Nomograms for predicting the incidence of white matter lesions: a real-world study based on elderly patients.

Abstract

Nowadays, there are still some doubts about the etiology, development, and preventive measures of white matter lesions. This research collected patient information in the Dryad Digital Repository database to identify predictors of patients with white matter lesions. We studied patients who underwent head MRI and blood tests during comprehensive physical examinations at Shin Takeo Hospital, Japan, between April 1, 2016, and October 30, 2017. We screened patients over the age of 60 and investigated white risk factors in material lesions in older patients and, randomly assigning patients to training and validation groups, we built nomograms on the basis of the training group and used other group data to verify its accuracy and consistency. A total of 854 patients were included in this study. Multivariate analysis was performed according to the data before randomization, and the results showed that the age of patients (OR=2.81, 95% CI: 1.74-4.54), high-density lipoprotein (OR=1.31, 95% CI: 1.02-1.68), diastolic blood pressure (OR=1.47, 95% CI: 1.04-2.08). Taking antihypertensive drugs (OR=1.78, 95% CI: 1.21-2.66) was significantly associated with white matter lesions. The area under the curve value (AUC) was 0.625 for the training group and 0.729 for the validation group. The clinical impact curve and calibration curve show that the model has good accuracy and clinical application value. Age, high-density lipoprotein (HDL), diastolic blood pressure (DBP), and the use of antihypertensive drugs are closely related to white matter lesions. Furthermore, our model may be a useful tool for predicting the incidence of white matter lesions.