Table 2 Study Characteristics of the eighteen identified PET/MRI hybrid studies in patients with Parkinson’s disease
Lead author (year); location | Population | Number of participants | Mean age ± SD years | PET tracer/ MR protocol | Summary of findings |
|---|---|---|---|---|---|
An et al. (2016) Korea* | PD UPDRS III score 7.6 ± 4.5 Hoehn & Yahr score 1.4 ± 0.5 HC Healthy normal volunteers | PD (n = 20) HC (n = 10) | 58.7 ± 7.3 | [18F]-FP-CIT Images acquired 110 min after injection [18F]-FDG Images acquired 40 min after injection Dual-echo UTE sequence (TE 0.07 and 2.46 ms, TR = 11.9 ms, flip angle = 10°, slice = 1.6, matrix = 192 × 192 × 192 with isotropic voxel size = 1.33 mm, FOV = 300 × 300 mm) T1-weighted 3D ultrafast GE sequence matrix = 208 × 256 × 256 with voxel size = 1.0 × 0.98 × 0.98 mm) | The MR-based attenuation correction including bones using UTE using level-set segmentation with inhomogeneity correction was demonstrated to be a feasible method with the tracers used The attenuation maps generated by MR using level-set segmentation and PET images were superior compared to corrected attenuation and scatter offered by the manufacturer of the PET/MRI |
Biundo et al. (2021) Italyϕ | PD with normal cognition UPDRS total score 60.1 ± 33.5 UPDRS III score 36.3 ± 21.5 MCI Dementia LBD | PD (n = 5) MCI (n = 22) Dementia (n = 13) LBD (n = 10) | 71.5 ± 5.7 | [18F]-flutemetamol Images acquired 0–10 and 90–110 min after injection UTE used for attenuation correction T1-weighted 3D MPRAGE (TE = 2.53 ms, TR = 1.9 ms, slice = 1 mm, matrix = 256 × 256, FOV = 250) 1 mm isotropic T2-weighted 3D, and 2D SWI were acquired for clinical evaluation to exclude secondary parkinsonisms, vascular brain damage and to allow visual rating scales assessment | The presence of amyloid beta does not differentiate the clinical profile between PD and LBD |
Brakedal et al. (2022) Norway | Newly diagnosed, drug-naïve PD (treatment with nicotinamide riboside versus placebo) UPDRS total score 43 ± 13.3 UPDRS III score 29 ± 9.35 | PD (treatment with nicotinamide riboside) (n = 15) PD (placebo) (n = 15) | 63.8 ± 9.5 | [18F]-FDG Imaging acquired 25–30 min after injection T1-weighted 3D MPRAGE (TE = 2.26 ms, TR = 2.4 s, TI = 900 ms, flip angle = 8°, matrix = 256 × 256 × 192, FOV = 256 × 256 × 192) MRS 31P spectroscopy 3D chemical shift imaging FID sequence For calculating field-strength dependent chemical shift difference, relative amplitudes and frequency separations of oxidised (NAD +) and reduced NAD (NADH) | NR may be a promising neuroprotective medication for PD. Confirmation in longitudinal studies is required to determine the effects of NR on cerebral metabolism and cellular nicotinamide adenine dinucleotide (NAD) levels |
Chen et al. (2019) USA | PD Unspecified clinical features and physical examinations AD MCI LBD HC | PD (n = 12) AD (n = 6) MCI (n = 2) LBD (n = 1) HC (n = 11) | 67.7 ± 7.9 | [18F]-florbetaben Imaging acquired 90–110 min after injection T1-weighted spoiled gradient recalled acquisition (TE = 3.092 ms, TR = 7.648 ms, TI = 400 ms, flip angle = 11°) T2-weighted FSE (TE = 100.896 ms, TR = 3484 ms, flip angle 111°) T2-FLAIR (TE = 161.488 ms, TR = 6000 ms, TI = 1773 ms, flip angle = 90°) | High-quality amyloid PET images can be generated by using deep-learning methods with ultra-low-dose PET data |
Choi et al. (2014) Korea* | PD Unspecified clinical features and physical examinations Non-parkinsonian tremor | PD (n = 14) Non-parkinsonian tremor (n = 2) | 61.3 ± 9.4 | [18F]-FP-CIT Images acquired 110 min after injection Dual-echo UTE sequence (TE 0.07 and 2.46 ms, TR = 11.9 ms, flip angle = 10°, slice = 1.6, matrix = 192 × 192 × 192 with isotropic voxel size = 1.33 mm, FOV = 300 × 300 mm) | UTE sequence-based attenuation correction caused spatial bias despite attenuation maps accounting for cortical bones The DAT BR in the caudate nucleus was considerably underestimated due to the CSF space being misclassified as air in the lateral ventricles |
Choi et al. (2016) Korea* | PD UPDRS III score 7.6 ± 4.5 Hoehn & Yahr score 1.4 ± 0.5 Non-parkinsonian tremor | PD (n = 16) Non-parkinsonian tremor (n = 7) | 60.8 ± 9.5 | [18F]-FP-CIT Images acquired 110 min after injection Dual-echo UTE sequence (TE 0.07 and 2.46 ms, TR = 11.9 ms, flip angle = 10°, slice = 1.6, matrix = 192 × 192 × 192 with isotropic voxel size = 1.33 mm, FOV = 300 × 300 mm) T1-weighted 3D MPRAGE (TE = 1.89 ms, TR = 1.67 ms, flip angle = 9°, matrix = 256 × 256, isotropic voxel = 1.0 mm, FOV = 250) | There is a close relationship between global structural changes and the degeneration of striatal dopaminergic circuits Although not statistically significant, there is a trend of positive correlation between anterior striatal dopaminergic degeneration with changes in the cerebellum and parahippocampal gyri. These may be associated with the non-motor circuits of the basal ganglia There was also a trend with the posterior striatal dopaminergic degeneration being negatively associated with grey matter density in the temporal and occipital cortex |
Delva et al. (2020); Belgium | PD UPDRS total 41.5 ± 13.7 UPDRS III (off medication) 24.0 ± 7.2 Hoehn & Yahr score (On medication) 1 ± 1 (Off medication) 2 ± 0 HC age- and gender- matched | PD (n = 30) HC (n = 20) | 60 ± 8.7 | [18F]-FE-PE2I Images acquired 50–70 min after injection Performed on medication because antiparkinsonian drugs are considered not to interfere significantly with DAT imaging MR protocol not reported | In patients with early stage PD, the axon terminals appear to be the most vulnerable part of nigrostriatal dopaminergic neurons |
Garon et al. (2022) Italyϕ | PD-MCI UPDRS total 49.7 ± 11.5 UPDRS III 28.4 ± 15.5 Hoehn & Yahr score 2.5 ± 1 | PD-MCI Amyloid beta-positive (n = 8) PD-MCI Amyloid beta-negative (n = 17) | 70.8 ± 5.5 | [18F]-FMM Images were acquired between 0– 10 and 90–110 min after injection T1-weighted 3D MPRAGE (TE = 2.53 ms, TR = 1.9 ms, slice = 1 mm, matrix = 256 × 256, FOV = 250) 1 mm isotropic T2-weighted 3D, and 2D SWI were acquired for clinical evaluation | Amyloid burden in the fronto-striatal network may be associated in the worsening of executive functions in PD-MCI Within this cohort, amyloid accumulation was not related to cognitive deficits and brain atrophic patterns that are typically found in AD nor with specific clinical features (disease duration, age of motor symptom onset, medication dosage) or demographics |
Hu et al. (2021) China | PD Unspecified stage and disease severity MSA | PD (n = 60) MSA (n = 30) | 56 ± 9.2 | [18F]-FDG Patients fasted for 6 h and stopped taking any medication for at least 12 h Image was acquired 1 h post injection T1-weighted (TE = 3.0 ms, TR = 7.9 ms, flip angle = 12°, matrix = 288 × 224, slice = 1 mm) T2-weighted (TE = 105 ms, TR = 5523 ms, refocusing flip angle = 142°, matrix = 384 × 240, slice = 5 mm, slice gap = 1.5 mm) T2-FLAIR (TE = 100 ms, TR = 9000 ms, TI = 2475 ms, refocusing flip angle = 160°, matrix = 256 × 192, slice = 5 mm, slice gap = 1.5 mm) fMRI SWI (TE = 4.0 ms, TR = 45.5 ms, flip angle = 15°, matrix = 384 × 320, slice = 3 mm) DWI (TE = 70.0 ms, TR = 6379 ms, matrix = 128 × 128, slice = 5 mm, slice gap = 1.5 mm) | The clinical-radiomics integrated model based on PET/MRI displayed higher performance (more accurate) compared to construction of radiomics signatures by the least absolute shrinkage and selection operator (LASSO) method to identify PD and MSA |
Kwon et al. (2016) Korea | Suspected PD as diagnosed by experienced neurologist | Suspected PD (n = 15) | 65.1 ± 10.4 | [18F]-FP-CIT Images acquired 110 min after injection Antiparkinsonian drugs were stopped 12 h prior to the scans Dual UTE sequence (TE 0.07 and 2.46 ms, TR = 11.94 ms, flip angle = 10°, matrix = 192 × 192, FOV = 300 × 300 mm) | PET/MRI was found to be comparable to PET/CT in discriminating parkinsonian disorders |
Mangalore et al. (2022) India | PD Disease status not reported FLD AD LBD PSP Corticobasal degeneration MND NPH | PD (n = 2) FLD (n = 16) AD (n = 9) LBD (n = 4) PSP (n = 1) Corticobasal degeneration (n = 2) MND (n = 1) NPH (n = 2) | Not reported | PET tracer and parameters not specified T1-weighted MPRAGE (TE = 2.4 ms, TR = 2200 ms, slice = 1 mm, FOV = 220 × 220 mm | This was bilateral frontal, parietal, and temporal atrophy (as with PSP) Patterns of atrophy correlated with areas of hypometabolism shown on PET |
Pan et al. (2022) China | PD UPDRS III 21.5 ± 11.8 Hoehn & Yahr score 1.9 ± 0.9 HC | PD (n = 30 HC (n = 38) | 56.3 ± 13.1 | [18F]-FP-DTBZ Participants were scanned during their off-state condition (12 h after last medication) T1-weighted 3D imaging (TE 3.2 ms, TR = 7.86 ms, flip angle = 10°, voxel size = 0.5 × 0.5 × 0.67 mm, FOV = 230 × 230 mm) | Multi-atlas-based PET/MRI image segmentation method was superior than the template-based method. This method has potential value for enhancing the efficiency and accuracy of routine clinical PD diagnosis |
Quarantelli et al. (2022) Italy | PD UPDRS III 20.4 ± 12.6 Hoehn & Yahr score 1.6 ± 0.6 ET | PD (n = 39) ET (n = 16) | 68.6 ± 8.1 | [18F]-DOPA Participants fasted for 4 h and discontinued antiparkinsonian medication for at least 12 h prior to scanning. 200 mg of carbidopa was administered 1 h before [18F]-DOPA injection to ensure maximal availability of DOPA to the brain and reduce bladder and kidney absorption Imaging was done 86.1 ± 21.8 min following the injection T2*-weighted single shot EPI (TE = 30 ms, TR = 2040 ms, 37 axial slices with 0.5 mm gap, flip angle = 90°, voxel = 3.5 × 3.5 × 3.0 mm, matrix = 64 × 64 mm) MPRAGE (TE = 2.3 ms, TR = 2300 ms, TI = 900 ms, 176 contiguous sagittal slices, flip angle = 8°, matrix = 256 × 248, voxel = 1 × 1 × 1 mm) | The functional connectivity pattern of the cortico-striatothalamic-cortical loop is influenced by nigrostriatal innervation in PD Decreasing metabolic values directly correlate with the connectivity of the sensorimotor cortex with the paracentral lobule and superior temporal regions. The clinical scores demonstrate that these in-turn could potentially cause loss of motor function There is an inverse correlation between the functional connectivity of the thalamus with the sensorimotor cortices that may represent ineffective compensatory mechanisms |
Rodriguez-Rojas et al. (2020) Spain | PD UPDRS III 33.0 ± 6.4 Hoehn & Yahr score 1.6 ± 0.6 | PD (n = 8) | 61.1 ± 10.3 | [18F]-FDG PET acquisition was done 40 min after injection Data was acquired with patients off-medication T1-weighted MPRAGE (TE = 3.34 ms, TR = 2300 ms, slice = 1 mm, matrix = 256 × 256, FOV = 256 × 256) SWI Parameters not specified T2-weighted TSE Parameters not specified | There were significant positive changes in glucose metabolism in distributed nodes o the cortical-subcortical networks following a incisionless transcranial intervention (MRgFUS-subthalamotomy). The procedure shifted metabolic patterns towards healthy values |
Shang et al. (2021) China | Early stage PD UPDRS III 26.88.0 ± 8.73 Hoehn & Yahr score 1.28 ± 0.45 HC matched for age, sex, education, and Montreal Cognitive Assessment (MoCA) scores | Early stage PD (n = 25) HC (24) | 65 ± 12.6 | [18F]-DOPA Patients were fasted for at least 6 h. Carbidopa was orally administered 1 h prior to injection of [18F]-DOPA Data acquisition was done 90 min following injection Transverse EPI sequence-based DTI (TE = 78 ms, TR = 4663 ms, flip angle = 90°, slice = 4 mm, matrix = 118 × 128 mm, FOV = 230 × 250 mm, acceleration factor = 2 T1-weighed spoiled GRE (TE = 3.2 ms, TR = 7.86 ms, TI = 810 ms, slice = 0.71 mm, flip angle = 10°, matrix = 348 × 384 mm, FOV = 232 × 256 mm) | PET and DTI abnormalities and asymmetry of the dopaminergic system were detected in the early stage PD patients On the more affected side, there were significant associations between DTI metrics (nigral FA, putaminal MD and FA of the nigrostriatal projection) and motor performance. These were also significantly medicated putaminal standardised uptake value |
Wimalarathne et al. (2022) China | Patients suspected of AD/PD | Patients suspected of AD/PD (n = 75) | 61.2 ± 8.9 | [11 C]-CFT Unspecified PET protocol [11 C]-PiB Unspecified PET protocol T1-weighted 3D GE sequence (TE 2.6 ms, TR = 6.9 ms, flip angle = 12°, matrix = 384 × 384, FOV = 240 × 240 mm) | Standard uptake value and image quality parameters were improved by time-of-flight reconstruction |
Wurster et al. (2022) Germany | Rare SNCA-Triplication Early onset PD | Early onset PD (n = 1) | 24 | [18F]-FDG PET data was acquired 1 h after injection MPRAGE (TE = 3.37 ms, TR = 2300 ms, TI = 900 ms, voxel = 1.0 × 1.0 × 1.0 mm) | Bilateral hypometabolism in the frontal, temporoparietal, occipital, precuneus, gyrus cinguli posterior regions were noted in this patient. Corresponding left-sided atrophy was noted in these regions. Conversely, there was increase glucose metabolism in the basal ganglia with normal volumetric appearance |
Zang et al. (2022) China | PD UPDRS III 59.71 ± 15.00 Hoehn & Yahr score 3.00 ± 0.83 HC age- and sex-matched | PD (n = 34) HC (n = 25) | 61.2 ± 5.5 | [18F]-FDG Patients fasted at least 6 h and did not take medication 12 h before examination Images were acquired on an average of 1 h after injection Resting-state fMRI SWI (TE = 30 ms, TR = 2000 ms, flip angle = 90°, slice = 3.5 mm, slice gap = 0.7 mm voxel = 3.5 × 3.5 × 3.5 mm, FOV = 230 × 230 mm High-resolution T1-weighted 3D imaging (TE = 3.8, TR = 7.9, FOV = 256 × 256 mm, Voxel = 1 × 1 × 1mm) 3D multi echo GE SWI (TE = 3.1/6.4/9.7/13.0/16.3/19.6 ms, TR = 29 ms, flip angle = 15°, matrix = 256 × 256 mm, voxel = 1 × 1 × 2mmacceleration factor = 2) | In PD patients, there was increased iron deposition in the substantia nigra, increased FDG uptake in the putamen bilaterally, and decrease in functional connectivity between the substantia nigra and the anterior putamen bilaterally There was also significant interaction between nigral iron deposition and nigral-putamen connectivity with putaminal FDG uptake |