Self-Perceived Pre-Motor Symptoms Load in Patients with Parkinson’s Disease: A Retrospective Study

Abstract

Background: Parkinson’s disease is characterized by motor and non-motor clinical features. The latter may present as pre-motor symptoms several years before the motor onset.

Objective: To analyze the association between pre-motor symptoms load and its lead-time in relation to the motor onset and time to diagnosis.

Methods: A cross-sectional study was carried including subjects with Parkinson’s disease from five different movement disorders clinics in Mexico. A structured questionnaire was applied to assess the presence of six self-perceived pre-motor symptoms (hyposmia, depression, anxiety, constipation, pain and sleep disorders).

Results: Overall frequency of pre-motor symptoms was 76.2% . Among the most prevalent symptoms were depression (38%), sleep disorders (37%) and anxiety (36.6%). The lead time to motor onset was greater for constipation (9.2±17.89 years) and pain (8.66±13.36 years). Patients with more than two pre-motor symptoms had a later age at motor onset when compared to patients without pre-motor symptoms (52.04±13.11 vs 56.55±12.97 years, p = 0.037). Late onset patients had a higher frequency of pre-motor symptoms (79% vs 65% in early onset, p = 0.002) and a higher load (1.75±1.37 vs 1.44±1.38, p = 0.033) in comparison to those with early onset. Female subjects reported a higher number of pre-motor symptoms (1.91±1.43 versus 1.48±1.29, p≤0.001). PIGD patients reported a greater frequency of pain (8%) compared to tremor (1%, p = 0.0064) and bradykinetic-rigid (0.61%, p = 0.0061). Anxiety lead-time was greater in tremor-dominant (10.83±15.77 years) compared to bradykinetic-rigid patients (3.48±12.56, p = 0.014).

Conclusions: Pre-motor symptoms load is associated to a later motor onset of PD. Pre-motor symptoms are more frequent in subjects with late onset Parkinson’s disease. Female subjects report a higher number of pre-motor symptoms, depression and anxiety being the most common.

Keywords

Parkinson’s disease pre-motor symptoms non-motor symptoms motor onset

INTRODUCTION

Parkinson’s disease (PD) is the second most common neurodegenerative disease after Alzheimer’s disease [1]. Subjects with PD have a wide-range of non-motor symptoms such as neuropsychiatric symptoms, hyposmia, autonomic dysfunction and sleep disorders. Prevalence of non-motor symptoms (NMS) in PD has been estimated at 21% at the time of diagnosis; however, it increases up to 88% after seven years of disease progression [2].

Nevertheless, associations between non-motor features and other clinical features of PD had been inconsistent among studies [3 –5].

Furthermore, several studies report the onset of pre-motor symptoms before fully manifested Parkinson‘s disease [6]. In the ONSET PD study, 109 newly diagnosed PD patients were evaluated for pre-motor symptoms using a NMS questionnaire. For this study, pre-motor time frames were pre-established (2, 2–10 and >10 years before motor onset). Anhedonia, apathy, memory complaints and inattention were more frequent during the two-year pre-motor period; followed by smell loss, mood disturbances, taste loss, excessive sweating, fatigue and pain in the 2–10 year pre-motor period. Constipation, dream-enacting behavior, excessive daytime sleepiness and postprandial fullness were among the first perceived symptoms (>10 years) [7].

Schrag et al. obtained similar results in a retrospective case-control study in 8,166 patients with PD. Constipation, hypotension, erectile dysfunction, urinary dysfunction, dizziness, fatigue, depression and anxiety were reported by up to five years prior to the time PD diagnosis was made [8].

The objective of the present study is to analyze the association between pre-motor symptoms load and its lead-time in relation to the motor onset and time to diagnosis in a sample of subjects with PD.

PATIENTS AND METHODS

A cross-sectional study was conducted including consecutive subjects attending the movement disorders outpatient clinic at five different centers in Mexico (Mexico City, Zacatecas, Merida, Monterrey and Guadalajara) between January 2014 and June 2015. Diagnosis of PD was determined using the Queen Square Brain Bank Criteria [9].

A structured questionnaire for the assessment of pre-motor symptoms (Supplementary Data) was applied to all participants by a neurologist with expertise in movement disorders. The questionnaire was designed to assess the presence of the following symptoms: hyposmia, depression, anxiety, constipation, pain and sleep disorders. Sleep disorders included the presence of insomnia, rapid eye movement behavior disorder (RBD), restless leg syndrome and excessive daytime sleepiness. Questions were based on the content of other validated NMS instruments including the Nonmotor Symptoms Scale (NMSS) [10] and the Movement Disorders Society-Unified Parkinson’s Disease Rating Scale (MDS-UPDRS) part I [11]. For study purposes, the term “premotor symptom” was defined as a non-motor symptom preceding self-perceived (subjective) motor features.

Demographic and clinical data was collected including gender, current age, age at motor onset, motor onset asymmetry, motor subtype predominance (tremor-dominant, rigid/bradykinetic, postural imbalance and gait disorder) and age at diagnosis. PD was classified as early onset PD (EOPD ≤45 years) and late onset PD (LOPD >45 years).

Lead-time was defined as the time elapsed, in years, from self-perceived pre-motor symptom onset to the motor symptoms onset. The lead-time was calculated for each pre-motor symptom individually. In addition, time from the perceived pre-motor symptoms to the time of PD diagnosis was collected.

Finally, pre-motor symptoms load was defined as the number of pre-motor symptoms reported by the patient. In addition, pre-motor load was categorized in four groups: a) No pre-motor symptoms, b) one pre-motor symptom, c) two pre-motor symptoms, and d) more than two pre-motor symptoms.

All patients gave full written consent for participation, and the study was approved by the Local Ethics Committee and Institutional Review Board.

STATISTICAL ANALYSIS

Demographic and qualitative data were calculated and expressed in terms of frequencies, means, and standard deviations. For each variable, normality was tested. Comparisons between the pre-motor load and gender, motor asymmetry and motor subtype were performed using a Chi-square test. Comparison of current age, pre-motor lead time and time from motor onset to diagnosis relative to pre-motor load was performed using ANOVA. When needed, post hoc analysis was performed using Tukey test and Bonferroni correction. A p level of <0.05 was considered significant. Statistical analyses were performed using SPSS, version 17 (SPSS, Inc., Chicago Illinois).

RESULTS

A total of 500 consecutive PD patients (270 men, 230 women) were included in the study. The mean age of the sample was 62.99±12.03 years with a mean disease duration of 8.66±6.20 years. A total of 110 were categorized as EOPD and 390 as LOPD. Three hundred and eighty one patients (76.2%) reported at least one pre-motor symptom before the onset of PD. General demographic data according to the number of pre-motor symptoms reported is shown in Table 1.

Pre-motor symptom prevalence

From the 381 patients with a history of pre-motor symptoms, 131 (26.2%) reported only one NMS, 113 (22.6%) reported two and 137 (27.4%) had three or more NMS. None of the patients presented with all six history symptoms. The associations between each of the self-perceived pre-motor symptoms are shown in Table 2.

The most prevalent pre-motor symptoms were depression (38.0%, n = 190), sleep disorders (37.0%, n = 188), anxiety (36.6%, n = 183) and constipation (32.8%, n = 164). Hyposmia and pain were reported in 22.8% (n = 114) and 1.2% (n = 6), respectively. Regarding sleep disorders; RBD was the most frequently reported (61.2%, n = 106), followed by insomnia (33.5%, n = 58), restless leg syndrome (RLS) (4.0%, n = 7) and excessive daytime sleepiness (EDS) (1.15%, n = 2).

When categorizing by age of onset of PD, 72 (65.45%) of the subjects with EOPD reported a positive history of pre-motor symptoms compared to 309 (79.23%) of the subjects with LOPD (p = 0.002). No differences in the frequency of individual pre-motor symptoms were found except for the case of pain. Despite a low prevalence of pain in the study, a greater proportion of PIGD-dominant patients reported a history of pain (8%) in contrast to tremor (1%, p = 0.006) and bradykinetic-rigid predominant groups (0.61%, p = 0.006).

Pre-motor symptoms load

The mean number of pre-motor symptoms reported was 1.68±1.37. Although, pre-motor symptom frequency was similar between female and male patients (79.56% versus 73.33%, respectively, p = 0.102); female subjects reported a higher number of pre-motor symptoms (1.91±1.43 versus 1.48±1.29, p≤0.001). Depression and anxiety were more common in female subjects (27.40% versus 50.43%, p≤0.001 and 27.03% versus 39.56%, p = 0.002; respectively).

Similarly to the pre-motor frequency, a statistical significant difference for pre-motor symptoms load was found between EOPD and LOPD (1.44±1.38 versus 1.75±1.37, p = 0.033, respectively).

In regards to the age of motor onset and pre-motor symptoms load, a statistical significant difference was found only between subjects without pre-motor symptoms versus those with more than two pre-motor symptoms (52.04±13.11 versus 56.55±12.97 years, p = 0.037). Neither disease asymmetry, nor the initial motor subtype showed differences between pre-motor load groups.

No statistically significant difference was found between the time from motor onset to PD diagnosis according to the different pre-motor symptoms load groups.

Pre-motor symptoms lead-time

The mean pre-motor lead-time was longer for constipation (9.25±17.89 years), pain (8.66±13.36 years) and anxiety (8.18±14.94 years). The mean lead-time for sleep disorders was 6.65±13.99 years. For insomnia the lead time was of 8.70±14.79 years, while for RBD was 6.11±13.18 years. RLS and EDS had a lead-time of 11.4±18.48 and 22.50±30.41 years, respectively.

On the other hand, hyposmia and depression had the shortest mean lead-time, 4.75±12.18 and 3.81±9.65 years, respectively. Lead times for pre-motor symptoms to motor onset are illustrated in Fig. 1.

When individual pre-motor symptoms lead time were evaluated separately in relation to each motor phenotype; anxiety symptoms had a greater lead time (10.83±15.77 years) in tremor-dominant patients compared to those with a bradykinetic-rigid predominance (3.48±12.56 years, p = 0.014). No other difference in the lead-time was found for the rest of the pre-motor symptoms.

DISCUSSION

The Braak hypothesis of pathology progression in PD proposes a spread of α-synuclein originating in nondopaminergic structures of the brainstem and olfactory bulb, with a later progression to more caudal brain regions [12]. In this model, patients develop the cardinal motor symptoms of PD when stages 3 and 4 are reached. Consequently, during stages 1 and 2, the patient may be considered as asymptomatic but some non-motor symptoms can be already present.

The prevalence of pre-motor symptoms has been widely described; both at the time of motor onset and as the motor disease progresses [13 –16]. Relatively little is known about the time span from self-perceived onset of non-motor symptoms to manifested Parkinson’s disease [8].

We carried out a cross-sectional study to analyze the relation between pre-motor symptoms load and its lead-time with the motor symptom onset and time to diagnosis.

More than three quarters of the subjects with PD referred at least one pre-motor symptom. The mean number of pre-motor symptoms referred was almost two.

The Tuebinger evaluation of Risk factors for Early detection of Neurodegeneration (TREND) Study analyzed the number of total prodromal features known to precede motor manifestations in a cohort of subjects in risk of PD. The number of prodromal symptoms in the TREND study ranged between two and seven [17], our findings are in line with the lower limit.

Mood disorders, sleep-related disorders and constipation were the most prevalent NMS in our population. Interestingly, hyposmia did not represent a major component. This finding is in opposition to the ONSET-PD study, where smell loss was reported in 46% of the patients with PD. It should be mentioned that patients in the ONSET-PD study were older by approximately three years in comparison to our study [7]. Conversely, Schrag et al. reported a prevalence of less than 1% of anosmia in their prediagnostic study [8]. The discrepancies between studies may be the result of different methods applied for smell assessment (smell identification tests versus interview). In our study, hyposmia was only assessed by a questionnaire, and it has been reported that self-perceived smell loss symptoms are difficult to establish [18]. It is also important to highlight that Schrag et al. used primary care health codes to identify the pre-motor symptoms in previously healthy subjects; in other words, they studied the symptoms for which patients sought general practitioner consultation prior to the diagnosis of PD.

Regarding neuropsychiatric symptoms; both depression and anxiety were frequently reported as pre-motor symptoms. In our study, more patients reported self-perceived depression symptoms before the motor onset compared to a recently premotor study (38% vs 13%) [19]. This difference could be result of different assessment methods. Wu et al. verified pre-existing diagnosis using ICD-9, while we applied a questionnaire. Anxiety also represented a more prevalent pre-motor symptom in our population compared to a recent study (36.6% vs 8%) [8].

We found a lower frequency of RBD (21.2%) as a pre-motor symptom than expected. It has been reported that up to 38% of subjects who develop PD after a 5-year follow-up had been previously diagnosed with RBD by polysomnography [20]. One possible explanation for this low frequency may be the fact that the presence of a bed-partner to witness the dream-enactment behaviors was not accounted for. In addition, the diagnosis of RBD in our study was not established by a polysomnography study. Nevertheless, it has been reported that a Single-Question Screen for RBD yields a positive predictive value of 87.9% [21].

Constipation is one of the most prevalent non-motor symptoms in PD patients as well [22]; a similar pre-motor prevalence was found in our patients compared to previous reports (32.8% vs 40.4%, respectively) [19].

Pain did not represent a major pre-motor symptom in our study; however, it has been reported by up to 20% (unexplained) and 10% (chest pain) [9]. This low prevalence could be due to a non-specific question of generalized pain, as other studies reported some specific forms of pain, such as shoulder or chest pain [7, 8].

In regards to the perceived timespan (given date) for the onset of pre-motor symptoms related to motor onset and diagnosis; constipation, pain and anxiety were the first pre-motor symptoms to manifest in our sample. A recent systematic review and meta-analysis of constipation as a premotor symptom reported that constipation can predate PD diagnosis by more than ten years [23]; which is in line with our results.

The lead-time for all sleep disorders (6.65±13.99 years) and specifically for RBD (6.11±13.18 years) were not as high as what the ONSET-PD reported (>10 years), but more similar to Wu et al. findings (>3 years) [19].

Depressive symptoms occurred around four years before the motor onset, whereas the ONSET-PD reported a lead time of 2 to 10 years. On the other hand, the study in Taiwanese population reported more than three years of depressive symptoms before PD diagnosis [19]. However, Schrag et al. reported a lead time of less than 2 years for this symptom [8].

We found a statistical significant difference when comparing the load of pre-motor symptoms between early and late onset PD patients. This association has been previously established for patients with manifested PD, with a higher number of NMS reported in LOPD [24].

Also, we found a larger pre-motor symptom load in female patients. On this matter, gender differences between NMS had been established. Women tend to report depression and anxiety more frequently, while men tend to report hyposmia/anosmia more often [25]. The larger pre-motor load in women could be the result of the construct of the questionnaire, which included several symptoms known to be more frequent in women [26].

When age at motor onset and age at PD diagnosis were compared with the pre-motor symptoms load; patients who had more than two pre-motor symptoms presented with a more advanced age than patients who did not report any self-perceived pre-motor symptom. It has been suggested that PD patients with a younger age at onset have a slower disease progression and subjects with LOPD have a faster and more extensive neurodegeneration [27].

Several studies have addressed NMS in regards to motor subtypes in manifested PD. However, the association between the non-motor symptoms and these subtypes differ between studies. Some authors have reported a higher prevalence of mild cognitive impairment, orthostatic hypotension and RBD at baseline in diffuse/malignant phenotype (non-tremor dominant patients) [28]. van Rooden et al. reported PD subtypes with a prominent involvement of PIGD, cognitive impairment, autonomic dysfunction, daytime sleepiness, psychosis and depression (non-dopaminergic features) [29].

To our knowledge, the ONSET-PD is the only study that has assessed the motor phenotypes to pre-motor symptoms; although it did not find an association between them, constipation and insomnia occurred more frequently in patients with akinetic/rigid subtype in comparison to the tremor-dominant subtype [7]. In our study an association between PIGD and pre-motor pain was found but due to the low prevalence in our sample it is not possible to draw any conclusions. Tremor-dominant subjects reported a longer lead-time between anxiety symptoms and motor onset. More studies need to be conducted in order to elucidate these associations.

Our study has several limitations. First, premotor symptom is defined as a non-motor symptom preceding the motor features onset. Due to the study methodology, we cannot rule out the possibility that a patient may have in fact an unnoticed minor motor symptom at the time the premotor symptom was recalled. This issue should be taken into account when interpreting the study results.

Second, our study is subject to risk of bias. Recall bias is expected due to the study design. Moreover, enrolled patients have different PD duration and recalling the exact time when non-motor symptoms presented may be less consistent. Furthermore, measurement of lead-time is subjective, both in non-motor and motor onset. To reduce this bias, the intent of questions in the survey instrument was masked to the patient.

Interviewer bias may also be a concern, but since we used a standardized questionnaire no systematicdifference between how information was solicited, recorded or interpreted was expected. Finally, we enrolled consecutive patients which reduced the risk of selection bias by including all available subjects during the study period.

Third, our questionnaire has not been formally validated although it should be highlighted that it was constructed in a very similar way as the ONSET-PD tool and based on other validated instruments such as the NMSS and the MDS-UPDRS part I.

In addition, a higher than expected frequency of EOPD was found. This finding may suggest a referral bias. The study was carried out in secondary and tertiary hospitals; where younger subjects with PD are much likely to be referred and treated. Consequently, generalizing our results to a primary care setting should be done cautiously.

In conclusion, pre-motor symptoms in Parkinson’s disease are frequent. Nonetheless, they do not present in all patients before motor onset. A greater load of pre-motor symptoms is related to LOPD. Neither frequency of pre-motor symptoms, nor their load appears to be related to motor clinical phenotype. Methodological issues in the way these variables are measured may raise important differences, thus the development of a standardized questionnaire for assessing pre-motor symptoms is needed.

CONFLICT OF INTEREST

The authors have no conflict of interest to report.

Footnotes

Appendix

References

Jankovic

(2008) Parkinson’s disease: Clinical features and diagnosis. J Neurol Neurosurg Psychiatry, 79, 368–376.

Barone

, Antonini

, Colosimo

, Marconi

, Morgante

, Avarello

, Bottacchi

, Cannas

, Ceravolo

, Cicarelli

, Gaglio

, Giglia

, Iemolo

, Manfredi

, Meco

, Nicoletti

, Pederzoli

, Petrone

, Pisani

, Pontieri

, Quatrale

, Ramat

, Scala

, Volpe

, Zappulla

, Bentivoglio

, Stocchi

, Trianni

, Dotto

; PRIAMO study group (2009) The PRIAMO study: A multicenter assessment of nonmotor symptoms and their impact on quality of life in Parkinson’s disease. Mov Disord, 11, 1641–1649.

Herman

, Weiss

, Brozgol

, Wilf-Yarkoni

, Giladi

, & Hausdorff

(2015) Cognitive function and other non-motor features in non-demented Parkinson‘s disease motor subtypes. J Neural Transm, 122, 1115–1124.

Kwon

, Kang

, Kim

, Lee

, Jang

, Kim

, Lee

, & Koh

(2014) Non-motor symptoms and cognitive decline in de novo Parkinson’s disease. Can J Neurol Sci, 41, 597–602.

Zhang

, Gu

, An

, Wang

, & Chan

(2014) Non-motor symptoms in treated and untreated Chinese patients with early Parkinson’s disease. Tohoku. J Exp Med, 232, 129–136.

Noyce

, Bestwick

, Silveira-Moriyama

, Hawkes

, Giovannoni

, Lees

, & Schrag

(2012) Meta-analysis of early nonmotor features and risk factors for Parkinson’s disease. Ann Neurol, 72, 893–901.

Pont-Sunyer

, Hotter

, Caig

, Compta

, Katzenschlager

, Mas

, Hofeneder

, Brücke

, Bayés

, Wenzel

, Infante

, Zach

, Pirker

, Posada

, Álvarez

, Ispierto

, De Fàbregues

, Callén

, Palasí

, Aguilar

, Martí

, Valldeoriola

, Salamero

, Poewe

, & Tolosa

(2014) The onset of nonmotor symptoms in Parkinson’s disease (The ONSET PD Study). Mov Disord, 30, 1–9.

Schrag

, Horsfall

, Walters

, Noyce

, & Petersen

(2015) Prediagnostic presentations of Parkinson’s disease in primary care: A case-control study. Lancet Neurol, 14, 57–64.

Gibb

& Lees

(1998) The relevance of the Lewy body to the pathogenesis of idiopathic Parkinson’s disease. J Neurol Neurosurg Psychiatry, 5, 745–752.

10.

Chaudhuri

, Martinez-Martin

, Brown

, Sethi

, Stocchi

, Odin

, Ondo

, Abe

, Macphee

, Macmahon

, Barone

, Rabey

, Forbes

, Breen

, Tluk

, Naidu

, Olanow

, Williams

, Thomas

, Rye

, Tsuboi

, Hand

, & Schapira

(2009) The metric properties of a novel non-motor symptoms scale for Parkinson’s disease: Results from an international pilot study. Mov Disord, 22, 1901–1911.

11.

Martinez-Martin

, Rodriguez-Blazquez

, Alvarez-Sanchez

, Arakaki

, Bergareche-Yarza

, Chade

, Garretto

, Gershanik

, Kurtis

, Martinez-Castrillo

, Mendoza-Rodriguez

, Moore

, Rodriguez-Violante

, Singer

, Tilley

, Huang

, Stebbins

, & Goetz

(2013) Expanded and independent validation of the Movement Disorder Society– Unified Parkinson’s Disease Rating Scale (MDS-UPDRS). J Neurol, 260, 228–236.

12.

Braak

, Del Tredici

, Rüb

, de Vos

, Jansen Steur

, & Braak

(2003) Staging of brain pathology related to sporadic Parkinson’s disease. Neurobiol Aging, 24, 197–211.

13.

Zis

, Martínez-Martín

, Saubier

, Rizos

, Sharma

, Worth

, Sophia

, Silverdale

, & Chaudhuri

(2015) Non-motor symptoms burden in treated and untreated early Parkinson’s disease patients: Argument for non-motor subtypes. Eur J Neurol, 22, 1145–1150.

14.

Chaudhuri

, Sauerbier

, Rojo

, Sethie

, Schapira

AHV

, Brown

, Antonini

, Stocchi

, Odin

, Bhattacharya

, Tsuboi

, Abe

, Rizos

, Rodriguez-Balzquez

, & Martines-Marin

(2015) The burden of non-motor symptoms in Parkinson’s disease using a self-completed non-motor questionnaire: A simple grading system. Parkinsonism Relat Disord, 21, 287–291.

15.

Zis

, Rizos

, Martínez-Martin

, Pal

, Silverdale

, Sharma

, Sauerbier

, & Chaudhuri

(2014) Non-motor symptoms profile and burden in drug naïve versus long-term Parkinson's disease patients. J Parkinsons Dis, 4, 541–547.

16.

Breen

, & Drutyte

(2013) Non-motor symptoms of Parkinson's disease: The patients’ perspective. J Neural Transm, 120, 531–535.

17.

Gaenslen

, Wurster

, Brockmann

, Huber

, Godau

, Faust

, Lerche

, Eschweiler

, Maetzler

, & Berg

(2014) Prodromal features for Parkinson‘s disease—baseline data from the TREND study. Eur J Neurol, 21, 766–772.

18.

Haehner

, Boesveldt

, Berendse

, Mackay-Sim

, Fleischmann

, Silburn

, Mellick

, Herting

, Reichmann

, & Hummel

(2009) Prevalence of smell loss in Parkinson’s disease-a multicenter study. Parkinsonism Relat Disord, 15, 490–494.

19.

, Liao

, Chen

, Chang

, & Lin

(2015) Risk of premotor symptoms in patients with newly diagnosed PD: A nationwide, population-based, case-control study in Taiwan. PLoS One, 10, 1–9.

20.

Schenck

, Bundlie

, & Mahowald

(2013) Delayed emergence of a parkinsonian disorder or dementia in 81% of older men initially diagnosed with idiopathic rapid eye movement sleep behavior disorder: A 16-year update on a previously reported series. Sleep Med, 14, 744–748.

21.

Postuma

, Arnulf

, Hogl

, Iranzo

, Miyamoto

, Dauvilliers

, Oertel

, Ju

, Puligheddu

, Jennum

, Pelletier

, Wolfson

, Leu-Semenescu

, Frauscher

, Miyamoto

, Cochen De Cock

, Unger

, Stiasny-Kolster

, Fantini

, & Montplaisir

(2012) A single-question screen for rapid eye movement sleep behavior disorder: A multicenter validation study. Mov Disord, 27, 913–916.

22.

Pagano

, Tan

, Haider

, Bautista

, & Tagliati

(2015) Constipation is reduced by beta-blockers and increased by dopaminergic medications in Parkinson‘s disease. Parkinsonism Relat Disord, 21, 120–125.

23.

Adams-Carr

, Bestwick

, Shribman

, Lees

, Schrag

, Noyce

(2015) Constipation preceding Parkinson’s disease: A systematic review and meta-analysis. J Neurol Neurosurg Psychiatry, doi: 10.1136/jnnp-2015-311680[Epub ahead of print].

24.

Spica

, Pekmezovic

, Svetel

, & Kostic

(2103) Prevalence of non-motor symptoms in young-onset versus late-onset Parkinson’s disease. J Neurol, 260, 131–137.

25.

Fereshtehnejad

, Romenets

, Anang

, Latreille

, Gagnon

, & Postuma

(2015) New clinical subtypes of Parkinson disease and their longitudinal progression: A prospective cohort comparison with other phenotypes. JAMA Neurol, 72, 863–873.

26.

Liu

, Umbach

, Peddada

, Xu

, Trösyer

, Huang

, & Chen

(2015) Potential sex differences in nonmotor symptoms in early drug-naïve Parkinson disease. Neurology, 84, 2107–2115.

27.

Wickremaratchi

, Ben-Shlomo

, & Morris

(2009) The effect of onset age in the clinical features of Parkinson’s disease. Eur J Neurol, 16, 450–456.

28.

Selikhova

, Williams

, Kempster

, Holton

, Revesz

, & Lees

(2009) A clinico-pathological study of subtypes in Parkinson’s disease. Brain, 132, 2947–2957.

29.

van Rooden

, Colas

, Martínez-Martín

, Visser

, Verbaan

, Marinus

, Chaudhuri

, Kok

, & van Hilten

(2011) Clinical subtypes of Parkinson's disease. Mov Disord, 26, 51–58.