Journal of Neurology

, Volume 260, Issue 1, pp 131–137

Prevalence of non-motor symptoms in young-onset versus late-onset Parkinson’s disease

Authors

  • Vladana Špica
    • Institute of Neurology CCS, School of MedicineUniversity of Belgrade
  • Tatjana Pekmezović
    • Institute of Epidemiology, School of MedicineUniversity of Belgrade
  • Marina Svetel
    • Institute of Neurology CCS, School of MedicineUniversity of Belgrade
    • Institute of Neurology CCS, School of MedicineUniversity of Belgrade
Original Communication

DOI: 10.1007/s00415-012-6600-9

Cite this article as:
Špica, V., Pekmezović, T., Svetel, M. et al. J Neurol (2013) 260: 131. doi:10.1007/s00415-012-6600-9

Abstract

Non-motor symptoms (NMS) of Parkinson’s disease (PD) have only recently been increasingly recognized for their impact on a patient’s quality of life. In this study, we applied the validated, comprehensive self-completed NMS questionnaire for PD (NMS Quest) to 101 patients with young-onset PD (onset between 21 and 45 years, YOPD) and 107 patients with late-onset PD (onset of PD ≥ 55 years, LOPD). The mean total NMS (NMSQ-T) was 11.9 ± 6.0 (range: 0 to of a maximum of 26) in LOPD and 7.7 ± 5.8 (range: 0 to of a maximum of 26) in YOPD (p < 0.05). Compared to YOPD, dribbling of saliva, loss of taste/smell, nocturia, forgetfulness, loss of interest, hallucinations, lack of concentration, anxiety, change in libido and difficulty in sexual activities, were significantly more prevalent in LOPD. The only NMS more prevalent in YOPD were restless legs and sweating, although such findings might be associated with drug effects. Among the nine NMS Quest domains, in both LOPD and YOPD patients the three most prevalent domains were depression/anxiety, urinary and sexual. Also, in both groups, hallucinations/delusions had the lowest frequency. In the multivariate linear regression model, the Hoehn and Yahr (HY) stage of the disease and activities of daily living scores in YOPD patients, while only the HY stage in LOPD patients appeared to be statistically significant predictors of increasing number of NMS. In contrast to a previous suggestion that YOPD patients might have an increased risk for NMS, we found a higher prevalence of NMS in LOPD patients than in those with YOPD.

Keywords

Non-motor symptomsYoung-onset Parkinson’s disease

Introduction

Quinn et al. [1] applied the term “young-onset Parkinson’s disease” (YOPD) to those with an age of onset between 21 and 40 years (approximately 5–10 % of all PD patients), but later studies included onset up to age 50–55 years [2]. Although the age at PD onset is still a variable of uncertain significance, the majority of previous studies reveal that YOPD has (a) a slower disease progression [3], (b) an increased risk of dystonia at onset of PD and during treatment [4], (c) a lower rate of dementia [5], (d) an increased rate of depression [6], (e) less frequent gait disturbances [7], and (f) an increased rate of levodopa-induced motor complications [8]. Finally, heredity accounts for a greater proportion of YOPD cases than of older-onset cases [2].

Non-motor symptoms (NMS) of PD, although an essential part of the disease, have only recently been increasingly recognized for their impact on a patient’s quality of life [9]. Both dopaminergic and non-dopaminergic mechanisms contribute to their development, which may be an explanation for poor or only a partial response to dopaminergic drugs [10]. It has been reported that they were not well recognized in clinical practice: in over 50 % of consultations, the NMS of PD were not identified by neurologists [11]. The lack of their recognition may deprive patients from adequate planning of care and treatment [12].

In this study, we applied the validated, comprehensive self-completed NMS questionnaire for PD (NMS Quest) [13] to 101 patients with YOPD and 107 patients with late-onset PD (LOPD) in order to test a hypothesis on the existence of differences in prevalence of NMS between YOPD and LOPD, as well as different contribution of selected demographic and clinical factors to the total number of NMS in these subgroups of PD patients.

Patients and methods

This cross-sectional study comprised 101 consecutive outpatients with YOPD (onset of the disease between 21 and 45 years), and 107 outpatients with LOPD (onset of PD ≥ 55 years), matched according to sex, Hoehn and Yahr (HY) staging [14], and the daily levodopa equivalent unit (LEU) dose [15]. They were all recruited from the Department for Movement Disorders, Clinic for Neurology CCS (Belgrade, Serbia). The diagnosis was established according to the UK PD Brain Bank Criteria [16]. At the inclusion to the study, all the patients were “stable” under dopaminergic treatment for at least 3 months and had a Mini-Mental State Examination (MMSE) [17] score >24. Parkin and glucocerebrosidase mutations were excluded in all the patients. The ethical committee of the Medical School University of Belgrade approved the study.

All the tests were conducted during the “on” period. The unified Parkinson’s disease rating scale (UPDRS) (total and part III) was used to evaluate motor disability [18]. Patients were also administered the Hamilton depression (HDRS) [19] and anxiety rating scales (HARS) [20].

After obtaining routine demographic and clinical data (Table 1), PD patients completed the revised NMS Quest (with the explanation of the questions, before completion of the participating neurologist and of caregivers when necessary; i.e., dialect-related issues) [13]. This version is a 30-item self-completed questionnaire featuring response as “present/absent” for each item, with reference to the previous month. The questions were then grouped to nine domains: gastrointestinal tract (seven items), urinary tract (two items), sexual function (two items), cardiovascular (two items), apathy/attention/memory (three items), hallucinations/delusions (two items), depression/anxiety (two items), sleep/fatigue (five items), and miscellany (five items) [13]. The Kuder-Richardson’s Formula 20 (KR-20) was used for calculation of a coefficient of reliability of NMS Quest scale (Kuder–Richardson’s ρ).
Table 1

Demographic and clinical characteristics of patients with young-onset (21 to ≤45 years) and typical-onset (≥55 years) of Parkinson’s disease (PD)

 

Young-onset PD (n = 101)

Typical-onset PD (n = 107)

p

Age (years)a

51.8 (8.8)

69.1 (6.0)

0.001

Males:females

67:24

73:34

0.443

Education (years)a

11.9 (3.1)

11.8 (4.2)

0.771

Age at onset (years)a

40.6 (4.9)

62.0 (5.2)

0.001

Duration of the disease (years)a

11.1 (6.8)

7.1 (4.4)

0.001

Asymmetrical onset, n (%)

98 (97 %)

103 (96 %)

0.532

Initially affected: arm/leg

96/5

93/14

0.640

Duration of treatment (months)a

108.9 (74.8)

89.2 (65.7)

0.001

LEU (mg/day)a

647.5 (354.8)

614.8 (262.7)

0.450

Hoehn and Yahr stagea

2.2 (0.7)

2.3 (0.6)

0.203

UPDRS IIIa

34.6 (15.2)

35.1 (12.3)

0.821

UPDRS totala

56.8 (26.8)

56.6 (21.9)

0.947

ADLa

80.3 (14.1)

75.7 (12.8)

0.016

MMSEa

28.4 (2.6)

27.5 (2.7)

0.011

HDRSa

8.7 (7.1)

11.3 (7.3)

0.012

HARSa

6.9 (6.0)

7.4 (5.0)

0.591

Antiparkinsonian treatment, n (%)

 Levodopa/benserazide

89 (88 %)

101 (94 %)

0.108

 Dopamine agonist

80 (79 %)

60 (56 %)

0.001

 MAO B inhibitor

4 (4 %)

9 (8 %)

0.185

 COMT inhibitor

4 (4 %)

3 (3 %)

0.468

 Amantadine sulphate

67 (66 %)

32 (30 %)

0.001

Patients with dyskinesia, n (%)

60 (59 %)

40 (37 %)

0.001

Patients with motor fluctuations, n (%)

70 (69 %)

49 (46 %)

0.001

Patients on antidepressive treatment, n (%)b

19 (19 %)

33 (31 %)

0.020

PD Parkinson’s disease, LEU Daily levodopa equivalent unit, UPDRS Unified Parkinson’s disease rating scale, ADL Activities of daily living, HDRS Hamilton depression rating scale, HARS Hamilton anxiety rating scale, MAO Monoamino oxidase, COMT Catechol-O-methyl transferase

aValues are means (standard deviation)

bAntidepressants: SSRI (young-onset PD 12 patients, typical-onset PD 30 patients), tricyclic antidepressants (young-onset PD 7 patients, typical-onset PD 3 patients)

Prevalence of NMS was calculated for each symptom for both groups (YOPD and LOPD) separately by computing the number of positive responses and express as percentages. Differences in investigated parameters were assessed by using ANOVA for continuous variables and χ2 for categorical variables. Spearman rank correlation coefficient was used to check significance of the different relationships. The two separate backward stepwise linear regression analyses (for YOPD and LOPD groups) were used to examine how the various clinical characteristics (age at onset, duration of disease, HY stage, activities of daily living (ADL) (independent variables) contribute to the total number of NMS (dependent variable).

Results

The two groups (YOPD and LOPD) were well matched for sex, HY stage of PD, and the LEU dose (Table 1). The duration of the disease was not matched due to suggestion that YOPD patients had a slower disease progression [3, 7]. Indeed, YOPD patients matched for the HY stage with LOPD patients had significantly longer duration of the disease (11.1 ± 6.8 and 7.1 ± 4.4 years, respectively; p = 0.0001). They also had longer time interval from initial symptoms to therapy administration (25.4 ± 22.1 and 14.9 ± 14.0 months, respectively; p < 0.0001) and longer duration of treatment (108.9 ± 74.8 and 89.2 ± 65.7 months, respectively; p < 0.0001). Patients with PD onset ≤45 years had better initial response to levodopa (p = 0.003), more prevalent levodopa-induced dyskinesia (p = 0.001) and motor fluctuations (p = 0.0001), but less frequent hallucinations (p = 0.019), when compared to LOPD patients. Patients with LOPD had lower MMSE score (p = 0.011), but higher HDRS score (p = 0.012). In line with this observation, the larger proportion of LOPD patients received an antidepressive drug when compared to YOPD (19 and 33 patients, respectively; p = 0.02), while YOPD more frequently received dopamine agonist and amantadine (Table 1).

All patients completed the NMS Quest correctly. The internal consistency reliability, measured by Kuder-Richardson’s ρ, was 0.876. The mean total NMS (NMSQ-T) was 11.9 ± 6.0 (range: 0 to of a maximum of 26) in LOPD and 7.7 ± 5.8 (range: 0 to of a maximum of 26) in YOPD (p < 0.05). Only six (5.6 %) and nine (8.9 %) patients with LOPD and YOPD declared no NMS at all, respectively (Table 2). We grouped patients according to the number of reported NMS and found no difference in any of groups (Table 2). However, χ2 analysis for trend showed that a significantly higher number of NMS was associated with LOPD group (χ2 for trend = 3.994, p = 0.045). Compared to YOPD, dribbling of saliva, loss of taste/smell, nocturia, forgetfulness, loss of interest, hallucinations, lack of concentration, anxiety, change in libido and difficulty in sexual activities, were significantly more prevalent in LOPD [with an additional trend (p ≤ 0.1) for constipation] (Table 3). The only NMS more prevalent in YOPD were restless legs and sweating (Table 3).
Table 2

Distribution of non-motor symptoms in patients with young and late-onset Parkinson’s disease depending on the their number per patient

Number of NMS

YOPD (n = 101)

LOPD (n = 107)

pa

ORb

0

9

6

0.357

1.0

1 to ≤10

61

55

0.193

1.35

11 to ≤20

29

41

0.142

2.12

≥21

2

5

0.247

3.75

NMS non-motor symptoms, YOPD young-onset Parkinson’s disease, LOPD late-onset Parkinson’s disease, OR odds ratio

aχ2 test, bχ2 for trend = 3.994, p = 0.045

Table 3

Frequency of positive answers on 30-item NMS Quest in patients with young- and late-onset Parkinson’s disease

 

Young-onset PD (n = 101) (%)

Late-onset PD (n = 107) (%)

p

1. Dribbling

29 (28.7)

61 (57.0)

0.001

2. Taste/smell

29 (28.7)

44 (41.1)

0.042

3. Swallowing

27 (26.7)

30 (28.0)

0.478

4. Vomiting

10 (9.9)

9 (8.4)

0.447

5. Constipation

45 (44.6)

47 (43.9)

0.619

6. Bowel incontinence

3 (2.9)

1 (0.9)

0.289

7. Bowel emptying incomplete

12 (11.9)

22 (20.6)

0.066

8. Urgency

42 (41.6)

43 (40.2)

0.417

9. Nocturia

46 (45.5)

65 (60.7)

0.020

10. Pains

22 (21.8)

28 (26.2)

0.282

11. Weight

8 (7.9)

10 (9.3)

0.454

12. Forgetfulness, memory

27 (26.7)

42 (39.3)

0.038

13. Loss of interest

23 (22.8)

47 (43.9)

0.001

14. Hallucinations

16 (15.8)

28 (26.2)

0.049

15. Concentrating

34 (33.7)

59 (55.1)

0.001

16. Sad, blues

52 (51.5)

53 (49.5)

0.443

17. Anxiety

45 (44.6)

68 (63.6)

0.004

18. Sex drive

41 (40.6)

60 (56.1)

0.018

19. Sex difficulty

31 (30.7)

51 (47.7)

0.009

20. Dizzy

23 (22.8)

25 (23.4)

0.526

21. Falling

24 (23.8)

23 (21.5)

0.411

22. Daytime sleepiness

19 (18.8)

22 (20.6)

0.444

23. Insomnia

34 (33.7)

40 (37.4)

0.339

24. Intense, vivid dreams

28 (27.7)

30 (28.0)

0.542

25. Acting out during dreams

21 (20.8)

23 (21.5)

0.519

26. Restless legs

38 (37.6)

21 (19.6)

0.003

27. Swelling

18 (17.8)

14 (13.1)

0.225

28. Sweating

22 (21.8)

13 (11.2)

0.047

29. Double vision

2 (1.9)

4 (3.7)

0.369

30. Delusions

2 (1.9)

3 (2.8)

0.527

All values present number of patients with a symptom with the percentage in brackets

The NMSQuest has nine domains (Table 4) and among these, in both LOPD and YOPD patients, the three most prevalent domains were depression/anxiety, urinary, and sexual. Also, in both groups, hallucinations/delusions had the lowest frequency (Table 4).
Table 4

Frequency and distribution of positive answers classified by NMS Quest domains among patients with young- and typical-onset Parkinson’s disease (PD)

Domain

Number of items

Positive answers in young-onset PD patients (n = 101)

Positive answers in late-onset PD patients (n = 107)

p

% on the maximum

Median

Mean ± SD

Range

% on the maximum

Median

Mean ± SD

Range

Digestive

7

31

1

1.53 ± 1.47

0–5

34

2

2.02 ± 1.46

0–6

0.018

Urinary

2

44

1

0.88 ± 0.89

0–2

51

1

1.01 ± 0.86

0–2

0.295

Sexual

2

40

0

0.71 ± 0.86

0–2

52

1

1.04 ± 0.95

0–2

0.011

Cardiovascular

2

24

0

0.47 ± 0.70

0–2

23

0

0.45 ± 0.69

0–2

0.862

Memory

3

28

0

0.83 ± 1.08

0–3

46

1

1.38 ± 1.19

0–3

0.001

Hallucinations/delusions

2

18

0

0.18 ± 0.38

0–1

15

0

0.29 ± 0.49

0–2

0.072

Depression/

anxiety

2

48

1

0.96 ± 0.85

0–2

57

1

1.13 ± 0.87

0–2

0.154

Sleep disorder

5

28

1

1.39 ± 1.39

0–5

25

1

1.27 ± 1.33

0–5

0.543

Miscellany

5

18

0

0.72 ± 0.98

0–4

32

0

0.64 ± 0.89

0–2

0.602

SD Standard deviation

The correlation analysis revealed that in both YOPD and LOPD groups, statistically significant positive strong correlations were registered between the NMSQ-T and HY stage [ρ (Spearman correlation coefficient) = 0.607, p = 0.001 and ρ = 0.556, p = 0.001, respectively). Statistically significant moderate correlations were observed between the NMSQ-T and duration of PD in both groups (YOPD: ρ = 0.499, p = 0.001 and LOPD: ρ = 0.359, p = 0.001, respectively). In YOPD group, the NMSQ-T correlated moderately with actual age (ρ = 0.480, p = 0.001) while in LOPD group this relationship was weak (ρ = 0.261, p = 0.007), although statistical significance was registered for both coefficients. In the multivariate linear regression model (NMSQ-T as a dependent variable), in both groups, the HY stage of the disease and ADL scores only in YOPD group, appeared to be statistically significant predictors of increasing number of NMS (Table 5).
Table 5

Predictors of the NMS prevalence

 

Multivariate adjusted β

Standard error

95 % CI

p

Young-onset PD

 Hoehn and Yahr stage

2.51

1.07

0.38–4.63

0.015

 ADL

0.23

0.09

0.04–0.42

0.021

Older-onset PD

 Hoehn and Yahr stage

4.22

1.06

2.07–6.30

0.001

ADL Activities of daily living

Discussion

The main finding of our study, in contrast to a suggestion of Calne and Kumar [21] that YOPD patients might have an increased risk for NMS, was the higher prevalence of NMS in LOPD patients than in those with YOPD (Tables 2 and 3). Although the influence of genetic forms of YOPD was substantially limited through exclusion of parkin and glucocerebrosidase mutations, possible role of other mutations, proven to be rare in our population, could not be excluded. The frequency of NMS among advanced PD patients and correlation between caregivers and patients varied with the instrument used: in one study [22], it was greater with NMSQ than with the NMS Assessment Scale for PD [23]. Martinez-Martin et al. [24] observed that an onset of PD later than 50 years age, duration of disease, and disease severity were independent variables associated with developing NMS. In a smaller study on the prevalence and differences of NMS in patients with YOPD with and without mutations in the Parkin gene, and LOPD, Kaji et al. [25] found that anxiety, change in libido, and RLS were significantly more common in LOPD compared with YOPD without mutations, whereas, loss of taste/smell was the only NMS more common in YOPD. This is partially different from the distribution pattern of NMS in our study. One of the possible explanations is the small number of patients in the study of Kaji et al. [25] (27 patients with YOPD and 27 patients with LOPD).

Ten NMS were significantly more common in patients with LOPD (dribbling of saliva, loss of taste/smell, nocturia, forgetfulness, loss of interest, hallucinations, lack of concentration, anxiety, change in libido, and difficulty in sexual activities), while only RLS and excessive sweating were more common in the YOPD group (Table 2). Distinguishing between RLS and RLS-like symptoms in patients with motor fluctuations (more prevalent in YOPD patients) may be difficult, but in our study the same methodological approach to this issue was applied in both groups [26]. Peralta et al. [27] reported that PD patients with RLS had an earlier onset of PD than patients who scored negative for RLS, although other studies failed to observe such an association [2830]. However, another explanation for the difference we found may not be a younger-onset of PD, but longer duration of the disease (Table 1), shown to be one of the significant predictors of RLS development in PD [29], as well as association with drug effects. For instance, hyperhidrosis was suggested to be a function of plasma levodopa fluctuations: peak-dose dyskinesias were associated with increased sweating, while “drenching sweats” were associated with end-of-dose off periods [31]. Excessive sweating in our YOPD group may be a consequence of a higher prevalence of both levodopa-induced dyskinesia and motor fluctuations when compared to the LOPD group (Table 1). Decreased olfactory function has been reported in >70 % of PD patients [32]. We found smaller prevalence of “smell/taste” deficits (28.7 % in YOPD and 41.1 % in LOPD), which is similar to the findings of Chaudhuri et al. [13] (26 %). Such small estimates may be due to the fact that we used only one question of the NMS Quest for the evaluation of the issue, and the fact that patients usually did not report changes in smell [33]. We have no good explanation for the particularly interesting difference in “smell/taste” deficits between these two groups (Table 3), except the suggestion that registration of symptoms in these two groups may differ due to differences in priorities, needs, and expectations of YOPD patients from those of their older counterparts [21].

Galagher et al. [34] demonstrated that an autonomic dysfunction, psychiatric complications, pain, fatigue, and sleep disorders (the majority of them being more frequent in our LOPD group when compared to the YOPD group) were major correlates of poor health-related quality of life (QoL) in PD. However, QoL proved to be significantly worse in YOPD patients than in LOPD patients [35] who were otherwise not different concerning disease duration and severity. Therefore, Schrag et al. [35] suggested that other, non-disease-related factors may be particularly important in determining “the greater impact of PD on younger patients.”

Braak et al. [36], based on a pattern of distribution of α-synuclein-immunopositive Lewy neuritis and Lewy bodies, introduced a six-stage pathological process in PD, beginning in the anterior olfactory nucleus and in the lowest part of the brainstem, while the typical motor symptoms of PD emerged later, at Braak stages 3 and 4, when pathological process affected substantia nigra. In the final stages 5 and 6, associated with the Lewy bodies in the limbic regions and the entire neocortex, patients may experience for instance, cognitive decline and hallucinations [37], which were more prevalent in our LOPD patients (Table 3). One may hypothesize that progression of neurodegeneration in LOPD is faster and more extensive in comparison to YOPD patients, in accordance with previous clinical observations [3, 5, 7].

Our finding that NMSQ-T associated best in both groups with HY stages (more NMS declared with progression of PD) (Table 5) was consistent with previous reports [13]. In general, our data are concordant with earlier reports on NMS frequency in the PD population, although some differences in the prevalence of individual symptoms and/or domains exist [13, 24, 37]. Such differences were mainly explained by the methodological issues as well as cultural and geographic differences [37].

Limitations of this study include: (1) the relatively small sample size; (2) the data are based on a self-completed questionnaire; (3) relative recruitment disproportion favoring inclusion of patients in earlier stages of PD; and (4) the identification of NMS was done only in the “on” period. Although both groups in our study were matched according to the Hoehn and Yahr staging system [14] and had comparable scores of the UPDRS part III, YOPD patients received slightly higher daily levodopa equivalent doses (Table 1). This statistically nonsignificant difference may, at least partially, reflect higher expectations and demands of YOPD patients [34].

A better description of the YOPD phenotype is necessary for strategically tailored treatment and care in this particular group of PD patients. Despite a similar disease severity, the YOPD patients rated their quality of life (HRQoL) as worse in comparison to older-onset ones [35]. We must bear in mind that the needs and expectations of YOPD patients are different from those of their older counterparts [21] and that further research is necessary to elucidate if the decreased HRQoL in such patients is at least in part due to specific psychosocial factors, including role expectations (e.g., marital, family, socio-economic functioning) [35].

Acknowledgments

This study was supported by a grant from the Ministry of Science and Technology (MST), Republic of Serbia (projects no. 175090).

Conflicts of interest

Tha authors declared that they have no conflict of interest.

Ethical standard

All human studies must state that they have been approved by the appropriate ethics committe and have therefore been performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki.

Copyright information

© Springer-Verlag 2012