| Objective1.To clarify the factors influencing the development of dysphagia in stroke patients in both domestic and abroad through bibliometric analysis.2.To develop an assessment model that can predict the risk of prolonged dysphagia in ischemic stroke patients at admission,and to validate and evaluate the model efficacy.By prospectively collecting patient data and establishing a model based on the early clinical symptoms of stroke patients at the time of admission,a predictive model to predict whether stroke patients are at risk for prolonged dysphagia was constructed to provide clinical assistance in the rational selection of nutritional support modalities and early implementation of swallowing therapy,ultimately improving the prognosis and quality of life of patients.Methods1.Bibliometric analysis:The literature related to factors influencing post-stroke dysphagia was retrieved by bibliometric analysis,and for those meeting the inclusion criteria,basic information such as title,author,journal name,year of publication,and country was obtained;relevant influencing factors(P<0.05)were extracted by reviewing the article abstracts or full text,and word frequency analysis was performed to determine the factors to be included in the model-building study.2.Model construction and evaluation:A general information questionnaire was developed based on the proposed inclusion factors identified by the bibliometric analysis,and the availability and usefulness of the relevant factors were assessed in the context of clinical practice.Patients with ischemic stroke who were hospitalized at the First affiliated Hospital of Soochow University from October 2019 to December 2020 were selected for the study using a convenience sampling method,and those who met the inclusion criteria were included in the study for data collection.Patients were followed up 30 days after stroke onset to assess their prognosis,including swallowing function,nutritional status,independence of somatic function,and comorbidities.Patients were grouped according to their swallowing outcomes at 30 days after stroke,and the factors influencing prolonged dysphagia in patients with ischemic stroke were initially determined using univariate analysis;factors with significance in the univariate analysis were included in the multiple logistic regression analysis to establish a risk prediction model for prolonged dysphagia after ischemic stroke,and the ROC curve and Hosmer-Lemeshow(H-L)test were used to evaluate the discrimination and calibration performance of the model,respectively.The bootstrap method was applied to internally validate the model and evaluate the reproducibility of the model;the clinical application value of the model was evaluated by plotting DCA curves and clinical impact curves;finally,the R software was applied to produce nomogram to visualize the presentation of the model.Results1.A bibliometric analysis of factors influencing dysphagia after ischemic strokeA total of 164 relevant documents were included in bibliometric analysis,including 151 in English and 13 in Chinese;43 factors were extracted after reviewing the abstracts or full texts,with a total of 9 classifications and 525 total word frequencies.The top five factors in order were:age(51 times,9.71%),severity of stroke:NIHSS score(48 times,9.14%),brain lobe where the lesion was located(46 times,8.76%),pre-stroke functional level:mRS/BI(38 times,7.24%),and lesion area/volume(27 times,5.14%).2.Construction and evaluation of the predictive modelA total of 350 patients with dysphagia after ischemic stroke were included.165 cases had swallowing disorders after 30 days and 185 cases recovered from dysphagia.The incidence of prolonged dysphagia was 47.1%.Six factors were eventually entered into the logistic regression model,and the results showed that the independent predictors of longterm dysphagia in patients with ischemic stroke were:age(OR=2.009,P=0.041),baseline NIHSS score(OR=1.106,P=0.008),thrombolysis/thrombectomy(OR=0.275,P<0.001),admission FOIS score(OR=0.313,P<0.001),prehospital mRS(OR=3.645,P=0.008),and homocysteine(OR=1.030,P=0.015).The predictive model was:logit P=0.820+0.698×age(0=≤70 years,1=>70 years)+0.101 × baseline NIHSS score-1.291 × thrombolysis/thrombectomy(0=no,1=yes)-1.160×admission FOIS score(1=grade 1,4=grade 4)+1.293×prehospital mRS(0=<3,1=≥3)+0.030 × homocysteine.ROC analysis showed that the C-statistic of the model was 0.938(95%CI:0.912-0.963)and the best cut-off value was 0.653,at which point the sensitivity of the model for risk assessment was 86.06%and the specificity was 90.81%;the H-L test(χ2=4.916,P=0.767)showed that the model had good consistency.DC A analysis showed that when the predicted probability for ischemic stroke patients calculated by the model was 0.1 or more,the net benefit of intervention was higher than "all intervention" or "no intervention";the results of the clinical impact curve analysis showed that the model has some clinical value.3.Prognostic analysis of model-predicted patients in the high-risk and low-risk groupsThe predicted probability of patients’ risk of long-term dysphagia was calculated by the model,and this value≥0.653 was classified as a high-risk group,and vice versa as a low-risk group.Compared with the low-risk group,patients in the high-risk group had lower nutritional scores[10.0(8.0,11.0)],more mRS≥3(156 patients,98.1%),a higher incidence of pneumonia(95 patients,59.7%),and more referrals to rehabilitation hospitals(132 patients,83.0%),all differences were statistically significant(P<0.05).Conclusions1.A bibliometric analysis revealed that recovery from dysphagia after ischemic stroke is influenced by multiple factors.2.Age,baseline NIHSS score,thrombolysis/thrombectomy,admission FOIS score,prehospital mRS and homocysteine levels as independent predictors of long-term dysphagia after ischemic stroke.3.This study developed a nomogram for predicting the risk of prolonged dysphagia after ischemic stroke,which can assist clinical assessment of patients’ swallowing function and provide a reference for clinical selection of swallowing rehabilitation treatment timing and enteral nutrition support modality. |
