Translate this page into:
Role of 6-Fluoro-18F-l-3,4-DOPA MR-PET in Clinical Workup in Movement Disorders Cases: An Exploratory Study in a Tertiary Care Neuropsychiatric Setup
Address for correspondence: Dr. Keerti Sitani, Department of Neuroimaging and Interventional Radiology, National Institute of Mental Health and Neurosciences, Bengaluru - 560 029, Karnataka, India. E-mail: keerti.sitani@gmail.com
-
Received: ,
Accepted: ,
This article was originally published by Wolters Kluwer - Medknow and was migrated to Scientific Scholar after the change of Publisher.
Abstract
Introduction:
6-Fluoro-18F-l-3,4-dihydroxyphenylalanine (18F-DOPA) is primarily used to differentiate nondopamine deficient symptomatic subjects from Parkinsonian disorders and various etiologies from one another, which otherwise can be challenging. The current study aimed to determine (18F)-DOPA positron emission tomography (PET) patterns in Idiopathic Parkinson’s disease (IPD), nondopamine deficient symptomatic (NDDS) and other movement disorders.
Materials and Methods:
Sixty patients referred from neuropsychiatric outpatient department were retrospectively analyzed who underwent (18F)-DOPA magnetic resonance (MR)-PET scans. Out of 60 subjects, 36 were diagnosed with IPD, 5 were cases of other movement disorders, and 19 were normal on PET imaging (NDDS). To evaluate and quantify the tracer uptake from the PET scans, a region of interest (ROI) was drawn on the least radiotracer uptake region, and subsequently, on bilateral occipital cortex, caudate (head, body, tail), putamen (anterior, posterior), periaqueductal gray matter and red nucleus using structural MR images (MRI) for better delineation of anatomical regions. Tracer uptake ratios from each selected region were generated as SUVmax of ROI/occipital. The statistical analysis was performed using SPSS 29.0 version.
Results:
The posterior putamen showed a significant difference between IPD and other patients with an SUV cutoff of 1.69 and 100% specificity and 99.2% accuracy for normal controls. Whereas, we observed an SUV cutoff of 1.75 and an accuracy and specificity of 100% in patients with other movement disorders. The absolute values from the caudate nucleus were demonstrating a statistical significance between other movement disorders compared to normal controls with a cutoff of 1.69 and an accuracy of 81.2%. The ratio of posterior putamen/occipital ratio was significant between IPD and normal controls, with a cutoff of 3.89 and an accuracy of 95.6%. Right caudate to occipital ratio (RCOR) was significant for differentiating between other movement disorders and normal controls with a cutoff of 3.01 and an accuracy of 95.6%. One novel finding was that occipital lobe uptake was significantly reduced in patients with IPD compared to other movement disorders.
Conclusion:
The findings from the current study demonstrated that hybrid imaging using 18F-DOPA PET/MRI could efficiently differentiate between various causes of Parkinson’s disease with diagnostic accuracy and necessary cut-offs.
Keywords
6-Fluoro-18F-l-3
4-dihydroxyphenylalanine
cutoff
idiopathic Parkinson’s disease
movement disorders
Highlights
Unique study in view of a neuropsychiatric setup where both neurology and psychiatric patients are included
The presence of MRI with PET was significant in ruling out other structural causes of Parkinson’s disease as well as in delineating the anatomy better for creating an ROI for the red nucleus. Organic disorders can have misleading findings on stand-alone PET/CT such as in vascular PD and NBIA
The paper concluded that absolute values for ROI are more sensitive for differentiating healthy and PD patients, whereas ratios are more specific for differentiating the healthy and PD patients, where absolute values can be an easy tool for differentiating in an outpatient department (OPD) setup
This paper was helpful in creating cutoffs for differentiating between various causes of Parkinson’s disease from healthy controls with diagnostic accuracy
One important novel observational finding in this study was a decreased tracer uptake in occipital in patients with idiopathic Parkinson’s disease compared to other movement disorders-which has not been reported in the literature to the best of our knowledge.
Introduction
Movement disorders can be hypokinetic (e.g., parkinsonism), hyperkinetic, or dystonic and are commonly attributed to the altered functioning of the basal ganglia nuclei or their connections.[1] Parkinson’s disease (PD) most commonly presents around the 60 s and is characterized by bradykinesia, rigidity, postural instability, and tremor. However, few cases may present early, but the risk increases with age. The annual mortality rate from PD averages 2 per 100,000 population. Almost 70% of the deaths occurred between 65 and 79 years of age.[2] Pathologically it shows a neuronal loss in the substantia nigra with dopaminergic denervation of the striatum. Neuronal degeneration in the substantia nigra preferentially affects the ventrolateral cell group projecting to the posterolateral putamen. It is accompanied by the formation of Lewy bodies composed of aggregated α-synuclein.[3]
Positron emission tomography (PET) allows measurements of various physiological parameters, including regional brain blood flow and metabolism, various brain receptors, neurotransmitter systems, postsynaptic signal transduction, and abnormal protein deposition. Each of these measures can reveal various insights into the pathophysiology of the clinical manifestations of movement disorders, including PD and other related Parkinsonian syndromes.[4] In the past decade, molecular imaging has provided essential diagnostic tools and methods for assessing novel therapeutics, hence is a powerful means for revealing changes along the various pathophysiological stages of the disease.[5]
Although molecular imaging using PET is good for assessing the dopaminergic function, however, the structural delineation is poor due to the limited spatial resolution of PET; hence, the integration of PET with magnetic resonance imaging (MRI) in a single hybrid scanner would overcome this limitation, as MRI offers the enormous potential of obtaining high-resolution information on brain anatomy, structural volumetric of the gray matter, and structural connectivity of the white matter, which can be used to improve the spatial resolution of data acquired with PET.[6]
To the best of our knowledge, very few studies have been performed using a hybrid simultaneous PET-MRI in patients with movement disorders. Furthermore, this is the first study in which we have reported the cutoff values for absolute standardized uptake values (SUV) and ratios from the regions showing dopaminergic dysfunction in PD hence the main objectives of our study are to visually see the 6-Fluoro-18F-l-3,4-dihydroxyphenylalanine (F18-DOPA) PET/MRI patterns in various movement disorders (Idiopathic PD [IPD], other movement disorders and nondopamine deficient symptomatic NDDS), assess the additional role of MRI along with PET in imaging of the movement disorders, to determine the cutoff values in 18F-DOPA PET scans in IPD, movement disorders, and NDDS and to assess the utility of absolute value versus ratio.
Materials and Methods
These data comprise referral cases between 2019 and 2022; the main burden of the referrals included patients presenting with movement disorders in Neurology and Psychiatry outpatient department (OPD). The psychiatry OPD included patients on psychiatric medications, presenting with movement abnormalities to rule out the presence of PD.
A total of 60 patients were retrospectively visually analyzed by two trained nuclear medicine physicians and two trained radiologists with a minimum of 3 years of clinical experience.
After visual analysis, out of 60 patients (35 males and 25 females, age 24–73 years of age, mean age - 48.5 years), 36 showed the involvement of either unilateral or bilateral posterior putamen and were classified as IPD, 19 were normal on FDOPA PET hence were classified as the nondopamine deficient state (NDDS). Rest 5 were found to have caudate nucleus and midbrain involvement in the PET scan without the involvement of putamen and subsequently classified as other movement disorders.
MRI were also analyzed to rule out any vascular causes for PD, including infarcts and microbleeds, and also to find any other imaging signature findings such as hummingbird or morning glory signs to confirm the differential diagnosis of other movement disorders.
Positron emission tomography/magnetic resonance imaging protocol
Magnetic resonance imaging protocol
MRI scans were performed on 3.0 T mMR Biograph (Siemens Healthcare, Germany), an integrated simultaneous MR-PET system. Routine MRI sequences for clinical diagnosis were used, including T1-MPRAGE, 3D FLAIR, ASL, and PCASL. T1 MPRAGE images were used for the co-registration of PET scans to delineate anatomical structures better.
Positron emission tomography protocol
5–7 mci of radioactivity was injected, and a PET scan was acquired 90 min postinjection. The scan was done using 2FWHM, and reconstruction was done using OSEM, five subsets, 21 iterations, a Gauss filter, and matrix 344.
Imaging analysis
In the analysis of absolute values, we correlated the side of the pathology with the visual analysis of the PET image and divided right, left, and bilateral IPD, and similarly in other movement disorders. For a correlational study with IPD and other movement disorders, the healthy controls were also divided into right and left groups. Only the SUV values of the region of interests (ROIs) from the affected side were considered for further analysis. The SUV values from the unaffected side were not taken into account.
SUV values were calculated by placing a point ROI, as shown in Figure 1, in the following regions and were entered into a database: Caudate (left and right) and further subdivided into head caudate, body caudate, and tail caudate; putamen (right and left); putamen was also subdivided into anterior and posterior; red nucleus (right and left); and periaqueductal gray matter (PAGM) (right and left). The SUV was quantified, and 14 ROI SUV maps (LBM masks) were noted. These values were considered absolutes for calculating the ROI and occipital ratio (right and left). Right and left occipital lobe values were entered, and ratios were generated which were as follows: right caudate: occipital ratio (RCOR), left caudate nucleus (head/body/tail): Occipital lobe; right anterior putamen: occipital lobe; left anterior putamen; occipital ratio; right posterior putamen; occipital ratio (PPOR); left PPOR; right periaqueductal gray matter: occipital ratio; left periaqueductal gray: occipital ratio; right red nucleus: occipital ratio; left red nucleus: occipital ratio.
We had 38 healthy controls as normative data for the right and left ROIs. In IPD, we had 32 unilateral and 4 bilateral; in other movement disorders, we had 4 unilateral and one bilateral. A total of 9 datasets were received in other movement disorders, for a total of 40.
The occipital values were looked at separately, as they were used as the standard normative denominator for generating ratio values in the DOPA scan because it presumed that the occipital lobe does not show any uptake for 18F-DOPA.
Statistical analysis
We used the SPSS 29.0 version (IBM corporation, NY, USA) for statistical analysis. Because our data were not normally distributed, we used the Kruskal–Wallis test, a nonparametric alternative to ANOVA, to test hypotheses and find significant differences between the ROIs of healthy, IPD, and other movement disorders. Outliers were excluded from each ROI for generating the cutoff values to differentiate the groups.
The Bonferroni correction for multiple tests has adjusted all the significance values.
For the receiver operating characteristic (ROC) curve and cutoff value, we kept criteria of minimum >80% and >70% of sensitivity and specificity, respectively for generating cut-off value for each ROI for differentiating the groups. All the other statistical values are shown in the Table 1.
| Category | Absolute | Cut-off | Sensitivity | Specificity | AUROC |
|---|---|---|---|---|---|
| Other movement disorders vs NDDS | Caudate nucleus | 1.69 | 88.9% | 75% | 81.9% |
| IPD vs NDDS | Anterior putamen | 1.79 | 84.4% | 70% | 75.7% |
| Posterior putamen | 1.69 | 88.2% | 100% | 99.2% | |
| IPD vs other movement disorders | Anterior putamen Posterior putamen |
1.79 1.75 |
89.9% 100% |
70% 100% |
85.7% 100% |
| Occipital lobe | 0.655 | 80% | 84.3% | 75.7% | |
| Category | Ratio | Cut-off | Sensitivity | Specificity | AUROC |
| IPD vs NDDS | PPOR | 3.35 | 86.6% | 86.6% | 95.6% |
| Other movement disorders vs NDDS | RCOR RNOR |
3.02 2.1 |
92.1% 83.8% |
85.7% 75% |
95.5% 80.2% |
IPD: Idiopathic Parkinson’s disease, NDDS: Non dopamine deficient state, PPOR: Posterior putamen/occipital ratio, RCOR: Right caudate occipital ratio, AUROC: Area under the receiver operating characteristic, RNOR: Red nucleus occipital ratio
Results
Demographic and clinical features
The patients in the IPD group presented with tremors (most common symptom), gait disturbances (second most common), memory disturbances, speech disturbances, sleep disturbances, RBD- REM sleep and behavioural disorders (2), non-specific behavioural symptoms (anxiety and paranoia) and urinary incontinence (1). Other movement disorders presented mainly with tremors and gait difficulties.
NDDS group presented with the most common tremors, slowness of gait, and nonspecific behavioral symptoms.
F-l-3,4-dihydroxyphenylalanine
Idiopathic Parkinson’s disease
Out of 36 patients with IPD, the involvement is shown in Figure 2. The right-side involvement was decided when the right posterior putamen showed reduced fluorodeoxyglucose uptake as compared to the left side and vice versa, as shown in the master chart category IPD.
Out of 5 patients with other movement disorders, the 18 F-DOPA pattern is shown in Figure 3. For patient details and their uptake patterns kindly, refer to master chart category other movement disorders.
Magnetic resonance imaging
Idiopathic Parkinson’s disease
MRI was useful in this category to rule out any other structural causes for PD that can be seen. The main MRI findings of the patients are noted in Table 2.
| MRI (IPD) | MRI (other movement disorders) |
|---|---|
| Normal-12 | caudate putamen volume loss, with humming bird morphology in midbrain, fronto- temporal atrophy-1 (patient 2 master chart) |
| Diffuse parenchymal atrophy-4 | No structural abnormality-3 |
| Small vessel disease- 4 | Signal changes involving the bilateral globus pallidus, subthalamic nucleus and substantia nigra-1 (patient 5 master chart) |
| Gliosis -1 | |
| ASL abnormalities-3 | |
| Putaminal atrophy-1 | |
| Prominent mineralization of bilateral globus pallidus-1 |
MRI: Magnetic resonance imaging, IPD: Idiopathic Parkinson’s disease
Other movement disorders
MRI was especially useful in this group of patients to make a diagnosis since MRI features are relied on for the same, as reported in Table 1.
Statistical results
With a P ≤ 0.001, the caudate heads show the most significant difference between IPD and normal, while other movement disorders and normal show a P = 0.006. The caudate body also shows the most significant difference between IPD and normal, with a P = 0.0. The red nucleus shows a significant difference between IPD and normal, with a P = 0.018.
With a P ≤ 0.001, the anterior putamen shows the most significant difference between IPD and normal. IPD and other movement disorders show a significant P = 0.001. The posterior putamen shows a significant difference between IPD and normal with a P ≤ 0.001; IPD and other movement disorders show a significant P ≤ 0.001.
The cutoff values for the caudate nucleus had a diagnostic accuracy of 81.9% in differentiating other movement disorders and NIDDS; the posterior putamen had the highest diagnostic accuracy in differentiating between IPD and NIDDS with a cutoff of 1.69, the posterior putamen has the highest diagnostic accuracy in differentiating between other movement disorders and IPD, the sensitivity, specificity, and area under the ROC (AUROC) is mentioned in Table 1.
In ratios, the cutoff for differentiating between NIDDS and IPD was 3.5 for PPOR. To differentiate between NIIDS and other movement disorders, a cutoff value of 3.02 was noted in RCOR with a diagnostic accuracy of 95.5% – the details about the sensitivity, specificity, and AUROC as reported in Table 1.
This statistical analysis concludes that absolute values for ROI are more sensitive for differentiating healthy and PD patients, whereas ratios are more specific for differentiating the healthy and PD patients.
Discussion
F18-DOPA PET-MRI can be used for monitoring and evaluating the disease progression, dopamine receptor mapping, and evaluation of levodopa-induced dyskinesias that may occur during chronic PD treatment with L-DOPA.[7] In recent studies considering the role of 18F-DOPA for movement disorders, it was suggested to be the best diagnostic tool for PD and other movement disorders.[8] However, no specific cutoff values are still cited for the same. Studies have been done to assess the intrastriatal dopamine gradient in cases of movement disorders. In our study, we aimed to differentiate the cases of IPD, Atypical parkinsons disease (APD), and NDDS based on F18-DOPA uptake and calculate the cutoff values for absolute and ratios from the SUV values using much more extensive regions than any of the previously reported studies.
It is seen that F18-DOPA uptake follows a regional selectivity in cases of PD. The regional selectivity of PD is relatively specific. A study by Fearney et al.,[9] on a data set of 36 patients concluded that patients with IPD mostly affect the ventrolateral parts of the substantia nigra and its projections.[10] Furthermore, a study performed by Matuskey et al.[10] reported a significantly decreased density of synapses in the substantia nigra, red nucleus, and locus ceruleus which can explain the significant decrease in radiotracer uptake in the red nucleus in the patients with idiopathic PD however no such analysis was done in this study.
In a previously done retrospective study by Jamini et al.[11] in 117 patients with movement disorders, circular regions of interest (141.5 mm2 each for the caudate, anterior and putamen, and occipital region) were placed over the mean image. The striatal-to-occipital ratios (SORs) were generated for each structure using bilaterally averaged occipital ROI data. The main ratios considered were SOR, intra-striatal ratios caudate to anterior putamen, caudate to posterior putamen, caudate to whole putamen, anterior to posterior putamen. In this study, significance was seen in the regional striatal-to-occipital ratio (RSOR) (SOR of caudate/SOR of posterior putamen and whole putamen) of caudate to posterior putamen and whole putamen, between IPD and APD. SOR was seen as significant between IPD and healthy controls.
Another study by Stormezand et al.[12] did a subregional analysis of F18-DOPA uptake on a total of 58 patients. Their data set included 28 cases of IPD, 13 APD, and 17 healthy controls. They used RSORs and multiple in-line VOIs from the caudate nucleus to the posterior part of the putamen. A linear regression fit and associated R-squared values determined the linearity of anteroposterior decline. ROC curves were used to assess the diagnostic performance of these measurements. Data contralateral to the clinically most affected side were used for analysis. Their results showed the highest AUC for the caudate nucleus-to-posterior putamen ratio. For differentiating IPD from other movement disorders, the highest AUC was found for the caudate nucleus-to-anterior putamen ratio. They concluded that subregional analysis of the striatum in 18F-DOPA PET might provide additional diagnostic information in patients screened for a presynaptic dopaminergic deficit. Still, they did not provide a cutoff for the subregion SUV values.
It should be noted that both the studies cited above used a computed tomography (CT) along with PET, but in our study, MRI was used, which helped in better delineation of anatomy than CT, and hence we were able to use more extensive ROIs involving the deeper structures in the midbrain.
To the best of our knowledge, our study is the only known study to have used PET/MRI, extensive ROIs and also to have computed the cutoffs for the regions. The cutoffs reached in our study, for other movement disorders versus NIDDS were 1.69 for caudate nucleus, 1.69 and 1.79 for anterior putamen and posterior putamen, respectively in IPD versus NIDDS, diagnostic accuracy for posterior putamen more than anterior putamen, 1.79 and 1.75 respectively for IPD versus other movement disorders, diagnostic accuracy being highest for posterior putamen. Also in ratios, the cutoff for differentiating between NIDDS and IPD was 3.5 for PPOR and to differentiate between NIIDS and other movement disorders, cutoff of 3.02 was noted in RCOR with a diagnostic accuracy of 95.5%. The previously noted gradient in other studies was also noted in our study.
One of the other new findings in our study was a significant difference in the radiotracer uptake in the occipital lobe between IPD and other movement disorders; our study is the first known study to have concluded this.
More studies can be performed in the future to look into the causes of occipital lobe differences between the IPD and other movement disorders and to validate the cutoffs generated.
Limitations
The study’s main limitation was the skewness of the data, where the right side was more involved than the left side. Our sample size was also relatively small, considering the paucity of PET studies in NDDS disorders. This study has significant clinical implications in the differential diagnosis of the subgroup of movement disorders.
Conclusion
This study showed the regions significant in the differential diagnosis of PD and generated cutoff values for the same. Furthermore, MRI with PET was useful in vividly delineating the anatomy and ruling out other causes of PD. Hence, F18 DOPA SUV value cutoffs can be utilized to differentiate between various etiologies of movement disorders in a neuropsychiatric setup. However, more studies are required to further validate the results.
Conflicts of interest
There are no conflicts of interest.
Nil.
References
- Neuropathology of Parkinson disease. Parkinsonism Relat Disord. 2018;46(Suppl 1):S30-3.
- [Google Scholar]
- Hybrid PET-MRI applications in movement disorders. Int Rev Neurobiol. 2019;144:211-57.
- [Google Scholar]
- Nuclear imaging in Parkinson’s disease: The past, the present, and the future. J Neurol Sci. 2022;436:120220.
- [Google Scholar]
- Current status and future challenges of brain imaging with (18)F-DOPA PET for movement disorders. Hell J Nucl Med. 2016;19:33-41.
- [Google Scholar]
- Ageing and Parkinson’s disease: Substantia nigra regional selectivity. Brain. 1991;114:2283-301.
- [Google Scholar]
- Synaptic changes in Parkinson disease assessed with in vivo imaging. Ann Neurol. 2020;87:329-38.
- [Google Scholar]
- Utility of intrastriatal ratios of FDOPA to differentiate idiopathic Parkinson’s disease from atypical parkinsonian disorders. Nucl Med Commun. 2013;34:426-31.
- [Google Scholar]
- Intrastriatal gradient analyses of 18F-FDOPA PET scans for differentiation of parkinsonian disorders. Neuroimage Clin. 2020;25:102161.
- [Google Scholar]