Study On The Change Track Of Vocal Function And Its Influencing Factors In Stroke Patients

PurposeIn this study,the Latent class growth model(LCGM)was used to fit potential categories and characteristics of the trajectory of vocal function change in stroke patients and to further investigate the specific factors influencing the existence of different subgroups of vocal function in stroke patients,to provide basis for clinical intervention measures at different treatment stages and early identification of high-risk patients with vocal disorders.MethodsAdopting longitudinal survey research,and selected 322 stroke patients who met the inclusion criteria in the Department of Neurology of a tertiary care hospital in Suzhou from June 2021 to June 2022,and Voice Impairment Index-10 scale(VHI-10)was used to investigate the stroke patients within 48 hours after onset(acute onset,T1),2 weeks after onset(in-hospital stabilization,T2),1 month after onset(initial recovery,T3),3 months after onset(sustained recovery,T4),and 3 months after onset(sustained recovery,T4).T1),2 weeks after onset(in-hospital stabilization,T2),1 month after onset(initial recovery,T3),3 months after onset(continuous recovery,T4),and 6 months after onset(adaptation,T5)for continuous dynamic assessment of vocal function.Firstly,the vocal function of stroke patients at different disease stages was dynamically described by nonparametric test;Secondly,the latent category of the vocal function change track of stroke patients was fitted by the latent category growth model.After determining the best fitting model,the category of individuals was determined by a posteriori probability.Finally,taking the vocal function change track subgroup as the dependent variable,the statistically significant independent variables in the single factor analysis were included into the binary logistic regression equation to analyze the influencing factors of the vocal function change track sub-group classification.Results(1)Characteristics of vocal function changes in stroke patients at different disease stages:The overall trend of vocal disorders in stroke patients is to recover quickly,but there are significant differences at different time points.The total score of VHI and the scores of each dimension decreased significantly within 1 month after the onset of the disease,and the decrease tended to be gradual within 1 to 3 months after the onset of the disease(2)Model fitting results:This study centers on the VHI total score data of 309 patients with≥ 3 times of participation.Based on the LCGM statistical method,the fitting results are the best when the change track of vocal function of stroke patients is determined as two subgroups.At this time,it has a lower BIC value of 9027.881,the highest entropy value of 0.988,and statistically significant LMR and BLRT values(all P<0.01).Combined with intercept and slope value,C1 group was named as "high risk group of phonation disorders",with 30 cases,accounting for 9.7%of the total;Group C2 was named as "low risk group of phonation disorders".There were 279 cases in this group,accounting for 90.3%of the total.(3)Factors influencing the trajectory of vocal function changes in different subgroups:univariate analysis of the two groups of patients was performed in terms of age(χ2=10.197,P=0.006),smoking history(χ2=4.746,P=0.029),comorbid chronic diseases(χ2=6.514,P=0.038),NIHSS score(χ2=13.420,P<0.001),BI index(χ2=6.908,P=0.032),swallowing function(χ2=16.491,P<0.001)and depressive symptoms(χ2=26.458,P<0.001)were statistically significant in terms of comparative differences.(4)Binary Logistic Regression Analysis of the Changes of Vocal Function in Different Subgroups:Multifactorial logistic regression equation analysis of combined chronic diseases≥2(OR=5.904,95%CI:1.205～28.932),age range 45-59 years(OR=6.427,95%CI:2.253～18.334),history of smoking(OR=3.423,95%CI:1.345～8.709),and 5 indicators of abnormal swallowing function(OR=6.758,95%CI:1.623～28.147),and the presence of depressive symptoms(OR=6.656,95%CI:2.146～20.642)were risk factors for entering the high-risk group for vocal disorders in stroke patients.ConclusionThe vocal function of stroke patients shows a rapid recovery trend within 6 months of onset,and there are two different developmental trajectories overall:the low risk group and the high risk group of vocal disorders.Five indicators,including smoking history,age,the number of concurrent chronic diseases,swallowing function,and depressive symptoms,have been identified as important indicators for distinguishing and predicting the developmental trajectory of vocal function in different subgroups of patients,It can provide reference for early identification and screening of high-risk patients with vocal disorders in the future.