| Object1.To investigate the current status and risk factors of frailty in elderly ischemic stroke patients.2.To develop a nomogram risk prediction model for frailty in elderly ischemic stroke patients,and to internally validate and clinically evaluate the model in the hope of providing an individualized assessment tool for clinical medical staff to recognize frailty in elderly ischemic stroke patients intuitively,concisely,and efficiently.MethodsFrom December 2021 to July 2022,by convenience sampling,a total of 340 elderly ischemic stroke patients from the three departments of a 3-A-grade hospital were recruited.Patients were divided into frailty and non frailty groups by whether they developed weakness.The following questionnaires were used: general demographic questionnaire,modified Rankin Scale(m RS),mini nutritional assessment short form(MNA-SF),health questionnaire depression symptom cluster scale(PHQ-9),social support rating scale(SSRS),mini mental state examination(MMSE),national institutes of health stroke scale(NIHSS),activities of daily living(ADL),and FRAIL scale.The demography characteristics of the research objects are described by means of mean and standard deviation(normal distribution),median and quartile(non normal distribution),frequency and percentage,continuous variables are tested by nonparametric rank sum test,and categorical variable are tested byχ~2 conduct univariate analysis,using multivariate logistic regression to screen independent influencing factors,calculate OR values and 95% CI;Finally,based on the results of logistic regression analysis,a visual nomogram prediction model for frailty in elderly ischemic stroke patients was established using R software.The internal validation of the nomogram prediction model was conducted using the 1000 Bootstrap method to test its feasibility.The predictive performance of the nomogram model was evaluated through discrimination,calibration,and clinical practicality.Results1.In this study,a total of 340 questionnaires were distributed,and 335 questionnaires were returned with a 98.53% response rate,including 331 valid questionnaires with a 97.35% effective response rate.Frailty stroke occurred in 196 of 331 elderly ischemic stroke patients,nonfrail in 135,and frailty in 59.21%.The rank order of the frail scale entries was fatigue 258 cases(77.94%),inability to walk unaided up one flight of stairs 242 cases(73.11%),inability to walk 500 m independently 182 cases(54.98%),suffering from more than five diseases 116 cases(35.04%),and unplanned loss of body mass > 5% 100 cases(30.21%).2.The results of univariate analysis showed that age,education,marital status,family history of stroke,number of strokes,motor impairment,cognitive impairment,language impairment,swallowing disorder,comorbidity of multiple diseases,polypharmacy,m RS score,MNA-SF score,PHQ-9 score,MMSE score,social support total score,objective score,subjective score,NIHSS score,ADL score were significantly different between the frailty and non frailty groups(P<0.05).3.Multivariate logistic regression analysis of the following variables: number o f strokes(OR=2.216,95%CI:1.012-4.853,P=0.047),dyskinesia(OR=5.788,95%CI:2.145-15.616,P=0.001),multiple diseases(OR=4.332,95%CI:2.372-7.912,P<0.001),mal nutrition(OR=0.445,95%CI:0.215-0.924,P=0.030),Poor neurological function(OR=2.219,95%CI:1.331-3.701,P=0.002),ADL descend(OR=0.971,95%CI:0.945-0.999,P=0.040)were independent influencing factors of frailty in elderly ischemic stroke pati ents,the difference was statistically significant(P<0.05).4.The constructed nomogram prediction model for frailty ischemic stroke in the elderly included number of strokes,dyskinesia,multiple diseases,MNA-SF score,m RS score,and ADL score.The nomogram prediction model had a total score of 450 points,and when the total score was > 140 points,the risk of developing frailty was > 60%,and the higher the score,the higher the risk of developing frailty.The model had an AUC of 0.937,an optimal threshold value of 0.698,sensitivity of 0.801,and specificity of 0.963,and internal validation using the 1000 bootstrap method showed an area under the ROC curve of 0.937,indicating good discrimination.Calibration plots showed good agreement between the nomograms in predicting the probability of developing frailty and the actual probability of developing it in patients with ischemic stroke.The decision curve showed that the nomogram predicted net benefit in model patients better and had clinical usefulness.Conclusions1.Frailty accounts for about 59.21% of elderly ischemic stroke patients,with a high incidence.2.The independent predictors of frailty in older ischemic stroke patients included six items: number of strokes,dyskinesia,multiple diseases,Poor neurological function,malnutrition,and ADL descend.3.The established nomogram prediction model of frailty status in older ischemic stroke patients showed good performance in discrimination,calibration,and clinical utility.It can individualize prediction of the incidence of frailty stroke in elderly ischemic stroke patients and provides a basis for clinical caregivers to recognize frailty elderly ischemic stroke patients at an intuitive,concise,and efficient early stage. |
PDF Full Text Request