This meta-analysis aimed to evaluate diagnostic accuracy of Neuromelanin-sensitive MRI (NM-MRI) in distinguishing Parkinson’s disease (PD) and related syndromes from healthy controls.

28 studies comprising 2,268 patients were included. NM-MRI showed high diagnostic accuracy for differentiating PD from healthy controls, with pooled sensitivity and specificity of 88.0% and 84.4%, respectively. Subgroup analyses indicated that volumetric and areal assessments of the substantia nigra were highly accurate, with an area under the curve (AUC) of 0.869 and 0.857, respectively. Signal intensity performed less effectively, with an AUC of 0.848. Subjective approaches yielded the highest sensitivity (90.8%). 

NM-MRI demonstrates high overall diagnostic accuracy in differentiating PD from healthy controls, with pooled sensitivity of 88% and specificity of 84%. Among evaluated parameters, volumetric and areal analyses of the substantia nigra proved robust, while subjective approaches yielded the highest sensitivity but with limited reproducibility.