Machine-learning model for the prediction of acute orthostatic hypotension after levodopa administration.

BACKGROUND: Levodopa could induce orthostatic hypotension (OH) in Parkinson's disease (PD) patients. Accurate prediction of acute OH post levodopa (AOHPL) is important for rational drug use in PD patients. Here, we develop and validate a prediction model of AOHPL to facilitate physicians in identifying patients at higher probability of developing AOHPL. METHODS: The study involved 497 PD inpatients who underwent a levodopa challenge test (LCT) and the supine-to-standing test (STS) four times during LCT. Patients were divided into two groups based on whether OH occurred during levodopa effectiveness (AOHPL) or not (non-AOHPL). The dataset was randomly split into training (80%) and independent test data (20%). Several models were trained and compared for discrimination between AOHPL and non-AOHPL. Final model was evaluated on independent test data. Shapley additive explanations (SHAP) values were employed to reveal how variables explain specific predictions for given observations in the independent test data. RESULTS: We included 180 PD patients without AOHPL and 194 PD patients with AOHPL to develop and validate predictive models. Random Forest was selected as our final model as its leave-one-out cross validation performance [AUC_ROC 0.776, accuracy 73.6%, sensitivity 71.6%, specificity 75.7%] outperformed other models. The most crucial features in this predictive model were the maximal SBP drop and DBP drop of STS before medication (ΔSBP/ΔDBP). We achieved a prediction accuracy of 72% on independent test data. ΔSBP, ΔDBP, and standing mean artery pressure were the top three variables that contributed most to the predictions across all individual observations in the independent test data. CONCLUSIONS: The validated classifier could serve as a valuable tool for clinicians, offering the probability of a patient developing AOHPL at an early stage. This supports clinical decision-making, potentially enhancing the quality of life for PD patients.

Distinctive Pattern of Metal Deposition in Neurologic Wilson Disease: Insights From 7T Susceptibility-Weighted Imaging.

BACKGROUND AND OBJECTIVES: Noninvasive and accurate biomarkers of neurologic Wilson disease (NWD), a rare inherited disorder, could reduce diagnostic error or delay. Excessive subcortical metal deposition seen on susceptibility imaging has suggested a characteristic pattern in NWD. With submillimeter spatial resolution and increased contrast, 7T susceptibility-weighted imaging (SWI) may enable better visualization of metal deposition in NWD. In this study, we sought to identify a distinctive metal deposition pattern in NWD using 7T SWI and investigate its diagnostic value and underlying pathophysiologic mechanism. METHODS: Patients with WD, healthy participants with monoallelic ATP7B variant(s) on a single chromosome, and health controls (HCs) were recruited. NWD and non-NWD (nNWD) were defined according to the presence or absence of neurologic symptoms during investigation. Patients with other diseases with comparable clinical or imaging manifestations, including early-onset Parkinson disease (EOPD), multiple system atrophy (MSA), progressive supranuclear palsy (PSP), and neurodegeneration with brain iron accumulation (NBIA), were additionally recruited and assessed for exploratory comparative analysis. All participants underwent 7T T1, T2, and high-resolution SWI scanning. Quantitative susceptibility mapping and principal component analysis were performed to illustrate metal distribution. RESULTS: We identified a linear signal intensity change consisting of a hyperintense strip at the lateral border of the globus pallidus in patients with NWD. We termed this feature "hyperintense globus pallidus rim sign." This feature was detected in 38 of 41 patients with NWD and was negative in all 31 nNWD patients, 15 patients with EOPD, 30 patients with MSA, 15 patients with PSP, and 12 patients with NBIA; 22 monoallelic ATP7B variant carriers; and 41 HC. Its sensitivity to differentiate between NWD and HC was 92.7%, and specificity was 100%. Severity of the hyperintense globus pallidus rim sign measured by a semiquantitative scale was positively correlated with neurologic severity (ρ = 0.682, 95% CI 0.467-0.821, p < 0.001). Patients with NWD showed increased susceptibility in the lenticular nucleus with high regional weights in the lateral globus pallidus and medial putamen. DISCUSSION: The hyperintense globus pallidus rim sign showed high sensitivity and excellent specificity for diagnosis and differential diagnosis of NWD. It is related to a special metal deposition pattern in the lenticular nucleus in NWD and can be considered as a novel neuroimaging biomarker of NWD. CLASSIFICATION OF EVIDENCE: The study provides Class II evidence that the hyperintense globus pallidus rim sign on 7T SWI MRI can accurately diagnose neurologic WD.