Personalized Screening Tool for Early Detection of Sarcopenia in Stroke Patients: A Machine Learning-Based Comparative Study

Personalized Screening Tool for Early Detection of Sarcopenia in Stroke Patients: A Machine Learning-Based Comparative Study | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Personalized Screening Tool for Early Detection of Sarcopenia in Stroke Patients: A Machine Learning-Based Comparative Study Huan Yan, Juan Li, Yujie Li, Lihong Xian, Huan Tang, Xuejiao Zhao, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5354644/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 20 Feb, 2025 Read the published version in Aging Clinical and Experimental Research → Version 1 posted 8 You are reading this latest preprint version Abstract Background Sarcopenia often occurs in stroke patients and contributes to worse recovery and a higher risk of death. There is no standardized tool for screening sarcopenia in stroke patients. The objective of this study is to explore the factors influencing sarcopenia in stroke patients, develop a risk prediction model, and evaluate its predictive accuracy. Methods Demographic and clinical characteristics of 794 stroke patients were collected. LASSO regression analysis was used for variable selection, and the selected variables were analyzed using multivariate regression. Logistic Regression (LR), Random Forest (RF), and XGBoost were used to construct prediction models, with the optimal model selected for external validation. Bootstrap resampling was used for internal validation of the training cohort, and another 159 stroke patients were collected for external validation. The performance of models was evaluated using the AUC, calibration curve, and Decision Curve Analysis (DCA). Results Based on LASSO and multivariate logistic regression analysis, seven variables were selected. The AUC value for the LR model was 0.805, surpassing that of the RF model (0.796) and the XGBoost model (0.780). The LR model also outperformed RF and XGBoost in terms of accuracy, precision, recall, specificity, and F1-score. In external validation, the LR model achieved an AUC of 0.816, and the calibration curve along with the DCA curve demonstrated that the model has nice accuracy and clinical applicability. Conclusions In this study, we developed a model and presented it as a nomogram to detect the risk of sarcopenia in stroke patients, and such early screening may benefit these patients. Stroke Sarcopenia Machine Learning Predictive model Figures Figure 1 Figure 2 Figure 3 Figure 4 1 Introduction Stroke is the leading cause of disability in adults, and about 80% of stroke patients will suffer from limb dysfunction( 1 ). Experiencing neurological damage and feeding difficulties after stroke will directly or indirectly lead to structural, metabolic, and functional abnormalities of muscle tissues, which will then cause muscle atrophy and structural changes, leading to the development of secondary sarcopenia, which is also known as stroke-related sarcopenia and is mainly manifested by loss of muscle mass and strength, and somatic dysfunction( 2 , 3 ). The prevalence of sarcopenia is about 15% in healthy older adults and can be as high as 56% in stroke patients( 4 ). The early symptoms of sarcopenia are not easy to detect and often go unnoticed, leading to a lack of attention. In the comorbid state of stroke, the condition becomes even more harmful to patients, resulting in higher risks such as infections, malnutrition, and prolonged hospital stays( 5 ). Prevention is the key to managing sarcopenia, and the basis for prevention comes from early screening. Early and standardized screening of patients with brain injuries can help accelerate their recovery process( 6 ). Currently, the screening for sarcopenia primarily relies on calf circumference and the Strength, Assistance with walking, Rise from a chair, Climb stairs, and Falls (SARC-F) scale scores( 7 ). Although this scale is widely used among the elderly, the complex conditions brought on by neurologic damage in stroke patients may result in a lack of relevance and accuracy and limited screening effectiveness( 8 ). In addition, the skeletal muscle index assessed by computed tomography (CT) serves as a key diagnostic indicator for sarcopenia and plays an important role in its prevention and management. However, its widespread use faces challenges due to high costs, exposure to ionizing radiation, and the complexity of the procedure( 9 , 10 ). Overall, the assessment of sarcopenia in clinical practice remains a challenging task. At present, predictions of sarcopenia are mainly focused on elderly populations in the community, while research on stroke patients is limited. Traditional screening tools are often based on a single standard and do not take into account the individual characteristics of each patient. Therefore, using machine learning in combination with easily accessible data in the clinic can effectively compensate for the shortcomings of traditional methods, enabling personalized predictions. This approach can help identify high-risk sarcopenia patients earlier and with greater precision. This study aimed to construct three predictive models using machine learning and present the optimal model that has been selected and validated visually, making it easier and more personalized to identify the high risk of sarcopenia in stroke patients. 2 Materials and Methods 2.1 Study design and population This study was conducted in the departments of rehabilitation, neurologyand and neurosurgery of a hospital. The study was approved by the Ethics Committee of the hospital with the approval number “Ethical Approval Word (Research) [2023] No. 060”. In this study, the training cohort included 794 stroke patients who were retrospectively enrolled between January 2021 and December 2023. A prospective approach was used for external validation, with 195 patients enrolled at the hospital between January and May 2024. The inclusion criteria for this study were as follows: aged 18 years or older; stroke clearly diagnosed by CT or MRI; and no significant intellectual or cognitive dysfunction. The exclusion criteria were: patients with psychiatric diseases; patients who had suffered from sarcopenia before stroke (recalled pre-stroke SARC-F score ≥ 4); patients who suffered from other neurological disorders; and those with incomplete clinical data. The diagnostic criteria for this study are based on the standards established by the Asian Working Group for Sarcopenia (AWGS) in 2019( 11 ). Muscle mass was represented by the skeletal muscle index, with a threshold value of < 7.0 kg/m 2 for men and < 5.7 kg/m 2 for women; muscle strength was assessed using grip strength, and muscle strength was considered to be decreased when it was < 28 kg for men and < 18 kg for women; and somatic function was assessed using a 5-sit-up test (5STS) ≥ 12 s as the cut-off value reflecting decreased somatic function. Sarcopenia was diagnosed when decreased muscle mass was combined with decreased muscle strength and/or 5STS ≥ 12s. 2.2 Data collection Data were gathered from the electronic medical record system, including both medical and nursing records that included: age, gender, history of smoking, history of drinking, osteoporosis, number of strokes, limb dysfunction, stroke location, diabetes, stroke type, times since stroke, tube feeding, hospitalization days, Barthel Index (BI) score, Nutritional Risk Screening 2002 (NRS2002) score, BMI, National Institutes of Health Stroke Scale (NIHSS) score, C-reactive protein, serum total cholesterol and serum albumin, for a total of 20 items. During data collection, two researchers used a standardized survey and diagnostic criteria to gather information. After collection, both researchers cross-checked the data. If the results were inconsistent, then revisit to the medical records for verification until accurate. 2.3Statistical analysis Statistical analysis was performed using SPSS (Version 25.0, IBM Corp., USA) and R (Version 4.2.1, R Foundation for Statistical Computing, Austria) software. Categorical data were presented as n (%) and analyzed using the χ 2 test. Categorical data were expressed as n (%) and analyzed using the χ 2 test. Continuous data were first tested for normality. Data following a normal distribution were presented as mean ± standard deviation and analyzed using the independent samples t-test, while non-normally distributed data were presented as median and interquartile range (IQR) and analyzed using the Mann-Whitney U test. The glmnet package was utilized to perform LASSO regression analysis to select predictive factors, which were then subjected to multivariate logistic regression analysis.Prediction models were constructed using logistic regression, random forest, and XGBoost. The models' predictive performance was evaluated using AUC, accuracy, specificity, sensitivity, and F1 score. Model calibration and accuracy were assessed using the Hosmer-Lemeshow (HL) test and calibration curves. The clinical utility of the model was evaluated using DCA. A P -value of < 0.05 was regarded as statistically significant. 3 Results 3.1 Demographic and Clinical Characteristics of Patients Table 1 presents the demographic and clinical characteristics of the patients. The results of this study revealed that the prevalence of sarcopenia was 37.53% (298/794) in the training cohort and 38.36% (61/159) in the validation cohort. Except for age and diabetes, the other factors showed no significant differences in the baseline characteristics between the two cohorts ( P > 0.05). Table 1 Demographic and clinical characteristics of patients Variable Training cohort ( n = 794) Validation cohort ( n = 159) P value Sarcopenia (n = 298 ) Non-Sarcopenia (n = 496) Sarcopenia (n = 61 ) Non-Sarcopenia (n = 98 ) Age(year) 69.76 ± 12.25 65.69 ± 11.21 68.59 ± 13.75 62.78 ± 10.80 0.032 Gender, n (%) 0.828 Male 163(54.70) 224(45.16) 49(50.00) 30(49.18) Female 135(45.30) 272(54.84) 49(50.00) 31(50.82) History of smoking, n (%) 0.168 No 151(50.67) 261(52.62) 56(57.14) 36(59.02) Yes 147(49.33) 235(47.38) 42(42.86) 25(40.98) History of drinking, n (%) 0.633 No 144(48.32) 259(52.22) 54(55.10) 30(49.18) Yes 154(51.68) 237(47.78) 44(44.90) 31(50.82) Osteoporosis, n (%) 0.654 No 140(46.98) 250(50.40) 50(51.02) 25(40.98) Yes 158(53.02) 246(49.60) 48(48.98) 36(59.02) Number of strokes, n (%) 0.481 1 time 85(28.52) 134(27.02) 27(27.55) 13(21.31) 2 times 77(25.84) 124(25.00) 28(28.57) 9(14.75) 3 times 56(18.79) 123(24.80) 22(22.45) 23(37.70) More than 3 times 80(26.85) 115(23.19) 21(21.43) 16(26.23) Limb dysfunction, n (%) 0.919 No 103(34.56) 300(60.48) 61(62.24) 19(31.15) Yes 195(65.44) 196(39.52) 37(37.76) 42(68.85) Stroke location, n (%) 0.434 Left side 92(30.87) 169(34.07) 31(31.63) 25(40.98) Right side 109(36.58) 173(34.88) 34(34.69) 14(22.95) Both sides 97(32.55) 154(31.05) 33(33.67) 22(36.07) Diabetes, n (%) 0.029 No 118(39.60) 256(51.61) 70(71.43) 20(32.79) Yes 180(60.40) 240(48.39) 28(28.57) 41(67.21) Stroke type, n (%) 0.653 Ischemic 139(46.64) 240(48.39) 51(52.04) 28(45.90) Hemorrhagic 159(53.36) 256(51.61) 47(47.96) 33(54.10) Times since stroke, n (%) 0.481 < 1 month 51(17.11) 90(18.15) 19(19.39) 12(19.67) 1–3 months 62(20.81) 100(20.16) 14(14.29) 9(14.75) 3–6 months 69(23.15) 107(21.57) 20(20.41) 14(22.95) 6–12 months 62(20.81) 92(18.55) 21(21.43) 14(22.95) > 12 months 54(18.12) 107(21.57) 24(24.49) 12(19.67) Tube feeding, n (%) 0.15 No 188(63.09) 419(84.48) 77(78.57) 36(59.02) Yes 110(36.91) 77(15.52) 21(21.43) 25(40.98) Hospitalization days 5.69 ± 1.83 5.58 ± 1.78 5.70 ± 1.73 5.15 ± 1.82 0.404 Barthel score 48.06 ± 29.74 51.60 ± 28.30 51.40 ± 24.83 55.18 ± 24.58 0.245 NRS2002 score 4.43 ± 2.82 4.38 ± 2.92 4.80 ± 2.94 3.72 ± 2.74 0.914 BMI(kg/m²) 19.94 ± 2.05 21.38 ± 3.06 20.19 ± 2.34 21.42 ± 3.16 0.644 NIHSS score 7.24 ± 5.39 5.84 ± 3.81 7.93 ± 4.71 5.72 ± 3.39 0.588 C-reactive protein(mg/L) 8.21 ± 4.43 6.16 ± 3.89 8.41 ± 4.54 6.64 ± 4.25 0.289 Serum total cholesterol(mg/dL) 197.38 ± 60.60 204.35 ± 58.88 192.45 ± 59.70 205.25 ± 61.07 0.789 Serum albumin(g/dL) 4.49 ± 0.60 4.53 ± 0.57 4.52 ± 0.81 4.66 ± 0.96 0.243 3.2 Results of variables selection The results of the LASSO regression analysis are shown in Fig. 1 . When λ .1 se = 0.033, seven variables were selected from 20 variables: age, limb dysfunction, diabetes, tube feeding, BMI, NIHSS, and C-reactive protein. Multivariate logistic regression analysis of the seven variables (Table 2 ) revealed that all were statistically significant ( P < 0.05). These factors were used to construct the prediction model. Table 2 Multivariate logistic regression analysis of the seven variables Variable β SE z Wald χ2 P OR 95% CI Age 0.032 0.007 4.318 18.648 < 0.001 1.032 1.017 ~ 1.047 Limb dysfunction 1.116 0.173 6.444 41.529 < 0.001 3.053 2.174 ~ 4.287 Diabetes 0.522 0.172 3.044 9.263 0.002 1.685 1.204 ~ 2.359 Tube feeding 1.063 0.198 5.379 28.937 < 0.001 2.894 1.965 ~ 4.263 BMI -0.222 0.036 -6.115 37.393 < 0.001 0.801 0.746 ~ 0.860 NIHSS score 0.060 0.020 3.048 9.289 0.002 1.062 1.022 ~ 1.103 C-reactive protein 0.139 0.021 6.562 43.056 < 0.001 1.149 1.102 ~ 1.198 3.3 Model Construction and Evaluation Results Table 3 shows that the LR model performed best across all five metrics. Figure 2 displays AUC values: LR at 0.805, RF at 0.796, and XGBoost at 0.780. While all models performed well but LR was superior, so it was chosen for external validation. In external validation, the LR model achieved an AUC of 0.816 (Fig. 3 A). The HL test resulted in a p-value of 0.128 ( P > 0.05), indicating good model fit. The calibration curve in Fig. 3 B shows that the bias-corrected line closely aligns with the ideal line, with a mean squared error of 0.01, demonstrating good calibration. Figure 3 C displays the DCA curve, where the net benefit of the model surpasses both the None and All lines, indicating strong clinical utility. Based on the model’s excellent predictive performance, a nomogram was constructed (Fig. 4 ) to visually present the stroke-related sarcopenia risk scoring system. Table 3 Comparison of the performance of the three models Accuracy Precision Recall Specificity F1-score Logistic Regression 0.747 0.764 0.767 0.912 0.762 Random Forest 0.738 0.667 0.607 0.818 0.635 XGBoost 0.728 0.632 0.621 0.790 0.626 4 Discussion Secondary sarcopenia in stroke patients has a high prevalence, averaging 42%( 4 ), which is notably higher than in other chronic diseases. Previous studies have reported that the prevalence of sarcopenia ranges from 7–29.3% in diabetes patients, 15.5% in COPD patients, and 32.5% in cancer patients( 12 – 14 ), all of which are lower than in stroke populations. This study found a sarcopenia prevalence of 37.53% and 38.36% in stroke patients, further confirming its high occurrence. The high prevalence of sarcopenia in stroke patients highlights the urgency of early detection and prevention. Other studies suggest that factors such as hyperlipidemia and atrial fibrillation may increase the risk of sarcopenia in stroke survivors( 15 ). This study identified age, limb dysfunction, diabetes, tube feeding, BMI, NIHSS score, and C-reactive protein as independent risk factors for SRS. Consistent with the definition of sarcopenia and early research findings, the prevalence of sarcopenia increases with age. With age increasing, muscle mass, quality, and strength gradually decline, this is also true for stroke patients. Eighty-five percent of stroke patients experience upper limb dysfunction from the onset, yet only one-third regain some limb function, and even then, the recovery is often incomplete( 16 ). Nerve damage weakens the central nervous system’s regulation of muscles. When limbs lack necessary muscle contractions and activity, muscle fibers gradually degrade, leading to a reduction in muscle mass and strength. However, the relationship between the severity of neurological damage (assessed in this study using the NIHSS scale) and sarcopenia in stroke patients remains unclear. Some studies suggest that sarcopenia may induce endothelial dysfunction( 17 ), which is associated with neurological deterioration( 18 , 19 ), the underlying mechanisms are still not fully understood. In this study, the sarcopenia group had an average NIHSS score of 7.24 ± 5.39 in the training cohort and 7.93 ± 4.71 in the validation cohort, both higher than the non-sarcopenia group. Therefore, we hypothesize that a higher NIHSS score correlates with an increased risk of sarcopenia. Previous studies have confirmed the complex and close relationship between diabetes and sarcopenia, particularly in patients with type 2 diabetes. This connection primarily occurs through mechanisms such as chronic inflammation, insulin resistance, oxidative stress, and neuropathy, all of which contribute to the decline in muscle mass and function( 20 , 21 ). Dysphagia is one of the most common complications in stroke patients, with over 50% experiencing this condition( 22 ). When oral intake becomes difficult or unsafe, tube feeding is an important method for providing enteral nutrition. While this approach supplies the necessary calories for basic metabolism, many patients may still experience inadequate nutritional intake, potentially delaying the recovery of swallowing function( 23 ). Prolonged disuse of oral and pharyngeal muscles can lead to degeneration and atrophy, affecting not only swallowing muscles but also increasing the risk of sarcopenia throughout the body. After a stroke, patients are often in a hypermetabolic state, requiring more energy for recovery. If nutrition from a nasogastric tube doesn’t meet these needs, weight and BMI may decrease( 24 ). This study shows that stroke patients with lower BMI have a higher risk of developing sarcopenia. CRP is a non-specific inflammation marker used to assess systemic inflammation. Previous evidence has shown that sarcopenia patients have higher CRP levels than non-sarcopenic individuals( 25 ). This chronic inflammation accelerates muscle breakdown, limits recovery, and negatively affects neurological and metabolic functions. This study used LR, RF, and XGBoost to build and validate a sarcopenia risk prediction model for stroke patients, incorporating seven variables: age, limb dysfunction, diabetes, tube feeding, BMI, NIHSS score, and CRP. The LR model performed best, with the highest AUC, specificity, and sensitivity, indicating better accuracy in predicting sarcopenia risk. A nomogram can visually represent the risk of sarcopenia and aid healthcare professionals in dynamically assessing it. It helps reduce the incidence of sarcopenia and improves patients' quality of life. This study has some limitations. First, there is no unified diagnostic standard for sarcopenia, with different cutoff values used in Asia and Europe, so caution is needed when applying these results to different populations. Second, as a single-center study, while the model showed good clinical utility, further validation in multi-center, large-sample studies is required. Third, the retrospective study may have missed certain relevant variables, such as cognitive impairment and psychological status, which could be considered in future research. 5 Conclusion We developed and compared three machine learning models, ultimately establishing a nomogram including seven predictive factors: age, limb dysfunction, diabetes, tube feeding, BMI, NIHSS score, and CRP. This may be a practical tool for early screening of sarcopenia in stroke patients. The nomogram is simple to use and can help to identify patients at high risk for sarcopenia to aid in early intervention. Declarations Acknowledgements The authors would like to express gratitude to the study participants and clinical staff for their support and contributions to this study. Authorship contributions Huan Yan: Data curation, Investigation, Methodology, Software, Writing – original draft, Writing – review & editing. Juan Li: Conceptualization, Formal analysis, Funding acquisition, Supervision, Visualization, Writing – review & editing. Yujie Li: Data curation, Investigation, Methodology, Software, Writing – original draft. Lihong Xian and Huan Tang: Methodology, Software, Validation, Visualization, Writing – original draft. Xuejiao Zhao and Ting Lu: Methodology, Software, Validation, Visualization, Writing – original draft. Funding This work was supported by the National Natural Science Foundation of China(grant numbers 72364005). Data availability The data that support the findings of this study are available from the corresponding author, upon reasonable request. Ethical approval The study was approved by the Ethics Committee of the Guizhou Provincial People’s Hospital with the approval number “Ethical Approval Word (Research) [2023] No. 060”. All participants signed informed consent. Conflict of interest The authors declare no competing interests. References Feigin VL, Stark BA, Johnson CO, Roth GA, Bisignano C, Abady GG, et al. Global, regional, and national burden of stroke and its risk factors, 1990–2019: a systematic analysis for the Global Burden of Disease Study 2019. The Lancet Neurology. (2021) 20(10):795–820. DOI: 10.1016/S1474-4422(21)00252-0 Aydin T, Kesiktaş FN, Oren MM, Erdogan T, Ahisha YC, Kizilkurt T, et al. Sarcopenia in patients following stroke: an overlooked problem. International Journal of Rehabilitation Research. (2021) 44(3):269–75. DOI: 10.1097/MRR.0000000000000487 Yao R, Yao L, Rao A, Ou J, Wang W, Hou Q, et al. Prevalence and risk factors of stroke-related sarcopenia at the subacute stage: A case control study. Front Neurol. (2022) 13:899658. DOI: 10.3389/fneur.2022.899658 Su Y, Yuki M, Otsuki M. Prevalence of stroke-related sarcopenia: A systematic review and meta-analysis. Journal of Stroke and Cerebrovascular Diseases. (2020) 29(9):105092. DOI: 10.1016/j.jstrokecerebrovasdis.2020.105092 Mas MF, González J, Frontera WR. Stroke and Sarcopenia. Curr Phys Med Rehabil Rep. (2020) 8(4):452–60. DOI: 10.1007/s40141-020-00284-2 Yang G, Xie W, Li B, Zhao G, Li J, Xiao W, et al. Casual associations between brain structure and sarcopenia: A large‐scale genetic correlation and mendelian randomization study. Aging Cell. (2024) :e14252. DOI: 10.1111/acel.14252 Dent E, Morley JE, Cruz-Jentoft AJ, Arai H, Kritchevsky SB, Guralnik J, et al. International Clinical Practice Guidelines for Sarcopenia (ICFSR): Screening, Diagnosis and Management. The Journal of nutrition, health and aging. (2018) 22(10):1148–61. DOI: 10.1007/s12603-018-1139-9 Cai G, Ying J, Pan M, Lang X, Yu W, Zhang Q. Development of a risk prediction nomogram for sarcopenia in hemodialysis patients. BMC Nephrol. (2022) 23(1):319. DOI: 10.1186/s12882-022-02942-0 Vogele D, Otto S, Sollmann N, Haggenmüller B, Wolf D, Beer M, et al. Sarcopenia – Definition, Radiological Diagnosis, Clinical Significance. Rofo. (2023) 195(05):393–405. DOI: 10.1055/a-1990-0201 Chianca V, Albano D, Messina C, Gitto S, Ruffo G, Guarino S, et al. Sarcopenia: imaging assessment and clinical application. Abdom Radiol. (2021) 47(9):3205–16. DOI: 10.1007/s00261-021-03294-3 Chen L-K, Woo J, Assantachai P, Auyeung T-W, Chou M-Y, Iijima K, et al. Asian Working Group for Sarcopenia: 2019 Consensus Update on Sarcopenia Diagnosis and Treatment. Journal of the American Medical Directors Association. (2020) 21(3):300-307.e2. DOI: 10.1016/j.jamda.2019.12.012 Izzo A, Massimino E, Riccardi G, Della Pepa G. A Narrative Review on Sarcopenia in Type 2 Diabetes Mellitus: Prevalence and Associated Factors. Nutrients. (2021) 13(1):183. DOI: 10.3390/nu13010183 Sepúlveda‐Loyola W, Osadnik C, Phu S, Morita AA, Duque G, Probst VS. Diagnosis, prevalence, and clinical impact of sarcopenia in COPD: a systematic review and meta‐analysis. J cachexia sarcopenia muscle. (2020) 11(5):1164–76. DOI: 10.1002/jcsm.12600 Jang MK, Park S, Raszewski R, Park CG, Doorenbos AZ, Kim S. Prevalence and clinical implications of sarcopenia in breast cancer: a systematic review and meta-analysis. Support Care Cancer. (2024) 32(5):328. DOI: 10.1007/s00520-024-08532-0 Nozoe M, Kanai M, Kubo H, Yamamoto M, Shimada S, Mase K. Prestroke Sarcopenia and Stroke Severity in Elderly Patients with Acute Stroke. Journal of Stroke and Cerebrovascular Diseases. (2019) 28(8):2228–31. DOI: 10.1016/j.jstrokecerebrovasdis.2019.05.001 Tang C, Zhou T, Zhang Y, Yuan R, Zhao X, Yin R, et al. Bilateral upper limb robot-assisted rehabilitation improves upper limb motor function in stroke patients: a study based on quantitative EEG. Eur J Med Res. (2023) 28(1):603. DOI: 10.1186/s40001-023-01565-x He N, Zhang Y, Zhang L, Zhang S, Ye H. Relationship Between Sarcopenia and Cardiovascular Diseases in the Elderly: An Overview. Front Cardiovasc Med. (2021) 8:743710. DOI: 10.3389/fcvm.2021.743710 Martin AJ, Price CI. A Systematic Review and Meta-Analysis of Molecular Biomarkers Associated with Early Neurological Deterioration Following Acute Stroke. Cerebrovasc Dis. (2018) 46(5–6):230–41. DOI: 10.1159/000495572 Geng H-H, Wang Q, Li B, Cui B-B, Jin Y-P, Fu R-L, et al. Early neurological deterioration during the acute phase as a predictor of long-term outcome after first-ever ischemic stroke. Medicine. (2017) 96(51):e9068. DOI: 10.1097/MD.0000000000009068 Chen H, Huang X, Dong M, Wen S, Zhou L, Yuan X. The Association Between Sarcopenia and Diabetes: From Pathophysiology Mechanism to Therapeutic Strategy. DMSO. (2023) Volume 16:1541–54. DOI: 10.2147/DMSO.S410834 Sanz-Cánovas J, López-Sampalo A, Cobos-Palacios L, Ricci M, Hernández-Negrín H, Mancebo-Sevilla JJ, et al. Management of Type 2 Diabetes Mellitus in Elderly Patients with Frailty and/or Sarcopenia. IJERPH. (2022) 19(14):8677. DOI: 10.3390/ijerph19148677 Dziewas R, Michou E, Trapl-Grundschober M, Lal A, Arsava EM, Bath PM, et al. European Stroke Organisation and European Society for Swallowing Disorders guideline for the diagnosis and treatment of post-stroke dysphagia. European Stroke Journal. (2021) 6(3):LXXXIX–CXV. DOI: 10.1177/23969873211039721 Braun RG, Arata J, Gonzalez-Fernandez M. Dysphagia and Enteral Feeding After Stroke in the Rehabilitation Setting. Physical Medicine and Rehabilitation Clinics of North America. (2024) 35(2):433–43. DOI: 10.1016/j.pmr.2023.07.001 Zielińska-Nowak E, Cichon N, Saluk-Bijak J, Bijak M, Miller E. Nutritional Supplements and Neuroprotective Diets and Their Potential Clinical Significance in Post-Stroke Rehabilitation. Nutrients. (2021) 13(8):2704. DOI: 10.3390/nu13082704 Bano G, Trevisan C, Carraro S, Solmi M, Luchini C, Stubbs B, et al. Inflammation and sarcopenia: A systematic review and meta-analysis. Maturitas. (2017) 96:10–5. DOI: 10.1016/j.maturitas.2016.11.006 Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 20 Feb, 2025 Read the published version in Aging Clinical and Experimental Research → Version 1 posted Editorial decision: Revision requested 23 Dec, 2024 Reviews received at journal 10 Dec, 2024 Reviewers agreed at journal 09 Dec, 2024 Reviewers agreed at journal 11 Nov, 2024 Reviewers invited by journal 07 Nov, 2024 Editor assigned by journal 05 Nov, 2024 Submission checks completed at journal 30 Oct, 2024 First submitted to journal 29 Oct, 2024 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-5354644","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":375333150,"identity":"01d3ba09-e364-42f9-930b-dbc50f6b4ca0","order_by":0,"name":"Huan Yan","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA0klEQVRIiWNgGAWjYBACNvb24x8+VEjIMbY3HyBOCx/PmTTGGWdsjJl7jiUQp0VOIsGMmbMlLbF9Ro4BkQ6TSEh7zNhw2Ji3IefjjTcMdnK6DYS08Dw8bly447CcZMPZzZZzGJKNzQ4Q0sKekCA988xhY8PG3m3SPAwHErcR1MKQYCDN23Y4cf9hnmdEauFIMANqSUtsbONhI1ILz5lkQ1AgM/awGVvOMSDCL/Lt7QcfgKNy/uOHN95U2MkR1IICJHiIjBpkLaTqGAWjYBSMghEBALsPRcBrfVD3AAAAAElFTkSuQmCC","orcid":"","institution":"Zunyi Medical University","correspondingAuthor":true,"prefix":"","firstName":"Huan","middleName":"","lastName":"Yan","suffix":""},{"id":375333151,"identity":"056f106b-0ae7-4a79-a19e-3b1ccfdcb9a3","order_by":1,"name":"Juan Li","email":"","orcid":"","institution":"Guizhou Provincial People's Hospital","correspondingAuthor":false,"prefix":"","firstName":"Juan","middleName":"","lastName":"Li","suffix":""},{"id":375333152,"identity":"f6bde8b1-019d-4a5b-8faf-f37137cee0d0","order_by":2,"name":"Yujie Li","email":"","orcid":"","institution":"Guizhou University of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Yujie","middleName":"","lastName":"Li","suffix":""},{"id":375333153,"identity":"b3680ca8-8ee5-4c5b-8c89-1a3b2cd094b6","order_by":3,"name":"Lihong Xian","email":"","orcid":"","institution":"Zunyi Medical University","correspondingAuthor":false,"prefix":"","firstName":"Lihong","middleName":"","lastName":"Xian","suffix":""},{"id":375333154,"identity":"461a4a7f-25fa-4a33-8867-69351799cdca","order_by":4,"name":"Huan Tang","email":"","orcid":"","institution":"Guizhou Provincial People's Hospital","correspondingAuthor":false,"prefix":"","firstName":"Huan","middleName":"","lastName":"Tang","suffix":""},{"id":375333155,"identity":"b2f61552-d3be-44cb-bce4-9ee3ded1b520","order_by":5,"name":"Xuejiao Zhao","email":"","orcid":"","institution":"Guizhou University of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Xuejiao","middleName":"","lastName":"Zhao","suffix":""},{"id":375333156,"identity":"f3f30cd7-9f2e-4a0f-a09c-e0f49641c820","order_by":6,"name":"Ting Lu","email":"","orcid":"","institution":"Guizhou University of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Ting","middleName":"","lastName":"Lu","suffix":""}],"badges":[],"createdAt":"2024-10-29 13:23:24","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5354644/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5354644/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s40520-025-02945-5","type":"published","date":"2025-02-20T15:56:57+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":69442717,"identity":"df1d28b4-a2f5-4d59-88b6-980bd4bb5ed9","added_by":"auto","created_at":"2024-11-20 11:28:16","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":701000,"visible":true,"origin":"","legend":"\u003cp\u003eVariable selection using LASSO regression analysis. \u003cstrong\u003eA\u003c/strong\u003e Cross-validation curve for selecting independent variables. \u003cstrong\u003eB\u003c/strong\u003e The variation characteristics of the coefficient of variables.\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-5354644/v1/d4979a7a339fc3bbddbe6549.png"},{"id":69442723,"identity":"431fd8d9-ca74-4423-bc75-b6e1bb41c380","added_by":"auto","created_at":"2024-11-20 11:28:25","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":115420,"visible":true,"origin":"","legend":"\u003cp\u003eThe AUC of the predictive model\u003c/p\u003e","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-5354644/v1/169cacc9b5122ccd8e70844b.png"},{"id":69442719,"identity":"9a1ee23c-c7f0-46fb-9106-b29cda575d63","added_by":"auto","created_at":"2024-11-20 11:28:18","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":386383,"visible":true,"origin":"","legend":"\u003cp\u003eExternal validation results of the LR model.\u003cstrong\u003e A \u003c/strong\u003eROC curve and AUC value of the model.\u003cstrong\u003e B \u003c/strong\u003eCalibration curve.\u003cstrong\u003e C \u003c/strong\u003eDecision Curve Analysis (DCA).\u003c/p\u003e","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-5354644/v1/84103b00b6daefe115e2f17e.png"},{"id":69442721,"identity":"9cf80297-72f4-4288-bb09-c20ff41c442a","added_by":"auto","created_at":"2024-11-20 11:28:22","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":161287,"visible":true,"origin":"","legend":"\u003cp\u003eNomogram for predicting the risk of sarcopenia in patients with stroke\u003c/p\u003e","description":"","filename":"Figure4.png","url":"https://assets-eu.researchsquare.com/files/rs-5354644/v1/6190d5a62fb620a843edcd9a.png"},{"id":77052467,"identity":"088dccd3-751c-401f-b776-271dfef22988","added_by":"auto","created_at":"2025-02-24 16:06:20","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2078939,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5354644/v1/7a8b89a2-8955-4dd5-9eb2-a53657b35b50.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Personalized Screening Tool for Early Detection of Sarcopenia in Stroke Patients: A Machine Learning-Based Comparative Study","fulltext":[{"header":"1 Introduction","content":"\u003cp\u003eStroke is the leading cause of disability in adults, and about 80% of stroke patients will suffer from limb dysfunction(\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e). Experiencing neurological damage and feeding difficulties after stroke will directly or indirectly lead to structural, metabolic, and functional abnormalities of muscle tissues, which will then cause muscle atrophy and structural changes, leading to the development of secondary sarcopenia, which is also known as stroke-related sarcopenia and is mainly manifested by loss of muscle mass and strength, and somatic dysfunction(\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e). The prevalence of sarcopenia is about 15% in healthy older adults and can be as high as 56% in stroke patients(\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e). The early symptoms of sarcopenia are not easy to detect and often go unnoticed, leading to a lack of attention. In the comorbid state of stroke, the condition becomes even more harmful to patients, resulting in higher risks such as infections, malnutrition, and prolonged hospital stays(\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e).\u003c/p\u003e \u003cp\u003ePrevention is the key to managing sarcopenia, and the basis for prevention comes from early screening. Early and standardized screening of patients with brain injuries can help accelerate their recovery process(\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e). Currently, the screening for sarcopenia primarily relies on calf circumference and the Strength, Assistance with walking, Rise from a chair, Climb stairs, and Falls (SARC-F) scale scores(\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e). Although this scale is widely used among the elderly, the complex conditions brought on by neurologic damage in stroke patients may result in a lack of relevance and accuracy and limited screening effectiveness(\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e). In addition, the skeletal muscle index assessed by computed tomography (CT) serves as a key diagnostic indicator for sarcopenia and plays an important role in its prevention and management. However, its widespread use faces challenges due to high costs, exposure to ionizing radiation, and the complexity of the procedure(\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e). Overall, the assessment of sarcopenia in clinical practice remains a challenging task.\u003c/p\u003e \u003cp\u003eAt present, predictions of sarcopenia are mainly focused on elderly populations in the community, while research on stroke patients is limited. Traditional screening tools are often based on a single standard and do not take into account the individual characteristics of each patient. Therefore, using machine learning in combination with easily accessible data in the clinic can effectively compensate for the shortcomings of traditional methods, enabling personalized predictions. This approach can help identify high-risk sarcopenia patients earlier and with greater precision. This study aimed to construct three predictive models using machine learning and present the optimal model that has been selected and validated visually, making it easier and more personalized to identify the high risk of sarcopenia in stroke patients.\u003c/p\u003e"},{"header":"2 Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Study design and population\u003c/h2\u003e \u003cp\u003eThis study was conducted in the departments of rehabilitation, neurologyand and neurosurgery of a hospital. The study was approved by the Ethics Committee of the hospital with the approval number \u0026ldquo;Ethical Approval Word (Research) [2023] No. 060\u0026rdquo;. In this study, the training cohort included 794 stroke patients who were retrospectively enrolled between January 2021 and December 2023. A prospective approach was used for external validation, with 195 patients enrolled at the hospital between January and May 2024. The inclusion criteria for this study were as follows: aged 18 years or older; stroke clearly diagnosed by CT or MRI; and no significant intellectual or cognitive dysfunction. The exclusion criteria were: patients with psychiatric diseases; patients who had suffered from sarcopenia before stroke (recalled pre-stroke SARC-F score\u0026thinsp;\u0026ge;\u0026thinsp;4); patients who suffered from other neurological disorders; and those with incomplete clinical data. The diagnostic criteria for this study are based on the standards established by the Asian Working Group for Sarcopenia (AWGS) in 2019(\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e). Muscle mass was represented by the skeletal muscle index, with a threshold value of \u0026lt;\u0026thinsp;7.0 kg/m\u003csup\u003e2\u003c/sup\u003e for men and \u0026lt;\u0026thinsp;5.7 kg/m\u003csup\u003e2\u003c/sup\u003e for women; muscle strength was assessed using grip strength, and muscle strength was considered to be decreased when it was \u0026lt;\u0026thinsp;28 kg for men and \u0026lt;\u0026thinsp;18 kg for women; and somatic function was assessed using a 5-sit-up test (5STS)\u0026thinsp;\u0026ge;\u0026thinsp;12 s as the cut-off value reflecting decreased somatic function. Sarcopenia was diagnosed when decreased muscle mass was combined with decreased muscle strength and/or 5STS\u0026thinsp;\u0026ge;\u0026thinsp;12s.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Data collection\u003c/h2\u003e \u003cp\u003eData were gathered from the electronic medical record system, including both medical and nursing records that included: age, gender, history of smoking, history of drinking, osteoporosis, number of strokes, limb dysfunction, stroke location, diabetes, stroke type, times since stroke, tube feeding, hospitalization days, Barthel Index (BI) score, Nutritional Risk Screening 2002 (NRS2002) score, BMI, National Institutes of Health Stroke Scale (NIHSS) score, C-reactive protein, serum total cholesterol and serum albumin, for a total of 20 items. During data collection, two researchers used a standardized survey and diagnostic criteria to gather information. After collection, both researchers cross-checked the data. If the results were inconsistent, then revisit to the medical records for verification until accurate.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3Statistical analysis\u003c/h2\u003e \u003cp\u003eStatistical analysis was performed using SPSS (Version 25.0, IBM Corp., USA) and R (Version 4.2.1, R Foundation for Statistical Computing, Austria) software. Categorical data were presented as n (%) and analyzed using the \u003cem\u003eχ\u003c/em\u003e2 test. Categorical data were expressed as n (%) and analyzed using the \u003cem\u003eχ\u003c/em\u003e2 test. Continuous data were first tested for normality. Data following a normal distribution were presented as mean\u0026thinsp;\u0026plusmn;\u0026thinsp;standard deviation and analyzed using the independent samples t-test, while non-normally distributed data were presented as median and interquartile range (IQR) and analyzed using the Mann-Whitney U test. The glmnet package was utilized to perform LASSO regression analysis to select predictive factors, which were then subjected to multivariate logistic regression analysis.Prediction models were constructed using logistic regression, random forest, and XGBoost. The models' predictive performance was evaluated using AUC, accuracy, specificity, sensitivity, and F1 score. Model calibration and accuracy were assessed using the Hosmer-Lemeshow (HL) test and calibration curves. The clinical utility of the model was evaluated using DCA. A \u003cem\u003eP\u003c/em\u003e-value of \u0026lt;\u0026thinsp;0.05 was regarded as statistically significant.\u003c/p\u003e \u003c/div\u003e"},{"header":"3 Results","content":"\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Demographic and Clinical Characteristics of Patients\u003c/h2\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e presents the demographic and clinical characteristics of the patients. The results of this study revealed that the prevalence of sarcopenia was 37.53% (298/794) in the training cohort and 38.36% (61/159) in the validation cohort. Except for age and diabetes, the other factors showed no significant differences in the baseline characteristics between the two cohorts (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026gt;\u0026thinsp;0.05).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eDemographic and clinical characteristics of patients\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e \u003cp\u003eTraining cohort (\u003cem\u003en\u003c/em\u003e\u0026thinsp;=\u0026thinsp;794)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e \u003cp\u003eValidation cohort (\u003cem\u003en\u003c/em\u003e\u0026thinsp;=\u0026thinsp;159)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u003cem\u003eP\u003c/em\u003e value\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSarcopenia\u003c/p\u003e \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;298 )\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNon-Sarcopenia\u003c/p\u003e \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;496)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eSarcopenia\u003c/p\u003e \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;61 )\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNon-Sarcopenia\u003c/p\u003e \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;98 )\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge(year)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e69.76\u0026thinsp;\u0026plusmn;\u0026thinsp;12.25\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e65.69\u0026thinsp;\u0026plusmn;\u0026thinsp;11.21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e68.59\u0026thinsp;\u0026plusmn;\u0026thinsp;13.75\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e62.78\u0026thinsp;\u0026plusmn;\u0026thinsp;10.80\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.032\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGender, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e0.828\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e163(54.70)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e224(45.16)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e49(50.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e30(49.18)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFemale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e135(45.30)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e272(54.84)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e49(50.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e31(50.82)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHistory of smoking, \u003cem\u003en\u003c/em\u003e (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e0.168\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e151(50.67)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e261(52.62)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e56(57.14)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e36(59.02)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e147(49.33)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e235(47.38)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e42(42.86)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e25(40.98)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHistory of drinking, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e0.633\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e144(48.32)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e259(52.22)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e54(55.10)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e30(49.18)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e154(51.68)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e237(47.78)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e44(44.90)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e31(50.82)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOsteoporosis, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e0.654\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e140(46.98)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e250(50.40)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e50(51.02)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e25(40.98)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e158(53.02)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e246(49.60)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e48(48.98)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e36(59.02)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNumber of strokes, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\" morerows=\"4\" rowspan=\"5\"\u003e \u003cp\u003e0.481\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1 time\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e85(28.52)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e134(27.02)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e27(27.55)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e13(21.31)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2 times\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e77(25.84)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e124(25.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e28(28.57)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e9(14.75)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e3 times\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e56(18.79)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e123(24.80)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e22(22.45)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e23(37.70)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMore than 3 times\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e80(26.85)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e115(23.19)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e21(21.43)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e16(26.23)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLimb dysfunction, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e0.919\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e103(34.56)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e300(60.48)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e61(62.24)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e19(31.15)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e195(65.44)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e196(39.52)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e37(37.76)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e42(68.85)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eStroke location, \u003cem\u003en\u003c/em\u003e (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003e0.434\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLeft side\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e92(30.87)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e169(34.07)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e31(31.63)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e25(40.98)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRight side\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e109(36.58)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e173(34.88)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e34(34.69)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e14(22.95)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBoth sides\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e97(32.55)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e154(31.05)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e33(33.67)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e22(36.07)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDiabetes, \u003cem\u003en\u003c/em\u003e (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e0.029\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e118(39.60)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e256(51.61)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e70(71.43)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e20(32.79)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e180(60.40)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e240(48.39)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e28(28.57)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e41(67.21)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eStroke type, \u003cem\u003en\u003c/em\u003e (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e0.653\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIschemic\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e139(46.64)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e240(48.39)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e51(52.04)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e28(45.90)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHemorrhagic\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e159(53.36)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e256(51.61)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e47(47.96)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e33(54.10)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTimes since stroke, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\" morerows=\"5\" rowspan=\"6\"\u003e \u003cp\u003e0.481\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;1 month\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e51(17.11)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e90(18.15)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e19(19.39)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e12(19.67)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u0026ndash;3 months\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e62(20.81)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e100(20.16)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e14(14.29)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e9(14.75)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e3\u0026ndash;6 months\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e69(23.15)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e107(21.57)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e20(20.41)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e14(22.95)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e6\u0026ndash;12 months\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e62(20.81)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e92(18.55)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e21(21.43)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e14(22.95)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u0026gt;\u0026thinsp;12 months\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e54(18.12)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e107(21.57)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e24(24.49)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e12(19.67)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTube feeding, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e0.15\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e188(63.09)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e419(84.48)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e77(78.57)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e36(59.02)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e110(36.91)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e77(15.52)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e21(21.43)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e25(40.98)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHospitalization days\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5.69\u0026thinsp;\u0026plusmn;\u0026thinsp;1.83\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e5.58\u0026thinsp;\u0026plusmn;\u0026thinsp;1.78\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e5.70\u0026thinsp;\u0026plusmn;\u0026thinsp;1.73\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e5.15\u0026thinsp;\u0026plusmn;\u0026thinsp;1.82\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.404\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBarthel score\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e48.06\u0026thinsp;\u0026plusmn;\u0026thinsp;29.74\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e51.60\u0026thinsp;\u0026plusmn;\u0026thinsp;28.30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e51.40\u0026thinsp;\u0026plusmn;\u0026thinsp;24.83\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e55.18\u0026thinsp;\u0026plusmn;\u0026thinsp;24.58\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.245\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNRS2002 score\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4.43\u0026thinsp;\u0026plusmn;\u0026thinsp;2.82\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4.38\u0026thinsp;\u0026plusmn;\u0026thinsp;2.92\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4.80\u0026thinsp;\u0026plusmn;\u0026thinsp;2.94\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3.72\u0026thinsp;\u0026plusmn;\u0026thinsp;2.74\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.914\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBMI(kg/m\u0026sup2;)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e19.94\u0026thinsp;\u0026plusmn;\u0026thinsp;2.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e21.38\u0026thinsp;\u0026plusmn;\u0026thinsp;3.06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e20.19\u0026thinsp;\u0026plusmn;\u0026thinsp;2.34\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e21.42\u0026thinsp;\u0026plusmn;\u0026thinsp;3.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.644\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNIHSS score\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e7.24\u0026thinsp;\u0026plusmn;\u0026thinsp;5.39\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e5.84\u0026thinsp;\u0026plusmn;\u0026thinsp;3.81\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e7.93\u0026thinsp;\u0026plusmn;\u0026thinsp;4.71\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e5.72\u0026thinsp;\u0026plusmn;\u0026thinsp;3.39\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.588\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eC-reactive protein(mg/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e8.21\u0026thinsp;\u0026plusmn;\u0026thinsp;4.43\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e6.16\u0026thinsp;\u0026plusmn;\u0026thinsp;3.89\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e8.41\u0026thinsp;\u0026plusmn;\u0026thinsp;4.54\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e6.64\u0026thinsp;\u0026plusmn;\u0026thinsp;4.25\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.289\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSerum total cholesterol(mg/dL)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e197.38\u0026thinsp;\u0026plusmn;\u0026thinsp;60.60\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e204.35\u0026thinsp;\u0026plusmn;\u0026thinsp;58.88\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e192.45\u0026thinsp;\u0026plusmn;\u0026thinsp;59.70\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e205.25\u0026thinsp;\u0026plusmn;\u0026thinsp;61.07\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.789\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSerum albumin(g/dL)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4.49\u0026thinsp;\u0026plusmn;\u0026thinsp;0.60\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4.53\u0026thinsp;\u0026plusmn;\u0026thinsp;0.57\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4.52\u0026thinsp;\u0026plusmn;\u0026thinsp;0.81\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e4.66\u0026thinsp;\u0026plusmn;\u0026thinsp;0.96\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.243\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Results of variables selection\u003c/h2\u003e \u003cp\u003eThe results of the LASSO regression analysis are shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. When λ .1 se\u0026thinsp;=\u0026thinsp;0.033, seven variables were selected from 20 variables: age, limb dysfunction, diabetes, tube feeding, BMI, NIHSS, and C-reactive protein. Multivariate logistic regression analysis of the seven variables (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e) revealed that all were statistically significant (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05). These factors were used to construct the prediction model.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eMultivariate logistic regression analysis of the seven variables\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eβ\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eSE\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cem\u003ez\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cem\u003eWald χ2\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u003cem\u003eP\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u003cem\u003eOR\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003e\u003cem\u003e95% CI\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.032\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.007\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e4.318\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e18.648\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e1.032\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e1.017\u0026thinsp;~\u0026thinsp;1.047\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLimb dysfunction\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.116\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.173\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e6.444\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e41.529\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e3.053\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e2.174\u0026thinsp;~\u0026thinsp;4.287\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDiabetes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.522\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.172\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3.044\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e9.263\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.002\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e1.685\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e1.204\u0026thinsp;~\u0026thinsp;2.359\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTube feeding\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.063\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.198\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e5.379\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e28.937\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e2.894\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e1.965\u0026thinsp;~\u0026thinsp;4.263\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBMI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.222\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.036\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-6.115\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e37.393\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.801\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.746\u0026thinsp;~\u0026thinsp;0.860\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNIHSS score\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.060\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.020\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3.048\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e9.289\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.002\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e1.062\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e1.022\u0026thinsp;~\u0026thinsp;1.103\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eC-reactive protein\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.139\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.021\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e6.562\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e43.056\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e1.149\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e1.102\u0026thinsp;~\u0026thinsp;1.198\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e3.3 Model Construction and Evaluation Results\u003c/h2\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e shows that the LR model performed best across all five metrics. Figure\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e displays AUC values: LR at 0.805, RF at 0.796, and XGBoost at 0.780. While all models performed well but LR was superior, so it was chosen for external validation. In external validation, the LR model achieved an AUC of 0.816 (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA). The HL test resulted in a p-value of 0.128 (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026gt;\u0026thinsp;0.05), indicating good model fit. The calibration curve in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB shows that the bias-corrected line closely aligns with the ideal line, with a mean squared error of 0.01, demonstrating good calibration. Figure\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC displays the DCA curve, where the net benefit of the model surpasses both the None and All lines, indicating strong clinical utility. Based on the model\u0026rsquo;s excellent predictive performance, a nomogram was constructed (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e) to visually present the stroke-related sarcopenia risk scoring system.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eComparison of the performance of the three models\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAccuracy\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePrecision\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRecall\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSpecificity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eF1-score\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLogistic Regression\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.747\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.764\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.767\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.912\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.762\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRandom Forest\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.738\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.667\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.607\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.818\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.635\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eXGBoost\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.728\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.632\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.621\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.790\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.626\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"4 Discussion","content":"\u003cp\u003eSecondary sarcopenia in stroke patients has a high prevalence, averaging 42%(\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e), which is notably higher than in other chronic diseases. Previous studies have reported that the prevalence of sarcopenia ranges from 7\u0026ndash;29.3% in diabetes patients, 15.5% in COPD patients, and 32.5% in cancer patients(\u003cspan additionalcitationids=\"CR13\" citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e), all of which are lower than in stroke populations. This study found a sarcopenia prevalence of 37.53% and 38.36% in stroke patients, further confirming its high occurrence. The high prevalence of sarcopenia in stroke patients highlights the urgency of early detection and prevention. Other studies suggest that factors such as hyperlipidemia and atrial fibrillation may increase the risk of sarcopenia in stroke survivors(\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e). This study identified age, limb dysfunction, diabetes, tube feeding, BMI, NIHSS score, and C-reactive protein as independent risk factors for SRS.\u003c/p\u003e \u003cp\u003eConsistent with the definition of sarcopenia and early research findings, the prevalence of sarcopenia increases with age. With age increasing, muscle mass, quality, and strength gradually decline, this is also true for stroke patients. Eighty-five percent of stroke patients experience upper limb dysfunction from the onset, yet only one-third regain some limb function, and even then, the recovery is often incomplete(\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e). Nerve damage weakens the central nervous system\u0026rsquo;s regulation of muscles. When limbs lack necessary muscle contractions and activity, muscle fibers gradually degrade, leading to a reduction in muscle mass and strength. However, the relationship between the severity of neurological damage (assessed in this study using the NIHSS scale) and sarcopenia in stroke patients remains unclear. Some studies suggest that sarcopenia may induce endothelial dysfunction(\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e), which is associated with neurological deterioration(\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e), the underlying mechanisms are still not fully understood. In this study, the sarcopenia group had an average NIHSS score of 7.24\u0026thinsp;\u0026plusmn;\u0026thinsp;5.39 in the training cohort and 7.93\u0026thinsp;\u0026plusmn;\u0026thinsp;4.71 in the validation cohort, both higher than the non-sarcopenia group. Therefore, we hypothesize that a higher NIHSS score correlates with an increased risk of sarcopenia. Previous studies have confirmed the complex and close relationship between diabetes and sarcopenia, particularly in patients with type 2 diabetes. This connection primarily occurs through mechanisms such as chronic inflammation, insulin resistance, oxidative stress, and neuropathy, all of which contribute to the decline in muscle mass and function(\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e). Dysphagia is one of the most common complications in stroke patients, with over 50% experiencing this condition(\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e). When oral intake becomes difficult or unsafe, tube feeding is an important method for providing enteral nutrition. While this approach supplies the necessary calories for basic metabolism, many patients may still experience inadequate nutritional intake, potentially delaying the recovery of swallowing function(\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e). Prolonged disuse of oral and pharyngeal muscles can lead to degeneration and atrophy, affecting not only swallowing muscles but also increasing the risk of sarcopenia throughout the body. After a stroke, patients are often in a hypermetabolic state, requiring more energy for recovery. If nutrition from a nasogastric tube doesn\u0026rsquo;t meet these needs, weight and BMI may decrease(\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e). This study shows that stroke patients with lower BMI have a higher risk of developing sarcopenia. CRP is a non-specific inflammation marker used to assess systemic inflammation. Previous evidence has shown that sarcopenia patients have higher CRP levels than non-sarcopenic individuals(\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e). This chronic inflammation accelerates muscle breakdown, limits recovery, and negatively affects neurological and metabolic functions.\u003c/p\u003e \u003cp\u003eThis study used LR, RF, and XGBoost to build and validate a sarcopenia risk prediction model for stroke patients, incorporating seven variables: age, limb dysfunction, diabetes, tube feeding, BMI, NIHSS score, and CRP. The LR model performed best, with the highest AUC, specificity, and sensitivity, indicating better accuracy in predicting sarcopenia risk. A nomogram can visually represent the risk of sarcopenia and aid healthcare professionals in dynamically assessing it. It helps reduce the incidence of sarcopenia and improves patients' quality of life.\u003c/p\u003e \u003cp\u003eThis study has some limitations. First, there is no unified diagnostic standard for sarcopenia, with different cutoff values used in Asia and Europe, so caution is needed when applying these results to different populations. Second, as a single-center study, while the model showed good clinical utility, further validation in multi-center, large-sample studies is required. Third, the retrospective study may have missed certain relevant variables, such as cognitive impairment and psychological status, which could be considered in future research.\u003c/p\u003e"},{"header":"5 Conclusion","content":"\u003cp\u003eWe developed and compared three machine learning models, ultimately establishing a nomogram including seven predictive factors: age, limb dysfunction, diabetes, tube feeding, BMI, NIHSS score, and CRP. This may be a practical tool for early screening of sarcopenia in stroke patients. The nomogram is simple to use and can help to identify patients at high risk for sarcopenia to aid in early intervention.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors would like to\u0026nbsp;express gratitude\u0026nbsp;to the study participants and clinical staff for their support and contributions to this study.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthorship contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eHuan Yan:\u0026nbsp;\u003c/strong\u003eData curation, Investigation, Methodology, Software, Writing \u0026ndash; original draft, Writing \u0026ndash; review \u0026amp; editing. \u003cstrong\u003eJuan Li:\u0026nbsp;\u003c/strong\u003eConceptualization, Formal analysis, Funding acquisition, Supervision, Visualization, Writing \u0026ndash; review \u0026amp; editing. \u003cstrong\u003eYujie Li:\u003c/strong\u003eData curation, Investigation, Methodology, Software, Writing \u0026ndash; original draft. \u003cstrong\u003eLihong Xian and Huan Tang:\u0026nbsp;\u003c/strong\u003eMethodology, Software, Validation, Visualization, Writing \u0026ndash; original draft. \u003cstrong\u003eXuejiao Zhao and\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eTing Lu:\u0026nbsp;\u003c/strong\u003eMethodology, Software, Validation, Visualization, Writing \u0026ndash; original draft.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eFunding\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by the\u0026nbsp;National Natural Science Foundation of China(grant numbers 72364005).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe data that support the findings of this study are available from the corresponding author, upon reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthical approval\u0026nbsp;\u003c/strong\u003eThe study was approved by the Ethics Committee of the Guizhou Provincial People\u0026rsquo;s Hospital with the approval number \u0026ldquo;Ethical Approval Word (Research) [2023] No. 060\u0026rdquo;. All participants signed informed consent.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflict of interest\u0026nbsp;\u003c/strong\u003eThe authors declare no competing interests.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eFeigin VL, Stark BA, Johnson CO, Roth GA, Bisignano C, Abady GG, et al. Global, regional, and national burden of stroke and its risk factors, 1990\u0026ndash;2019: a systematic analysis for the Global Burden of Disease Study 2019. The Lancet Neurology. (2021) 20(10):795\u0026ndash;820. DOI: 10.1016/S1474-4422(21)00252-0\u003c/li\u003e\n\u003cli\u003eAydin T, Kesiktaş FN, Oren MM, Erdogan T, Ahisha YC, Kizilkurt T, et al. Sarcopenia in patients following stroke: an overlooked problem. International Journal of Rehabilitation Research. (2021) 44(3):269\u0026ndash;75. DOI: 10.1097/MRR.0000000000000487\u003c/li\u003e\n\u003cli\u003eYao R, Yao L, Rao A, Ou J, Wang W, Hou Q, et al. Prevalence and risk factors of stroke-related sarcopenia at the subacute stage: A case control study. Front Neurol. (2022) 13:899658. DOI: 10.3389/fneur.2022.899658\u003c/li\u003e\n\u003cli\u003eSu Y, Yuki M, Otsuki M. Prevalence of stroke-related sarcopenia: A systematic review and meta-analysis. Journal of Stroke and Cerebrovascular Diseases. (2020) 29(9):105092. DOI: 10.1016/j.jstrokecerebrovasdis.2020.105092\u003c/li\u003e\n\u003cli\u003eMas MF, Gonz\u0026aacute;lez J, Frontera WR. Stroke and Sarcopenia. Curr Phys Med Rehabil Rep. (2020) 8(4):452\u0026ndash;60. DOI: 10.1007/s40141-020-00284-2\u003c/li\u003e\n\u003cli\u003eYang G, Xie W, Li B, Zhao G, Li J, Xiao W, et al. Casual associations between brain structure and sarcopenia: A large‐scale genetic correlation and mendelian randomization study. Aging Cell. (2024) :e14252. DOI: 10.1111/acel.14252\u003c/li\u003e\n\u003cli\u003eDent E, Morley JE, Cruz-Jentoft AJ, Arai H, Kritchevsky SB, Guralnik J, et al. International Clinical Practice Guidelines for Sarcopenia (ICFSR): Screening, Diagnosis and Management. The Journal of nutrition, health and aging. (2018) 22(10):1148\u0026ndash;61. DOI: 10.1007/s12603-018-1139-9\u003c/li\u003e\n\u003cli\u003eCai G, Ying J, Pan M, Lang X, Yu W, Zhang Q. Development of a risk prediction nomogram for sarcopenia in hemodialysis patients. BMC Nephrol. (2022) 23(1):319. DOI: 10.1186/s12882-022-02942-0\u003c/li\u003e\n\u003cli\u003eVogele D, Otto S, Sollmann N, Haggenm\u0026uuml;ller B, Wolf D, Beer M, et al. Sarcopenia \u0026ndash; Definition, Radiological Diagnosis, Clinical Significance. Rofo. (2023) 195(05):393\u0026ndash;405. DOI: 10.1055/a-1990-0201\u003c/li\u003e\n\u003cli\u003eChianca V, Albano D, Messina C, Gitto S, Ruffo G, Guarino S, et al. Sarcopenia: imaging assessment and clinical application. Abdom Radiol. (2021) 47(9):3205\u0026ndash;16. DOI: 10.1007/s00261-021-03294-3\u003c/li\u003e\n\u003cli\u003eChen L-K, Woo J, Assantachai P, Auyeung T-W, Chou M-Y, Iijima K, et al. Asian Working Group for Sarcopenia: 2019 Consensus Update on Sarcopenia Diagnosis and Treatment. Journal of the American Medical Directors Association. (2020) 21(3):300-307.e2. DOI: 10.1016/j.jamda.2019.12.012\u003c/li\u003e\n\u003cli\u003eIzzo A, Massimino E, Riccardi G, Della Pepa G. A Narrative Review on Sarcopenia in Type 2 Diabetes Mellitus: Prevalence and Associated Factors. Nutrients. (2021) 13(1):183. DOI: 10.3390/nu13010183\u003c/li\u003e\n\u003cli\u003eSep\u0026uacute;lveda‐Loyola W, Osadnik C, Phu S, Morita AA, Duque G, Probst VS. Diagnosis, prevalence, and clinical impact of sarcopenia in COPD: a systematic review and meta‐analysis. J cachexia sarcopenia muscle. (2020) 11(5):1164\u0026ndash;76. DOI: 10.1002/jcsm.12600\u003c/li\u003e\n\u003cli\u003eJang MK, Park S, Raszewski R, Park CG, Doorenbos AZ, Kim S. Prevalence and clinical implications of sarcopenia in breast cancer: a systematic review and meta-analysis. Support Care Cancer. (2024) 32(5):328. DOI: 10.1007/s00520-024-08532-0\u003c/li\u003e\n\u003cli\u003eNozoe M, Kanai M, Kubo H, Yamamoto M, Shimada S, Mase K. Prestroke Sarcopenia and Stroke Severity in Elderly Patients with Acute Stroke. Journal of Stroke and Cerebrovascular Diseases. (2019) 28(8):2228\u0026ndash;31. DOI: 10.1016/j.jstrokecerebrovasdis.2019.05.001\u003c/li\u003e\n\u003cli\u003eTang C, Zhou T, Zhang Y, Yuan R, Zhao X, Yin R, et al. Bilateral upper limb robot-assisted rehabilitation improves upper limb motor function in stroke patients: a study based on quantitative EEG. Eur J Med Res. (2023) 28(1):603. DOI: 10.1186/s40001-023-01565-x\u003c/li\u003e\n\u003cli\u003eHe N, Zhang Y, Zhang L, Zhang S, Ye H. Relationship Between Sarcopenia and Cardiovascular Diseases in the Elderly: An Overview. Front Cardiovasc Med. (2021) 8:743710. DOI: 10.3389/fcvm.2021.743710\u003c/li\u003e\n\u003cli\u003eMartin AJ, Price CI. A Systematic Review and Meta-Analysis of Molecular Biomarkers Associated with Early Neurological Deterioration Following Acute Stroke. Cerebrovasc Dis. (2018) 46(5\u0026ndash;6):230\u0026ndash;41. DOI: 10.1159/000495572\u003c/li\u003e\n\u003cli\u003eGeng H-H, Wang Q, Li B, Cui B-B, Jin Y-P, Fu R-L, et al. Early neurological deterioration during the acute phase as a predictor of long-term outcome after first-ever ischemic stroke. Medicine. (2017) 96(51):e9068. DOI: 10.1097/MD.0000000000009068\u003c/li\u003e\n\u003cli\u003eChen H, Huang X, Dong M, Wen S, Zhou L, Yuan X. The Association Between Sarcopenia and Diabetes: From Pathophysiology Mechanism to Therapeutic Strategy. DMSO. (2023) Volume 16:1541\u0026ndash;54. DOI: 10.2147/DMSO.S410834\u003c/li\u003e\n\u003cli\u003eSanz-C\u0026aacute;novas J, L\u0026oacute;pez-Sampalo A, Cobos-Palacios L, Ricci M, Hern\u0026aacute;ndez-Negr\u0026iacute;n H, Mancebo-Sevilla JJ, et al. Management of Type 2 Diabetes Mellitus in Elderly Patients with Frailty and/or Sarcopenia. IJERPH. (2022) 19(14):8677. DOI: 10.3390/ijerph19148677\u003c/li\u003e\n\u003cli\u003eDziewas R, Michou E, Trapl-Grundschober M, Lal A, Arsava EM, Bath PM, et al. European Stroke Organisation and European Society for Swallowing Disorders guideline for the diagnosis and treatment of post-stroke dysphagia. European Stroke Journal. (2021) 6(3):LXXXIX\u0026ndash;CXV. DOI: 10.1177/23969873211039721\u003c/li\u003e\n\u003cli\u003eBraun RG, Arata J, Gonzalez-Fernandez M. Dysphagia and Enteral Feeding After Stroke in the Rehabilitation Setting. Physical Medicine and Rehabilitation Clinics of North America. (2024) 35(2):433\u0026ndash;43. DOI: 10.1016/j.pmr.2023.07.001\u003c/li\u003e\n\u003cli\u003eZielińska-Nowak E, Cichon N, Saluk-Bijak J, Bijak M, Miller E. Nutritional Supplements and Neuroprotective Diets and Their Potential Clinical Significance in Post-Stroke Rehabilitation. Nutrients. (2021) 13(8):2704. DOI: 10.3390/nu13082704\u003c/li\u003e\n\u003cli\u003eBano G, Trevisan C, Carraro S, Solmi M, Luchini C, Stubbs B, et al. Inflammation and sarcopenia: A systematic review and meta-analysis. Maturitas. (2017) 96:10\u0026ndash;5. DOI: 10.1016/j.maturitas.2016.11.006\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"aging-clinical-and-experimental-research","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"acer","sideBox":"Learn more about [Aging Clinical and Experimental Research](http://link.springer.com/journal/40520)","snPcode":"40520","submissionUrl":"https://submission.nature.com/new-submission/40520/3","title":"Aging Clinical and Experimental Research","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Stroke, Sarcopenia, Machine Learning, Predictive model","lastPublishedDoi":"10.21203/rs.3.rs-5354644/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5354644/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e \u003cb\u003eBackground\u003c/b\u003e Sarcopenia often occurs in stroke patients and contributes to worse recovery and a higher risk of death. There is no standardized tool for screening sarcopenia in stroke patients. The objective of this study is to explore the factors influencing sarcopenia in stroke patients, develop a risk prediction model, and evaluate its predictive accuracy.\u003c/p\u003e \u003cp\u003e \u003cb\u003eMethods\u003c/b\u003e Demographic and clinical characteristics of 794 stroke patients were collected. LASSO regression analysis was used for variable selection, and the selected variables were analyzed using multivariate regression. Logistic Regression (LR), Random Forest (RF), and XGBoost were used to construct prediction models, with the optimal model selected for external validation. Bootstrap resampling was used for internal validation of the training cohort, and another 159 stroke patients were collected for external validation. The performance of models was evaluated using the AUC, calibration curve, and Decision Curve Analysis (DCA).\u003c/p\u003e \u003cp\u003e \u003cb\u003eResults\u003c/b\u003e Based on LASSO and multivariate logistic regression analysis, seven variables were selected. The AUC value for the LR model was 0.805, surpassing that of the RF model (0.796) and the XGBoost model (0.780). The LR model also outperformed RF and XGBoost in terms of accuracy, precision, recall, specificity, and F1-score. In external validation, the LR model achieved an AUC of 0.816, and the calibration curve along with the DCA curve demonstrated that the model has nice accuracy and clinical applicability.\u003c/p\u003e \u003cp\u003e \u003cb\u003eConclusions\u003c/b\u003e In this study, we developed a model and presented it as a nomogram to detect the risk of sarcopenia in stroke patients, and such early screening may benefit these patients.\u003c/p\u003e","manuscriptTitle":"Personalized Screening Tool for Early Detection of Sarcopenia in Stroke Patients: A Machine Learning-Based Comparative Study","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-11-20 11:28:09","doi":"10.21203/rs.3.rs-5354644/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2024-12-23T09:33:34+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-12-10T18:22:21+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"268739107612455248435653941535218248931","date":"2024-12-09T15:22:30+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"52897397803301401845954044172813513161","date":"2024-11-11T09:06:13+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-11-07T13:06:54+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-11-05T19:46:05+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-10-30T14:24:46+00:00","index":"","fulltext":""},{"type":"submitted","content":"Aging Clinical and Experimental Research","date":"2024-10-29T13:17:55+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"aging-clinical-and-experimental-research","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"acer","sideBox":"Learn more about [Aging Clinical and Experimental Research](http://link.springer.com/journal/40520)","snPcode":"40520","submissionUrl":"https://submission.nature.com/new-submission/40520/3","title":"Aging Clinical and Experimental Research","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"4b90e25c-8448-4ed6-9451-9aae2e3bc2b3","owner":[],"postedDate":"November 20th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-02-24T15:59:04+00:00","versionOfRecord":{"articleIdentity":"rs-5354644","link":"https://doi.org/10.1007/s40520-025-02945-5","journal":{"identity":"aging-clinical-and-experimental-research","isVorOnly":false,"title":"Aging Clinical and Experimental Research"},"publishedOn":"2025-02-20 15:56:57","publishedOnDateReadable":"February 20th, 2025"},"versionCreatedAt":"2024-11-20 11:28:09","video":"","vorDoi":"10.1007/s40520-025-02945-5","vorDoiUrl":"https://doi.org/10.1007/s40520-025-02945-5","workflowStages":[]},"version":"v1","identity":"rs-5354644","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5354644","identity":"rs-5354644","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}