近期皮质下小梗死患者外周动脉僵硬度与脑小血管病总负荷程度的相关性研究OA

Objective To investigate the correlation between peripheral arterial stiffness and total cerebral small vessel disease(CSVD)burden in patients with recent small subcortical infarction(RSSI),and to establish a prediction model for total CSVD burden in RSSI patients.Methods Consecutive RSSI patients admitted to the Department of Neurology,the Third Affiliated Hospital of Nanjing Medical University(the Second People's Hospital of Changzhou)from March 2025 to November 2025 were retrospectively enrolled.Clinical data were collected,including demographic characteristics(age,sex),body mass index(BMI),admission blood pressure(systolic and diastolic blood pressure),National Institutes of Health stroke scale(NIHSS)score at admission,medical history(hypertension,diabetes mellitus,hyperlipidemia,coronary heart disease,atrial fibrillation),smoking history,drinking history,and laboratory tests on admission or the next day(total cholesterol,triglyceride,high-density lipoprotein cholesterol,low-density lipoprotein cholesterol,urea nitrogen,creatinine,uric acid,homocysteine[Hcy]).According to total CSVD burden scores,all RSSI patients were divided into mild(0-1points),moderate(2points),and severe(3-4points)groups.Brachial-ankle pulse wave velocity(baPWV)and ankle-brachial index were measured using an arterial stiffness detector(ADS-6000B).Clinical characteristics were compared among the three groups,post hoc multiple comparisons and collinearity analysis were performed for variables with significant differences,variables with collinearity(variance inflation factor≥10)were excluded,and the remaining variables were entered into multivariate ordinal Logistic regression to identify independent risk factors for total CSVD burden(mild,moderate,severe)of RSSI patients.Mediation analysis was performed with baPWV as the independent variable,total CSVD burden(mild,moderate,severe)as the dependent variable,and Hcy as the mediator.Variables with significant intergroup differences were selected by backward stepwise multivariate ordinal Logistic regression to construct a prediction model for total CSVD burden,and the parallel lines test was used(P>0.05 indicated that the proportional odds assumption was satisfied).Restricted cubic splines were used to explore nonlinear relationships between continuous variables and the outcome(P≤0.05 indicated a nonlinear relationship).Ordinal Logistic regression was performed,with cumulative probability of moderate and severe CSVD burden used as the primary clinical endpoint to establish a prediction model for RSSI patients with moderate to severe CSVD burden(the cumulative probability of moderate-to-severe CSVD burden)prediction.The model was visualized using a nomogram and its performance was evaluated with calibration curves and decision curve analysis(DCA).Receiver operating characteristic(ROC)curves of the total CSVD burden level of the RSSI patients were plotted to assess the model's discrimination in predicting the total moderate and severe CSVD burden,with area under the curve(AUC)values of0.5-0.7 as low,>0.7-0.8 as moderate,>0.8-0.9 as excellent,and>0.9 as outstanding.Fit a baseline ordinal Logistic regression model excluding baPWV,and calculate the integrated discrimination improvement(IDI)using the nricens package;IDI>0 and 95％CI not crossing 0 suggested adding baPWV improved the model discrimination compared with the baseline model.Results(1)A total of 101 RSSI patients were enrolled,including 67 males and 34 females,aged 45-91 years with a median age of 67(61,75)years.The NIHSS score was between 0-8,with median NIHSS score at admission was 2(1,2).The total CSVD burden score was between0-4,and median total CSVD burden score was 2(1,3).There were 28 patients in the mild group,53 in the moderate group,and 20 in the severe group.(2)Significant differences were observed among the three groups in age,prevalence of hypertension,admission systolic blood pressure,Hcy,and baPWV(all P≤0.05).Post hoc analysis showed that age,hypertension prevalence,and admission systolic blood pressure were higher in the severe group than those in the mild group;admission systolic blood pressure was higher in the moderate group than that in the mild group;and Hcy and baPWV differed significantly in all pairwise comparisons between groups(all P≤0.05).(3)Multivariable ordinal Logistic regression showed that admission systolic blood pressure(OR,1.028,95％CI 1.007-1.051,P=0.010),Hcy(OR,1.186,95％CI 1.057-1.334,P=0.004),and baPWV(OR,1.409,95％CI 1.142-1.738,P=0.001)were independent influencing factors for total CSVD burden in the RSSI patients.(4)Mediation analysis showed that the total effect of baPWV on CSVD burden was 0.436,and the indirect effect mediated by Hcy was 0.136,accounting for 31.19％of the total effect.(5)Using the total CSVD burden in RSSI patients as dependent variable,enter all variables with significant statistical differences between the three groups(age,hypertension,admission systolic blood pressure,Hcy,and baPWV[all variance inflation factors≤10])into backward stepwise ordinal Logistic regression.Admission systolic blood pressure,Hcy,and baPWV were ultimately included in the model,and the proportional odds assumption was satisfied(P=0.796).The ordinal Logistic regression prediction model was established as following:logit(P[Y≤1])=10.160-(0.025×admission systolic blood pressure+0.164×Hcy+0.298×baPWV);logit(P[Y≤2])=13.752-(0.025×admission systolic blood pressure+0.164×Hcy+0.298×baPWV).(6)Based on the ordinal Logistic regression model(with cumulative probability of moderate and severe CSVD burden used as the primary clinical endpoint),establish the alignment diagram prediction model.After Bootstrap correction,the AUC obtained from predicting RSSI patients with moderate and severe CSVD was 0.821(95％CI 0.749-0.908),indicating good model discrimination and calibration.IDI analysis showed that adding baPWV significantly improved model performance(IDI=0.065,95％CI 0.003-0.192,P=0.019).DCA demonstrated high net benefit within a threshold range of 0.2-0.7,the model demonstrated a high net benefit.Conclusions Admission systolic blood pressure,Hcy,and baPWV are independently associated with total CSVD burden in RSSI patients,and Hcy partially mediates the relationship between baPWV and total CSVD burden.The prediction model based on admission systolic blood pressure,Hcy,and baPWV can effectively stratify moderate-to-severe CSVD burden in RSSI patients and shows favorable clinical utility.