I-123 DaTscan SPECT Brain Imaging in Parkinsonian Syndromes:

Utility of the Putamen-to-Caudate Ratio

Manuela Matesan , Santhosh Gaddikeri, Katelan Longfellow, Robert Miyaoka, Saeed Elojeimy, Shana Elman,
Shu-Ching Hu, Satoshi Minoshima, David Lewis
From the Department of Radiology, University of Washington, Seattle, WA (MM, RM, SE, DL); Department of Diagnostic Radiology and Nuclear Medicine at Rush University
Medical Center, Chicago, IL (SG); CHI Franciscan Health, Tacoma, WA (KL); University of New Mexico, Albuquerque, NM (SE); Department of Neurology, University of
Washington, Seattle, WA (S-CH); and Department of Radiology, University of Utah, Salt Lake City, UT (SM).

ABSTRACT
BACKGROUND AND PURPOSE: Computer-based analysis of Dopamine transporter imaging (DaTscan) can aid in image
interpretation. In this study, we examined the distribution of putamen-to-caudate ratios (PCRs) obtained by using a clinically
available semiquantification method.
METHODS: Medical records of 32 patients (M:16) with a diagnosis of Parkinson’s disease (PD) (n = 22) or Parkinson’s
plus syndromes (PPS) (n = 10) based on clinical follow-up, were retrospectively reviewed. Single photon emission tomography
(SPECT) imaging was performed 4 hours after intravenous injection of 3–5 mCi [I-123]-ioflupane. Semiquantitative evaluation
using DaTQUANT software was performed. Utility of PCR with a cutoff of .7 and .8 in the diagnosis of nigrostriatal degeneration
was assessed. PD and PPS groups based on clinical assessment and caudate-to-background ratio (CBR) were assessed separately.
RESULTS: Minimum PCR for both hemispheres was .74 ± .09 (Mean ± SD, range: .58-.89), with 65.63% patients (21/32)
having PCR > .7. Mean PCR in mild nigrostriatal degeneration was .77 ± .08 (range: .62-.89) and in advanced nigrostriatal
degeneration was .73 ± .09 (range: .58-.89). Mean PCR in PD group was .73 ± .09 (range: .58-.89) and in PPS group was .75 ±
.10 (range: .61-.88).
CONCLUSIONS: Although PCR can intrinsically be a useful indication of disease, this ratio obtained in our analysis by using
one of the clinically available automatic semiquantitative methods has large variability and might not be a reliable numeric
marker in interpretation of [I-123]ioflupane studies. This may be due to difficulty in separating caudate from putamen on SPECT
images, as well as the nonuniform decreased Ioflupane uptake in both putamen and caudate.

Keywords: I-123-Ioflupane, DaTscan, quantification.

Acceptance: Received March 19, 2018, and in revised form May 14, 2018. Accepted for publication May 21, 2018.
Correspondence: Address correspondence to Manuela Matesan MD, PhD, Department of Radiology, University of Washington, 1959 NE Pacific Street,
Box 357115, Seattle 98195-0001. E-mail: mmatesan@u.washington.edu.
Acknowledgment and disclosures: There are no known conflicts of interest associated with this publication and there has been no financial support
for this work.
J Neuroimaging 2018;00:1-6.
DOI: 10.1111/jon.12530

Introduction (VOI), and more advanced automated systems using VOI or

123
I- Ioflupane single photon emission tomography (SPECT)
imaging is used to evaluate the integrity of the nigrostriatal
pathway primarily through binding to the presynaptic plasma
membrane dopamine transporter, though the radiotracer can
also bind to the serotonergic raphe nuclei.1 By binding to the
dopamine transporter, 123 I- Ioflupane imaging can provide a
visual display of the density of the dopamine neurons in the
substantia nigra, and can help differentiate neurodegenerative
causes of parkinsonism in which dopamine neurons are de-
creased such as Parkinson’s disease (PD) and Parkinson’s plus
syndromes (PPS) from disorders that have intact dopamine
neurons such as essential tremor, vascular parkinsonism, drug-
induced parkinsonism, and psychogenic Parkinsonism.2–4
Commonly 123 I-Ioflupane SPECT study interpretation is
made qualitatively by visual assessment, which is sufficient for
diagnosis of nigrostriatal system degeneration in most of the
cases.5 As an aid to visual interpretation, various semiquantita-
tive methods have been proposed. These methods can be clas-
sified into four categories: classic manual delineation of regions
of interest (ROI), manual delineation of volumes of interest
voxel-based mathematic systems.6–9
Semiquantitative methods can generate several numeric
parameters such as striatum-to-background ratio, caudate-
to-background ratio (CBR) or putamen-to-background ratio,
putamen-to-caudate ratio (PCR), asymmetry between left and
right putamen and caudate, and caudate-putamen index.9,10
The ratio between caudate and putamen has been suggested
to be particularly valuable since it is background-, age- and
camera-independent (cross-camera calibration not needed).10
In this study, we assessed the distribution of PCR in a group
of patients with Parkinson’s or Parkinson plus disease and the
capacity to predict the presence of nigrostriatal degeneration
based on these ratio values.
Materials and Methods
Subjects
This is a retrospective study that included 32 patients with an
average age of 67 ± 10.2 years, 16 male and 16 female. Based
on the evaluation by a movement disorder specialist, there

Copyright ◦ 2018 by the American Society of Neuroimaging

C 1
were 22 patients with a diagnosis of Parkinsons’s disease and
10 patients with PPS. The PPS group included 5 patients with
multisystem atrophy, 3 patients with progressive supranuclear
palsy, 1 patient with corticobasal degeneration, and 1 patient
with nonspecified PPS. Only 1 patient had autopsy performed,
which confirmed the clinical diagnosis of progressive supranu-
clear palsy. The study was approved by the Institutional Review
Board.
Scanning
Patients received between 111 and 185 MBq (3-5 mCi) [I-123]-
ioflupane and SPECT/CT images were acquired at approxi-
mately 4 hours postinjection using a dual-head gamma camera
(BrightView XCT Philips) with a low-energy high-resolution
collimator. Two to three drops of Lugol’s solution were given
orally 1 hour before the tracer injection. The patient’s head was
positioned in a head holder. A total of 128 views of 30 seconds
each over 360 degrees were acquired on a 128 × 128 matrix.
The field of view included the entire brain, and the rotational
radius used was approximately 13 cm. Data reconstruction was
performed using filtered backprojection with Butterworth filter
.55, power of 8, and no attenuation correction was applied.
Data Analysis
All subjects included in this study had a positive DaTscan by
visual interpretation.
Semiquantitative analysis was performed using
DaTQUANT software (GE Healthcare, Little Chalfont,
England). DaTQUANT uses a fully automated registration
of the patient SPECT scan to a template followed by inter-
rogation of uptake in the striatal regions using a volume of
interest (VOI). The template is based on the large European
multicenter database of healthy controls for the [123 I]-FP-CIT
SPECT (ENC-DAT trial).11
In this study, we examined the distribution of PCRs ob-
tained by using a clinically available semiquantification method
(DaTQUANT software). We also assessed the utility of PCR ra-
tio when .7 and respective .8 are used as cutoff values in the
diagnosis of nigrostriatal degeneration. The normal PCR range
for health control group was based upon the latest update of
DaTQUANT normal database (ie, .91 ± .06, right; .90 ± .07,
left) that is also used in daily clinical interpretation. Analysis
was separately made by using two different ways to classify the
patients: one based on the clinical diagnosis (PD and PPS) and
another one based on the CBR values (mild nigrostriatal degen-
eration [CBR > 2] and severe nigrostriatal system degeneration
[CBR ࣘ 2]).
Results
For the 32 patients included in the study, the values obtained
from [I-123]-ioflupane scans quantification using DaTQUANT
are presented in Table 1, Figures 1(A) and (B), and Figures 2(A)
and (B).
PCR for both hemispheres ranged from .58 to .95. However,
because even only one site of abnormal (low) putamen uptake
(left or right) satisfies the criteria for a positive study for nigros-
triatal degeneration, only the lowest value for each patient was
included in the analysis. The range of PCR for the lowest value
was .58-.89 with a mean of .74 ± .09. A total of 21 patients out
of 32 (65.63%) had lowest PCR (left or right) above .7. In the
2 Journal of Neuroimaging Vol 00 No 0 xxxx 2018
group with mild nigrostriatal degeneration (CBR more than 2;
n = 8 patients), there were 7 patients (87.5%) with PCR > .7 and
in the group with advanced nigrostriatal degeneration (CBR
less than 2; n = 24), there were 14 patients (58.33%) with PCR
> .7. In PD group (mean CBR 1.62 ± .88; n = 22), there were
15 patients (68.18%) with PCR more than >.7. In Parkinson’s
plus disease (mean CBR: 1.5 ± .75; n = 10), there were 6
patients (60%) with PCR more than .7. A separate analysis
with a cutoff value for PCR higher than .8 is also included in
Table 1.
Two illustrative examples, one for a patient with PD and
one with PPS, are presented in Figures 3 and 4, respectively.
Discussion
Several parameters that analyze striatal uptake can be obtained
by applying quantification software to [I-123]-ioflupane images.
Among these, caudate-to-putamen ratio (CPR) was shown to
be particularly useful in the interpretation since it is not only
age-independent but also camera-independent10,12,13 and is less
affected by the reconstruction methods. CPR was shown to
be a relevant measure to discriminate early PD from control
subjects.14,15
CPR values obtained on non-attenuation correction (AC)
images from healthy controls in the European Normal Control
Database of DaTscan (ENC-DaT) showed a normal range of
1.13 ± .13 (mean ± SD) for the right and 1.09 ± .13 for the
left.10 This corresponds to a PCR cutoff for normal of >.72
and .74, respectively (when the mean CPR plus 2SD is used
for the normal range).11 Alternatively, Brass Datscan (Hermes
Medical Solutions, Stockholm, Sweden) analysis uses a refer-
ence number for mean CPR of less than 1.29 (corresponding
to a PCR mean of .78).10 A normal value of PCR higher than
.7 has been previously suggested as a normal reference value;16
however, more recently data from the PPMI trial (Parkinson
Progression Marker Initiative) suggest a slightly higher normal
value of more than .9 ± .06 (for 2 SD, normal reference of .78)
on the right and more than .9 ± .07 (for 2 SD, normal reference
of .76) on the left.
Our study assessed the utility of PCR values in a group
of 32 patients with [I-123]-ioflupane study read as positive for
nigrostriatal system degeneration by visual interpretation and
validated by diagnosis of a movement disorder specialist. There
were 21 out of 32 patients (65.63%) with PCR higher than .7;
when .8 was used as cutoff for normal PCR, there were 10 out
of 32 (31%) with PCR higher than .8.
There are a couple of possible explanations for these results.
Prior studies showed that the preferential decrease of posterior
putamen uptake might not be found in patients with progressive
supranuclear palsy or corticobasal degeneration and decrease
uptake in the caudate might also lead to PCR values previously
defined as “normal range.”17 Indeed, as seen in Table 1, the
results did show that a smaller percentage of the patients in
PD group has PCRs higher than .8 (22.7%) compared to PPS
group of patients (50%). In addition, our study included a high
number of patients with CBR lower than 2 (24 patients) who
have probably more advanced nigrostriatal system degenera-
tion and loss of preferential decreased in putamen compared
to caudate. Other contributing factor is the limited spatial reso-
lution on SPECT images, which makes the separation between
the caudate and putamen difficult. However, Soderlund et al
Table 1. Putamen-to-Caudate Ratio (PCR) Values Obtained for the 32 Patients Included in the Study. Only the Lowest PCR Value for 1 Patient
(Either Left or Right Side) Was Included in the Analysis. The Values Were Obtained Using the Semiquantification Method Provided by
DaTQUANT Software

PCR Range; Percentage of Patients Percentage of Patients with

Patients No. of Patients Average ± SD with PCR more than .7 PCR more than .8

Entire population 32 .58-.89; .74 ± .09 65.63% (21 patients) 31% (10 patients)
Parkinson’s disease 22 .58-.89; .73 ± .09 68.185% (15 patients) 22.7% (5 out of 22 patients)
Parkinson’s plus syndromes 10 .61-.88; .75 ± .1 60% (6 patients) 50% (5 out of 10 patients)
Patients with mild nigrostriatal 8 .62-.89; .77 ± .08 87.5% (7 patients) 37.5% (3 out of 8 patients)
degeneration CBR>2
Patients with severe nigrostriatal 24 .58-.89, .73 ± .09 58.33% (14 patients) 29.1% (7 out of 24 patients)
degeneration CBR ࣘ2

CBR = caudate-to-background ratio; SD = standard deviation.

Fig 1. (A) Scattergram of putamen-to-caudate ratios for the group of patients with high caudate uptake (caudate-to-background ratio more
than 2) versus low caudate uptake (caudate-to-background value less than 2) and (B) for the Parkinson’s patients versus Parkinson’s plus
patients.

have shown that ratio methods using SPECT data may be less
affected by issues related to limited spatial resolution and corre-
spond to improved interobserver variability.10 Nevertheless, it
is clear that full width half maximum (FWHM) spatial measures
of low or no activity zones in SPECT using filtered backprojec-
tion may not be optimal for separation of caudate and putamen
regions, particularly if there are variances in between-subject
anatomy and size.
We hypothesized that PCR values would have decreased
utility in patients with CBR lower than 2, who probably have
more advanced nigrostriatal system degeneration and loss of
preferential decrease in putamen compared to caudate. Sub-
group analysis comparing patients with CBR more or less than
2 (high and low caudate uptake groups) showed that a majority
of patients (58.3%) with CBR < 2 did demonstrate a PCR > .7,
supporting our expectation that PCR would perform poorly in

Matesan et al: DaTscan: Putamen-to-Caudate Ratio 3

Fig 2. (A) Distribution of PCR values obtained for patients with caudate to background values more (8 patients) and less than 2 (14 patients)
and (B) distribution of PCR values for patients with Parkinson’s disease (22 patients) and Parkinson’s plus syndromes (10 patients).

Fig 3. A 64-year-old patient with Parkinson’s plus syndrome (multisystem atrophy-cerebellar type with ataxia and autonomic dysfunction
diagnosed by a movement disorder specialist) with significantly abnormal DaTscan by visual interpretation but high values of putamen-
to-caudate ratio. Oblique transverse [I-123]-ioflupane images without (A) and with (B) quantification display of the region of interest show
decrease uptake worse in the bilateral putamen (worse in the posterior putamen) and also right caudate. (C) Axial noncontrast CT image
showed generalized volume loss within the posterior fossa (pons, middle cerebellar peduncles, and cerebellar hemispheres), in a distribution
suggestive of multisystem atrophy-cerebellar subtype. Chronic posterior circulation ischemia was excluded by brain MRI/MR angiography.
Putamen-to-caudate ratio values are high (.94 right and .88 on the left)—bottom row.

4 Journal of Neuroimaging Vol 00 No 0 xxxx 2018

Fig 4. A 72-year-old male with Parkinson disease, abnormal [I-123]-ioflupane scan by visual interpretation and high putamen-to-caudate ratio.
Oblique transverse [I-123]-ioflupane images without (A) and with (B) quantification show severely decreased uptake in the bilateral putamen
and mild decreased uptake in bilateral caudate (worse on the right). Quantification showed decreased overall caudate-to-background ratio;
however, the putamen-to-caudate ratio values are higher than .8 (.81 both left and right).

the group of patients with advanced nigrostriatal degeneration.
However, an even higher percentage of patients with CBR > 2
(87.5%) had a PCR > .7, suggesting that PCR has poor perfor-
mance irrespective of disease severity.
Limitations of the Study
Our study is limited by the small sample size. In addition, while
all patients had a clinical diagnosis of PD or PPS based on
clinical assessment by a movement disorder specialist, only 1
patient had pathologic confirmation.18
Although our analysis split the patients with CBR more or
less than 2 (high and low caudate uptake groups), there were
mixed results in previously published studies regarding correla-
tion between disease severity and the values of [I-123]-ioflupane
uptake, probably depending on the quantification methods.14,19
Only one semiquantification software (DaTQUANT) was
used in our study, and further analysis using other methods
(like BRASS or SBRquant) might bring more information re-
garding this parameter. Even if there is a gradient of decreased
uptake between the putamen and caudate, SPECT resolution
is limited, making it difficult to accurately delineate the border
between the caudate and putamen, which would also affect the
quantification.
It is known that the reconstruction methods have an im-
pact in quantification;12,13 however, since PCR is a ratio, it is
less influenced by the reconstruction methods (our processing
methods used filter backprojection and ENC-DAT data for in-
stance used OSEM method of processing). On the other hand,
our images were reconstructed without attenuation or scatter
correction. This can lead to a spatially variant bias in the im-
ages that can impact the PCR ratio. Thus, while using a ratio
helps to mitigate differences between imaging systems and re-
construction methods, it is not a complete solution.
Thus, although PCR can intrinsically be a useful indication
of disease, this ratio obtained by using the automatic semiquan-
titative assessment methods (in particular investigated here—
DaTQUANT) has large variability and might not be a reliable
numeric marker in interpretation of [I-123]ioflupane studies.
This is likely due to small ROI size, associated difficulty in ac-
curately separating caudate from putamen using SPECT scans,
as well as the nonuniform decreased [I-123]-ioflupane uptake in
the putamen. In addition, in the advanced nigrostriatal degener-
ation, there might be a nonuniform decreased [I-123]-ioflupane
uptake in both putamen and caudate that can further contribute
to the large variability of the results.
References
1. Qamhawi Z, Towey D, Shah B, et al. Clinical correlates of
raphe serotonergic dysfunction in early Parkinson’s disease. Brain
2015;138:2964-73.
2. Meyer PT, Hellwig S. Update on SPECT and PET in parkinson-
ism - part 1: imaging for differential diagnosis. Curr Opin Neurol
2014;27:390-7.
3. Zijlmans JC, Daniel SE, Hughes AJ, et al. Clinicopathological in-
vestigation of vascular parkinsonism, including clinical criteria for
diagnosis. Mov Disord 2004;19:630-40.
4. Brooks DJ. Molecular imaging of dopamine transporters. Ageing
Res Rev 2016;30:114-21.
5. Djang DS, Janssen MJ, Bohnen N, et al. SNM practice guideline
for dopamine transporter imaging with 123I-ioflupane SPECT 1.0.
J Nucl Med 2012;53:154-63.
6. Linke R, Gostomzyk J, Hahn K, et al. [123I]IPT binding to the
presynaptic dopamine transporter: variation of intra- and interob-
server data evaluation in parkinsonian patients and controls. Eur J
Nucl Med 2000;27:1809-12.
7. Matesan MC, Cross DJ, Lewis DH, et al. Differential alterations of
dopamine transporter in the striatum and midbrain in patients with
parkinsonian syndrome. Clin Nucl Med 2015;40:191-4.

Matesan et al: DaTscan: Putamen-to-Caudate Ratio 5

 8. Badiavas K, Molyvda E, Iakovou I, et al. SPECT imaging evalu-
ation in movement disorders: far beyond visual assessment. Eur J
Nucl Med Mol Imaging 2011;38:764-73.
9. Yokoyama K, Imabayashi E, Sumida K, et al. Computed-
tomography-guided anatomic standardization for quantitative as-
sessment of dopamine transporter SPECT. Eur J Nucl Med Mol
Imaging 2017;44:366-72.
10. Soderlund TA, Dickson JC, Prvulovich E, et al. Value of semiquan-
titative analysis for clinical reporting of 123I-2-beta-carbomethoxy-
3beta-(4-iodophenyl)-N-(3-fluoropropyl)nortropane SPECT stud-
ies. J Nucl Med 2013;54:714-22.
11. Varrone A, Dickson JC, Tossici-Bolt L, et al. European multicen-
tre database of healthy controls for [123I]FP-CIT SPECT (ENC-
DAT): age-related effects, gender differences and evaluation of dif-
ferent methods of analysis. Eur J Nucl Med Mol Imaging 2013;40:
213-27.
12. Dickson JC, Tossici-Bolt L, Sera T, et al. The impact of reconstruc-
tion method on the quantification of DaTSCAN images. Eur J Nucl
Med Mol Imaging 2010;37:23-35.
13. Zaknun JJ, Schucktanz H, Aichner F. Impact of instrumen-
tation on DaTSCAN imaging: how feasible is the concept of
6 Journal of Neuroimaging Vol 00 No 0 xxxx 2018
cross-systems correction factor? Q J Nucl Med Mol Imaging
2007;51:194-203.
14. Tissingh G, Booij J, Bergmans P, et al. Iodine-123-N-omega-
fluoropropyl-2beta-carbomethoxy-3beta-(4-iod ophenyl)tropane
SPECT in healthy controls and early-stage, drug-naive Parkinson’s
disease. J Nucl Med 1998;39:1143-8.
15. Seibyl JP, Marek KL, Quinlan D, et al. Decreased single-photon
emission computed tomographic [123I]beta-CIT striatal uptake
correlates with symptom severity in Parkinson’s disease. Ann Neu-
rol 1995;38:589-98.
16. Moncayo V, Barron B, Halkar R. Spectrum of findings on 123I
Ioflupane (DatScan) Imaging: normal, abnormal, artifacts and in-
terpretative Pitfalls. Presented at RSNA, 2014, Chicago.
17. Brooks DJ. Diagnosis and management of atypical parkinsonian
syndromes. J Neurol Neurosurg Psychiatry 2002;72(Suppl 1):I10-
6.
18. Matesan MC, Elojeimy S, Minoshima S. 123I-FP-CIT and 99mTc-
HMPAO in pathologically confirmed progressive supranuclear
palsy. Clin Nucl Med 2016;41:e514-6.
19. Tinaz S, Chow C, Kuo PH, et al. Semiquantitative analysis of
dopamine transporter scans in patients with Parkinson disease. Clin
Nucl Med 2018;43:e1-7.