Effect of age on substantia nigra subregions using neuromelanin-sensitive magnetic resonance imaging overlapping on quantitative susceptibility mapping.

Background: The signal intensity (SI) of neuromelanin and iron concentration in substantia nigra (SN) increase with age, but may vary within the SN subregions. We aimed to investigate the effect of age on neuromelanin-sensitive magnetic resonance imaging (NM-MRI) measurements and iron concentration on quantitative susceptibility mapping (QSM) in SN subregions. Methods: A total of 95 healthy volunteers aged 18-84 years underwent NM-MRI scans using deep learning reconstruction, with and without a magnetization transfer (MT) pre-pulse, and a three-dimensional (3D) enhanced susceptibility-weighted angiography sequence that was used to construct the QSM. Two radiologists conducted manual measurements twice of the lateral, central, and medial parts of bilateral dorsal and ventral SN and cerebral peduncles (CP) using 2-mm2 regions of interest (ROIs) on NM-MRI and QSM. Signal contrast ratios (CRs) and MT ratios (MTR) as well as the QSM value of SN subregions were measured. Intra-rater and inter-rater reliability of NM-MRI SI and QSM value in SN subregions was examined using intraclass correlation coefficient (ICC). Polynomial regression analysis was performed to explore age-related CRs, MTR, and QSM values in SN subregions. Partial Pearson's correlation analysis using age and sex as covariates was conducted between any two imaging parameters (CRs, MTR, and QSM value) in SN subregions. Statistical significance was set at P<0.05 with Bonferroni correction. Results: CRs (R2=0.270, P<0.001) and QSM value (R2=0.235, P<0.001) as well as MTR (R2=0.241, P<0.001) in dorsal SN respectively exhibited positive and negative linear associations with age; CRs (R2=0.257, P<0.001) in ventral SN fitted a quadratic curve function with age; and QSM value (R2=0.061, P=0.020) in ventral SN showed a mild increasing trend with age. After controlling for age and sex, the correlation between CRs and MTR was negative in the dorsal (R=-0.350, P<0.001) and ventral (R=-0.312, P=0.003) SN, whereas the correlation between CRs and QSM value presented a positive trend only in dorsal SN (R=0.221, P=0.040>0.017) which could not withstand the Bonferroni correction. Conclusions: Effect of age on NM-MRI measures varied across the SN subregions. The relationship between CRs and MTR in the dorsal and ventral SN as well as between CRs and QSM value in dorsal SN could be referential for pathological conditions.