Classification of Scans Without Evidence of Dopamine Deficit (SWEDD) According to the Olfactory Function

INTRODUCTION

Scans without evidence of dopaminergic deficit (SWEDD) refers to patients with clinical Parkinson’s disease (PD) have normal dopamine transporter (DAT) imaging [1]. SWEDD has been considered to be a heterogeneous disorder. Many subjects with SWEDD are confirmed to have other underlying conditions including dystonia, essential tremor, and other secondary parkinsonism [2–4]. Due to the fact that both SWEDD and PD patients have similar clinical manifestations and that DAT scan is usually not available in the majority of clinical situations, studies have been sought to distinguish these two disorders clinically [5]. For example, gait and handwriting have been shown to be more impaired in PD than in SWEDD [6, 7]. However, these motor symptoms are influenced by patients’ conditions and disease severity, thus limit the ability to compare PD to SWEDD [6, 7]. Non-motor symptoms are now widely used distinctive clinical features to differentiate various parkinsonism and reflect the stages of PD [8–10]. Furthermore, distribution of non-motor symptoms vary between SWEDD and PD [10, 11]. While most non-motor symptoms are dominant in PD, some symptoms are more dominant in SWEDD patients, suggesting that the underlying pathology must be distinct and heterogeneous in SWEDD [9, 10]. Olfactory dysfunction is representative non-motor biomarkers of neurodegeneration that can be present as a premotor signs up to 10 years before parkinsonian motor symptoms, and occur in 80% to 100% of symptomatic PD [12]. Generally, anosmia is known to be more prominent in PD than in SWEDD [10–12]. However, not all patients with SWEDD have normal range in olfactory function. In two studies, 19.4% and 23.8% of patients with SWEDD are found to have olfactory dysfunction exceeding control patients, suggesting that SWEDD is a heterogeneous group of disorder distinct from PD and control with different underlying pathophysiology [11, 12]. These findings raise questions of whether SWEDD with olfactory dysfunction and those with normal olfaction have different non-motor symptom profiles. In addition, subgroup of SWEDD could be possibly made according to olfactory performance that reflects the presence of neurodegeneration. Therefore, the objective of this study was to determine the differences of olfactory dysfunction and non-motor symptoms in patients with SWEDD, PD, and healthy controls. Subgroup classification of SWEDD was made according to olfactory performance and non-motor symptoms distribution of each group were compared.

METHODS

This cross-sectional study involved 94 subjects who visited the movement disorder clinic of a university affiliated hospital for a two-year duration. Fluorinated N-3-fluoropropyl-2-α-carboxymethoxy-3-α-(4-iodophenyl) nortropane (FP-CIT) positron emission tomography (PET) scans were performed for all subjects except controls. PD was diagnosed based on the UK Parkinson’s Disease Society Brain Bank clinical diagnostic criteria [13]. Patients who had parkinsonism with normal scans were categorized as SWEDD based on previous studies [14, 15]. FP-CIT PET scans were reviewed and described by visual inspection of an experienced nuclear medicine physicians. Twenty-six age-matched healthy elderly subjects without any neurological or psychiatric history were included as controls. Clinical information including age, sex, disease duration, and education duration was obtained. Data from complete physical and neurological examinations and laboratory tests were also obtained. Magnetic resonance imaging of the brain was performed for all patients with PD and SWEDD. Patients who had a previous history of stroke or other neurological or psychiatric disorders, who were taking medication that may cause parkinsonism, and those who had secondary causes of dementia were excluded. We performed detailed neuropsychological tests to the all patients by experienced neuropsychologist. Patients with PD-dementia according to the clinical diagnostic criteria of the Movement Disorder Society were also excluded [16]. None of the patients had a history of anti-parkinsonian or anti-dementia medication use. No patient in this study had ever taken anti-psychotics, antidepressants, or anxiolytics prior to the evaluation of cognitive status and non-motor symptoms. All subjects were evaluated for their motor symptoms using the Unified Parkinson’s Disease Rating Scale (UPDRS) Part III and the modified Hoehn and Yahr (H&Y) stage score. Local ethics committee approved this study.

Olfaction testing was performed using the Korean version of the Sniffin’s stick test II (KVSS II) [17]. KVSS II includes three components of olfactory function: olfactory threshold, odor discrimination, and odor identification. This test was performed using pens with 16-different liquid odorants familiar to Koreans. Scores ranged from 0 to 16 for each component. The sum of the three test scores comprised the total score of KVSS II (ranging from 0 to 48). Total scores of 0–20, 20–27, and 27–48 indicated anosmia, hyposmia, and normosmia, respectively. In this study, the SWEDD group was divided into SWEDD-normal olfaction (SWEDD-N, KVSS II total score 28–48) and SWEDD-olfactory dysfunction (SWEDD-D, KVSS II total score 0–27). Non-motor symptoms were evaluated using the Korean version of non-motor symptoms scale for PD (K-NMSS). This has been validated for Koreans [18]. K-NMSS contains the following nine domains: cardiovascular function, sleep/fatigue, mood/cognition, perceptual problems/hallucinations, attention/memory, gastrointestinal tract, urinary function, sexual function, and miscellaneous other non-motor symptoms (pain, taste or smell, weight change, and excessive sweating). Symptom severity was rated on a scale of 0 to 3. Frequency was rated on a scale of 1 to 4. A composite score (defined as the product of frequency and severity) ranging from 0 to 12 was calculated. Total K-NMSS score ranged from 0 to 360.

Statistical analyses were performed using SPSS version 15.0 (SPSS Inc., Chicago, IL, USA). One-way analysis of variance (ANOVA) with Bonferroni post-hoc testing was used to compare the means among groups. Pearson’s χ2 test was used to compare the frequencies of categorical variables. For group comparisons of olfactory function and non-motor symptoms, analysis of covariance (ANCOVA) with Bonferroni post-hoc test controlling for age, mini-mental state examination (MMSE), UPDRS motor score, and disease duration was conducted. Statistical significance was considered when p-value was less than 0.05.

RESULTS

A total of 40 SWEDD, 28 PD, and 26 age-matched healthy controls participated in this study. The mean ages of patients in the SWEDD, PD, and control groups (SWEDD: 67.0±9.4 years; PD: 64.4±10.1 years; controls: 64.1±7.7 years) were not statistically (p > 0.05) different. The distributions of sex and education duration were not statistically (p > 0.05) different among the three groups either. Disease durations in PD and SWEDD groups were similar to each other. The mean MMSE scores in the controls were 28.1±2.2, which were significantly (p = 0.006) higher than those in the other two groups (SWEDD, 25.3±4.0; PD, 25.8±3.4). Motor symptoms of patients in the PD group (UPDRS part III 35.0±9.8, H&Y stage 2.0±0.5) were significantly (p < 0.001) more severe than those in the SWEDD group (UPDRS part III 16.9±9.9, H&Y stage score 1.7±0.5) or the control group (UPDRS part III 2.0±3.8, H&Y stage score 0.0±0.0, Table 1). Overall, olfaction and non-motor scores of SWEDD group were placed between PD and control. The KVSS II total scores of the PD group, SWEDD group, and controls were 14.9±6.9, 23.7±7.9, and 30.0±2.8 (p < 0.001), respectively. The total K-NMSS scores of PD subjects (48.1±29.7) was significantly (p < 0.001) higher than those of SWEDD patients (20.2±16.7) or controls (16.4±14.0). Of the K-NMSS subdomain, patients with PD had more deficits in sleep/fatigue, mood/cognition, attention/memory, gastrointestinal function, urinary function, pain, and taste or smell than patients with SWEDD. The difference in K-NMSS scores was not significant between the SWEDD and control groups (Table 2). On subgroup analysis, SWEDD-D group was older than the SWEDD-N group. However, there was no statistical significance in age between the two groups (SWEDD-D, 70.6±7.8; SWEDD-N, 64.5±9.3; p = 0.08). The SWEDD-D group had significantly (p < 0.001) lower MMSE scores than the SWEDD-N group (SWEDD-D: 23.8±4.1; SWEDD-N: 27.6±2.6), although neither of them had dementia. Disease duration and H&Y stage score were similar to each other among SWEDD-N, SWEDD-D, and PD groups (Table 3). The KVSS II total score of the SWEDD-D group was 18.5±5.5, which was significantly (p < 0.001) lower than that (31.4±3.1) in SWEDD-N group and that (30.0±2.8) in controls. KVSS II subscores of olfactory threshold test, odor discrimination test, and odor identification test in the SWEDD-N and controls were also significantly (p < 0.01) higher than those in PD and SWEDD-D groups (Figure 1). The K-NMSS score in the PD group (48.1±29.7) was the highest, followed by that in SWEDD-D (28.2±16.5) and in SWEDD-N (8.3±7.3) (p < 0.001) (Table 4). According to K-NMSS subscores, PD group had more problems than SWEDD-N group in sleep/fatigue, mood/cognition, attention/memory, gastrointestinal, urinary, pain, and taste or smell subdomains. In the subdomains of pain and taste or smell, the PD group had more severe symptoms than the SWEDD-D group. Significantly different scores in mood/cognition and attention/memory subdomains were observed between the SWEDD-N and SWEDD-D groups (Table 4).

DISCUSSION

In this study, olfactory dysfunction was more severe in SWEDD compared to that in normal controls but less severe than that in patients with PD. The frequency and severity of non-motor symptoms were increased in PD compared to those in SWEDD or controls. Anosmia and hyposmia is a widely accepted biomarker that reflects underlying neurodegeneration. Olfactory dysfunction can be present several years prior to motor symptoms. It has been found in over 80% of patients with early PD [19, 20]. Olfaction test has been used to differentiate between SWEDD and PD patients. In which, the University of Pennsylvania Smell Test score in SWEDD patients was reportedly to be higher than in PD but similar to that of controls [11, 12]. We found that the KVSS II score was also lower in PD patients than that in the SWEDD group. However, the KVSS II score of the SWEDD group was lower than that of controls, which was different from previous studies, [11, 12] KVSS II scores in the SWEDD group had wide ranges, indicating that SWEDD might comprise a group of heterogeneous disorders. On subgroup analysis of SWEDD, the KVSS II score of SWEDD-D was similar to that in PD patients but higher than that in the SWEDD-N group and the controls. The SWEDD-D group had more non-motor feature complains than the SWEDD-N and controls as well. These findings suggest the possibilities that patients with SWEDD-D might have different underlying pathophysiology compared to patients with SWEDD-N. SWEDD-D might have a close pathological relationship with PD, which differs from that of SWEDD-N or controls. This is in accordance with the findings of Silveira-Moriyama et al. indicating that SWEDD is a group distinct from PD but not free of pathologic changes [11]. Further, we asserted that those diverse clinical groups could be divided by olfactory function, a sensitive marker of neurodegeneration. As olfactory dysfunction often precedes motor symptoms by more than a decade [21] and Lewy body pathology begins in the olfactory bulb and lower medulla, [22] there is a possibility that patients with SWEDD-D are prone to developing PD or other neurodegenerative disorders in the future. Therefore, SWEDD may not be considered as a simple radiologic termination, rather a heterogeneous disorder with various underlying etiologies. Our results revealed that the severity and frequency of non-motor symptoms in SWEDD patients were lower than those of PD patients but similar to those of controls. The total scores of K-NMSS of PD were the highest in the subject group. Of subdomain scores, sleep/fatigue, mood/cognition, attention/memory, gastrointestinal, urinary, pain, and taste or smell problems were more severe in PD patients. Non-motor symptoms were more widely affected in our study than those in a previous study [9]. However, non-motor symptoms in SWEDD and control groups were not different in our study. Non-motor symptoms are more frequent in SWEDD patients compared to those of controls in a recent study [10]. Although patients with SWEDD and PD in our study were in drug-naïve state (not taking anti-parkinsonian medications), the different parameters used to evaluate non-motor symptoms and the relatively small sample size might explain such discrepant results. Subgroup analysis of SWEDD revealed that non-motor symptoms of SWEDD-D were more frequent and severe than those of SWEDD-N but less severe than those of PD. The scores for mood/cognition and attention/memory subdomains of K-NMSS were increased in the SWEDD-D group than those in the SWEDD-N group. In terms of sleep/fatigue, mood/cognition, attention/memory, gastrointestinal, urinary, pain, and taste or smell subdomains, the SWEDD-N group had significantly milder and less symptoms than the PD group, while the SWEDD-D group and PD group had similar patterns of non-motor symptoms except in pain and taste or smell subdomains. These findings suggest that olfaction in SWEDD may predict the severity and frequency of non-motor symptoms, similar degree of non-motor severity with PD in SWEDD-D group. The strength of our study was that newly diagnosed drug-naïve parkinsonism patients were included. As the evaluation was performed without any anti-parkinsonian medications, dopaminergic effects on non-motor symptoms could be excluded. In addition, age and sex-matched healthy controls were included, which allowed us to compare the effect of aging on non-motor symptoms. Lastly, all participants underwent otorhinolaryngological evaluation before KVSS II test to exclude structural and other causes of anosmia. Through this, the reliability of the test result was raised. To our knowledge, this is the first study that SWEDD patients were classified into subgroup according to olfactory function test. In this study, several different clinical symptoms between the two subgroups of SWEDD compared to PD and controls were observed. However, this study has several limitations that need to be considered. First, all patients were evaluated with FP-CIT PET/CT scans and categorized into either the SWEDD group or the PD group. Although most patients with SWEDD did not progress to PD, a minority of SWEDD patients may exhibit reduced dopamine uptake over long-term follow-up [5, 23–25]. Second, as the number of subjects in each group was relatively small and the design of the study was cross-sectional, the present study has limitations in generalizing differences of each group and elucidating long-term prognosis. Third, in our study, 40% of SWEDD patients had abnormal scores of KVSS II. A recent study revealed that 23 SWEDD patients of 41 patients with two years of follow up had similar olfactory function to that of controls and had better olfaction than that of patients with PD [25]. The cross-sectional design of the study might also cause the differences. Fourth, although we excluded subjects with dementia, mildly impaired cognition might be associated with olfactory dysfunction at the time of examination [26]. For this, we performed detailed neurocognitive test by experienced clinical psychologists within several days to olfactory testing to exclude cognition and mood problems. Fifth, even though critical clinical diagnostic criteria were used, the lack of genetic study and pathological confirmation might have impacted the diagnosis of SWEDD. Lastly, even an experienced professional nuclear medicine physicians performed visual analyses of FP-CIT PET scans, visual analyses and automated quantitative analyses showed differences in 10% of discrepant cases [27]. Therefore, a small number of patients in the present study might not be properly categorized. In conclusion, olfactory dysfunction in SWEDD was significantly different from that in PD or control group. Olfaction might be one of the valuable clinical biomarkers for differentiating SWEDD from PD. In addition, heterogeneous SWEDD was divided into two subgroup according to olfactory function test. This might help us elucidate the distinct pathophysiology between subgroups. Non-motor symptoms were less common in SWEDD than those in PD. Mood and memory problems were more frequent in SWEDD with olfactory dysfunction than those in SWEDD with normal olfaction. Long-term longitudinal follow-up studies are warranted in the future to reveal the associations between olfactory function and non-motor symptoms in patients with SWEDD. The relationship between initial olfactory function and the prognosis of SWEDD also requires future investigation.

AUTHORS’ ROLE

Research project: Do-Young Kwon, Ji Ho Choi.

CONFLICT OF INTEREST

None.

Financial Disclosures of all authors (for the preceding 12 months):

DY Kwon, YS Oh, JH Choi: nothing to declare regarding financial disclosures.