Parkinson’s disease (PD), the second-most common neurodegenerative disease, is characterized by motor and nonmotor symptoms. PD is often misdiagnosed; inappropriate treatment due to misdiagnosis has undesired consequences, as does delayed diagno- sis. Unfortunately, most people with PD receive a diagnosis only after motor symptoms have emerged, by which time 40% to 60% of dopamine neurons have already been lost. Advances in imaging tech- niques have provided clinicians with increasingly sophisticated tools. In 2011, the US Food and Drug Administration approved ioflupane I-123 injection (DaTscanTM) for striatal dopamine transporter visu- alization using single-photon emission computed tomography (SPECT) imaging, which provides an effective tool for assessing striatal dopaminergic deficiency. Among patients with suspected parkinso- nian syndromes, of which PD is one, the diagnostic sensitivity and specificity of DaTscan SPECT imag- ing are high. In clinical studies that were part of the DaTscan new drug application, no serious drug- related adverse events reported by the 1236 partici- pants were attributed to DaTscan. The introduction of DaTscan imaging and its utility necessitate the devel- opment of clinical recommendations for appropriate use; thus, a multidisciplinary panel of experts was convened to develop clinical criteria and algorithms to help guide clinicians and managed care organiza- tions in the application of DaTscan SPECT imaging. Based on the consensus of this expert panel, appro- priate use of DaTscan SPECT imaging includes cases where: (1) PD diagnosis is uncertain; (2) tremor of uncertain etiology is present; and (3) nonmotor and/ or supportive symptoms and features associated with PD are present but the classical motor syndrome is absent or atypical.
Am J Manag Care. 2004;20:S97-S109 Parkinson’s disease (PD) is a common neurological illness associated with significant patient and payer burden. Its accurate diagnosis is challenging and misdiagnosis is common.1-3 Additionally, overdiagnosis of PD has been observed when only clinical symptoms were used to assess uncertain cases.4 Although experts in movement disorders are able to achieve greater diagnostic accuracy than their primary care or general neurology counterparts, they are still uncertain about the diagnosis in many patients. Adding to this, access to experts may be limited, thus further reducing opportunities for diagnosis.5 Imaging technology has improved considerably in recent years. The 2011 approval by the US Food and Drug Administration (FDA) of the radiopharmaceutical ioflupane I-123 injection (hereafter referred to as DaTscanTM) for striatal dopamine transporter (DaT) visualization using single-photon emission computed tomography (SPECT) offers a new biomarker that can assist in differentiating conditions associated with nigrostriatal degeneration (eg, PD, multiple system atrophy [MSA], progressive supranuclear palsy [PSP], and other parkinsonian syndromes) from those movement disorders without this degeneration (eg, essential tremor [ET]).6 This publication is based on a roundtable discussion involving a multidisciplinary panel of experts with experience in movement disorders, nuclear imaging, managed care, and patient advocacy, and provides clinical criteria and algorithms to help guide managed care organizations (MCOs) and practitioners in the appropriate use of DaTscan imaging.
Overview of Parkinson’s Disease
PD is second only to Alzheimer’s disease (AD) as a cause of neurodegenerative disease.7 More than 1 million Americans currently have PD and approximately 60,000 new cases are diagnosed each year.8 PD is characterized by a spectrum of motor and nonmotor dysfunction.1,9 The motor-related disorders of PD include resting tremor, rigidity, postural instability, and bradykinesia.9 Common nonmotor symptoms in PD include sleep disorders, olfactory dysfunction, and neuropsychiatric and autonomic symptoms.10 Key pathologic characteristics of PD include neuronal loss, particularly dopaminergic neuronal loss, appearing most prominently in the substantia nigra, as well as various protein deposits (cytoplasmic inclusions), Lewy bodies, and Lewy neurites.9,10
Diagnosis and Differential Diagnosis of Parkinson’s Disease
PD is diagnosed based on clinical information regarding symptoms and findings from a physical examination. Although tests may be used to rule out other conditions, there are no tests that specifically confirm a diagnosis of PD.11 The most commonly used diagnostic criteria are the 1992 UK PD Society (UK PDS) Brain Bank criteria (see Sidebar 1).12,13 The UK PDS clinical diagnostic criteria include a 3-step algorithm. The first step involves evaluation of symptoms that indicate possible PD. Step 2 involves consideration of factors that would rule out PD. Step 3 involves evaluation for the presence of additional supportive criteria to support a diagnosis of PD.13
Use of the “levodopa challenge test” for confirmation of a PD diagnosis remains an area of controversy. Falsepositive and false-negative responses to dopaminergic challenge may be as high as 40%.14 The previously mentioned UK PDS Brain Bank clinical diagnostic criteria count excellent (ie, 70% to 100%) response to levodopa as one of the diagnostic criteria for PD, and the American Academy of Neurology (AAN) practice parameter, published in 2006, recommends the levodopa (or apomorphine) challenge when a PD diagnosis is in doubt.12,13 However, these recommendations preceded the availability of DaT SPECT imaging. Newer guidelines, such as the 2013 European Federation of Neurological Societies and the European Section of the Movement Disorder Society (EFNS/MDSES) diagnostic recommendations, do not support the use of drug challenges for the diagnosis of de novo parkinsonian patients because there is insufficient evidence to support their role in differentiating PD and other parkinsonian syndromes.14 Taken together, particularly in light of the recent evidence-based EFNS/MDS-ES recommendations, dopamine challenge should be regarded as a far from optimal means of supporting diagnosis and ought to be seen primarily as a therapeutic rather than a diagnostic tool in PD.
The motor and nonmotor symptoms experienced by patients are not exclusive to PD. Other neurodegenerative conditions, including MSA, PSP, dementia with Lewy bodies (DLB), and corticobasal ganglionic degeneration, can also manifest with parkinsonism.9 ET, with which PD is often confused, is far more common than PD, affecting an estimated 5% of people in the United States 65 years or older.15 Conditions that are frequently mistaken for PD are listed in Table 1.9,16-21 Currently, most people with PD receive a diagnosis only after motor symptoms have appeared, by which time between 40% and 60% of dopamine neuronal markers have already been lost.10,22-24 The importance of nonmotor symptoms in early diagnosis has become better recognized in recent years. Barone et al surveyed 1072 patients with PD and found that 98.6% experienced at least 1 nonmotor symptom. Additionally, the researchers found that 3 domains of nonmotor symptoms— apathy, attention/memory, and psychiatric symptoms (eg, depression and anxiety)—were often not associated with disease duration or dopaminergic treatment, suggesting that these symptoms can start very early in the disease course and may function as early diagnostic markers.1
Misdiagnosis of PD is extremely common, particularly in the primary care setting, where PD diagnoses may be incorrect nearly half of the time.3 The diagnosis of PD may be especially challenging for primary care physicians (PCPs), as it may be difficult for nonspecialist physicians to maintain a high level of knowledge about PD because a typical PCP may see a new case of PD relatively infrequently.25 With regard to specialist diagnosis, a study of diagnostic accuracy of PD in patients with clinically uncertain parkinsonian syndromes found that movement disorder specialists fail to achieve an accurate diagnosis in about 47% of cases, while general neurologists misdiagnose as much as 75% of the time in this patient population.4,26 The earlier the patient is in the disease process, the more likely it is that the diagnosis will be mistaken.25 Table 2 provides an overview of studies on the misdiagnosis of parkinsonian syndromes and ET.3,4,18,26-29
The consequences of delayed diagnosis or misdiagnosis are serious. Over time, delayed treatment can result in untreated symptoms that may result in prolonged disability. Also, patients with PD who have been misdiagnosed may receive inappropriate therapies with consequent risk of adverse events, while possibly undergoing additional and inappropriate testing.26,30,31
Role of Imaging in Disease Detection of Nigrostriatal Degeneration
A number of imaging modalities and diagnostic agents have been studied for the assessment of patients with parkinsonism, including magnetic resonance imaging (MRI), blood flow measures (Tc-99m hexamethylpropyleneamine oxime SPECT), 18F-fluorodopa positron emission tomography, 123I-β-CIT SPECT, and DaTscan SPECT. Many of these imaging techniques/diagnostic agents are currently used in the research setting rather than in routine clinical practice. DaTscan (ioflupane I-123 injection) is approved by the FDA and the European Medicines Agency (under the brand name DaTSCAN) and is indicated for striatal DaT visualization using SPECT brain imaging to assist in the evaluation of adult patients with suspected parkinsonian syndromes. At present, MRI is a common imaging technique used in the diagnosis (or exclusion) of many conditions; however, MRI does not visualize nigrostriatal degeneration and is not specific for PD or parkinsonian syndromes.14,32
DaTscan is a radiolabeled analogue of cocaine that binds to presynaptic dopamine receptors. It was administered in a total of 1236 subjects (1176 patients and 65 healthy volunteers) in 11 clinical studies that were part of the DaTscan new drug application. Adverse events were mild and infrequent, and no serious drug-related adverse reactions were attributed to DaTscan.33 The injected substance is a radioactive material, and it is considered a scheduled, controlled substance in the United States.6 DaTscan images of normal and abnormal striatum activity are shown in Figure 1.6
DaTscan Clinical Effectiveness Studies
A phase 3 trial of DaTscan imaging involved 158 patients with a diagnosis of parkinsonism, 27 patients with ET, and 35 healthy volunteers at 6 centers where the images were read by the individual centers (“institutional on-site read”) and by an expert panel of 5 nuclear medicine physicians (“blinded read”), all assessors being blinded to patients’ clinical data. Positive percent agreement (statistical equivalent to sensitivity) for a parkinsonism diagnosis was 97% and negative percent agreement (statistical equivalent of specificity) for ET was 100% for the institutional assessors, whereas sensitivity was 95% and specificity was 93% for the expert panel. In this study, patients had established clinical diagnoses, and for that reason an expert panel for clinical diagnosis (reference standard) was not considered necessary.34
The effectiveness of DaT imaging to achieve an accurate differential diagnosis was the subject of a study by Jennings et al in which the diagnostic accuracy of movement disorder specialists (MDSs) was compared with DaT imaging in patients with suspected PD. The authors found that while MDSs were able to identify people who had PD (92% sensitivity), they also misidentified a large proportion of people who did not have PD (30% specificity). By contrast, DaT imaging allowed for exactly the same degree of sensitivity but with a 100% specificity rate.35 A similar study conducted in the European Union by Marshall et al observed diagnostic sensitivity of 93% and specificity of 46% among neurologists compared with a sensitivity of 78% and specificity of 97% using DaTscan imaging. This study used 3 blinded imaging experts and a clinical expert panel with a review of videotapes to establish the clinical diagnosis at 3 years follow-up (reference standard) after DaTscan imaging. It is important to point out that overdiagnosis of PD (as reflected by high sensitivity and low specificity of clinical diagnosis in the Jennings and Marshall studies when imaging information was not provided), and subsequent inappropriate treatment, is common in the clinical setting.4
DaTscan Clinical Utility Studies
Several studies in both the United States and Europe have demonstrated the impact of DaTscan on clinician diagnostic thinking and patient management. In a recent global randomized study (United States and Europe) involving 273 patients with clinically uncertain parkinsonian syndromes, patients were randomized to receive either DaTscan imaging (135 patients) or no imaging (138 patients; the control group) to evaluate the impact on clinical disease management. The follow-up period after DaTscan imaging was 12 months, with visits at weeks 4, 12, and 52. At 4 weeks, 45% of the DaTscan group experienced a change in diagnosis versus 9% of the control group (P <.001). At 12 weeks, 46% and 12% experienced diagnosis changes (P <.001). At 1 year, 54% of patients in the DaTscan group received a change in diagnosis compared with 23% in the control group (P <.001). In addition, a significantly greater mean change from baseline in confidence of diagnosis was observed in the DaTscan group versus the control group at all visits (P <.001). In the DaTscan group, a change in diagnosis at 1 year was significantly more common among general neurologists (75%) compared with MDSs (47%; P = .022). Compared with those in the control group, significantly more patients in the DaTscan group had at least 1 change in clinical management at 12 weeks (31% vs 50%; P = .002) and 1 year (22% vs 41%; P <.001). Examples of changes in clinical management included initiation or escalation of dopaminergic therapy in patients with abnormal scans and discontinuation of dopaminergic therapy in patients with normal scans.26
The diagnostic accuracy of DaTscan was also assessed in a multicenter, open-label follow-up study in patients from the Clinically Uncertain Parkinson Syndrome study (n = 118).30 At the 2-year follow-up (n = 85), the rate of agreement between the follow-up diagnosis and the initial clinical diagnosis at baseline before DaTscan imaging was 56% (suggesting a change in clinical diagnosis in 44% of patients).28
The impact of DaTscan on diagnosis and clinical management was also evaluated in an independent study of cases from clinical practice by Seifert et al (2013). As part of the study, physicians ordering DaTscan imaging for their patients were contacted retrospectively and asked to complete a brief survey with questions regarding diagnosis, confidence in diagnosis, and clinical management of the patient before and after DaTscan imaging. A total of 112 patients were included in the study. After DaTscan imaging, there was a change in diagnosis in 31% of patients (35 of 112). Physicians felt that the scan influenced their diagnosis in 68% of patients (76 of 112), and physicians were more confident in the diagnosis. Clinical management was affected in 58% of patients (65 of 112). This study had certain methodological limitations (ie, it was a retrospective survey), which may explain some differences from the 2 previous prospective studies.36
Guidelines for Diagnostic Use of DaTscan SPECT Imaging
The clinical value of DaTscan in patient diagnosis and management has been recognized in guidelines developed by clinicians and payer organizations. The 2013 EFNS/ MDS-ES guidelines, the most recent of the major guidelines for PD imaging, state that “DaTscan SPECT is indicated in the presence of significant diagnostic uncertainty and particularly in patients presenting atypical tremor manifestations,” and grade the level of evidence as “A” (see Sidebar 2).12,14,25,37,38 The guidelines further note that “All other SPECT imaging studies do not fulfill registration standards and [other agents] cannot be recommended for routine clinical use.”14
The US-based Society of Nuclear Medicine and Molecular Imaging (SNMMI), which has developed technical guidelines specific to DaTscan SPECT, has identified situations for which DaTscan imaging is ideally suited. These situations include visualization of striatal DaT in the evaluation of adult patients with suspected parkinsonian syndromes, early diagnosis of presynaptic parkinsonian syndromes, and differentiating ET from tremor due to presynaptic parkinsonian syndromes, which include PD, MSA, and PSP. The SNM guidelines also recommend DaTscan imaging for differentiating presynaptic parkinsonian syndromes from parkinsonism without presynaptic dopaminergic loss, as with drug-induced parkinsonism or psychogenic parkinsonism, and also for differentiating DLB from AD.37
The current AAN practice parameter for PD was published in 2006 (before approval of DaTscan in the United States) and no comprehensive update has been developed. However, at the time, DaT SPECT imaging was cited as “possibly useful,” and it was noted that insufficient evidence was available to determine whether these modalities could distinguish PD from other kinds of parkinsonism.12
In the United Kingdom, the National Institute for Health and Clinical Excellence, which, in addition to clinical recommendations, is focused on national health policy in the context of a governmental single-payer system, recommended that DaTscan imaging be “considered for people with tremor where ET cannot be clinically differentiated from parkinsonism” and “available to specialists with expertise in its use and interpretation.”38 Also from the United Kingdom, the Scottish Intercollegiate Guidelines Network stated that DaTscan SPECT “should be considered as an aid to clinical diagnosis in patients where there is uncertainty between PD and nondegenerative parkinsonism/tremor disorders.”25
Proposed Algorithms for the Appropriate Use of DaTscan Imaging in the Diagnosis of Parkinson’s Disease and ET
Although clinical guidelines broadly define the scope of indicated use of DaTscan imaging in the diagnosis of PD and ET, they offer little detail regarding its appropriate and inappropriate use and where it belongs in the diagnostic work flow.
The following information is based on discussion among members of an expert panel consisting of specialists in the areas of movement disorders, nuclear imaging, and managed care. The panelists recognized the utility of DaTscan and engaged in discussions with the goal of providing guidance regarding the appropriate use of DaTscan SPECT imaging. The suggestions offered by the expert panel, and provided here, should be regarded as general guidance, in recognition of the fact that cases will arise that fall outside this clinical framework.
Appropriate Use of DaTscan SPECT Imaging
Table 3 provides recommendations from the expert panel regarding appropriate versus inappropriate use of DaTscan imaging. In general, appropriate use includes cases where: (1) PD diagnosis is uncertain; (2) tremor of uncertain etiology is present; and (3) nonmotor and/ or supportive symptoms and features associated with PD are present but the classical motor syndrome is absent or atypical. Thus, appropriate use would include, for example, a patient with at least 1 motor symptom plus 1 nonmotor symptom such as hyposmia (impairment of sense of smell), rapid eye movement sleep behavior disorder, parasympathetic or sympathetic dysfunction, or neuropsychiatric symptoms.39 Inappropriate use includes cases where a certain diagnosis has been ascertained, it is not a PD diagnosis, and greater diagnostic clarity is needed. DaTscan imaging is currently perceived as inappropriate for disease monitoring due to a lack of regulatory validation in appropriately designed prospective trials (ie, it is not licensed for such an indication). DaTscan is currently being used in clinical trials to assess patient study eligibility (ie, whether patients have a DaT deficiency) and monitor changes in DaT density (eg, in the Parkinson Progression Marker Initiative 5-year follow-up study)40 (see Sidebar 3).
Table 4 offers criteria for a functional definition of atypical/unusual parkinsonism. The full and appropriate clinical utility of DaTscan SPECT imaging beyond its established diagnostic role remains to be elucidated. In addition to the criteria already noted for the appropriate and inappropriate use criteria in Table 3, it is important to further describe those patients for whom DaTscan SPECT may be clinically useful. A provisional list includes the following:
For patients with suspected DLB, DaTscan imaging may be appropriate, based on a case-by-case determination, particularly if a patient is experiencing autonomic disturbances, has demonstrated sensitivity to neuroleptic medications, or is receiving medications for PD. The DLB indication and differential diagnosis from AD are included in the European Union label,41 but they have not received approval from the FDA.
Figure 2 describes the diagnostic process for a patient with a parkinsonian syndrome, which includes a work-up consisting of a neurological examination with a mental status examination, a medication review (for ruling out drug-induced parkinsonism), and a review of systems (a history of cerebrovascular disease). Upon completion of these steps, it should be determined whether or not DaTscan imaging is indicated (Tables 3 and 4). For those patients deemed appropriate for DaTscan imaging, Figure 3 describes its use for detecting substantia nigra degeneration and the possible conditions associated with the presence or absence of such degeneration.
DaTscan Use by Physician Specialty
Although the panel recommends that DaTscan be ordered by a specialist with experience in diagnosing PD (see Sidebar 4), a PCP may be the first person to evaluate a patient with suspected early PD (Figure 4). Where the clinical or differential diagnosis is straightforward and confirmed, DaTscan imaging is not appropriate, and the next step will be to move toward disease management. Arriving at a certain diagnosis, however, is likely to require referral to an “appropriate expert,” who may be an expert PCP (eg, gerontologist/geriatrician), general neurologist, or MDS (movement disorder specialist). A non-expert PCP may refer patients to an MDS or general neurologist. The practitioner to whom a referral is made can then establish whether a diagnosis is straightforward, and if not, whether the other criteria indicate the appropriateness of DaTscan imaging.
Managed Care Considerations
Generally, MCOs concentrate their efforts on the most prevalent diseases affecting Americans (eg, diabetes, hypertension, AD) and focus less on rarer diseases such as PD, in large part due to limited resources and financial priorities. Thus, most MCOs do not have specific policies in place for the use of imaging agents in PD.
When assessing the utility of DaTscan imaging, it is important to note the clinical benefits that imaging with DaTscan may offer for a disease that is, in many cases, challenging to diagnose.2 DaTscan imaging has been shown to confer more accurate diagnoses, resulting in frequent diagnostic and therapeutic revisions for patients with suspected PD, and major clinical guidelines recommend DaTscan imaging.12,14,25,37,38 An earlier and more accurate diagnosis of PD may have substantial benefits. Not only does it ensure that a patient’s condition is correctly identified, but also that the patient is receiving appropriate treatment, while avoiding inappropriate and potentially harmful treatment. These factors are key in achieving improved outcomes and controlling the cost of care.
Additional research should add to the body of evidence regarding DaTscan imaging in the diagnosis of degenerative parkinsonism and nondegenerative mimics (eg, ET). Potential research topics include:
Such information would help MCOs further clarify the appropriate use of DaTscan in the diagnosis of PD and any promising new indications, and develop specific policies for its coverage.Author affiliations: Cleveland Clinic Lou Ruvo Center For Brain Health, Las Vegas, NV, and Cleveland Clinic Lerner College of Medicine of Case Western Reserve University, Cleveland, OH (JLC); Health Net, Huntington Beach, CA (MJF); Medical Affairs, GE Healthcare-Life Sciences, Princeton, NJ (IDG); GE Healthcare-Life Sciences, Princeton, NJ (CRJ); Health Economics and Outcomes Research, GE Healthcare, Chalfont St Giles, UK (MKJ); Department of Medical Imaging and Departments of Medicine and Biomedical Engineering, University of Arizona, Tucson, AZ (PHK); National Parkinson Foundation, Miami, FL (JAO); Movement Disorders Center, Neuroscience Institute, Central DuPage Hospital, Winfield, IL (MR); Movement Disorders Center, Neurological Institute, University Hospitals, and Department of Neurology, Case Western Reserve University School of Medicine, Cleveland, OH (DER); SelectHealth, Murray, UT (KLS); The Parkinson and Movement Disorder Institute, Orange Coast Memorial Medical Center, Fountain Valley, CA (DT).
Funding source: This meeting and publication were supported by GE Healthcare, Princeton, NJ.
Author disclosure: Drs Cummings and Kuo report serving as a consultant/ paid advisory board and receipt of honoraria from GE Healthcare. Dr Kuo also reports receipt of grants and lecture fees from GE Healthcare. Drs Grachev and Jarecke and Ms Johnson report employment with GE Healthcare. Dr Jarecke also reports stock ownership with GE Healthcare. Ms Oberdorf reports receipt of honoraria from GE Healthcare. Drs Fine, Rezak, Riley, Schaecher, and Truong report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this supplement.
Authorship information: Concept and design (JLC, MJF, IDG, CRJ, MKJ, PHK, MR, KLS, DT); acquisition of data (DER); analysis and interpretation of data (IDG, JAO, DT); drafting of the manuscript (JLC, MJF, CRJ, PHK, JAO, MR, KLS, DT); critical revision of the manuscript for important intellectual content (JLC, MJF, IDG, CRJ, MKJ, PHK, JAO, MR, DER, KLS, DT); provision of study materials or patients (IDG, MKJ); obtaining funding (IDG); administrative, technical, or logistic support (IDG); supervision (MKJ); participant at roundtable meeting (DER); and contribution at the roundtable for presenting information (CRJ).
Address correspondence to: Jeffrey L. Cummings, MD, ScD, Cleveland Clinic Lou Ruvo Center for Brain Health, 888 W Bonneville Ave, Las Vegas, NV 89106. E-mail: CumminJ@ccf.org.REFERENCES