Cardiac ¹⁸F‐Dopamine PET Distinguishes PD with Orthostatic Hypotension from Parkinsonian MSA

ABSTRACT

Background

Parkinson's disease with orthostatic hypotension (PD + OH) can be difficult to distinguish clinically from the parkinsonian form of multiple system atrophy (MSA‐P). Previous studies examined cardiac sympathetic neuroimaging to differentiate PD from MSA but without focusing specifically on PD + OH versus MSA‐P, which often is the relevant differential diagnostic issue.

Objective

To investigate the utility of cardiac sympathetic neuroimaging by ¹⁸F‐dopamine positron emission tomographic (PET) scanning for separating PD + OH from MSA‐P.

Methods

Cardiac ¹⁸F‐dopamine PET data were analyzed from 50 PD + OH and 68 MSA‐P patients evaluated at the NIH Clinical Center from 1990 to 2020. Noradrenergic deficiency was defined by interventricular septal ¹⁸F‐dopamine‐derived radioactivity <6000 nCi‐kg/cc‐mCi in the 5′ frame with mid‐point 8′ after initiation of 3′ tracer injection.

Results

¹⁸F‐Dopamine PET separated the PD + OH from the MSA‐P group with a sensitivity of 92% and specificity of 96%.

Conclusion

Cardiac ¹⁸F‐dopamine PET scanning efficiently distinguishes PD + OH from MSA‐P.

Parkinson's disease (PD) and multiple system atrophy (MSA) are in the disease family termed synucleinopathies, because PD is characterized by deposition of the protein alpha‐synuclein in Lewy bodies (LBs) in brainstem neurons, and MSA features alpha‐synuclein deposition in glial cytoplasmic inclusions (GCIs). ¹ MSA occurs in two forms, parkinsonian (MSA‐P) and cerebellar (MSA‐C).

Orthostatic hypotension (OH) is a characteristic but not a universal feature of MSA. ² , ³ OH also occurs in a substantial minority of PD patients, even early in the disease, ⁴ , ⁵ despite United Kingdom PD Society Brain Bank criteria that list early severe autonomic involvement as exclusionary for PD. ⁶ Overlap of PD + OH and MSA‐P in terms of the occurrence of OH makes the clinical distinction of the two conditions challenging. Such a distinction is important because of the different genetic predispositions, likely pathophysiological mechanisms, disease progression, prognoses, and responses to treatments. The clinical diagnosis of MSA‐P often is not straightforward. In a clinicopathologic study, of autopsy‐confirmed MSA patients, the median sensitivity for the initial clinical diagnosis was only 56%. ⁷ A prominent response to dopaminergic therapy is not exclusive to PD and may be evident in up to half of MSA patients. ⁸

Such findings have incited efforts to develop clinical laboratory means to distinguish MSA‐P from PD. In this regard, numerous studies have agreed on the utility value of ¹²³I‐metaiodobenzylguanidine (¹²³I‐MIBG) scanning, ⁹ , ¹⁰ although usually with overlaps between the groups. Other imaging biomarkers to distinguish PD from MSA include the hot cross bun sign and hyperintense putaminal rim on structural magnetic resonance imaging; however, these signs have low sensitivity and specificity. ¹¹ , ¹²

In the few studies that have stratified PD groups in terms of the presence or absence of OH in the PD groups, direct comparison of PD + OH and MSA‐P has not been the focus. ¹³ , ¹⁴ , ¹⁵ Cardiovascular autonomic function tests performed in patients with MSA‐P and PD + OH have limited value in discriminating between the two diseases because both feature prominent baroreflex‐sympathoneural failure. ¹⁶

Our group has consistently noted differences between PD + OH and MSA based on cardiac sympathetic neuroimaging by ¹⁸F‐dopamine positron emission tomographic (PET) scanning, but without formal calculations of diagnostic sensitivity and specificity in separating PD + OH from MSA‐P.

Here we report the results of ¹⁸F‐dopamine PET scanning in relatively large groups of PD + OH and MSA‐P patients. Data from a group with PD and no OH (PD No OH) are included for comparison purposes.

Methods

Data were reviewed from all individuals who participated in research protocols of the Autonomic Medicine Section in the Division of Intramural Research, National Institute of Neurological Disorders and Stroke (NINDS), National Institutes of Health (NIH), between 1990 and 2020 and underwent ¹⁸F‐dopamine PET scanning done as part of the evaluation. The subjects participated in protocols approved by the Institutional Review Board of the NINDS after having given written informed consent. In this study, we analyzed data from patients referred for PD (with or without OH) or MSA.

The clinical categorization was based on the history and neurological examination and was supported by the results of specialized tests as described below.

MSA‐P

Subjects were classified as having MSA based on the consensus criteria. ³ In this study, MSA‐P was considered if there were symptoms and signs of parkinsonism at the time of evaluation at the NIH, regardless of concurrent cerebellar ataxia. MSA‐P patients underwent cardiac ¹⁸F‐dopamine PET scanning and had data for interventricular septal concentrations of myocardial ¹⁸F‐dopamine‐derived radioactivity.

PD

Subjects were classified as having PD based on the UK PD Society Brain Bank criteria, ⁶ with the exception that early severe OH was not considered to be an exclusion criterion for PD. In most patients, the clinical impression of parkinsonism was supplemented by evidence of nigrostriatal dopaminergic deficiency, based on low putamen/occipital cortex ratios of ¹⁸F‐DOPA‐derived radioactivity. ¹⁷

Neurogenic OH

Cardiac autonomic function testing to identify neurogenic OH included continuous blood pressure recording with the performance of the Valsalva maneuver and head‐up tilt table testing. OH was defined by a sustained fall in systolic blood pressure of ≥20 mm Hg between lying supine and 5 minutes of upright posture. A neurogenic cause of OH was diagnosed based on abnormal blood pressure responses to the Valsalva maneuver. ¹⁸

¹⁸F‐Dopamine PET Scanning

PET scans were acquired on a GE Advance Tomograph (GE Healthcare) prior to January 2016 and on a Siemens PET/CT scanner after the GE Advance scanner was retired in January 2016. Phantoms were used to calibrate the scanners so that the attenuation‐corrected values were consistent across scanners. This helps minimize the variability in the data obtained from different scanners. The ¹⁸F‐DA PET scanning was done as described previously. ¹⁹

In subjects who underwent follow‐up PET scans, we analyzed the data from the first scan.

Data Analysis and Statistics

PET images were analyzed with Pixelwise modeling computer software (PMOD 2.61; PMOD Technologies LLC, Zurich, Switzerland).

Mean ± SEM concentrations of ¹⁸F‐dopamine–derived radioactivity in the cardiac interventricular septum were compared among the PD + OH, MSA‐P, and PD No OH groups. Radioactivity <6000 nCi‐kg/cc‐mCi at 8 minutes (8ʼ Radioactivity) was used as the cutoff value to identify cardiac noradrenergic deficiency. ²⁰

Analyses of variance (ANOVA) for groups with unequal variances and Dunnett's post hoc test were performed to assess differences in 8ʼ Radioactivity among the three groups (GraphPad Prism v. 9.0, GraphPad Software LLC, San Diego, CA).

The sensitivity and specificity of cardiac ¹⁸F‐DA PET scanning to differentiate the different groups were calculated. Receiver operating characteristic (ROC) curves were constructed to assess the efficiency of ¹⁸F‐DA PET scanning in separating PD + OH versus MSA‐P, and PD versus MSA‐P. Areas under the ROC curves were calculated.

Results

There were 50 PD + OH, 68 MSA‐P, and 60 PD No OH patients. PD + OH patients were older than MSA‐P and PD No OH patients (Table 1). The mean duration of disease from onset of motor symptoms to PET scanning was shorter in the MSA‐P than either of the 2 PD groups (Table 1).

TABLE 1.

Demographic data and interventricular septal myocardial ¹⁸F‐dopamine‐ (¹⁸F‐DA‐) derived radioactivity (nCi‐kg/cc‐mCi) in Parkinson's disease with orthostatic hypotension (PD + OH), PD without OH (PD No OH), and the parkinsonian variant of multiple system atrophy (MSA‐P)

	PD + OH (N = 50)	PD No OH (N = 60)	MSA‐P (N = 68)
Age, years	67.4 ± 1.1	60.4 ± 2	61.3 ± 1.1
Women, N (%)	15 (30)	17 (28.3)	26 (38.2)
Disease duration, years	8.0 ± 1.1	7.0 ± 1.0	3.8 ± 0.3
18F‐DA‐derived radioactivity,	3272 ± 261	5691 ± 449	9903 ± 282
Cardiac noradrenergic deficiency, N (%)	46 (92)	33 (55)	3 (4.4)

Ages, disease durations, and ¹⁸F‐DA‐derived radioactivity are expressed as means ± standard error of the mean.

Abbreviations: MSA, multiple system atrophy; OH, orthostatic hypotension; PD, Parkinson's disease.

Values for 8ʼ Radioactivity varied substantially as a function of subject group (F = 95, P < 0.0001). The two PD groups had lower mean 8ʼ Radioactivity than did the MSA‐P group (P < 0.0001 for both PD groups, Table 1).

Of the 50 patients with PD + OH, 46 (92%) had low 8ʼ Radioactivity, whereas only 3 out of the 68 (4.4%) patients with MSA‐P had this finding. In the PD No OH group, 33 of 61 (54%) patients had low 8' Radioactivity. The sensitivity for distinguishing PD + OH from MSA‐P was therefore 92%, and the specificity was 96%. In the PD No OH group the sensitivity was 54% and specificity 96%. Analysis of the PD group as a whole yielded 72% sensitivity and 96% specificity.

The receiver operating characteristic (ROC) curve efficiently distinguished the PD + OH group from the MSA‐P group (Fig. 1). The area under the ROC curve was 0.97 (P < 0.0001).

FIG. 1.

Individual data (left panel) and receiver operating characteristic (ROC) curve (right panel) for cardiac ¹⁸F‐dopamine‐derived radioactivity in patients with PD + OH or MSA‐P. Note that across all patients ¹⁸F‐dopamine‐derived radioactivity is distributed in two distinct populations, one with normal radioactivity (>6000 nCi‐kg/cc‐mCi) and one with low radioactivity. Patients with post‐mortem data are denoted by black circles. A patient with MSA‐P and low ¹⁸F‐dopamine‐derived radioactivity is indicated by (*). The ROC curve demonstrates high sensitivity and specificity for distinguishing PD + OH from MSA‐P (area 0.97, P < 0.001).

Six patients had autopsy confirmation of their diagnosis (black circles in Fig. 1). Three had PD + OH and 3 had MSA‐P. As reported previously, ²¹ one of the MSA‐P patients (denoted by (*) in Fig. 1) had low ¹⁸F‐DA‐derived radioactivity and intra‐axonal alpha‐synuclein deposition in arrector pili muscles in a skin biopsy during life; the same patient had markedly decreased myocardial norepinephrine content post‐mortem. ²²

Discussion

We report that ¹⁸F‐dopamine PET scanning is highly sensitive and specific for differentiating PD + OH from MSA‐P. In this retrospective study of relatively large patient groups, sensitivity was 92% and specificity was 96%, and the area under the ROC curve was 0.97.

The utility of cardiac ¹²³I‐MIBG SPECT to distinguish synucleinopathies with OH has been reported in only two small studies. ¹⁴ , ¹⁵ One study included 21 PD patients with “autonomic failure” and 7 MSA patients and demonstrated reduced cardiac uptake of ¹²³I‐ MIBG in all the PD patients, whereas uptake in the MSA patients did not differ from controls. ¹⁵ The other study included 10 MSA patients, 10 PD patients with “dysautonomia,” and 8 PD patients without dysautonomia. ¹⁴ The authors identified dysautonomia based on a series of tests as described by Ewing and colleagues. ²³ With a cut‐off value of 1.3 for the delayed heart/mediastinum ratio, 8 of 10 PD patients with dysautonomia had evidence of cardiac noradrenergic deficiency, whereas no patients with MSA or PD without dysautonomia had an abnormal study. ¹⁴ Our findings generally agree with these reports.

Other promising biomarkers include the analysis of the deposition of alpha‐synuclein in the cerebrospinal fluid or skin. ²⁴ , ²⁵ In our study, one MSA‐P patient had both low in vivo ¹⁸F‐dopamine‐derived radioactivity and low post‐mortem myocardial norepinephrine content. The same patient, however, had increased intra‐neuronal alpha‐synuclein deposition in a skin biopsy obtained during life, raising the possibility of a “hybrid” condition involving both glial cytoplasmic inclusions in the brain and Lewy body pathology in the sympathetic nervous system. We therefore propose that in patients with probable MSA‐P clinically who have abnormal cardiac sympathetic neuroimaging, differential diagnostic accuracy may be improved by analyzing skin biopsies for intra‐neuronal alpha‐synuclein deposition. Search of the literature has not revealed any cases of MSA in which increased intra‐neuronal alpha‐synuclein has been reported, although alpha‐synuclein deposition may occur in somatic nerve fibers. ²⁶

The separation of PD + OH from MSA‐P, while impressive, was imperfect. The finding that a few PD + OH patients had interventricular septal myocardial ¹⁸F‐DA‐derived radioactivity within the normal range might be explained by a spatio‐temporal sequence of loss of sympathetic innervation, with early involvement of the inferolateral portion and late involvement of the anteroseptal base of the heart. ²⁷ Regarding MSA‐P, a minority of patients might have “hybrid” pathology involving alpha‐synuclein deposition both in catecholaminergic neurons and in glial cells. This seems to have been the case in one of our MSA‐P patients as mentioned above, who had increased innervation‐adjusted alpha‐synuclein in a skin biopsy while alive ²¹ and had post‐mortem glial cytoplasmic inclusions and cardiac noradrenergic deficiency. ²²

Being retrospective in nature, our study has some limitations. There could have been observer or self‐selection biases in terms of patients tested; however, we included all consecutive patients seen at the NIH with the available physiological and neuroimaging data, and personnel who analyzed the ¹⁸F‐dopamine PET scanning were blinded as to the clinical diagnosis until the imaging analyses had been completed.

Post‐mortem neuropathological confirmation of the diagnosis was obtained in only six patients. The PD + OH group was older, with longer disease duration compared to the MSA‐P group; however, it has been previously reported that OH in PD is independently associated with age, ⁵ and we previously reported that ¹⁸F‐dopamine‐derived radioactivity decreases over time by a median of 4% per year in PD, whereas it does not decrease in MSA. ²⁸ Therefore, we would expect the results to be similar or more robust with longer disease duration.

The main limitation is that ¹⁸F‐dopamine PET scanning has been available only at the NIH Clinical Center. This has been the situation for a quarter century. We hope that recognition of the high sensitivity and specificity of this testing modality to distinguish PD + OH from MSA‐P will spur efforts to make ¹⁸F‐dopamine PET scanning available at other institutions. We would be happy to have investigators at other centers join in the IND covering ¹⁸F‐dopamine PET scanning, with a view toward a multi‐center collaborative study.

To conclude, ¹⁸F‐dopamine PET scanning is highly effective in differentiating PD + OH from MSA‐P. This may be considered in cases where accurate clinical distinction of PD + OH from MSA‐P is challenging.

Author Roles

(1) Research project: A. Conception, B. Organization, C. Execution; (2) Statistical Analysis: A. Design, B. Execution, C. Review and Critique; (3) Manuscript Preparation: A. Writing of the first draft, B. Review and Critique.

A.L.: 1A, 1B, 1C, 2A, 2B, 3A

G.L.: 1A, 1B, 1C, 2A, 2B, 2C, 3B

D.G.: 1A, 1B, 1C, 2C, 3B

Disclosures

Ethical Compliance Statement

Institutional Review Board of the NINDS had approved the protocols related to this study. Al the subjects were recruited after having given written informed consent. We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this work is consistent with those guidelines.

Funding Sources and Conflict of Interest

The research reported here was supported (in part) by the Division of Intramural Research, NINDS, NIH, Bethesda, MD, USA. The authors have no conflicts of interest.

Financial Disclosures for previous 12 months

None of the authors have any financial disclosures to make.