Quantification of brainstem depigmentation using neuromelanin-MRI in ageing and Parkinson's disease
Year: 2023
Type: dissertation
Study design: Observational Study
Abstract: Background Parkinson's disease (PD) is the most common form of movement disorder, predominantly affecting those above 60 years old with an upward trend of 2.5 million in 1990 to 6.1 million in 2016. PD diagnosis is currently administered by clinicians, whose skill and experience dependent. Neuromelanin-sensitive MRI (NM-MRI), a relatively new non-invasive imaging tool offers great potential as a quantitative biomarker in detecting neuromelanin depigmentation, a cardinal characteristic of PD. However, its clinical utility remains unclear. This study aims to investigate the clinical usefulness of dedicated NM-MRI in quantifying brainstem depigmentation in PD. The specific objectives were (i) to assess and optimise the quantification methods and acquisition protocols, (ii) to assess physiological ageing effects on Substantia nigra pars compacta (SNpc) depigmentation using a control cohort from adulthood-old age, (iii) to determine the diagnostic potential and accuracy of NM-MRI in Parkinson disease, and (iv) to explore whether NM- MRI may be sensitive to detect pathology in cases with uncertain parkinsonism alternative to dopamine transporter single-photon emission computed tomography (DaTSCAN). Materials and Methods A total of 120 normal subjects, 65 PD patients, 36 clinically uncertain Parkinsonism patients, and a meta-analysis of 849 healthy controls and 1213 PD patients from various published studies were analysed. This research compared two 3 Tesla NM sensitive MR imaging sequences and two image processing methods. The SNpc contrast-to-background ratio (CBR) was measured using a semiautomated manual drawing and a recently developed automated template-based method. The following specific research questions were addressed: (i) to assess and optimise the quantification methods and acquisition protocols The first study investigated the test-retest reliability of the two abovementioned image processing methods. MRI scans from six participants over a few sessions were analysed. Four aspects of reliability were analysed; intrarater reliability, interrater reliability, short-term rescan reliability and medium-term rescan reliability. All four aspects consistently displayed data points lie within the upper and lower boundaries limit of agreement (LOA) in the Bland-Altman (BA) plots. Hence displaying high agreement of the CBR in dorsal and ventral SNpc for both processing methods. Meanwhile, the second study evaluated the reproducibility of two dedicated NM-sensitive MRI sequences processed using methods assessed in the first study. Data from 69 volunteers comprising 35 controls and 34 PD patients were scanned via both NM-MRI protocols within a single session and analysed. A strong positive linear correlation was observed between sequences in control and PD measured using both methods. (ii) to analyse a control cohort from adulthood-old age to assess physiological aging effects on SNpc depigmentation The third analysis elucidated the in-vivo lifespan trajectory of neuromelanin in adults aged 22 to 83 years old. Analysis of 94 healthy participants via automated and semiautomated methods showed a strong non-linear effect with an inverted U-shapedassociation beginning with a linear increase up to a plateau at about the mid-5th decade and decayed until old age. (iii) to determine the diagnostic potential and accuracy in Parkinson’s The fourth study was a meta-analysis conducted by systematic pooling of 19 articles that focused on NM-weighted imaging and recent potential MR free water imaging to evaluate cumulative diagnostic ability and performance. NM-MRI forest plot with over 2000 population depicted a significant large disease effect size (1.41, P < 0.00001). The diagnostic performance of seven articles showed high pooled sensitivity (91%) and specificity (83%), suggesting the potential of NM as a diagnostic biomarker. For free- water imaging, the significant positive disease effect size alongside high heterogeneity suggests that the findings may not be suitable for meta-analysis pooling. The systematic review yielded a consistent outcome supporting the potential application of NM weighted imaging in detecting pigmented SN neuronal loss in detecting the severity of PD and monitoring progression, identifying PD subtyping, distinguishing PD and atypical parkinsonism. Subsequently, a prospective case-control study involving single centre data from a multi-centre study, Parkinson's MR Imaging Repository (PaMIR) project with a moderately large cohort of 65 PD patients and 60 controls were scanned from two NM sequences using automated and semi-automated methods. The SNpc CBR of both sequences was normalised and a marked decrease was observed among Parkinson’s than controls. In the dorsal SNpc and ventral SN, normalised CBR for healthy controls was significantly higher compared to that of PD patients using both methods (p<0.0001). The ventral SNpc demonstrated the best performance in distinguishing healthy controls and PD patients in both methods. (iv) to explore whether NM- MRI may be sensitive to detect pathology in cases with uncertain parkinsonism alternative to DaTSCAN The final work explored NM-MRI among clinically uncertain parkinsonian syndrome using the data gathered from an ongoing sub- study of nigrosomal iron imaging in PD (N3iPD). A total of 36 clinically uncertain Parkinson's cases were diagnosed by DaTSCAN and scanned with NM-MRI. Seven of them were identified as negative DaTSCAN group while the remaining 29 were classified as positive DaTSCAN group. NM-MRI using both methods showed consistent reduced CBRs in the positive DaTSCAN group compared to those with negative DaTSCAN group especially in ventral SNpc. Conclusion In conclusion, NM-MRI was demonstrated as a useful non-invasive diagnostic biomarker of PD and based on preliminary data that there might be clinical value in using NM to detect earlier with clinically uncertain parkinsonism. The automated processing approach used limited human resources while the faster acquisition protocol produced higher image contrast in comparison to the counterpart, satisfying the demand of a busy clinical setting. A major limitation of this research is lack of gold standard data from postmortem study hence data with patients diagnosed using PD Brain Bank Criteria, MDS Parkinson Criteria and DaTSCAN may have a possibility to include false positive and negative data.
Author: Sapuan, Abdul Halim
Institution: University of Nottingham
Language: English
Subfield: Clinical Neurology
Field: Medicine
Domain: Health Sciences
Open Access status: green
OpenAlex ID: W7217037644
