Introduction

Parkinsonian syndrome (PS) is characterized by a combination of motor symptoms and signs resulting from dysfunction in the nigro-striatal dopaminergic system. It is diagnosed clinically based on the presence of akinesia, the mandatory symptom, associated with resting tremor and/or rigidity1,2. Approximately 80% of PS cases are attributed to degenerative PS, with Parkinson’s disease being the most prevalent cause3,4,5,6. Clinical diagnosis of PS may lack accuracy, particularly in cases with mild symptoms and/or in early disease stages. For this reason, molecular imaging of the dopaminergic system, such as 123I-FP-CIT (123I-N-ω-fluoropropyl-2β-carbomethoxy-3β- (4-iodophenyl) nortropane) SPECT (trade name DaTSCAN®), is increasingly incorporated into the diagnostic process.

123I-FP-CIT SPECT can detect presynaptic dopamine neuronal denervation in degenerative PS7,8, aiding diagnosis when clinical assessment is challenging. Indeed, it allows to differentiate degenerative PS, characterized by nigro-striatal dopaminergic cell loss, such as Parkinson’s disease (PD), multiple system atrophy (MSA), progressive supranuclear palsy (PSP) and corticobasal degeneration (CBD) from secondary PS (e.g., due to lesions or iatrogenic causes) or from essential tremor, a movement disorder without nigro-striatal dopaminergic cell loss9,10,11,12,13,14. In addition, its diagnostic utility extends to differentiating patients with suspected Lewy bodies dementia from those with other forms of dementia or cognitive impairment15,16. 123I-FP-CIT SPECT necessitates appropriate and justified indications since it is costly and increasingly use, involves radiation exposure and does not distinguish between the different degenerative PS themselves.

The primary objective of our study was to assess the appropriateness of the indications for 123I-FP-CIT SPECT prescriptions in a French Nuclear Medicine Unit. Secondary objectives included assessing whether the appropriateness of 123I-FP-CIT SPECT prescriptions was related to the prescriber’s expertise in movement disorders (MD) and investigating changes in 123I-FP-CIT SPECT requests over time.

Methods

Study population

In this retrospective study, we included 723 consecutive patients referred for 123I-FP-CIT SPECT to the Nuclear Medicine Unit of Avicenne University Hospital between February 2009 and May 2019. This latter was the date when the study was initiated, and data collection began. Various types of physicians referred patients to the Nuclear Medicine Unit: neurologists at the Avicenne University Hospital; private neurologists; neurologists with mixed activity (hospital and private); general practitioners; geriatricians; and others (internists, psychiatrists…). During the study period, the neuro-oncology team, previously managing the neurology department at Avicenne University Hospital, was replaced by MD specialists with over 5 years of experience in the field from September 1, 2017. No patients were excluded. Data on demographics, clinical, brain imaging, and 123I-FP-CIT SPECT results were collected from medical records when available. The study was conducted in accordance with the Declaration of Helsinki and its later amendments or comparable ethical standards, and approved by the institutional ethics committee (CLEA-2019-75). Data collection has also been approved by the Commission Nationale de l’Informatique et des Libertés (CNIL) in accordance with French regulations (loi n° 2018–493 du 20 juin 2018), and patients have been informed that their medical records may be used for research purposes after data anonymization. Consequently, no written consent was required for this retrospective study. This study was registered on Clinicaltrial.gov (NCT05698784).

Classification of prescription relevance

Regarding the relevance of indications, an expert committee comprising three MD specialists (BD, MM, AL) and two nuclear medicine physicians (MS, GP) defined three groups: « inappropriate », « uncertain » and « relevant ». In the “inappropriate” group, we classified (i) prescriptions lacking information to justify the request (no information), (ii) 123I-FP-CIT SPECT requested to confirm a clinically evident parkinsonian syndrome (PS), and (iii) those prescribed for isolated and unequivocal cognitive-behavioral disorders, without mention of PS or a differential diagnosis. In the “uncertain” group, clinical data were provided by prescribers, but insufficient or unclear written data made it challenging to determine whether the indication was inappropriate or relevant. In the “relevant” group, i.e. with an appropriate indication, prescriptions involved differential diagnosis between degenerative PS and secondary PS (e.g. post-neuroleptics), or other ambiguous disorders (essential tremor; dementia with associated motor symptoms like rigidity, tremor, myoclonus), clinically uncertain PS (such as isolated asymmetric resting tremor without akinesia or lead-pipe rigidity) and PD with an unusual course.

The choice of this classification was based primarily on the indications given in the guidelines for performing 123I-FP-CIT SPECT. Indeed, 123I-FP-CIT SPECT is indicated for the detection of loss of functional dopaminergic neuron terminals in the striatum in patients with clinically questionable parkinsonian syndrome, to aid in the differential diagnosis between essential tremor and degenerative parkinsonian syndromes, and to aid in the differential diagnosis between probable Lewy body dementia and Alzheimer’s disease17,18,19. As in our present study, some authors have underlined that besides the licensed indication, 123I-FP-CIT SPECT might be useful to help diagnose other conditions such as drug-induced parkinsonism, functional parkinsonism20,21, or other rarer degenerative parkinsonian conditions22. In our study, we therefore decided to also include these unlicensed indications in the « relevant » group, aiming at helping practitioners in prescribing 123I-FP-CIT SPECT appropriately in the various situations they may encounter.

Statistical analysis

Data were expressed as mean (standard deviation) for numerical variables and as count (percentage) for categorical variables. To compare demographic, clinical and paraclinical data between the three groups of prescription relevance (i.e. inappropriate, uncertain and relevant), one-way analysis of variance (ANOVA) was used for numerical variables, while Fisher’s exact test was employed for categorical variables. To identify which groups differed from each other, post hoc comparisons were conducted using the Tukey’s test for numerical variables and pairwise Fisher’s exact test followed by the Benjamini–Hochberg correction to account for multiple testing for categorical variables.

For all analyses, the significance level was set at p < 0.05. Statistical analyses were performed using R software (version 4.3.1, R Core Team).

Results

Clinical and paraclinical features of patients

This study included a total of 723 consecutive patients who underwent a 123I-FP-CIT SPECT between February 2009 and May 2019 at Avicenne University Hospital.

The indications of 123I-FP-CIT SPECT were considered as inappropriate in 271 cases (37.5%), uncertain in 178 patients (24.6%) and relevant in 274 cases (37.9%) (Table 1).

Table 1 Characteristics of patients having benefited of 123I-FP-CIT SPECT.
Full size table

Most of the patients were men (394 patients; 54.5%). The median age was 69.9 ± 11.4 years. Sex and age did not differ significantly between the three groups (p = 0.800 and p = 0.134 respectively).

Several items of information were missing from medical records: disease duration (n = 433, 59.9%), medical history (n = 217, 30.0%), treatments (n = 229, 31.7%), brain imaging results (n = 342, 47.3%) and final diagnoses (n = 317, 43.8%). Patients with and without missing data in each kind of data were compared in Table S1. Missing data were associated with physicians’ specialties, with hospital neurologists, especially MD specialists, providing more available information in medical records.

The mean disease duration, available in 291 patients (40.1%), was significantly longer in the group with relevant indication than in the uncertain group (4.9 ± 7.5 vs 2.4 ± 2.5 years; p = 0.011, Table 1).

Sixty-eight patients (33.8%) in the “relevant” group had been treated with neuroleptics, a much higher proportion than in the other two groups (9.9% in the “inappropriate” group and 12.4% in the “uncertain” group, p < 0.001). Consistently, a psychiatric history was more frequent in the group with relevant indication compared to the two other groups (45.8% vs 27.8% in the “inappropriate” group vs 30.3% in the “uncertain” group, p = 0.001). Dopaminergic therapy was more frequent in the group with inappropriate indication (21.4% vs 6.6% in the uncertain indication vs 8.3% in the relevant indication, p = 0.001). Of the 39 prescriptions with inappropriate indications in patients on dopaminergic therapy, 35 (89.7%) were requested despite clinically evident PS.

Brain imaging report (MRI or CT scan) was available in 381 patients and was considered normal in 17.1% of cases. The presence of cortico-subcortical atrophy was more frequent in the group with inappropriate indication compared to the group with uncertain indication (67.3% vs 50.0%, p = 0.023).

123I-FP-CIT SPECT prescriptions and prescribers

The main indication of 123I-FP-CIT SPECT was to confirm clinically evident PS in 220 patients, accounting for 30.4% of 123I-FP-CIT SPECT requests (Table 2). Moreover, the 123I-FP-CIT SPECT results were more frequently abnormal in the group with inappropriate indications (77.5% vs 56.2% for uncertain indications and 50.7% for relevant indications, p < 0.001). Indeed, out of the 220 patients with clinically evident PS, 172 (78.2%) showed dopaminergic denervation. More than half of patients with abnormal results demonstrated bilateral asymmetric dopaminergic denervation.

Table 2 Detailed distribution of 123I-FP-CIT SPECT indications.
Full size table

In the “relevant” group, the two main indications for 123I-FP-CIT SPECT prescriptions were to clarify clinical doubt about PS (35.0%) and to discriminate between iatrogenic and degenerative PS (27.0%) (Table 2).

A final diagnosis was established in 406 patients (56.2%), with the most frequent diagnoses being PD and atypical PS, found in 124 (30.7%) and 92 (22.8%) patients, respectively, without significant differences between the three relevance groups (Table 1). Final diagnoses of essential tremor were less frequent in the “inappropriate” group compared to the other two (1.3% vs 9.9% in the “uncertain” group vs 11.1% in the “appropriate” group, p = 0.001). The proportion of patients with a final diagnosis of iatrogenic PS was higher in the “appropriate” group than in the other two groups (23.3% vs 1.9% in the “inappropriate” group vs 3.7% in the “uncertain” group, p = 0.001).

Most requests were made by hospital non-MD specialists (n = 231, 32.0%) and by neurologists with a mixed private and hospital practice (n = 232, 32.1%); 73 requests (10.1%) were made by hospital MD specialists; 63 (8.7%) by private neurologists; 62 (8.6%) by geriatricians; 47 (6.5%) by general practitioners and 14 (1.9%) by others physicians (internists, psychiatrists, etc.) (Fig. 1). Prescriptions made by hospital MD specialists were significantly more often relevant (67.1%) compared to those made by other physicians, while 48.5% of prescriptions from hospital non-MD specialists were inappropriate. On the 11 inappropriate prescriptions made by the hospital MD specialists, 10 were due to the presence of a clinically evident PS and 1 was due to isolated cognitive-behavioral disorders.

Fig. 1
figure 1

Distribution of 123I-FP-CIT SPECT prescription relevance according to physicians’ specialty. Relative percentages of prescription relevance were calculated for each physician specialty. Comparisons of relevance distribution among physician specialties were conducting using pairwise Fisher’s exact tests, corrected using Benjamini–Hochberg method. HN: hospital neurologists, MD: movement disorders

Full size image

Changes in the percentages of relevant 123I-FP-CIT SPECT prescriptions over time in the neurology department of Avicenne university hospital

An important increase in the percentages of relevant 123I-FP-CIT SPECT prescriptions was observed when the neuro-oncology team at Avicenne University Hospital was replaced by MD specialists: from 37.5% in the last semester before the replacement of the neuro-oncology team to 81.0% in the last semester of observation (Fig. 2). In the meantime, the mean number of 123I-FP-CIT SPECT requests remained stable or even increased very slightly, around 3 per semester per neurologist in the department. During the study period, the number of patients consulting for MD increased drastically after the arrival of the new neurology team. There were fewer than 30 consultations per year before 2016, 45 in 2017, 94 in 2018 and 46 from January to May 2019.

Fig. 2
figure 2

Evolution of the percentages of relevant 123I-FP-CIT SPECT prescriptions and the neurologist mean requests at Avicenne University Hospital from February 2009 to May 2019. Notes. The vertical dotted line indicates the semester when the neuro-oncology team at Avicenne University Hospital was replaced by the new neurology team, which included MD specialists. The percentage of relevant indications was calculated as the number of relevant indications among the total number of requests made by the team in the semester. The neurologist mean requests was calculated as the total number of requests made by the team in the semester divided by the number of neurologists in the team during that semester. The number (n) in parentheses next to the semester on the x-axis represents the number of requests made by the team in the semester.

Full size image

Discussion

To our knowledge, our study describes the largest cohort of patients who underwent 123I-FP-CIT SPECT23. We found that 37.5% of 123I-FP-CIT SPECT indications were inappropriate, with 81.2% of them attributed to a clinically evident parkinsonian syndrome (PS). Among these cases, 78.2% exhibited dopaminergic denervation suggesting that 123I-FP-CIT SPECT was unnecessary given the clinically evident PS.

Hospital neurologists and those with a mixed practice were responsible for the majority of 123I-FP-CIT SPECT requests (74.1%). Of these, hospital MD specialists more frequently made relevant prescriptions (67.1%), while hospital non-MD specialists had 48.5% of requests deemed inappropriate. Finally, the replacement of the neuro-oncology team at Avicenne University Hospital by a team of MD specialists increased the percentage of relevant SPECT 123I-FP-CIT prescriptions from 37.5% to 81.0%. The average demand from neurologists remained stable in the meantime, while the number of patients consulting for MD increased accordingly. This suggests that MD specialists are more likely to make a clinical diagnosis.

In our study, we have arbitrarily chosen to include requests without indication in the “inappropriate” group. If we had removed these cases (n = 33), the results would have changed little with only a significantly longer disease duration in the “relevant” group compared to the “inappropriate” group, and a loss of the significance between “inappropriate” and “uncertain” groups for atrophy percentages. The 123I-FP-CIT SPECT was considered an inappropriate indication in 37.5% of cases, which represents a large proportion of prescriptions (N = 271). Similar findings were reported by Thiriez et al., who analyzed 516 consecutive patients’ files and observed a comparable rate of inappropriate indications for 123I-FP-CIT SPECT (37%)20. The most common indication in our study was for confirming a clinically evident PS (30.4%). Gajos et al. investigated 67 patients referred for 123I-FP-CIT SPECT by non-MD specialists to the Nuclear Medicine Unit of the University Hospital of Lodz in Poland24. All these patients were clinically examined by a MD specialist. The diagnosis was clinically established in 46.3% of cases, rendering the use of 123I-FP-CIT SPECT unnecessary. Moreover, 123I-FP-CIT SPECT was requested by general neurologists in 16 patients to diagnose a degenerative PS (23.9% of cases), despite evident clinical elements being present for diagnosis by MD specialists24. Another study assessed cerebral 99mTc-TRODAT-1 SPECT imaging of the dopamine transporter in 119 patients with clinically suspected PD or unclear parkinsonism21. There were differences from our results and those of Thiriez et al., such as collection of non-consecutive data and data from selected patient, which did not allow a direct data comparison. Moreover, clinically suspected PD was considered an appropriate indication, despite being beyond the licensed indication of dopaminergic transporter (DAT) imaging21. Physicians should keep in mind that abnormal 123I-FP-CIT SPECT (or other DAT imaging) alone cannot be used to establish the diagnosis of PD, as multiple other degenerative parkinsonian conditions should be considered in case of dopaminergic denervation. Reduced 123I-FP-CIT striatal uptake is expected in all degenerative PS with a loss of nigrostriatal dopamine neurons, including PD, MSA, PSP, CBD, and thus, 123I-FP-CIT SPECT does not formally differentiate among these pathologies20,21,23,25,26,27. Clinical assessment and judgement remain essential for PD and other degenerative PS diagnoses based on the MDS-UPDRS international criteria1,28,29,30,31. We would like to point out that we were surprised to find a normal 123I-FP-CIT SPECT in 21.8% of patients with « clinically evident PS ». However, it has recently been reported that some patients with neuropathologically confirmed degenerative parkinsonian syndromes could have normal DAT scan32. Authors suggested that this might be secondary to a non-presynaptic dopaminergic parkinsonism. Another less likely explanation could be that clinicians made a clinical misdiagnosis in patients who actually had slowness and/or rigidity not related to a parkinsonian syndrome, but rather a physiological slowness linked to age, arthrosis or oppositional resistance, for instance.

It should also be noted that in the relevant group, one-third of the patients have been treated with neuroleptics. Patients on neuroleptics often present a challenge, as it is difficult, and sometimes impossible, to clinically distinguish iatrogenic parkinsonism from degenerative PS, with patients sometimes presenting with both conditions. Therefore, 123I-FP-CIT SPECT is interesting because its normality indicates that the parkinsonism could be entirely due to neuroleptics, via a blockade of the post-synaptic dopaminergic (D2) receptors. On the other hand, if the 123I-FP-CIT SPECT is abnormal, it suggests at least degeneration of the nigro-striatal dopaminergic pathway, although the involvement of neuroleptics in the parkinsonian condition cannot be ruled out.

It should be noted that the average duration of illness was significantly longer in the group with a relevant indication than in the group with an uncertain indication. There are several explanations for this. Firstly, although it seems counter-intuitive, it could suggest that the prescription of the 123I-FP-CIT SPECT was more deliberate in the “relevant” indication, whereas in the “uncertain” indication, it might reflect less experience and therefore a more automatic prescription of 123I-FP-CIT SPECT. Secondly, it could be due to the fact that MD specialists usually test the levodopa responsiveness of parkinsonian symptoms before resorting to diagnostic imaging. Thirdly, it could be attributed to a slower disease progression, with unusual symptoms emerging gradually and later. Fourthly, in our study, 18% of the relevant indications were aimed at discriminating suspected essential tremor from parkinsonian tremor, with a mean duration of 8.92 ± 11.56 years. Similarly, half of the uncertain indications for which we have an evolution time are suspicions of isolated parkinsonism without a differential diagnosis, with a short disease duration of 1.97 ± 2.03 years. This may have contributed in part to the significant longer duration observed in the relevant group. Arjona et al. reported a similar finding, with the most delayed but most appropriate requests coming from MD specialists (median of 4.8 years)21.

Our results also demonstrated that the prescriber’s specialty affects the relevance of the 123I-FP-CIT SPECT request. Indeed, prescriptions from MD specialists were twice as relevant as those from non-MD specialists (67.1% vs 33.3%, respectively). This aligns with findings from Thiriez et al., where the indications for prescriptions were more appropriate for MD specialists (79%) than for other physicians (57%)20. At Avicenne University Hospital, we also observed that following the arrival of the new team of neurologists, including MD specialists, the percentages of relevant 123I-FP-CIT SPECT prescriptions increased from 37.5% to 81.0%, while the number of patients consulting for MD tripled. This could be attributed to the greater expertise and experience of these MD specialists in the clinical diagnosis of PS, especially PD.

It is also noteworthy that brain imaging abnormalities were mainly observed in patients with inappropriate indications for 123I-FP-CIT SPECT. These kinds of morphologic anomalies are especially retrieved in patients with cognitive disorders, underlining the idea that differentiating akinesia from reduced motor activity secondary to cognitive disorders is sometimes difficult for non-MD specialists but also MD specialists.

The main limitations of our study are related to its monocentric and retrospective design, leading to a large amount of missing data. However, to our knowledge, this study represents the largest investigation into the appropriateness of prescribing 123I-FP-CIT SPECT, drawing data from hundreds of patients over a 10-year period, thereby reducing selection bias. Another strength of our study is the assessment of prescriber distribution by specialty and practice type. To strengthen and validate our findings, a prospective multicenter study would be beneficial.

To conclude, it is now widely acknowledged that functional imaging of the dopaminergic system is a useful tool in several but specific clinical scenarios. However, it is worth noting that this test is both costly and involves exposure to radiation, underscoring the importance of its judicious use. Despite its limitations, our study clearly demonstrated that indications for 123I-FP-CIT SPECT are not always appropriate. Furthermore, our findings revealed that prescriptions for 123I-FP-CIT SPECT by MD specialists were more frequently relevant. This suggests that seeking the expertise of MD specialists should precede the realization of this exploration.