Establishment and validation of a predictive model for vascular Parkinsonism after multiple lacunar infarcts
Gao Yan, Yang Xue, Wang Hua, Ning Yu, Wang Lingling
2026, 47(3):
621-629.
doi:10.3969/j.issn.1006-7795.2026.03.026
Asbtract
(
16 )
PDF (2945KB)
(
5
)
References |
Related Articles |
Metrics
Objective To explore the related risk factors of vascular Parkinsonism (VP) in patients with multiple lacunar infarction (MLI), and to evaluate the application value of hematological indicators and neuroimaging features in predicting the risk of VP. Methods A total of 169 patients with MLI admitted to Dunhua Hospital, Jilin Province from May 2021 to December 2024 were included. They were divided into the MLI group and the vascular Parkinsonism after multiple lacunar infarction (MLI-VP) group based on whether they developed VP. By using a random allocation method, the patients were divided into a model construction set (n=84) and an internal evaluation set (n=85). The clinical data and imaging features of the two groups were compared, and the risk factors for the occurrence of VP in MLI patients were screened. Then, a risk prediction model was constructed, and its performance and clinical application value were evaluated. Results The results showed that neutrophil count, high density lipoprotein-cholesterol (HDL-C), low density lipoprotein-cholesterol (LDL-C), and basal ganglia involvement were significantly associated with the occurrence of MLI-VP (P<0.05). Neutrophil count (OR=1.730, 95%CI:1.046-3.105, P=0.047), decreased HDL-C (OR=0.166, 95%CI:0.034-0.581, P=0.011), elevated LDL-C (OR=5.884, 95%CI:1.520-31.075, P=0.019), and basal ganglia involvement (OR=3.941, 95%CI:1.019-15.732, P=0.046) were all independent risk factors for MLI-VP. Based on these variables, a nomogram was constructed, and the area under the ROC curve was 0.892 (95%CI:0.806-0.892). The calibration curve showed good fit (Hosmer-Lemeshow P>0.05), and the decision curve analysis confirmed the model's clinical utility. Conclusion Neutrophil count, HDL-C, LDL-C, and basal ganglia involvement are closely related to MLI-VP and are independent risk factors. Based on hematological indicators and imaging features, the prediction model shows good discriminative ability in this cohort, and it can be applied to early high-risk patients, providing a basis for implementing individualized precise intervention.