https://orcid.org/0009-0005-7909-9895
Figures
Abstract
This study aims to develop and validate a prediction model for nutritional risk in Post-Stroke Dysphagia (PSD). This study retrospectively analyzed data on stroke patients with dysphagia from January 2022 to December 2023. A stepwise logistic regression model was used to construct the prediction model, and internal validation was performed using the bootstrap resampling (1000 iterations), the nomogram was developed for clinical applications. The final prediction model incorporated the following factors: age, marital status, mechanical ventilation, dysphagia treatment, fasting duration, atrial fibrillation, oral care frequency, serum potassium levels, and National Institute of Health stroke scale (NIHSS) score. The model demonstrated strong discriminatory power, with area under the ROC curve (AUC) values of 0.916 in the development set and 0.878 in the validation set. Calibration curves and the Hosmer-Lemeshow (H-L) test further confirmed the strong correlation between predicted and observed nutritional risks.The prediction model developed in this study exhibits high accuracy, consistency, and practical applicability, making it a valuable tool for predicting nutritional risk in PSD patients. The code library: https://osf.io/p3hjm.
Citation: Han A, Zhang S, Yao Y, He J, Wang Y, Yang H (2025) Development and validation of a risk predication model for nutritional risk based on water swallowing test in post-stroke with dysphagia. PLoS One 20(10): e0330982. https://doi.org/10.1371/journal.pone.0330982
Editor: Maheshkumar Baladaniya, Neighborhood Physical Therapy, UNITED STATES OF AMERICA
Received: March 11, 2025; Accepted: August 9, 2025; Published: October 3, 2025
Copyright: © 2025 Han et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: This study data contains potentially identifying or sensitive patient information. Data is available from the Henan University of Science and Technology Institutional Data Access for corresponding authors who meet the criteria for access to confidential data. If you require the data, please contact Henan University of Science and Technology Institutional through the following email address: yxybgs@mail.haust.edu.cn.
Funding: Huimin Yang was supported by grants from the Henan Province science and technology joint construction project (No. LHGJ20230457) and Key scientific research project of Henan University (No. 24B320003). Huimin Yang play an important role in research design, publication decisions, and manuscript preparation.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Stroke is an acute neurological condition primarily resulting from disturbances in cerebral blood flow, often due to cerebrovascular diseases. It is classified into ischemic and hemorrhagic stroke [1]. According to the World Stroke Organization, over 12.2 million new stroke cases are reported globally each year, and approximately 6.5 million stroke-related deaths occur annually [2]. Stroke patients frequently experience various complications, with Post-Stroke Dysphagia (PSD) being one of the most common [3]. PSD is a functional disorder that impairs the safe and efficient transfer of food from the mouth to the stomach, with an incidence ranging from 22% to 70% [4]. The main causes of dysphagia are bulbar palsy resulting from bilateral cortical or brainstem tract injuries, and damage to the tongue and pharyngeal nerves, leading to symptoms such as excessive salivation, coughing, and ineffective swallowing.
Dysphagia not only increases the risk of complications like aspiration and lung infection, but also leads to reduced food intake and malnutrition in PSD patients due to neurological impairment and decreased daily activity [5,6]. Nutritional risk, distinct from malnutrition, refers to the potential impact of nutritional deficiencies on disease-related clinical outcomes, with higher incidence rates. This can result in poor functional recovery of hemiplegic limbs, more frequent infections, longer hospital stays, increased healthcare costs, reduced quality of life, and higher mortality [7,8].Therefore, nutritional risk screening is vital for improving PSD patient outcomes. The American Medical Review Committee recommends that hospitals complete nutritional screening within 24 hours of patient admission [9]. The American Association of Enteral and Parenteral Nutrition also emphasizes that nutritional risk screening is the first step in standardized nutrition management [10]. However, there are no uniform standards for screening tools. While commonly used tools like Nutritional Risk Screening 2002 (NRS-2002), Mini Nutritional Assessment-Short Form(MNA-SF), Malnutrition Universal Screening Tool (MUST), and Malnutrition Universal Screening Tool-Short Form (MNA-SF) are prevalent in clinics, their specificity is limited, and they are most effective for patients with clear awareness and good communication skills. These tools may not fully capture the nutritional risk factors of PSD patients, limiting screening accuracy and sensitivity.
Consequently, constructing a nutritional risk prediction model for PSD patients can enable early and accurate screening of high-risk groups, improving patient quality of life, reducing adverse health outcomes, and enhancing clinical management efficiency. This approach holds significant clinical value.
Objectives
The objective of this study was to establish a nutritional risk screening nomogram for PSD patients based on the characteristics of PSD patients.
Methods
Study design and patients
We retrospectively collected the data of 601 eligible patients with PSD from the First Affiliated Hospital of Henan University of Science and Technology on April 30, 2024, with data from January 2022 to June 2023 used as the development set and data from 259 PSD patients between July 2023 and December 2023 as the validation set.
The inclusion criteria included: (1) age ≥ 18 years old; (2) stroke patients met the Diagnostic criteria of Diagnostic Points of Various Major cerebrovascular Diseases in China 2019 [11]; (3) water swallowing test (WST) ≥ grade Ⅲ. The exclusion criteria included: (1) previous history of dysphagia; (2) patients with hypermetabolic diseases such as tumors or hyperthyroidism; (3) have a history of mental illness; (4) same patient in different departments at the same time (retain complete information). The study was approved by The First Affiliated Hospital of Henan University of Science and Technology (No.2024-03-K0035).
Diagnostic criteria
Water swallow test.
The Water Swallow Test (WST), first proposed by Toshio Koda in 1982 [12], is an assessment tool for rapid bedside evaluation of swallowing function. The patient is asked to sit and drink 30 ml of warm water at a normal speed. The grading system is as follows: Grade I: One attempt to drink, no choking. Grade II: No choking after drinking more than twice. Grade III: Coughing occurs after one attempt. Grade IV: Coughing occurs after two attempts. Grade V: Unable to drink all 30 ml, with obvious coughing. Grade I is considered normal; Grade II is suspicious; Grade III and above are abnormal.
Nutritional risk definition.
NRS-2002 [13] is an objective nutritional risk screening tool developed by the European Society of Parenteral and Enteral Nutrition (ESPEN) Expert Group in 2002. It has been proven effective and reliable for assessing patients’ nutritional status. The scale consists of three components: (1) impaired nutritional status score (up to 3 points); (2) the impact of disease severity on nutritional status (up to 3 points); (3) age score (1 point for age ≥ 70 years). The total score ranges from 0 to 7, with a score ≥3 indicating nutritional risk.
Sample size.
According to the Events Per Variable (EPV) principle [14], the number of outcome events should be at least 10 times the number of dependent variables to ensure the robustness of regression analysis. Based on domestic studies [15], the incidence of nutritional risk in PSD patients is 60.5%, so the required sample size is at least 607 cases. In total, 860 patients were included in the study.
Predictive variable selection.
Through literature reviews, group discussions, and consultations with neurology and nutrition experts, predictors of nutritional risk in PSD patients were identified, including: age, sex, medical insurance, marital status, residence, caregiver, WST, stroke type, National Institute of Health stroke scale (NIHSS), Activities of daily living (ADL) score, digestive history, hypertension, diabetes, atrial fibrillation, pulmonary infection, digestive complications, other complications, mechanical ventilation, use of ≥2 prescription drugs, nutritional support mode, dysphagia treatment, frequency of incidence, oral care frequency, surgery, C-reactive protein(CRP), body mass index (BMI), fasting duration, hospital stay, total protein level, hemoglobin, serum potassium, albumin, neutrophils, and serum sodium.
Data collection
The research team consisted of two graduate students and one specialist nurse from the Department of Neurology. Data were collected through the hospital’s electronic medical record system based on the identified influencing factors. To ensure data accuracy and standardization, all investigators received uniform training before data collection, and two people checked whether the data was consistent.
Data preprocessing
Cases with more than 20% missing data are excluded. For data with less than 20% missing data, multiple imputation was used to fill in the continuous variables; For missing categorical variables, the mode is used to fill in.
Statistical analysis
Data analysis was performed using SPSS software (Version 25.0) and R software (Version 4.4.1). Continuous variables with a normal distribution are expressed as mean ± standard deviation (SD), and inter-group comparisons were made using the two-sample independent t-test. Non-normally distributed variables are presented as the median with interquartile range (IQR), and the Mann-Whitney U test was used for between-group comparisons. Categorical variables are presented as counts (%), and comparisons between groups were performed using the Chi-square test and Mann-Whitney U test. A P value < 0.05 was considered statistically significant.
First, univariate logistic regression was performed, and variables with a P value < 0.1 were included in multivariate logistic regression for independent correlation analysis. Multivariable regression was conducted using different methods, such as the inclusion method, forward selection, and backward elimination, creating models based on various variable sets. The optimal model was selected according to the minimum Akaike Information Criterion (AIC) [16], and a prediction model was developed.
The receiver operating characteristic (ROC) curve was used to estimate sensitivity and specificity, the larger the area under the ROC curve (AUC), the better the prediction performance of the model. The Hosmer-Lemeshow (H-L) test assessed the model, and the calibration curve was used to correct prediction accuracy. The decision curve analysis (DCA) evaluated the clinical applicability of the model. To assess model stability and prevent overfitting, bootstrapping was applied for internal validation. Fig 1 shows the flowchart of the prediction model development process. The code library: https://osf.io/p3hjm. Fig 1 Shows a flowchart for building the prediction model development process.
Results
Participant characteristics
A total of 942 patients were initially selected for the study. After excluding 82 patients with more than 20% missing data, 860 patients remained: 601 in the development set and 259 in the validation set. The overall incidence of nutritional risk was 64.30%, with 63.06% in the development set and 67.18% in the validation set. The basic characteristics and univariate analysis results of the development set are presented in Table 1.
Risk factors for nutritional risk
In this study, nutritional risk was the dependent variable, and 34 risk factors were considered as independent variables. Univariate logistic regression in the development set identified 24 predictors of nutritional risk, including 4 demographic factors, 16 disease-related factors, and 4 laboratory indicators (Table 2). Risk factors with a P value < 0.1 in the univariate analysis were further analyzed using multivariate regression, and the final model inclusion was determined by the minimum AIC. The selected factors included age, marital status, NIHSS score, atrial fibrillation, mechanical ventilation, dysphagia treatment, oral care frequency, fasting duration, and serum potassium.
Model development
Based on multivariate analysis results, a nomogram model was developed using nine independent factors to predict nutritional risk in PSD patients (Fig 2). How to use Nomogram: Within 24 hours of the patient’s admission, the medical staff scored each predictor according to the actual situation of the patient, added the total score, and projected it into the Predicted value, which is the predicted probability of nutritional risk, and the clinical staff were divided into two groups: high risk and low risk according to the optimal cut-off value of 0.672. For example, a person aged 60 years (47.5 points), no swallowing function therapy (15 points), divorced (10 points), no mechanical ventilation (0 points), fasted for 1 day (5 points), had a history of atrial fibrillation (25 points), had 1 oral care per day (5 points), blood potassium concentration was 3.5 (41 points), NIHSS score was between 5 ~ 15 points (17.5 points), and the total score was 166, and the corresponding probability of nutritional risk may occur is 82%. The optimal cut-off values of this model were 0.672, 0.82 > 0.672, suggesting that the patient was a high-risk group and should be intervened as soon as possible. Fig 2. Nomogram prediction of nutritional risk in PSD.
As a predictive tool for PSD nutritional risk, nomogram assigns each factor a score based on a scoring system. The sum of these points produces an overall score that, when aligned with the scale, provides the likelihood that PSD nutritional risk will occur.
Internal and external model validation
R4.4.1 was used to generate the ROC curve of the prediction model. The AUC of the development set was 0.916 (95% CI:0.893–0.938) (Fig 3A). After 1000 bootstrap iterations for internal verification, the mean AUC of the model was 0.921 (95% CI: 0.893–0.916), indicating that the model has strong discriminant ability. The accuracy and specificity of the model in the developmental concentration were 0.842 and 0.881, respectively. The calibration plot (Fig 3C) and the H-L test (χ² = 9.979, P = 0.267) supported the prediction accuracy of the model. The DCA (Fig 3E) shows that this model provides a greater net benefit than a “treat all” or “treat none” strategy. When the optimal critical value of the model is 0.672 calculated based on the Youden index, the sensitivity is 0.869 with a specificity of 0.823. Fig 3. Performance evaluation of the PSD nutritional risk prediction model.
For external validation, 259 PSD patients observed between June and December 2023 were included. The model’s AUC was 0.878 (95% CI: 0.837–0.920) (Fig 3B). The H-L test result was χ² = 8.666, P = 0.371, and the accuracy and specificity were 0.819 and 0.891, respectively, confirming the model’s strong predictive power in the external validation set. Calibration curves (Fig 3D) and DCA (Fig 3F) further support the model’s discriminatory ability and clinical utility.
Discussion
In this study, 601 PSD patients were included in the development set, with a nutritional risk incidence of 63.06%, slightly higher than the 58.9% reported in the Yoon study [17]. This discrepancy may be attributed to differences in geographical factors and screening tools. The nutritional risk nomogram prediction model developed in this study is characterized by its simplicity, rapid operation, and strong predictive performance. The AUC values were 0.916 for the development set and 0.878 for the external validation set. The H-L test showed a P value greater than 0.05, indicating good model fit. Compared with the studies of Gao [18], Lin [19], Duan [20], Liu [21] and Zeng [22], the AUC value of this study was higher than that of previous studies (0.819, 0.800, 0.871, 0.860, 0.854), and the ability of the model to distinguish whether PSD patients had nutritional risk was significantly improved. The overall performance of a prediction model is affected by the choice of predictor variables, the size of the sample size, the handling of missing and outlier values, and the method used to construct the model [23]. The source of the predictor variables was not clearly indicated in previous studies, but this study was based on the results of literature review and expert consultation, and combined with the actual clinical situation, the predictors were finally included in the model construction, so that the scientific and clinical practicability of the model were guaranteed. In addition, the small sample size and low quality of the included data can lead to model overfitting and affect the accuracy of the prediction results [24]. Lin [19]included only 92 patients, while Liu [21] included the largest sample size of 325 in the existing study. In this study, the sample size was further expanded, 601 patients with PSD were included for model construction, and < 20% of the missing data were multi-imputed, which can increase the reliability and stability of the model to a certain extent. These results demonstrate that the model possesses strong discriminative ability and clinical utility in identifying nutritional risk, making it effective for preliminary screening of high-risk patients.
This study identified several factors influencing nutritional risk in PSD patients, including age, marital status, use of mechanical ventilation, dysphagia treatment, fasting duration, history of atrial fibrillation, oral care frequency, serum potassium levels and NIHSS score. The included predictive factors are all objective assessment indicators, which avoid the information bias introduced by traditional nutritional assessment tools that rely on patients’ recollections.
This study found that the risk of nutritional risk in PSD patients increases with age. Previous research has shown that older individuals tend to experience poorer recovery of various functions, including gastrointestinal decline, poor oral hygiene, reduced ability to eat independently, and increased susceptibility to gastrointestinal complications and hemiplegic limb dysfunction [17,25]. These factors contribute to a higher incidence of nutritional risk in the elderly. Additionally, this study revealed that PSD patients without spouses were approximately 3.62 times more likely to face nutritional risks than those with spouses, consistent with previous findings [26,27]. Spouses play a crucial role in providing emotional support, assisting with daily activities, and boosting recovery confidence, which can help reduce the incidence of nutritional risk.
The study found that PSD patients with a high NIHSS score are more prone to nutritional risk. Research indicates that individuals with severe neurological impairment often experience reduced daily functioning, difficulty swallowing, facial or limb weakness, and cognitive deficits, all of which significantly impair eating behaviors and food intake, leading to inadequate nutrition. Moreover, PSD patients are at a higher risk of coughing and aspiration during eating, which can result in complications such as lung infections, further increasing the incidence of nutritional risk. Numerous studies have demonstrated that dysphagia is closely associated with the nutritional status of patients [28–30]. In particular, PSD patients are more prone to aspiration, which can lead to complications like lung infections. dysphagia treatment, such as sensory neuromuscular electrical stimulation [31], cold stimulation [32], acid stimulation [33], and acupuncture [34], have been shown to help restore swallowing ability, improving nutritional status, reducing complications, shortening hospital stays, and ultimately enhancing patient prognosis. Additionally, PSD patients with a history of atrial fibrillation are at greater risk of nutritional imbalances. Atrial fibrillation increases the likelihood of severe stroke, and more severe strokes often result in greater nutritional deficits. This can lead to prolonged stays in the ultra-acute stroke unit and an increased risk of early infections, disability, and mortality [35].
Hemorrhagic stroke patients often require fasting in the early stages due to issues like intestinal edema, ulcers, and dysbiosis, with parenteral nutrition provided during this period. However, this can lead to complications such as hyperglycemia and infection [36]. Studies have also shown that patients receiving parenteral nutrition are more likely to require subsequent mechanical ventilation [37]. ventilation In this study, 63.71% of fasting patients underwent mechanical ventilation, which further increased the likelihood of nutritional risk [38–40]. Mechanical ventilation contributes to elevated intra-abdominal pressure, leading to mucosal damage, digestive tract bleeding, and digestive dysfunction, all of which impair nutrient absorption and increase nutritional risks. Additionally, the study found that more frequent daily oral care was associated with a higher nutritional risk, which aligns with previous research [41]. On the one hand, stroke patients with a small number of teeth, poor oral health, and weak daily activities often need frequent oral care, and such patients have more severe disease and are more prone to nutritional risks in hypermetabolic conditions [42]. On the other hand, Dysphagia can result in a loss of self-care ability and poor oral hygiene, which in turn affects appetite and food intake, leading to nutritional deficiencies and an increased risk of nutritional problems [43]. Lastly, lower serum potassium levels were found to correlate with a higher risk of nutritional issues. Stroke patients often experience potassium metabolism disorders, which place additional strain on the nervous system. Coupled with dysphagia and insufficient nutrition intake, this prolongs rehabilitation and increases the risk of stroke recurrence and nutritional complications [44].
Conclusions
The condition of PSD patients is complex and dynamic, with a high incidence of nutritional risk. This risk is closely associated with various factors, including age, marital status, mechanical ventilation, dysphagia treatment, fasting duration, atrial fibrillation, oral care frequency, serum potassium and NIHSS score. By constructing a prediction model based on the WST score, healthcare professionals can accurately screen for nutritional risk in PSD patients. This model is crucial for developing personalized nutrition interventions, improving patient nutritional status, enhancing quality of life, and ultimately improving clinical outcomes in PSD patients.
References
- 1. The Chinese Stroke Center Report. Brief Report on Stroke Center in China. Chin J Cerebrovasc Dis. 2024;21(8):565–76.
- 2. GBD 2019 Stroke Collaborators. Global, regional, and national burden of stroke and its risk factors, 1990-2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet Neurol. 2021;20(10):795–820. pmid:34487721
- 3. Martino R, Foley N, Bhogal S, Diamant N, Speechley M, Teasell R. Dysphagia after stroke: incidence, diagnosis, and pulmonary complications. Stroke. 2005;36(12):2756–63. pmid:16269630
- 4. Banda KJ, Chu H, Kang XL, Liu D, Pien L-C, Jen H-J, et al. Prevalence of dysphagia and risk of pneumonia and mortality in acute stroke patients: a meta-analysis. BMC Geriatr. 2022;22(1):420. pmid:35562660
- 5. Shimazu S, Yoshimura Y, Kudo M, Nagano F, Bise T, Shiraishi A, et al. Frequent and personalized nutritional support leads to improved nutritional status, activities of daily living, and dysphagia after stroke. Nutrition. 2021;83:111091. pmid:33388653
- 6. Huppertz VAL, Pilz W, Pilz Da Cunha G, de Groot LCPGM, van Helvoort A, Schols JMGA, et al. Malnutrition risk and oropharyngeal dysphagia in the chronic post-stroke phase. Front Neurol. 2022;13:939735. pmid:36247786
- 7. Arnold M, Liesirova K, Broeg-Morvay A, Meisterernst J, Schlager M, Mono M-L, et al. Dysphagia in Acute Stroke: Incidence, Burden and Impact on Clinical Outcome. PLoS One. 2016;11(2):e0148424. pmid:26863627
- 8. Sabbouh T, Torbey MT. Malnutrition in Stroke Patients: Risk Factors, Assessment, and Management. Neurocrit Care. 2018;29(3):374–84. pmid:28799021
- 9. Mueller C, Compher C, Ellen DM, American Society for Parenteral and Enteral Nutrition (A.S.P.E.N.) Board of Directors. A.S.P.E.N. clinical guidelines: Nutrition screening, assessment, and intervention in adults. JPEN J Parenter Enteral Nutr. 2011;35(1):16–24. pmid:21224430
- 10. Malone A, Mogensen KM. Key approaches to diagnosing malnutrition in adults. Nutr Clin Pract. 2022;37(1):23–34. pmid:34936131
- 11. Chinese Medical Association Branch of Neurology, Group of Cerebrovascular diseases B of N Chinese Medical Association. Diagnostic criteria of cerebrovascular diseases in China (version 2019). Chin Med Assoc J. 2019;52(9):710–5.
- 12. Hebert D, Lindsay MP, McIntyre A, Kirton A, Rumney PG, Bagg S, et al. Canadian stroke best practice recommendations: Stroke rehabilitation practice guidelines, update 2015. Int J Stroke. 2016;11(4):459–84. pmid:27079654
- 13. Kondrup J, Allison SP, Elia M, Vellas B, Plauth M, Educational and Clinical Practice Committee, European Society of Parenteral and Enteral Nutrition (ESPEN). ESPEN guidelines for nutrition screening 2002. Clin Nutr. 2003;22(4):415–21. pmid:12880610
- 14. Riley RD, Ensor J, Snell KIE, Harrell FE Jr, Martin GP, Reitsma JB, et al. Calculating the sample size required for developing a clinical prediction model. BMJ. 2020;368:m441. pmid:32188600
- 15. Zhong YZ, Hu C, Chen S, Tian YY, Ye ML, et al. The detection rate of nutritional risk in patients with stroke in China: a meta-analysis. Chin Evidence-Based Nurs. 2024;10(4):527–34.
- 16. Collins GS, Reitsma JB, Altman DG, Moons KGM. Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD): the TRIPOD statement. BMJ. 2015;350:g7594. pmid:25569120
- 17. Yoon J, Baek S, Jang Y, Lee CH, Lee ES, Byun H, et al. Malnutrition and Associated Factors in Acute and Subacute Stroke Patients with Dysphagia. Nutrients. 2023;15(17):3739. pmid:37686771
- 18. Gao Y-F, Kang C-B, Deng H-L, Gao Y. Identification of influencing factors and development of a nomogram for nutrition status after percutaneous endoscopic gastrostomy in stroke patients with swallowing disorders. WCJD. 2023;31(21):904–12.
- 19. Lin SY, Chen M, Zhou R, Huang Z, Hu WY. Predictive value of Kubota’s swallowing function score combined with fluorescent swallowing fluoroscopy score in malnutrition in patients with neurological dysphagia. China Modern Doctor. 2022;60(31):25–30.
- 20. Duan X, Hou JM, Peng YM, Zhou Y, Hong J, Liu HL. Analysis on related factors of hypoproteinemia occurrence in patients with dysphagia after stroke. Chin J Rehabil Med. 2022;51(21):3617–22.
- 21. Liu LL, Zhang FQ, Zeng X, Li Y, Wang LG, Zhang SJ. The prevalence of malnutrition among stroke survivors with bulbar palsy and the associated risk factors. Chinese Journal of Physical Medicine and Rehabilitation. 2024;45(12):006.
- 22. Zeng H, Cai A, Zhao W, Wu J, Ding Y, Zeng X. Factors and predictive model for malnutrition in poststroke disabled patients: A multicenter cross-sectional study. Nutrition. 2024;123:112423. pmid:38583267
- 23. Feng G, Xu H, Wan S, Wang H, Chen X, Magari R, et al. Twelve practical recommendations for developing and applying clinical predictive models. TIME. 2024;2(4):100105.
- 24. Strandberg R, Jepsen P, Hagström H. Developing and validating clinical prediction models in hepatology - An overview for clinicians. J Hepatol. 2024;81(1):149–62. pmid:38531493
- 25. Marin S, Serra-Prat M, Ortega O, Clavé P. Healthcare-related cost of oropharyngeal dysphagia and its complications pneumonia and malnutrition after stroke: a systematic review. BMJ Open. 2020;10(8):e031629. pmid:32784251
- 26. Cin A, Boyraz S, Öztürk V, Yaka E. Malnutrition in old patients with stroke. Türk Beyin Damar Hast Derg. 2019;25(3):155–63.
- 27.
Wang RB. Nutritional risk, post-stroke depression and poor prognosis of stroke patients and their correlation. Zhengzhou University. 2023.
- 28. Chen Y, Yang H, Lan M, Wei H, Chen Y. The controlling nutritional status score and risk factors associated with malnutrition in patients with acute ischemic stroke. Front Neurol. 2023;14:1067706. pmid:36970524
- 29. Fang W-J, Zheng F, Zhang L-Z, Wang W-H, Yu C-C, Shao J, et al. Research progress of clinical intervention and nursing for patients with post-stroke dysphagia. Neurol Sci. 2022;43(10):5875–84. pmid:35661279
- 30. Matsumoto A, Yoshimura Y, Nagano F, Bise T, Kido Y, Shimazu S, et al. Polypharmacy and Its Association with Dysphagia and Malnutrition among Stroke Patients with Sarcopenia. Nutrients. 2022;14(20):4251. pmid:36296943
- 31. Assoratgoon I, Shiraishi N, Tagaino R, Ogawa T, Sasaki K. Sensory neuromuscular electrical stimulation for dysphagia rehabilitation: A literature review. J Oral Rehabil. 2023;50(2):157–64. pmid:36357332
- 32. Cui F, Yin Q, Wu C, Shen M, Zhang Y, Ma C, et al. Capsaicin combined with ice stimulation improves swallowing function in patients with dysphagia after stroke: A randomised controlled trial. J Oral Rehabil. 2020;47(10):1297–303. pmid:32757479
- 33. Wang J, Chang E, Jiang Y. Effects of vitamin C stimulation on rehabilitation of dysphagia after stroke: a randomized trial. Eur J Phys Rehabil Med. 2022;58(4):558–64. pmid:35362717
- 34. Guo H, Pan X, Zheng Y, Yang X, Xu H, Zhang Y, et al. Current state of research on acupuncture for the treatment of post-stroke dysphagia: a scoping review. Front Neurosci. 2024;18:1391576. pmid:39211435
- 35. Han TS, Fry CH, Fluck D, Affley B, Gulli G, Barrett C, et al. Evaluation of anticoagulation status for atrial fibrillation on early ischaemic stroke outcomes: a registry-based, prospective cohort study of acute stroke care in Surrey, UK. BMJ Open. 2017;7(12):e019122. pmid:29247109
- 36. Xie WL, Wang SF, Li XG, Zhang Q. Establishment and validation of a risk prediction model for enteral nutrition-associated diarrhea in ICU patients. Chin J Nurs. 2022;57(19):2324–32.
- 37. Chen X-H, Zhao J-J, Chen C, Yao L. Establishment and validation of a predictive model for tracheotomy in critically ill patients and analysis of the impact of different tracheotomy timing on patient prognosis. BMC Anesthesiol. 2024;24(1):175. pmid:38760700
- 38. Silva PL, Ball L, Rocco PRM, Pelosi P. Physiological and Pathophysiological Consequences of Mechanical Ventilation. Semin Respir Crit Care Med. 2022;43(3):321–34. pmid:35439832
- 39. Liu GY, Zhang YY, Guo H, Yang DM. Establishment and validation of a risk prediction model for enteral nutrition-associated diarrhea in ICU patients. Chin J Nurs. 2022;37(2):97–101.
- 40. Shestopalov AE, Yakovleva AV, Yadgarov MY, Sergeev IV, Kuzovlev AN. Prevalence and Impact of Malnutrition Risk on Outcomes in Critically Ill Patients with Traumatic Brain Injury and Stroke: A Retrospective Cohort Study Using Electronic Health Records. Nutrients. 2024;16(15):2396. pmid:39125277
- 41. de Sire A, Ferrillo M, Lippi L, Agostini F, de Sire R, Ferrara PE, et al. Sarcopenic Dysphagia, Malnutrition, and Oral Frailty in Elderly: A Comprehensive Review. Nutrients. 2022;14(5):982. pmid:35267957
- 42. Todayama N, Hara R, Tabata T, Hatanaka Y, Mukai T, Someya M, et al. Systemic and Oral Characteristics of Convalescent Inpatients Requiring Oral-Health Management by a Dental Specialist during Hospitalization. Geriatrics (Basel). 2024;9(3):82. pmid:38920438
- 43. Yoshimura Y, Wakabayashi H, Nagano F, Matsumoto A, Shimazu S, Shiraishi A, et al. Coexistence of low body mass index and poor oral health negatively affects activities of daily living, swallowing, and cognition after stroke. Geriatr Gerontol Int. 2024;24(10):1045–52. pmid:39188241
- 44. Wang A, Cao S, Tian X, Zuo Y, Meng X, Chen P, et al. Lower Serum Potassium Levels at Admission are Associated with the Risk of Recurrent Stroke in Patients with Acute Ischemic Stroke or Transient Ischemic Attack. Cerebrovasc Dis. 2022;51(3):304–12. pmid:34856549