Predicting potential microbe-disease associations based on multi-source features and deep learning.

Studies have confirmed that the occurrence of many complex diseases in the human body is closely related to the microbial community, and microbes can affect tumorigenesis and metastasis by regulating the tumor microenvironment. However, there are still large gaps in the clinical observation of the microbiota in disease. Although biological experiments are accurate in identifying disease-associated microbes, they are also time-consuming and expensive. The computational models for effective identification of diseases related microbes can shorten this process, and reduce capital and time costs. Based on this, in the paper, a model named DSAE_RF is presented to predict latent microbe-disease associations by combining multi-source features and deep learning. DSAE_RF calculates four similarities between microbes and diseases, which are then used as feature vectors for the disease-microbe pairs. Later, reliable negative samples are screened by k-means clustering, and a deep sparse autoencoder neural network is further used to extract effective features of the disease-microbe pairs. In this foundation, a random forest classifier is presented to predict the associations between microbes and diseases. To assess the performance of the model in this paper, 10-fold cross-validation is implemented on the same dataset. As a result, the AUC and AUPR of the model are 0.9448 and 0.9431, respectively. Furthermore, we also conduct a variety of experiments, including comparison of negative sample selection methods, comparison with different models and classifiers, Kolmogorov-Smirnov test and t-test, ablation experiments, robustness analysis, and case studies on Covid-19 and colorectal cancer. The results fully demonstrate the reliability and availability of our model.

Effect of Intermittent Oro-Esophageal Tube Feeding in Bulbar Palsy After Ischemic Stroke: A Randomized Controlled Study.

BACKGROUND: Nasogastric tube feeding (NG) has been widely used in patients with bulbar palsy after ischemic stroke but is associated with a significant risk of complications including malnutrition and pneumonia. Intermittent oro-esophageal tube feeding (IOE) can help alleviate these concerns. This study explored the clinical effect of IOE versus NG on nutritional status, swallowing function, stroke-associated pneumonia, and depression in patients with bulbar palsy after ischemic stroke. METHODS: This randomized controlled study included 148 patients with bulbar palsy after ischemic stroke who underwent routine treatment and swallowing rehabilitation training in the Department of Rehabilitation Medicine between July 2017 and July 2019 in China. The participants were randomly divided into the IOE group (n=74) and NG group (n=74) with IOE and NG as nutritional supports, respectively. The primary outcome was nutritional status including (1) body mass index (kg/m2), (2) serum ALB (albumin, g/L), and (3) PA (prealbumin, mg/L). The secondary outcomes were (1) swallowing function including (i) Functional Oral Intake Scale (FOIS) and (ii) Penetration-Aspiration Scale, (2) pneumonia, (3) depression, and (4) adverse events. Statistical analyses for continuous outcomes were performed using t test, Mann-Whitney U test and Wilcoxon signed-rank test and categorical variables using χ2 test. SPSS 21.0 was used for all analysis. RESULTS: There were no significant baseline differences between the 2 groups. After the treatment, the IOE group demonstrated significantly better results compared with the NG group in ALB ([32.71±0.94] versus [32.28±0.81] g/L; P=0.003), PA ([278.15±13.81] versus [270.31±15.08] mg/L; P=0.001], body mass index ([19.77±1.03] versus [19.41±0.98] kg/m2; P=0.002], FOIS (P<0.001), Penetration-Aspiration Scale (P<0.001), stroke-associated pneumonia ([1, 4.05%] versus [26, 35.14%]; P<0.001), depression ([1, 1.35%] versus [44, 59.46%]; P<0.001) and overall less adverse events (reflux, fever, discomfort in the throat; P<0.001). CONCLUSIONS: In patients with dysphagia with bulbar palsy after ischemic stroke who received routine treatment and swallowing rehabilitation training, IOE is safer and more conducive to the improvement of nutritional status, swallowing function, stroke-associated pneumonia, and depression than NG. REGISTRATION: URL: https://www.chictr.org.cn; Unique identifier: ChiCTR-INC-17011741.

How enteral nutrition modes influence nasopharyngeal carcinoma survivors with late dysphagia after radiotherapy: a randomized controlled study.

BACKGROUND: Palliation to late dysphagia after radiotherapy for nasopharyngeal carcinoma (NPC) continues to be a challenge. Intermittent oro-esophageal tube feeding (IOE) is an emerging enteral nutrition mode that can address nutrition and quality of life issues related to nasogastric tube feeding (NGT). OBJECTIVES: This study aims to investigate the effect of IOE versus NGT in late dysphagia after radiotherapy for NPC. METHODS: This randomized controlled study included 82 NPC survivors with late dysphagia after radiotherapy. The subjects were randomized divided into the IOE and NGT groups (n1 = n2 = 41). Both groups received standard-of-care rehabilitation. Enteral nutrition supports were administered through IOE or NGT accordingly. This study lasted 2 weeks for each participant. The primary outcome was nutritional status including albumin, hemoglobin, total serum protein, and body mass index. The secondary outcomes were (i) the functional oral intake scale (FOIS), (ii) the penetration-aspiration scale (PAS), (iii) oral transit time (OTT), (iv) hyoid pause time (HPT), (v) pharyngeal transport time (PTT), and (vi) swallowing-quality of life (SWAL-QoL). RESULTS: Three cases quitted the study halfway and there were no significant baseline differences between the IOE (n = 40) and NGT (n = 39) groups. Both time and group effects were significant in all nutritional indicators. The time effect was significant in the FOIS levels, OTT and PTT, while the group effect was not. Either time or group effect were insignificant in the PAS levels and HPT. Both group and time effects were significant in the SWAL-QoL total scores (zGroup = 5.080, P < 0.001; zTime = 18.005, P < 0.001). The significance of time and group effects varied across different dimensions of the SWAL-QoL. CONCLUSIONS: Rehabilitation interventions can improve swallowing function among NPC survivors with late dysphagia after radiotherapy. In this population who received standard-of-care rehabilitation, IOE is more conducive to the improvement of nutritional status, and swallowing-related quality of life. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT06301763.

Analysis of risk factors and development of predictive model for malnutrition in patients with traumatic brain injury.

Malnutrition is a highly prevalent complication in patients with traumatic brain injury (TBI), and it is closely related to the prognosis of patients. Accurate identification of patients at high risk of malnutrition is essential. Therefore, we analyzed the risk factors of malnutrition in patients with TBI and developed a model to predict the risk of malnutrition. A retrospective collection of 345 patients with TBI, and they were divided into malnutrition and comparison groups according to the occurrence of malnutrition. Univariate correlation and multifactor logistic regression analyses were performed to determine patients' malnutrition risk factors. We used univariate and logistic regression (forward stepwise method) analyses to identify significant predictors associated with malnutrition in patients with TBI and developed a predictive model for malnutrition prediction. The model's discrimination, calibration, and clinical utility were evaluated using the receiver operating characteristic (ROC) curve, calibration plots, and decision curve analysis (DCA). A total of 216 patients (62.6%) developed malnutrition. Multifactorial logistic regression analysis showed that pulmonary infection, urinary tract infection, dysphagia, application of NGT, GCS score ≤ 8, and low ADL score were independent risk factors for malnutrition in patients with TBI (P < 0.05). The area under the curve of the model was 0.947. Calibration plots showed good discrimination of model calibration. DCA showed that the column line plot models were all clinically meaningful when nutritional interventions were performed over a considerable range of threshold probabilities (0-0.98). Malnutrition is widespread in patients with TBI, and the nomogram is a good predictor of whether patients develop malnutrition.

A survey of dysphagia services practice in China.

BACKGROUND: Dysphagia services have only recently been formalised within Chinese health services. There has been no comprehensive evaluation of all aspects of dysphagia services practice in China. AIMS: To provide a profile for the dysphagia practice in China by assessing various facets of dysphagia services. METHODS AND PROCEDURES: This is a survey study carried out from December 2022 to May 2023 online in China. One hundred fifty-nine respondents comprising doctors, nurses or therapists involved in the field of dysphagia rehabilitation completed a questionnaire via Sojump. The questionnaire included dichotomous (e.g., yes/no) and multiple-choice questions related to participants' general information and their provision of dysphagia management services. Results were analysed descriptively and the differences among different levels of hospitals were analysed. OUTCOMES AND RESULTS: Approximately two-thirds of hospitals (67.9%) started dysphagia services within the last 5 years. A significant difference in the initiation of dysphagia rehabilitation was observed across different hospital levels (χ2  = 32.70;p < 0.001). Routine dysphagia screening was a standard practice in most hospitals (71.7%), with the most frequently employed screening method being the water swallow test (95.6%). Clinical swallowing evaluation emerged as the preferred initial assessment method (80.5%) and the most utilised assessment method (91.8%). Video fluoroscopic swallow study and flexible endoscopic evaluation swallowing were used by 22.7% and 12.6% of respondents. A significant difference was observed in the choice of assessment methods across hospitals of different levels (χ2 = 43.397; p < 0.001). The majority of respondents (53.5%) primarily employed rehabilitation as the main intervention method (76%-100%). Most respondents indicated that they would reevaluate patients after treatment (74.8%) and reported routine follow-up with patients (60.4%). CONCLUSIONS AND IMPLICATIONS: Despite the relatively recent development of dysphagia management in China, the practice patterns uncovered in this study are broadly consistent with Chinese guidelines and consensus. However, the use of instrumental assessments remains low, resulting in a lack of objective evaluation of patient conditions and outcomes. There is a need for policy and resource support in the field of dysphagia rehabilitation in China. WHAT THIS PAPER ADDS: What is already known on this subject Assessment and treatment methods for dysphagia are various across different patients, medical facilities and settings. There has been no comprehensive evaluation of all aspects of dysphagia rehabilitation practice in China. What this study adds to existing knowledge The study assessed various facets of dysphagia rehabilitation in China through an online survey. We found that the relatively recent development of dysphagia management in China, but the practice patterns uncovered in this study are broadly consistent with Chinese guidelines and consensus. Variations in practice patterns between hospitals of different levels may be attributed to resource constraints and patient needs. Besides, while rehabilitation exercises are the primary intervention method for dysphagia in China, the assessment largely centres on clinical swallowing evaluation. The use of instrumental assessments remains low, resulting in a lack of objective evaluation of patient conditions and outcomes. What are the clinical implications of this work? The study assessed various facets of dysphagia rehabilitation in China, and the results indicate that the use of instrumental assessments needs to be improved and there is a need for policy and resource support in the field of dysphagia rehabilitation in China.

Prediction of lncRNA-disease association based on a Laplace normalized random walk with restart algorithm on heterogeneous networks.

BACKGROUND: More and more evidence showed that long non-coding RNAs (lncRNAs) play important roles in the development and progression of human sophisticated diseases. Therefore, predicting human lncRNA-disease associations is a challenging and urgently task in bioinformatics to research of human sophisticated diseases. RESULTS: In the work, a global network-based computational framework called as LRWRHLDA were proposed which is a universal network-based method. Firstly, four isomorphic networks include lncRNA similarity network, disease similarity network, gene similarity network and miRNA similarity network were constructed. And then, six heterogeneous networks include known lncRNA-disease, lncRNA-gene, lncRNA-miRNA, disease-gene, disease-miRNA, and gene-miRNA associations network were applied to design a multi-layer network. Finally, the Laplace normalized random walk with restart algorithm in this global network is suggested to predict the relationship between lncRNAs and diseases. CONCLUSIONS: The ten-fold cross validation is used to evaluate the performance of LRWRHLDA. As a result, LRWRHLDA achieves an AUC of 0.98402, which is higher than other compared methods. Furthermore, LRWRHLDA can predict isolated disease-related lnRNA (isolated lnRNA related disease). The results for colorectal cancer, lung adenocarcinoma, stomach cancer and breast cancer have been verified by other researches. The case studies indicated that our method is effective.

Prevalence and influencing factors of malnutrition in stroke patients with bulbar paralysis: a cross-sectional study in China.

Background: Although malnutrition has been shown to influence the clinical outcomes of Stroke Patients with Bulbar Paralysis (SPBP), the prevalence and influencing factors have yet to be uncovered. Objective: This study aims to assess the current prevalence and factors associated with malnutrition in SPBP. Methods: A multicenter cross-sectional investigation was conducted among SPBP in China from 2019 to 2021. Information was collected on basic information, health condition, diagnosis, treatment, neurological function, activities of daily living, swallowing function, and nutritional status. A multivariable logistic regression model was used to identify the factors that influenced nutritional status. ROC analysis was used to assess the predictive value of each independent influencing factor and the logit model. Results: In total, 774 SPBP were enrolled, and the prevalence of malnutrition was 60.59%. Pulmonary infection [aOR:2.849, 95%CI: (1.426, 5.691)], hemoglobin [aOR: 0.932, 95%CI: (0.875, 0.982)], serum albumin [aOR: 0.904, 95%CI: (0.871, 0.938)], total protein [aOR: 0.891, 95%CI: (0.819, 0.969)], prealbumin [aOR: 0.962, 95%CI: (0.932, 0.993)], and National Institute of Health Stroke Scale (NIHSS) scores [aOR: 1.228, 95%CI: (1.054, 1.431)] were independent factors associated with malnutrition in SPBP. ROC analysis revealed that the logit model had the best predictive value [area under the curve: 0.874, 95% CI: (0.812, 0.936); specificity: 83.4%; sensitivity: 79.3%; p < 0.05]. Subgroup analysis showed that the nutritional status in dysphagic SPBP was additionally influenced by swallowing function and nutrition support mode. Conclusion: The prevalence of malnutrition in SPBP was 60.59%. Pulmonary infection, hemoglobin level, and NIHSS score were the independent factors associated with malnutrition. Swallowing function and nutrition support mode were the factors associated with malnutrition in dysphagic SPBP.

Excavating important nodes in complex networks based on the heat conduction model.

Analyzing the important nodes of complex systems by complex network theory can effectively solve the scientific bottlenecks in various aspects of these systems, and how to excavate important nodes has become a hot topic in complex network research. This paper proposes an algorithm for excavating important nodes based on the heat conduction model (HCM), which measures the importance of nodes by their output capacity. The number and importance of a node's neighbors are first used to determine its own capacity, its output capacity is then calculated based on the HCM while considering the network density, distance between nodes, and degree density of other nodes. The importance of the node is finally measured by the magnitude of the output capacity. The similarity experiments of node importance, sorting and comparison experiments of important nodes, and capability experiments of multi-node infection are conducted in nine real networks using the Susceptible-Infected-Removed model as the evaluation criteria. Further, capability experiments of multi-node infection are conducted using the Independent cascade model. The effectiveness of the HCM is demonstrated through a comparison with eight other algorithms for excavating important nodes.

How stroke-related dysphagia relates to quality of life: the mediating role of nutritional status and psychological disorders, and the moderating effect of enteral nutrition mode.

Background: Although stroke-related dysphagia has been shown to influence quality of life (QOL), the underlying mechanisms have yet to be uncovered. Objective: This study aims to investigate the mediating role of nutritional status and psychological disorders in the relationship between stroke-related dysphagia and QOL in stroke patients and explore the moderating effect of enteral nutrition mode. Methods: In 2022, A questionnaire survey using stratified random sampling was conducted on 5,322 stroke patients with dysphagia, including Functional Oral Intake Scale (FOIS), Swallowing Quality of Life Questionnaire, Patient Health Questionnaire-9 (PHQ-9), and Generalized Anxiety Disorder-7 (GAD-7) to assess dysphagia, QOL and psychological disorders, respectively, for each participant. Records of serum albumin, Hemoglobin, Total serum protein, serum prealbumin and Body mass index were enrolled to assess nutritional status. Results: FOIS demonstrated a significant positive predictive effect on QOL. Nutritional status and psychological disorders (PHQ-9 and GAD-7) mediated the relationship between FOIS and QOL. Nutritional status-psychological disorders showed a chain mediation effect in the relationship between FOIS and QOL. The moderating effect of enteral nutrition mode was observed. Conclusion: The mediating role of nutritional status and psychological disorders with moderating effect of enteral nutrition mode in the relationship between dysphagia and QOL in stroke patients was found.

Acupuncture Therapy on Dysphagia in Patients with Parkinson's Disease: A Randomized Controlled Study.

OBJECTIVE: To explore the effect of acupuncture therapy on dysphagia in patients with Parkinson's disease. METHODS: This randomized controlled study lasted 42 days and included 112 patients with Parkinson's disease and dysphagia. Participants were randomly assigned to the experimental and control groups (56 cases each group) using the completely randomized design, all under routine treatment. The experimental group was given acupuncture therapy. The primary outcome was Penetration-Aspiration Scale (PAS). The secondary outcomes were (1) Standardized Swallowing Assessment (SSA), and (2) nutritional status including body mass index (BMI), serum albumin, prealbumin, and hemoglobin. Adverse events were recorded as safety indicators. RESULTS: One participant quitted the study midway. There were no significant differences in baseline assessment (P>0.05). After treatment, both groups showed significant improvement in PAS, SSA and nutritional status except for BMI of the control group. There were significant differences between the two groups in the PAS for both paste and liquid, SSA (25.18±8.25 vs. 20.84±6.92), BMI (19.97±3.34 kg/m2vs. 21.26 ±2.38 kg/m2), serum albumin (35.16 ±5.29 g/L vs. 37.24 ±3.98 g/L), prealbumin (248.33 ±27.72 mg/L vs. 261.39 ±22.10 mg/L), hemoglobin (119.09±12.53 g/L vs. 126.67±13.97 g/L) (P<0.05). There were no severe adverse events during the study. CONCLUSION: The combination of routine treatment and acupuncture therapy can better improve dysphagia and nutritional status in patients with Parkinson's disease, than routine treatment solely. (registration No. CLINICALTRIAL: gov NCT06199323).

The effect and mechanism of motor imagery based on action observation treatment on dysphagia in Wallenberg Syndrome: a randomized controlled trial.

BACKGROUND: The effect of motor imagery applied to dysphagia patients with Wallenberg syndrome has not yet been reported. AIM: This trial aimed to investigate the effect and mechanism of motor imagery based on action observation treatment in the rehabilitation of patients with dysphagia in Wallenberg syndrome. DESIGN: A randomized controlled trial. SETTING: The setting was in-patient. POPULATION: Thirty patients with dysphagia of Wallenberg syndrome. METHODS: The patients were divided into the experimental group and the control group. Both groups received conventional dysphagia treatment, and the experimental group underwent the addition of motor imagery based on action observation treatment to the control group once a day for 14 days. Overall swallowing function was assessed with specific scales before and after intervention. Meanwhile, the functional near infrared spectroscopy was used to detect changes in cerebral hemodynamics during the execution of volitional swallowing task and swallowing motor imagery. RESULTS: The standardized swallowing assessment score (P=0.030), Murray secretion scale score (P=0.044) and swallowing quality of life score (P=0.011) of the experimental group improved better than those of the control group. In addition, multiple brain regions of the cortical presented extensive activation (P<0.05) during the execution of swallowing motor imagery. Moreover, there were significant differences (P<0.05) in brain regions pre-motor and supplementary motor cortex, right primary motor cortex, and right primary somatosensory cortex of the experimental group before and after treatment. CONCLUSIONS: The present study indicated that motor imagery based on action observation treatment could improve swallowing function for patients with dysphagia of Wallenberg syndrome as an add-on training. As a top-down rehabilitation training, the mechanism of this therapy may be related to the selective activation of mirror neuron system. CLINICAL REHABILITATION IMPACT: Motor imagery based on action observation treatment can be implemented as part of the therapeutic for dysphagia of Wallenberg syndrome.

Effect of Simple Swallowing Training Program on Early Oropharyngeal Dysphagia in Community-Dwelling Older Adults: A Randomized Controlled Study.

OBJECTIVES: Oropharyngeal dysphagia (OD) in community-dwelling older adults continues to be a challenge due to its insidious onset. This study developed a simple swallowing training program (SSTP) to address these issues and conducted a randomized controlled trial to explore its effect on swallowing function and quality of life. DESIGN: Two-arm randomized controlled trial. SETTING: AND PARTICIPANTS: A total of 248 community-dwelling older adults with OD from were included in 2024 and randomly divided into intervention and control groups. METHODS: The SSTP was developed through expert consultation. A total of 248 community-dwelling older adults with OD were included in 2024 and randomly divided into intervention and control groups. The intervention group underwent the SSTP twice daily for 21 days, with weekends off, and the control group participants did light physical activities by themselves. The primary outcome was the Gugging Swallowing Screen (GUSS), and the secondary outcomes were the Eating Assessment Tool-10 (EAT-10), Swallowing Quality of Life questionnaire (SWAL-QoL), maximum tongue pressure, masticatory ability, bite force, and meal duration. Assessments were conducted on days 1 and 21, while meal duration was assessed every 3 days for lunch. RESULTS: Twenty-seven participants withdrew halfway. There were no significant differences in baseline assessments (P > .05). There were significant between-group and interactive effects in the GUSS [(19.07 ± 1.38) vs (17.28 ± 2.17), Fbetween-group = 6.893, Pbetween-group = .009, Finteractive = 59.504, Pinteractive<.001], EAT-10 {[4.00 (3.00, 5.00)] vs [9.00 (7.00, 10.00)], zbetween-group = -3.502, Pbetween-group<.001; zinteractive = -6.252, Pinteractive<.001}, SWAL-QoL {[166.00 (163.00, 171.50)] vs [154.00 (150.00, 158.00)], zbetween-group = 2.681, Pbetween-group = .007; zinteractive = 5.475, Pinteractive<.001}, maximum tongue pressure {[33.10 (26.48, 36.86)] vs [28.85 (19.21, 35.77)], zbetween-group = 3.377, Pbetween-group = .001; zinteractive = -6.208, Pinteractive<.001}, masticatory ability {[176.92 (133.10, 212.91) vs [163.33 (116.66, 189.32)], zbetween-group = 4.801, Pbetween-group<.001; zinteractive = 6.979, Pinteractive<.001}. Between-group, time, and interactive effects were significant in the meal duration [(23.39 ± 4.32) vs (27.64 ± 5.63), Fbetween-group = 8.692, Pbetween-group = .004, Ftime = 138.683, Ptime< 0.001, Pinteractive = 73.196, Pinteractive<.001]. CONCLUSIONS AND IMPLICATIONS: The SSTP can effectively improve early OD and swallowing-related quality of life in community-dwelling older adults.

[Acupuncture adjuvant treatment for dysphagia in patients with Parkinson's disease: a randomized controlled trial].

OBJECTIVE: To observe the effect of acupuncture on swallowing function and nutritional status of patients with Parkinson's disease (PD) dysphagia. METHODS: A total of 56 patients with PD dysphagia were randomly divided into an observation group and a control group, 28 cases in each one. Both groups were given conventional treatment and swallowing function rehabilitation training.On the basis, the observation group was treated with acupuncture (filiform needling and tongue picking acupuncture). The acupoints of filiform needling were Lianquan (CV 23), Shanglianquan (Extra), Yifeng (TE 17), etc.; and the tongue picking acupuncture was applied at Jinjin (EX-HN 12), Yuye (EX-HN 13) and posterior pharyngeal wall. Rehabilitation training and acupuncture were performed once a day for 5 consecutive days a week, 2 weeks as a course of treatment, 3 courses in total. The swallowing function [oral transit time (OTT), swallowing response time (SRT), pharyngeal transit time (PTT), laryngeal closure duration (LCD)] and nutritional indexes [body mass index (BMI), serum albumin (ALB), serum prealbumin (PA), hemoglobin (Hb)] in the two groups were evaluated before and after treatment, and the clinical effects were compared. RESULTS: After treatment, the paste and liquid OTT, SRT, PTT, LCD in the two groups were shorter than those before treatment (P<0.05), and the observation group was shorter than the control group (P<0.05). After treatment, the nutritional indexes (BMI, ALB, PA, Hb) in the two groups were higher than those before treatment (P<0.05), and the observation group was higher than the control group (P<0.05). The total effective rate of the observation group was 92.9% (26/28), which was better than 67.9% (19/28) of the control group (P<0.05). CONCLUSION: Acupuncture-assisted rehabilitation training can improve the swallowing function and nutritional status in patients with Parkinson's disease swallowing disorders.

How demographic and clinical characteristics contribute to the recovery of post-stroke dysphagia?

ABSTRACT: According to the analysis to find out how demographic and clinical characteristics influent the dysphagia outcome after stroke, furthermore, giving some insights to clinical treatment.One hundred eighty post-stroke dysphagia (PSD) patients were enrolled in this retrospective study at the stroke rehabilitation department. The outcome measurements are beside water swallow test at discharge and length of stay at hospital. Twenty-five demographic and clinical variables were collected for this study. Logistic regression and multilinear regression were utilized to estimate models to identify the risk and protect predictors of PSD outcome.Mouth-opening degree, drooling severity scale (DSS) level, mini-mental state exam (MMSE) level, Barthel index and Berg balance scale were significant different between recovered and unrecovered group. Type of stroke, MMSE degree, DSS and hemoglobin level shown significant predictive value for PSD outcome in logistic regression. In addition, obstructive sleep apnea (OSA) and DSS degree were important risk factors for PSD outcome. Gender, body mass index, drinking, hypertension, recurrent stroke, water swallow test level on admission, Berg balance scale, DSS and days between onset to admission shown significant predictive value for length of stay of PSD patients.PSD outcome was influenced by type of stroke, MMSE degree, DSS and hemoglobin level significantly and obstructive sleep apnea act as an important risk role for PSD recovery.