool, the Mann Assessment of Swallowing Ability (MASA), in the acute phase of stroke for oral intake status at discharge. Among 1,133 consecutive patients with acute stroke, 512 patients whose swallowing function was assessed using the MASA within 24 h of admission were included. Data including demographic information, stroke severity, MASA, Oral Health Assessment Tool, body mass index, and serum albumin level were collected. Predictive factors for oral intake were analyzed using a multiple logistic regression model, and the receiver operating characteristic (ROC) curve analysis was used to determine the cutoff values of the MASA score for determining oral intake at discharge. Oral intake at discharge was established in 69.1% of the cohort (354/512). The multiple logistic regression analysis identified a higher MASA score, younger age, and higher serum albumin level as significant predictors of oral intake at discharge. The cutoff value of the MASA score for oral intake was 136.5 points, with an area under the ROC curve of 0.87. These findings suggest that the MASA is a valid tool for predicting oral intake in patients with dysphagia during the acute phase of stroke.

RGX Ensemble Model for Advanced Prediction of Mortality Outcomes in Stroke Patients

Objective: This paper aims to address the clinical challenge of predicting the outcomes of stroke patients and proposes a comprehensive model called RGX to help clinicians adopt more personalized treatment plans. Impact Statement: The comprehensive model is first proposed and applied to clinical datasets with missing data. The introduction of the Shapley additive explanations (SHAP) model to explain the impact of patient indicators on prognosis improves the accuracy of stroke patient mortality prediction. Introduction: At present, the prediction of stroke treatment outcomes faces many challenges, including the lack of models to quantify which clinical variables are closely related to patient survival. Methods: We developed a series of machine learning models to systematically predict the mortality of stroke patients. Additionally, by introducing the SHAP model, we revealed the contribution of risk factors to the prediction results. The performance of the models was evaluated using multiple metrics, including the area under the curve, accuracy, and specificity, to comprehensively measure the effectiveness and stability of the models. Results: The RGX model achieved an accuracy of 92.18% on the complete dataset, an improvement of 11.38% compared to that of the most advanced state-of-the-art model. Most importantly, the RGX model maintained excellent predictive ability even when faced with a dataset containing a large number of missing values, achieving an accuracy of 84.62%. Conclusion: In summary, the RGX ensemble model not only provides clinicians with a highly accurate predictive tool but also promotes the understanding of stroke patient survival prediction, laying a solid foundation for the development of precision medicine.

Tongue Pressure and Grip Strength as Indicators of Persistent Dysphagia After Acute Stroke

This study aimed to identify the independent predictors of postacute stroke dysphagia at discharge using sarcopenia-related parameters. This single-center prospective observational study assessed consecutive inpatients diagnosed with cerebral infarction or cerebral hemorrhage upon admission to the stroke unit. Tongue pressure, grip strength, and body composition were evaluated within 48 h. Dysphagia was defined by a functional oral intake scale of ≤ 5. Patient characteristics were compared between non-dysphagia and dysphagia groups using Mann-Whitney or chi-squared tests. Logistic regression analysis was performed using age, sex, tongue pressure, grip strength, skeletal muscle mass index (SMI), and National Institutes of Health Stroke Scale (NIHSS) scores as explanatory variables, with dysphagia at discharge as the objective variable. A total of 302 patients (mean age: 69.4 ± 13.8 years, 67.5% male) were analyzed, with 64 having dysphagia at discharge (21.2%). The dysphagia group was significantly older (p<0.001), had higher NIHSS scores on admission (p<0.001), lower SMI (p = 0.002), lower grip strength (p<0.001), and lower tongue pressure (p<0.001) than the non-dysphagia group. Logistic regression revealed that age (OR: 1.042, p = 0.018), tongue pressure (OR: 0.954, p = 0.010), and grip strength (OR: 0.943, p = 0.048) on admission were independent predictors of dysphagia at discharge, while NIHSS scores (OR: 1.403, p = 0.106), sex, and SMI (OR: 1.403, p = 0.150) were not. Older age, reduced tongue pressure, and reduced grip strength are strong predictors of persistent poststroke dysphagia at discharge. Thus, muscle strength is a more valuable parameter than muscle mass in predicting persistent poststroke dysphagia.

AI Applications in Adult Stroke Recovery and Rehabilitation: A Scoping Review Using AI

Stroke is a leading cause of long-term disability worldwide. With the advancements in sensor technologies and data availability, artificial intelligence (AI) holds the promise of improving the amount, quality and efficiency of care and enhancing the precision of stroke rehabilitation. We aimed to identify and characterize the existing research on AI applications in stroke recovery and rehabilitation of adults, including categories of application and progression of technologies over time. Data were collected from peer-reviewed articles across various electronic databases up to January 2024. Insights were extracted using AI-enhanced multi-method, data-driven techniques, including clustering of themes and topics. This scoping review summarizes outcomes from 704 studies. Four common themes (impairment, assisted intervention, prediction and imaging, and neuroscience) were identified, in which time-linked patterns emerged. The impairment theme revealed a focus on motor function, gait and mobility, while the assisted intervention theme included applications of robotic and brain-computer interface (BCI) techniques. AI applications progressed over time, starting from conceptualization and then expanding to a broader range of techniques in supervised learning, artificial neural networks (ANN), natural language processing (NLP) and more. Applications focused on upper limb rehabilitation were reviewed in more detail, with machine learning (ML), deep learning techniques and sensors such as inertial measurement units (IMU) used for upper limb and functional movement analysis. AI applications have potential to facilitate tailored therapeutic delivery, thereby contributing to the optimization of rehabilitation outcomes and promoting sustained recovery from rehabilitation to real-world settings.

Predicting Swallowing Recovery in Subacute Stroke Patients via Temporal and Spatial Parameters of Videofluoroscopy

PURPOSE

The purpose of this study was to identify the temporal and spatial parameters of videofluoroscopic swallowing study (VFSS) that could predict the recovery of swallowing function in subacute stroke patients.

METHODS

We included 102 patients who were admitted to the Department of Rehabilitation Medicine between 2019 and 2022. Patients were classified into good and poor prognosis groups according to whether they had restored prestroke swallowing function or were able to consume sufficient nutrition via oral feeding to meet their body's needs. Univariate and multivariate regression analyses were used to identify the predictors. Calibration and discrimination were tested using the Hosmer-Lemeshow test and area under the curve (AUC), respectively.

RESULTS

Of the 102 included patients, 51 had a good prognosis for swallowing function within 6 months of onset. The final multivariate regression model included three significant factors: laryngeal closure duration (LCD) (OR: 0.998; 95% CI: 0.996-0.999; p < 0.05), maximum width of the upper esophageal sphincter opening (MWUESO) (OR: 1.251; 95% CI: 1.073-1.458; p < 0.05), and pharyngeal residual grade (PRG) (p < 0.05). The shorter LCD and larger MWUESO were positive predictors of good swallowing function outcomes, while higher PRG was a negative predictor of good outcomes. The AUC for PRG, MWUESO, and LCD were 0.767 (p < 0.05), 0.738 (p < 0.05), and 0.681 (p < 0.05), respectively.

CONCLUSION

Identifying prognostic factors for the recovery of swallowing function in patients with poststroke dysphagia is essential for developing treatment strategies. The findings of this study may provide an important reference for developing appropriate therapeutic interventions to promote the recovery of swallowing function in stroke patients.

The Effect of Oral Diet Training in Indwelling Nasogastric Tube Patients with Prolonged Dysphagia

Background: Patients with severe dysphagia are usually fed using a nasogastric tube (NGT). Many patients who receive long-term NGT feeding are unable to obtain sufficient nutrients orally immediately after NGT removal. Thus, a transitional period involving oral diet training is required to transition from NGT feeding to exclusive oral feeding. We aimed to investigate the therapeutic effect of oral diet training in indwelling NGT patients with prolonged dysphagia. Methods: A total of 175 patients who were fed using an NGT for more than 4 weeks were enrolled. Their swallowing function was evaluated by a videofluoroscopic swallowing study (VFSS). During the VFSS, patients received thick and thin barium while the NGT was inserted. Then, the patients underwent a VFSS without an NGT thirty minutes after NGT removal. If a patient had no aspiration with NGT inserted during the VFSS, oral diet training combined with NGT feeding was recommended. Results: Of the 49 indwelling NGT patients who were recommended to receive oral diet training, 39 (79.6%) transitioned to exclusive oral feeding. A transition period of 2-8 weeks was required for them to achieve full oral feeding. Patients who were eligible for oral feeding trials showed no significant aspiration during the VFSS with an NGT inserted and had sufficient cough function. Patients who required prolonged NGT feeding and who could not complete oral trials showed significant aspiration during the VFSS when an NGT was inserted. Conclusions: This study demonstrated that oral diet training combined with NGT feeding is safe in patients with prolonged dysphagia who have sufficient cough function and no aspiration during VFSS. We suggest that if the patient is a proper candidate for NGT removal, direct oral feeding training with an NGT inserted could be a useful therapeutic strategy during the transitional period from long-term NGT feeding to successful oral feeding.

Predictors of Complete Oral Intake in Patients With Stroke After Tracheostomy

BACKGROUND

Tracheostomy procedures inhibit swallowing, although details of subsequent recovery of oral intake remain unknown. This retrospective cohort study aimed to investigate factors influencing dysphagia improvement in patients with subacute stroke after tracheostomy.

METHODS AND RESULTS

The study included 117 patients who underwent tracheostomy after subacute stroke, cerebral hemorrhage, or endogenous subarachnoid hemorrhage and received care at 2 convalescent rehabilitation wards in urban and suburban Japan between 2015 and 2022. The primary outcome measure was the achievement of complete oral intake. Patient demographics, Functional Independence Measure scores, body mass index, food intake level scale scores, and the presence of severe white matter hyperintensities on imaging were retrospectively collected from medical records. Statistical analysis involved univariate logistic regression to identify potential predictors and multivariate logistic regression to refine the model while accounting for multicollinearity. In total, 47% of patients achieved complete oral intake on discharge. Sex, days from onset to admission, Functional Independence Measure motor and cognitive scores, body mass index, food intake level scale scores, and severe white matter hyperintensities were identified as potential predictors in the univariate analysis. However, multivariate logistic regression identified only food intake level scale scores (odds ratio [OR], 3.687 [95% CI, 1.519-8.949]; P=0.004) and severe white matter hyperintensities (OR, 0.302 [95% CI, 0.096-0.956]; P=0.042) as significant predictors of complete oral intake.

CONCLUSIONS

In patients with subacute stroke undergoing tracheostomy, the level of oral intake on admission and severe white matter hyperintensities on imaging may be better predictors of complete oral intake. However, prospective studies with larger sample sizes and more comprehensive data are warranted to confirm these findings.

Transition experiences of patients with post stroke dysphagia and family caregivers: A longitudinal, qualitative study

BACKGROUND

Stroke patients with dysphagia and family caregivers will experience multiple transitions during the whole process of the disease and various nursing needs will be generated. There is a lack of knowledge about their experiences at different transition stages. Thus, we aimed to explore the transition experiences of patients with post stroke dysphagia and family caregivers from admission to discharge home.

METHODS

A semi-structured interview based on Meleis's transition theory was used during hospitalization and telephone follow-up interviews were conducted in the first, third, and sixth month after the diagnosis of dysphagia. Interview transcripts were analyzed using the conventional content analysis method.

RESULTS

A total of 17 participants enrolled in the first face-to-face interview, 16 participants took part in the first month's telephone follow-up interview, 14 participants in the third month, and 12 participants in the sixth month. The transition experiences of patients with post stroke dysphagia and family caregivers could be summarized into three themes: (1)transition from onset to admission; (2)transition from discharge to other rehabilitation institutions; and (3)transition from discharge to home. Each theme had identified interrelated subthemes.

CONCLUSIONS

The experiences of patients with post stroke dysphagia and family caregivers during transition are a dynamic process with enormous challenges in each phase. Collaboration with health care professionals, follow-up support after discharge, and available community and social support should be integrated into transitional nursing to help patients facilitate their transition.

The Cortical and Subcortical Neural Control of Swallowing: A Narrative Review

Swallowing is a sophisticated process involving the precise and timely coordination of the central and peripheral nervous systems, along with the musculatures of the oral cavity, pharynx, and airway. The role of the infratentorial neural structure, including the swallowing central pattern generator and cranial nerve nuclei, has been described in greater detail compared with both the cortical and subcortical neural structures. Nonetheless, accumulated data from analysis of swallowing performance in patients with different neurological diseases and conditions, along with results from neurophysiological studies of normal swallowing have gradually enhanced understanding of the role of cortical and subcortical neural structures in swallowing, potentially leading to the development of treatment modalities for patients suffering from dysphagia. This review article summarizes findings about the role of both cortical and subcortical neural structures in swallowing based on results from neurophysiological studies and studies of various neurological diseases. In sum, cortical regions are mainly in charge of initiation and coordination of swallowing after receiving afferent information, while subcortical structures including basal ganglia and thalamus are responsible for movement control and regulation during swallowing through the cortico-basal ganglia-thalamo-cortical loop. This article also presents how cortical and subcortical neural structures interact with each other to generate the swallowing response. In addition, we provided the updated evidence about the clinical applications and efficacy of neuromodulation techniques, including both non-invasive brain stimulation and deep brain stimulation on dysphagia.

Systematic Review of Machine Learning Applied to the Secondary Prevention of Ischemic Stroke

Ischemic stroke is a serious disease posing significant threats to human health and life, with the highest absolute and relative risks of a poor prognosis following the first occurrence, and more than 90% of strokes are attributable to modifiable risk factors. Currently, machine learning (ML) is widely used for the prediction of ischemic stroke outcomes. By identifying risk factors, predicting the risk of poor prognosis and thus developing personalized treatment plans, it effectively reduces the probability of poor prognosis, leading to more effective secondary prevention. This review includes 41 studies since 2018 that used ML algorithms to build prognostic prediction models for ischemic stroke, transient ischemic attack (TIA), and acute ischemic stroke (AIS). We analyzed in detail the risk factors used in these studies, the sources and processing methods of the required data, the model building and validation, and their application in different prediction time windows. The results indicate that among the included studies, the top five risk factors in terms of frequency were cardiovascular diseases, age, sex, national institutes of health stroke scale (NIHSS) score, and diabetes. Furthermore, 64% of the studies used single-center data, 65% of studies using imbalanced data did not perform data balancing, 88% of the studies did not utilize external validation datasets for model validation, and 72% of the studies did not provide explanations for their models. Addressing these issues is crucial for enhancing the credibility and effectiveness of the research, consequently improving the development and implementation of secondary prevention measures.

Research trends and hotspots of post-stroke dysphagia rehabilitation: a bibliometric study and visualization analysis

Background

Post-stroke dysphagia (PSD) is one of the most prevalent stroke sequelae, affecting stroke patients' prognosis, rehabilitation results, and quality of life while posing a significant cost burden. Although studies have been undertaken to characterize the pathophysiology, epidemiology, and risk factors of post-stroke dysphagia, there is still a paucity of research trends and hotspots on this subject. The purpose of this study was to create a visual knowledge map based on bibliometric analysis that identifies research hotspots and predicts future research trends.

Methods

We searched the Web of Science Core Collection for material on PSD rehabilitation research from its inception until July 27, 2023. We used CiteSpace, VOSviewer, and Bibliometrix R software packages to evaluate the annual number of publications, nations, institutions, journals, authors, references, and keywords to describe present research hotspots and prospective research orientations.

Results

This analysis comprised 1,097 articles from 3,706 institutions, 374 journals, and 239 countries or regions. The United States had the most publications (215 articles), and it is the most influential country on the subject. "Dysphagia" was the most published journal (100 articles) and the most referenced journal (4,606 citations). Highly cited references focused on the pathophysiology and neuroplasticity mechanisms of PSD, therapeutic modalities, rehabilitation tactics, and complications prevention. There was a strong correlation between the terms "validity" and "noninvasive," which were the strongest terms in PSD rehabilitation research. The most significant words in PSD rehabilitation research were "validity" and "noninvasive brain stimulation," which are considered two of the most relevant hotspots in the field.

Conclusion

We reviewed the research in the field of PSD rehabilitation using bibliometrics to identify research hotspots and cutting-edge trends in the field, primarily including the pathogenesis and neurological plasticity mechanisms of PSD, complications, swallowing screening and assessment methods, and swallowing rehabilitation modalities, and this paper can provide in the follow-up research in the field of PSD rehabilitation. The results of this study can provide insightful data for subsequent studies in the field of PSD rehabilitation.

Proteomic advance of ischemic stroke: preclinical, clinical, and intervention

Ischemic stroke (IS) is the most common type of stroke and is characterized by high rates of mortality and long-term injury. The prediction and early diagnosis of IS are therefore crucial for optimal clinical intervention. Proteomics has provided important techniques for exploring protein markers associated with IS, but there has been no systematic evaluation and review of research that has used these techniques. Here, we review the differential proteins that have been found in cell- and animal- based studies and clinical trials of IS in the past 10 years; determine the key pathological proteins that have been identified in clinical trials; summarize the target proteins affected by interventions aimed at treating IS, with a focus on traditional Chinese medicine treatments. Overall, we clarify findings and problems that have been identified in recent proteomics research on IS and provide suggestions for improvements in this area. We also suggest areas that could be explored for determining the pathogenesis and developing interventions for IS.

The assessment of dysphagia after stroke: state of the art and future directions

Dysphagia is a major complication following an acute stroke that affects the majority of patients. Clinically, dysphagia after stroke is associated with increased risk of aspiration pneumonia, malnutrition, mortality, and other adverse functional outcomes. Pathophysiologically, dysphagia after stroke is caused by disruption of an extensive cortical and subcortical swallowing network. The screening of patients for dysphagia after stroke should be provided as soon as possible, starting with simple water-swallowing tests at the bedside or more elaborate multi-consistency protocols. Subsequently, a more detailed examination, ideally with instrumental diagnostics such as flexible endoscopic evaluation of swallowing or video fluoroscopy is indicated in some patients. Emerging diagnostic procedures, technical innovations in assessment tools, and digitalisation will improve diagnostic accuracy in the future. Advances in the diagnosis of dysphagia after stroke will enable management based on individual patterns of dysfunction and predisposing risk factors for complications. Progess in dysphagia rehabilitation are essential to reduce mortality and improve patients' quality of life after a stroke.

Risk factors independently associated with the maintenance of severe restriction of oral intake and alternative feeding method indication at hospital outcome in patients after acute ischemic stroke

BACKGROUND AND AIMS

From a clinical point of view, post-stroke patients present difficulties in swallowing management. The purpose of this research was to identify risk factors that were independently related to the maintenance of a severe restriction of oral intake in patients affected by acute ischemic stroke.

METHODS

The authors conducted a prospective observational cohort study of patients with dysphagia post-acute ischemic stroke who were admitted to an Emergency Room (ER). Demographic and clinical data were collected at ER admission. Swallowing data was based on The Functional Oral Intake Scale (FOIS) and was collected at two distinct moments: initial swallowing assessment and at the patient outcome. Patients were divided into two groups according to their FOIS level assigned on the last swallowing assessment (at hospital outcome): G1 with severe restriction of oral intake and indication of feeding tube - patients with FOIS levels 1 to 4; G2 without restriction of food consistencies in oral intake - patients with FOIS levels 5 to 7.

RESULTS

One hundred and six patients were included in our study. Results of the multivariate logistic regression model for the prediction of maintenance of a severe restriction of oral intake at hospital outcome in patients post-acute ischemic stroke indicated that increasing age (p = 0.006), and dysarthria (p = 0.003) were associated with higher chances of presenting severe restriction of oral intake at hospital outcome.

CONCLUSIONS

Patients with acute ischemic stroke in an Emergency Room may experience non-resolved severe dysphagia, indicating the need to prepare for the care/rehabilitation of these patients.

Trends in the incidence and prevalence of dysphagia requiring medical attention among adults in South Korea, 2006-2016: A nationwide population study

The prevalence of dysphagia is increasing, resulting in socioeconomic burden, but previous reports have only been based on a limited populations. Therefore, we aimed to investigate the nationwide incidence and prevalence of dysphagia requiring medical attention to provide adequate information for healthcare planning and resource allocation. In this nationwide retrospective cohort study, the data of adults aged ≥20 years recorded from 2006 to 2016 were sourced from the Korean National Health Insurance Service database. Medical claim codes based on ICD-10-CM were used to define dysphagia and possible causes. The annual incidence and prevalence of dysphagia were calculated. Cox regression was used to estimate dysphagia risk in people with possible dysphagia etiology. Survival analysis was performed to estimate the mortality and hazard ratio of dysphagia. The crude annual incidence of dysphagia increased continuously from 7.14 in 2006 to 15.64 in 2016. The crude annual prevalence of dysphagia in 2006 was 0.09% and increased annually to 0.25% in 2016. Stroke (odds ratio [OR]: 7.86, 95% confidence interval [CI]: 5.76-6.68), neurodegenerative disease (OR: 6.20, 95% CI: 5.76-6.68), cancer (OR: 5.59, 95% CI: 5.17-6.06), and chronic obstructive pulmonary disease (OR: 2.94, 95% CI: 2.71-3.18) were associated with a high risk of dysphagia. The mortality in the dysphagia group was 3.12 times higher than that in the non-dysphagia group (hazard ratio: 3.12, 95% CI: 3.03-3.23). The incidence and prevalence of dysphagia requiring medical attention are increasing annually. The increasing trend was conspicuous in the geriatric population. The presence of stroke, neurodegenerative disease, cancer, and chronic obstructive pulmonary disease is associated with a high risk of dysphagia. Therefore, adequate screening, diagnosis, and management of dysphagia in the older population must be emphasized in geriatric healthcare.

Demographic and clinical factors associated with recovery of poststroke dysphagia: A meta-analysis

BACKGROUND

Poststroke dysphagia (PSD) recovery depends on various factors. We aimed to provide evidence concerning predictive variables for the recovery of PSD.

METHODS

PubMed, Embase, Web of Science, China National Knowledge Infrastructure (CNKI), Wanfang Database, VIP database of Chinese periodicals, Chinese biomedical literature service system (SinoMed), and Cochrane Library databases were systematically searched up to September 21, 2022. According to the inclusion criteria, the literature searched in the database was screened. The methodological quality of included studies was assessed using the Newcastle-Ottawa Scale (NOS). Meta-analysis was performed to identify the factors prognostic for PSD.

RESULTS

Twenty-eight studies were eligible, and pooled analyses were allowed for 12 potential prognostic factors. We identified older age, higher National Institutes of Health Stroke Scale (NIHSS) score, lower activities of daily living (ADL) score, lower body mass index (BMI), severe dysphagia on admission, aspiration, brainstem stroke, severe cognitive impairment, and bilateral hemispheric stroke were negative factors for the recovery of PSD, while early intervention and Modified Rankin Scale (mRS) = 0 before onset were protective factors for the recovery of PSD. There was no significant association between stroke type and prognosis of PSD.

CONCLUSION

Prognostic factors of PSD summarized in this meta-analysis could be useful for developing reasonable treatment plan to better promote recovery of swallowing function after stroke.

Clinical and imaging predictors of dysphagia and swallowing ability recovery in acute ischemic stroke

OBJECTIVE

Dysphagia is one of the most common complications of acute ischemic stroke, and prediction of dysphagia is crucial for post-stroke treatment. We aimed to identify predictors of dysphagia and swallowing function recovery following ischemic stroke and to investigate dysphagia-associated lesion location.

METHODS

We prospectively enrolled patients with acute ischemic stroke confirmed on diffusion-weighted imaging. All patients received swallowing evaluation within 48 h after admission. Follow-up oral intake ability was measured on 7 and 30 days after stroke onset. Voxel-based lesion-symptom mapping was performed to determine locations associated with dysphagia.

RESULTS

Of 126 patients included in the final analysis, 23 patients (18.3%) were classified as initial dysphagia. The presence of facial palsy (P = 0.008) and larger white matter hyperintensity (WMH) volume (P = 0.003) was associated with initial dysphagia. Initial risk of aspiration assessed by Any2 score (P = 0.001) at baseline was identified as independent predictor for dysphagia at day 7. Patients with higher Any2 score (P < 0.001), aphasia (P = 0.013), and larger WMH volume (P = 0.010) were less likely to have a full swallowing function recovery at 1 month. Acute infarcts in right corona radiata and right superior longitudinal fasciculus were correlated with impaired recovery of swallowing ability at 1 month.

CONCLUSIONS

Initial risk of aspiration was identified as risk factor for short-term and long-term dysphagia. Aphasia and larger WMH volume were revealed to be significant predictors for swallowing function recovery at 1 month. Right corona radiata was identified as an essential brain area for dysphagia.

Predicting Failure to Recover Swallowing in Patients with Severe Post-stroke Dysphagia: The DIsPHAGIc Score

In post-stroke dysphagia, early identification of patients at highest risk of failing swallowing recovery (SR) would be useful to decide which of them should undergo percutaneous endoscopic gastrostomy. The studies on this subject were numerous but generally based on small statistical samples. In this retrospective study, 1232 patients with ischemic or hemorrhagic stroke (73.7 ± 13.0 years, 51% men) were assessed: 593 non-dysphagic, 351 partially dysphagic and 288 totally dysphagic. Among the latter, 45.1% could not recover oral intake. A score to assess the risk of failing SR was obtained from the group with total dysphagia, and further 210 patients with total post-stroke dysphagia were utilized for validation. A regular progression of stroke severity markers, complications and mortality was observed from non-dysphagic, to partially dysphagic, up to totally dysphagic patients. Among the latter, seven variables were independently associated with failure of SR, and formed the "DIsPHAGIc score": cerebral lesion Diameter ≥ 6 cm (+ 1), left frontal Ischemia (- 1), Partial anterior circulation syndrome (- 1), Hypoxia (+ 1), Antiplatelet drug (+ 1), GCS verbal reaction < 4 (+ 1), Internal capsule ischemia (- 1). The area under the ROC curve was 0.79 (95% CI 0.74-0.85). For total scores ≥ 2 there was a high risk of failing SR, with specificity 76.9%, sensitivity 72.1% and accuracy 74.7%. The application of the DIsPHAGIc score to the validation sample provided almost identical results. The evolution of post-stroke dysphagia towards irreversibility can be predicted by a simple, reproducible and robust scoring system based on 7 variables commonly available during hospitalization.

Clinical Predictors of Dysphagia Recovery After Stroke: A Systematic Review

Oropharyngeal dysphagia is common post-stroke and can have serious consequences for patients. Understanding dysphagia recovery is critically important to inform prognostication and support patients and professionals with care planning. This systematic review was undertaken to identify clinical predictors of dysphagia recovery post-stroke. Online databases (EMBASE, Scopus, Web of Science, PubMed, CINAHL, and Cochrane) were searched for studies reporting longitudinal swallowing recovery in adults post-stroke. Dysphagia recovery was defined as improvement measured on a clinical swallowing scale or upgrade in oral and/or enteral feeding status by the end of the follow-up period. The search strategy returned 6598 studies from which 87 studies went through full-text screening, and 19 studies were included that met the eligibility criteria. Age, airway compromise identified on instrumental assessment, dysphagia severity, bilateral lesions, and stroke severity were identified as predictors of persistent dysphagia and negative recovery in multiple logistic regression analysis. The available literature was predominated by retrospective data, and comparison of outcomes was limited by methodological differences across the studies in terms of the choice of assessment, measure of recovery, and period of follow-up. Future prospective research is warranted with increased representation of haemorrhagic strokes and uniform use of standardized scales of swallowing function.

Predictive Factors for Oral Intake Recovery After Acute Stroke: Analysis of a Japanese Nationwide Inpatient Database

OBJECTIVE

It remains unclear which factors contribute to dysphagia recovery after an acute stroke. This study aimed to identify factors associated with complete oral intake recovery in patients with post-stroke dysphagia.

METHODS

Data were obtained from the Diagnosis Procedure Combination database, which is a nationwide database of administrative claims and discharge abstracts in Japan. We selected patients admitted within 3 days of stroke onset and conducted multivariable logistic regression analysis. Total oral intake within 30 days of admission was the primary outcome measure.

RESULTS

A total of 151,302 patients were included, and total oral intake was observed in 48% of them within 30 days of admission. Total oral intake was significantly associated with age, sex, stroke subtype, consciousness disturbance, low body mass index, multiple comorbidities, and the modified Rankin Scale before stroke onset. In addition, males had worse outcomes than females (odds ratio, 0.75; 95% confidence interval, 0.73-0.77, p < 0.001), and intracerebral and subarachnoid hemorrhagic stroke subtypes were associated with non-recovery.

CONCLUSION

This study identified several prognostic factors for total oral intake in patients with acute stroke. These results may be useful for predicting a patient's dysphagia prognosis at the time of admission and designing a nutritional management plan for patients with acute stroke.

The Clinical Utility of Leukoaraiosis as a Prognostic Indicator in Ischemic Stroke Patients

Stroke constitutes a major cause of functional disability with increasing prevalence among adult individuals. Thus, it is of great importance for both clinicians and stroke survivors to be provided with a timely and accurate prognostication of functional outcome. A great number of biomarkers capable of yielding useful information regarding stroke patients' recovery propensity have been evaluated so far with leukoaraiosis being among them. Literature research of two databases (MEDLINE and Scopus) was conducted to identify all relevant studies published between 1 January 2012 and 25 June 2022 that dealt with the clinical utility of a current leukoaraiosis as a prognostic indicator following stroke. Only full-text articles published in English language were included. Forty-nine articles have been traced and are included in the present review. Our findings highlight the prognostic value of leukoaraiosis in an acute stroke setting. The assessment of leukoaraiosis with visual rating scales in CT/MRI imaging appears to be able to reliably provide important insight into the recovery potential of stroke survivors, thus significantly enhancing stroke management. Yielding additional information regarding both short- and long-term functional outcome, motor recovery capacity, hemorrhagic transformation, as well as early neurological deterioration following stroke, leukoaraiosis may serve as a valuable prognostic marker poststroke. Thus, leukoaraiosis represents a powerful prognostic tool, the clinical implementation of which is expected to significantly facilitate the individualized management of stroke patients.

Effects of Insular Cortex on Post-Stroke Dysphagia: A Systematic Review and Meta Analysis

Objective: To investigate the relationship of lobar and deep brain regions with post-stroke dysphagia (PSD). Method: The databases of Medline, Embase, Web of Science, and Cochrane Library were searched from the establishment to May 2022. Studies that investigated the effects of lesions in lobar and deep brain regions on swallowing function after stroke were screened. The primary outcomes were PSD-related brain regions (including aspiration-related and oral transit time-related brain regions). The secondary outcomes were the incidence rate of PSD. The brain regions with the most overlap in the included studies were considered to be most relevant to PSD, and were presented as percentages. Data were compared utilizing the t-tests for continuous variables and χ2 for frequency-based variables. Result: A total of 24 studies and 2306 patients were included. The PSD-related lobar and deep brain regions included the insular cortex, frontal lobe, temporal gyrus, basal ganglia, postcentral, precentral, precuneus, corona radiate, etc. Among these brain regions, the insular cortex was most frequently reported (taking up 54.2%) in the included studies. Furthermore, the total incidence rate of PSD was around 40.4%, and the incidence of male was nearly 2.57 times as much as that of female (χ2 = 196.17, p < 0.001). Conclusions: In lobar and deep brain regions, the insular cortex may be most relevant to PSD and aspiration, which may be a potentially promising target in the treatment of PSD.

Predicting Outcome in Patients with Brain Injury: Differences between Machine Learning versus Conventional Statistics

Defining reliable tools for early prediction of outcome is the main target for physicians to guide care decisions in patients with brain injury. The application of machine learning (ML) is rapidly increasing in this field of study, but with a poor translation to clinical practice. This is basically dependent on the uncertainty about the advantages of this novel technique with respect to traditional approaches. In this review we address the main differences between ML techniques and traditional statistics (such as logistic regression, LR) applied for predicting outcome in patients with stroke and traumatic brain injury (TBI). Thirteen papers directly addressing the different performance among ML and LR methods were included in this review. Basically, ML algorithms do not outperform traditional regression approaches for outcome prediction in brain injury. Better performance of specific ML algorithms (such as Artificial neural networks) was mainly described in the stroke domain, but the high heterogeneity in features extracted from low-dimensional clinical data reduces the enthusiasm for applying this powerful method in clinical practice. To better capture and predict the dynamic changes in patients with brain injury during intensive care courses ML algorithms should be extended to high-dimensional data extracted from neuroimaging (structural and fMRI), EEG and genetics.

Characteristics of dysphagia among different lesion sites of stroke: A retrospective study

Objective

This study aims to compare the characteristics of dysphagia among different lesion sites and explore the possible risk factors that are relevant to penetration and aspiration after stroke.

Materials and methods

Data on patients with post-stroke dysphagia were collected. Major measures of the videofluoroscopic swallowing study included pharyngeal transit duration (PTD), pharyngeal response duration (PRD), soft palate elevation duration (SED), stage transition duration (STD), hyoid bone anterior-horizontal displacement (HAD), hyoid bone superior-horizontal displacement (HSD), upper esophageal sphincter opening (UESO), Pharyngeal Residual Grade (PRG), and Penetration Aspiration Scale (PAS). Included patients were divided into supratentorial (deep or lobar intracerebral) and infratentorial stroke groups. The Kruskal–Wallis test, Spearman’s correlation analysis, and multivariate logistic regression analyses were used to test the difference and the correlation between those measures. Time-to-event endpoints (oral feeding) were analyzed by the Kaplan–Meier method.

Results

A total of 75 patients were included in this study. Significant differences were demonstrated in PTD, PRD, SED, STD, HAD, HSD, UESO, PAS, and PRG between supratentorial and infratentorial stroke groups (p < 0.05). The PRG score of the lobar intracerebral subgroup was significantly higher (p < 0.05) than that of the deep intracerebral and lobar + deep intracerebral stroke subgroups, while HSD was significantly shorter (p < 0.01). Spearman’s correlation analysis revealed that PAS was related to PTD, PRG, HAD, and UESO (p < 0.05). Multivariate logistic regression analysis demonstrated that HAD and PRG may be risk factors for penetration and aspiration (p < 0.05). Kaplan–Meier survival plot showed that there was a significant difference in time to oral feeding between supratentorial and infratentorial stroke groups (p < 0.01).

Conclusion

Infratentorial stroke may lead to worse swallowing function as compared with supratentorial stroke, and lobar intracerebral stroke may be worse than deep intracerebral stroke. Suitable preventive measures may be considered for patients with higher PRG scores and shorter HSD to avoid penetration and aspiration.

Swallowing function in the chronic stage following stroke is associated with white matter integrity of the callosal tract between the interhemispheric S1 swallowing representation areas

•

Swallowing performance was tested in dysphagic patients following stroke.

•

M1 and S1 callosal tracts relevant for swallowing was mapped in the HCP-dataset.

•

S1 and M1 swallowing tracts were overlapping between in house and HCP datasets.

•

Swallowing specific callosal tracts showed lower FA for patients compared to HCs.

•

Integrity of S1 callosal fibres (FA) was associated with swallowing performance.

Sensorimotor representations of swallowing in pre- and postcentral gyri of both cerebral hemispheres are interconnected by callosal tracts. We were interested in (1) the callosal location of fibers interconnecting the precentral gyri (with the primary motor cortex; M1) and the postcentral gyri (with the primary somatosensory cortex; S1) relevant for swallowing, and (2) the importance of their integrity given the challenges of swallowing compliance after recovery of dysphagia following stroke. We investigated 17 patients who had almost recovered from dysphagia in the chronic stage following stroke and age-matched and gender-matched healthy controls. We assessed their swallowing compliance, investigating swallowing of a predefined bolus in one swallowing movement in response to a ‘go’ signal when in a lying position. A somatotopic representation of swallowing was mapped for the pre- and postcentral gyrus, and callosal tract location between these regions was compared to results for healthy participants. We applied multi-directional diffusion-weighted imaging of the brain in patients and matched controls to calculate fractional anisotropy (FA) as a tract integrity marker for M1/S1 callosal fibers. Firstly, interconnecting callosal tract maps were well spatially separated for M1 and S1, but were overlapped for somatotopic differentiation within M1 and S1 in healthy participants’ data (HCP: head/face representation; in house dataset: fMRI-swallowing representation in healthy volunteers). Secondly, the FA for both callosal tracts, connecting M1 and S1 swallowing representations, were decreased for patients when compared to healthy volunteers. Thirdly, integrity of callosal fibers interconnecting S1 swallowing representation sites was associated with effective swallowing compliance. We conclude that somatosensory interaction between hemispheres is important for effective swallowing in the case of a demanding task undertaken by stroke survivors with good swallowing outcome from dysphagia.

Predictors of complete oral feeding resumption after feeding tube placement in patients with stroke and dysphagia: A systematic review

AIMS

The aim was to synthesise and discuss predictors of complete oral feeding resumption after feeding tube placement in stroke patients with dysphagia.

DESIGN

This was a systematic review, following the PRISMA 2020 checklist.

DATA SOURCES

Eight databases (PubMed, Web of Science, Embase, Cochrane, CINAHL, CNKI, WanFang and Vip) were searched for eligible studies from inception up to June 2021.

REVIEW METHODS

The JBI Manual for Evidence Synthesis was used to guide this systematic review. Any cross-sectional survey, longitudinal study, cohort study or case-control study that explored the recovery from tube feeding to complete oral feeding in patients with dysphagia after stroke was included. Qualitative studies, review articles, case reports and conference abstracts were excluded. Two reviewers independently screened and appraised the studies. The Newcastle-Ottawa scale was used for quality assessment. Content analysis was used to categorise factors predicting feeding tube removal in stroke patients with dysphagia.

RESULTS

This review included a total of 15 studies consisting of 1746 participants, of which 2 were case-control studies and 13 were cohort studies. Four studies were rated as having low risk of bias, and the other 11 had high risk of bias. The factors examined in the studies were categorised into demographic characteristics (age and sex), swallowing function (instrumental assessments and non-instrumental assessments), stroke characteristics (stroke severity, past stroke history and location of the stroke), functional status (cognitive function and physical function) and clinical measures (body mass index, geriatric nutritional risk index, white blood cell count and C-reactive protein level).

CONCLUSIONS

The major limitation of this review is the failure to identify predictors of different tube feeding types. Although the current evidence is insufficient to support or oppose the predictive effect of any single factor, these factors are still valuable data for clinical staff that provide information that researchers can use in developing prognostic models. Rigorously designed and high-quality research is needed to further explore the predictive value of these factors.

REGISTRATION

This review was registered prospectively with PROSPERO [CRD42021272552].

RELEVANCE TO CLINICAL PRACTICE

Healthcare providers should strengthen the monitoring of swallowing function in patients with stroke to promote complete oral feeding resumption. As the predictive value of the identified factors is still uncertain, large, well-designed, studies are needed to better clarify the importance of these predictors.

NO PATIENT OR PUBLIC CONTRIBUTION

No patient or public was involved in the design, analysis, preparation or writing of this review.

Prediction-Driven Decision Support for Patients With Mild Stroke: A Model Based on Machine Learning Algorithms

Background and Purpose: Treatment for mild stroke remains an open question. We aim to develop a decision support tool based on machine learning (ML) algorithms, called DAMS (Disability After Mild Stroke), to identify mild stroke patients who would be at high risk of post-stroke disability (PSD) if they only received medical therapy and, more importantly, to aid neurologists in making individual clinical decisions in emergency contexts.

Methods: Ischemic stroke patients were prospectively recorded in the National Advanced Stroke Center of Nanjing First Hospital (China) between July 2016 and September 2020. The exclusion criteria were patients who received thrombolytic therapy, age <18 years, lack of 3-month modified Rankin Scale (mRS), disabled before the index stroke, with an admission National Institute of Health stroke scale (NIHSS) > 5. The primary outcome was PSD, corresponding to 3-month mRS ≥ 2. We developed five ML models and assessed the area under curve (AUC) of receiver operating characteristic, calibration curve, and decision curve analysis. The optimal ML model was selected to be DAMS. In addition, SHapley Additive exPlanations (SHAP) approach was introduced to rank the feature importance. Finally, rapid-DAMS (R-DAMS) was constructed for a more urgent situation based on DAMS.

Results: A total of 1,905 mild stroke patients were enrolled in this study, and patients with PSD accounted for 23.4% (447). There was no difference in AUCs between the five models (ranged from 0.691 to 0.823). Although there was similar discriminative performance between ML models, the support vector machine model exhibited higher net benefit and better calibration (Brier score, 0.159, calibration slope, 0.935, calibration intercept, 0.035). Therefore, this model was selected for DAMS. In addition, SHAP approach showed that the most crucial feature was NIHSS on admission. Finally, R-DAMS was constructed and there was similar discriminative performance between R-DAMS and DAMS, but the former performed worse on calibration.

Conclusions: DAMS and R-DAMS, as prediction-driven decision support tools, were designed to aid clinical decision-making for mild stroke patients in emergency contexts. In addition, even within a narrow range of baseline scores, NIHSS on admission is the strongest feature that contributed to the prediction.

Can non-swallowing function assessment predict nasogastric tube removal in patients with poststroke dysphagia? A clinical study

Objective

This study aimed to predict nasogastric tube (NGT) removal in patients with poststroke dysphagia (PSD) by non-swallowing function assessment.

Methods

We enrolled 232 eligible patients and performed rehabilitation. The Fugl-Meyer assessment motor (FMM) and National Institute of Health Stroke Scale (NIHSS) scores were used to measure the motor and overall nervous system functions. Predictors for NGT removal in patients with PSD after rehabilitation were analyzed.

Results

Of the 232 included patients, the NGTs were removed from 78% of them, while 22% were dependent on a feeding tube after 4 weeks of rehabilitation. Compared to the preserved NGT group, older age, a higher rate of intubation or tracheostomy, and more severe baseline functions were found in the NGT removal group. Age [odds ratio (OR) = 0.907; 95% confidence interval (CI): 0.859-0.957; p = 0.000], difference in the FMM score after 4 weeks of rehabilitation (OR = 1.219; 95% CI: 1.145-1.299; p = 0.00), and item 9 of NIHSS (OR = 0.488; 95% CI: 0.252-0.946; p = 0.034) were predictors of NGT removal after rehabilitation.

Conclusion

We established a predictive model in patients with PSD using a non-swallowing assessment, which enabled us to predict swallowing recovery based on the non-swallowing function.

Neural correlates in the development of and recovery from dysphagia after supratentorial stroke: A prospective tractography study

•

Stroke may lead to unilateral or bilateral CBT changes regardless of dysphagia.

•

Poor fractional anisotropy of the unaffected sides relates to limited recovery.

•

Swallowing recovery may depend on the integrity of the unaffected CBT.

•

Bi-hemispheric reorganization of the CBT is associated with swallowing recovery.

Background

Methods

Results

Conclusions

Predicting functional outcome in patients with acute brainstem infarction using deep neuroimaging features

BACKGROUND AND PURPOSE

Acute brainstem infarctions can lead to serious functional impairments. We aimed to predict functional outcomes in patients with acute brainstem infarction using deep neuroimaging features extracted by convolutional neural networks (CNNs).

METHODS

This nationwide multicenter stroke registry study included 1482 patients with acute brainstem infarction. We applied CNNs to automatically extract deep neuroimaging features from diffusion-weighted imaging. Deep learning models based on clinical features, laboratory features, conventional imaging features (infarct volume, number of infarctions), and deep neuroimaging features were trained to predict functional outcomes at 3 months poststroke. Unfavorable outcome was defined as modified Rankin Scale score of 3 or higher at 3 months. The models were evaluated by comparing the area under the receiver operating characteristic curve (AUC).

RESULTS

A model based solely on 14 deep neuroimaging features from CNNs achieved an extremely high AUC of 0.975 (95% confidence interval [CI] = 0.934-0.997) and significantly outperformed the model combining clinical, laboratory, and conventional imaging features (0.772, 95% CI = 0.691-0.847, p < 0.001) in prediction of functional outcomes. The deep neuroimaging model also demonstrated significant improvement over traditional prognostic scores. In an interpretability analysis, the deep neuroimaging features displayed a significant correlation with age, National Institutes of Health Stroke Scale score, infarct volume, and inflammation factors.

CONCLUSIONS

Deep learning models can successfully extract objective neuroimaging features from the routine radiological data in an automatic manner and aid in predicting the functional outcomes in patients with brainstem infarction at 3 months with very high accuracy.

Status of dysphagia after ischemic stroke: A Korean nationwide study

OBJECTIVE

To identify the incidence of dysphagia after ischemic stroke and determine factors affecting the presence of dysphagia.

DESIGN

Retrospective case-control study. This was an interim analysis of a prospective multicenter Korean stroke cohort.

SETTING

Acute care university hospitals.

PARTICIPANTS

Patients (N=6000) with first-ever acute ischemic stroke. Patients were divided into two groups according to the presence or absence of dysphagia confirmed at 7 days after onset using the American Speech-Language-Hearing Association National Outcomes Measurement System (ASHA-NOMS) scale, which was determined after conducting screening or standardized tests.

INTERVENTIONS

Not applicable.

MAIN OUTCOME MEASURES

Age at stroke onset, body mass index (BMI), premorbid modified Rankin Scale (mRS), brainstem lesions, National Institutes of Health Stroke Scale (NIHSS), post-stroke mRS, and ASHA-NOMS swallowing level at post-stroke day 7 were evaluated.

RESULTS

Among ischemic stroke patients, 32.3% (N=1940) had dysphagia at 7 days after stroke onset. At discharge, 80.5% (N=1561) still had dysphagia. The prediction model for the presence of dysphagia identified age at onset, underweight (BMI < 18.5 kg/m²), premorbid mRS, brainstem lesions, and NIHSS as independent predictors. The odds ratio (OR) for the presence of dysphagia significantly increased with underweight (OR [95% confidence interval]: 1.6684 [1.27-2.20]), increased age at onset (1.0318 [1.03-1.04]), premorbid mRS (1.1832 [1.13-1.24]), brainstem lesions (1.6494 [1.39-1.96]), and NIHSS (1.2073 [1.19-1.23]).

CONCLUSIONS

The incidence of dysphagia after ischemic stroke was 32.3%. The prediction model for the presence of dysphagia identified age, low BMI, premorbid disabilities, brainstem lesions, and NIHSS as predictive factors.

Potential Prognostic Impact of Dopamine Receptor D1 (rs4532) Polymorphism in Post-stroke Outcome in the Elderly

Background: Single-nucleotide polymorphisms (SNPs) may affect post-stroke motor recovery, and some SNPs have been implicated in swallowing disturbances after stroke. Certain SNPs may also have altered influences according to different age. Objective: This post-hoc study investigated whether SNPs have different effects on dysphagia recovery between the elderly vs. young stroke patients. Methods: Analysis was conducted from a previous study including 218 stroke subjects with dysphagia. They were stratified into two groups, aged <65 and aged ≥65 years. The primary outcome was persistence of nil per mouth (NPM) at 3 months post-stroke onset. Association between outcome and nine different SNPs were investigated. Results: The elderly group (50%, n = 103) showed poorer swallowing outcomes than the young group. The minor allele of the dopamine receptor D1 (DRD1, rs4532) polymorphism showed potential association (p = 0.022) with an increased risk of NPM at 12 weeks post-stroke in the elderly, both in the additive (OR, 2.94; 95% CI, 1.17-7.37) and dominant models (OR, 2.93; 95% CI, 1.04-8.23) but did not reach statistical significance after Bonferonni correction. Logistic regression analysis showed that in those aged ≥65 years, models including the minor allele of rs4532 predicted the risk of the poor outcome with good accuracies even after adjustment of clinical factors, such as previous pneumonia episodes (AUROC, 0.86; 95% CI, 0.79-0.93) or the National Institutes of Health Stroke Scale (AUROC, 0.82; 95% CI, 0.67-0.92). In contrast, those aged <65 years seemed not to be affected by the presence of the rs4532 polymorphism, and models that included intubation history (AUROC, 0.81; 95% CI, 0.73-0.90) or previous pneumonia episodes (AUROC, 0.77; 95% CI, 0.68-0.87) showed modest levels of accuracies in predicting NPM at 12 weeks poststroke. Conclusions: Our study suggests a possible association between the rs4532 and post-stroke swallowing recovery, primarily in those aged ≥65 years. Certain SNPs may lead to less favorable outcomes in the elderly. The gene-age interaction should be considered in post-stroke swallowing recovery. Clinical Trial Registration: https://www.clinicaltrials.gov, Unique identifier [NCT03577444].

The Role of White Matter in the Neural Control of Swallowing: A Systematic Review

Background: Swallowing disorders (dysphagia) can negatively impact quality of life and health. For clinicians and researchers seeking to improve outcomes for patients with dysphagia, understanding the neural control of swallowing is critical. The role of gray matter in swallowing control has been extensively documented, but knowledge is limited regarding the contributions of white matter. Our aim was to identify, evaluate, and summarize the populations, methods, and results of published articles describing the role of white matter in neural control of swallowing.

Methods: We completed a systematic review with a multi-engine search following PRISMA-P 2015 standards. Two authors screened articles and completed blind full-text review and quality assessments using an adapted U.S. National Institute of Health's Quality Assessment. The senior author resolved any disagreements. Qualitative synthesis of evidence was completed.

Results: The search yielded 105 non-duplicate articles, twenty-two of which met inclusion criteria. Twenty were rated as Good (5/22; 23%) or Fair (15/22; 68%) quality. Stroke was the most represented diagnosis (n = 20; 91%). All studies were observational, and half were retrospective cohort design. The majority of studies (13/22; 59%) quantified white matter damage with lesion-based methods, whereas 7/22 (32%) described intrinsic characteristics of white matter using methods like fractional anisotropy. Fifteen studies (68%) used instrumental methods for swallowing evaluations. White matter areas commonly implicated in swallowing control included the pyramidal tract, internal capsule, corona radiata, superior longitudinal fasciculus, external capsule, and corpus callosum. Additional noteworthy themes included: severity of white matter damage is related to dysphagia severity; bilateral white matter lesions appear particularly disruptive to swallowing; and white matter adaptation can facilitate dysphagia recovery. Gaps in the literature included limited sample size and populations, lack of in-depth evaluations, and issues with research design.

Conclusion: Although traditionally understudied, there is sufficient evidence to conclude that white matter is critical in the neural control of swallowing. The reviewed studies indicated that white matter damage can be directly tied to swallowing deficits, and several white matter structures were implicated across studies. Further well-designed interdisciplinary research is needed to understand white matter's role in neural control of normal swallowing and in dysphagia recovery and rehabilitation.

Prediction of Progression to Severe Stroke in Initially Diagnosed Anterior Circulation Ischemic Cerebral Infarction

Purpose: Accurate prediction of the progression to severe stroke in initially diagnosed nonsevere patients with acute-subacute anterior circulation nonlacuna ischemic infarction (ASACNLII) is important in making clinical decision. This study aimed to apply a machine learning method to predict if the initially diagnosed nonsevere patients with ASACNLII would progress to severe stroke by using diffusion-weighted images and clinical information on admission. Methods: This retrospective study enrolled 344 patients with ASACNLII from June 2017 to August 2020 on admission, and 108 cases progressed to severe stroke during hospitalization within 3-21 days. The entire data were randomized into a training set (n = 271) and an independent test set (n = 73). A U-Net neural network was employed for automatic segmentation and volume measurement of the ischemic lesions. Predictive models were developed and used for evaluating the progression to severe stroke using different feature sets (the volume data, the clinical data, and the combination) and machine learning methods (random forest, support vector machine, and logistic regression). Results: The U-Net showed high correlation with manual segmentation in terms of Dice coefficient of 0.806 and R 2 value of the volume measurements of 0.960 in the test set. The random forest classifier of the volume + clinical combination achieved the best area under the receiver operating characteristic curve of 0.8358 (95% CI 0.7321-0.9269), and the accuracy, sensitivity, and specificity were 0.7780 (0.7397-0.7945), 0.7695 (0.6102-0.9074), and 0.8686 (0.6923-1.0), respectively. The Shapley additive explanation diagram showed the volume variable as the most important predictor. Conclusion: The U-Net was fully automatic and showed a high correlation with manual segmentation. An integrated approach combining clinical variables and stroke lesion volumes that were derived from the advanced machine learning algorithms had high accuracy in predicting the progression to severe stroke in ASACNLII patients.

Hyoid kinematic features for poor swallowing prognosis in patients with post-stroke dysphagia

Identification of prognostic factors for swallowing recovery in patients with post-stroke dysphagia is crucial for determining therapeutic strategies. We aimed at exploring hyoid kinematic features of poor swallowing prognosis in patients with post-stroke dysphagia. Of 122 patients who experienced dysphagia following ischemic stroke, 18 with poor prognosis, and 18 age- and sex-matched patients with good prognosis were selected and retrospectively reviewed. Positional data of the hyoid bone during swallowing were obtained from the initial videofluoroscopic swallowing study after stroke onset. Normalized hyoid profiles of displacement/velocity and direction angle were analyzed using functional regression analysis, and maximal or mean values were compared between the good and poor prognosis patient groups. Kinematic analysis showed that maximal horizontal displacement (P = 0.031) and velocity (P = 0.034) in forward hyoid motions were significantly reduced in patients with poor prognosis compared to those with good prognosis. Mean direction angle for the initial swallowing phase was significantly lower in patients with poor prognosis than in those with good prognosis (P = 0.0498). Our study revealed that reduced horizontal forward and altered initial backward motions of the hyoid bone during swallowing can be novel kinematic features indicating poor swallowing prognosis in patients with post-stroke dysphagia.

Role of rs6265 BDNF polymorphisms and post-stroke dysphagia recovery-A prospective cohort study

BACKGROUND

Previous studies suggest that specific genes may predispose some to increased risk of dysphagia in the geriatric population, but whether these genes may affect swallowing recovery after a stroke is unknown. This study investigated whether single-nucleotide polymorphisms (SNP) of the brain-derived neurotrophic factor (BDNF), catechol-O-methyl transferase, apolipoprotein E, interleukin-1 receptor antagonist, and dopamine, which have been linked to swallowing, could adversely affect the prognosis of post-stroke dysphagia.

METHODS

In this study, 218 subjects with confirmed post-stroke dysphagia were enrolled. The primary endpoint was failed recovery from nil per mouth (NPM) status with the first 3 months post-stroke.

KEY RESULTS

The Val/Val group from the rs6265, BDNF, showed higher score changes on the Functional Oral Intake Scale at 1 month. The proportion of patients with recovery from NPM status within the first 1 month was 60.8% in the Val/Val group, which was statistically higher than those in the Met allele groups (38.1%, P = .017). At 3 months, the BDNF rs6265 showed significant group differences in Modified Barium Swallow Impairment Profile^© score changes with the Val/Val allele leading to greater improvement. However, no single SNP was associated with increased risk of poor recovery with persistence of NPM at 3 months post-stroke.

CONCLUSIONS AND INFERENCES

Those with the dominant Val/Val phenotype of BDNF manifested with faster and greater improvement than the Met-phenotypes. Based on our results, the BDNF Val allele may play a positive role with faster score improvement and rapid recovery from NPM than the Met allele. Clinical Trials gov: NCT03577444 (https://clinicaltrials.gov/ct2/show/study/NCT03577444).

Role of bilateral corticobulbar tracts in dysphagia after middle cerebral artery stroke

BACKGROUND AND PURPOSE

The corticobulbar tract is a potential neural pathway involved in swallowing. The frontal operculum, insular cortex, corona radiata and internal capsule, which are frequently involved in middle cerebral artery (MCA) strokes, are locations in which lesions cause dysphagia. However, it is unclear whether the locations are linked to the corticobulbar tract or whether corticobulbar tract integrity is associated with dysphagia severity. This study aimed to assess the association between corticobulbar tract integrity and dysphagia severity after MCA stroke.

METHODS

Thirty dysphagic patients after MCA stroke and 27 healthy controls were examined. Diffusion tensor imaging (DTI)-derived parameters of the corticobulbar tract were compared between patient and control groups. Next, patients were divided into mild and moderate-to-severe dysphagia groups, and DTI-derived parameters of the corticobulbar tract were compared between the subgroups. Logistic regression analysis was used to determine the association between corticobulbar tract integrity and dysphagia severity.

RESULTS

The tract volume (TV) of the affected corticobulbar tract was lower in dysphagic patients than in healthy controls (P < 0.001). According to dysphagia severity, TV of the unaffected corticobulbar tract was higher in the mild dysphagia group than in the moderate-to-severe dysphagia group (P = 0.012). TV of the unaffected corticobulbar tract was independently associated with dysphagia severity according to the logistic regression model (adjusted odds ratio 0.817, 95% confidence interval 0.683-0.976).

CONCLUSIONS

The corticobulbar tract was affected after MCA stroke and may be associated with dysphagia. A higher corticobulbar TV in the unaffected hemisphere was indicative of better swallowing function in dysphagic patients after MCA stroke.