Depression in Parkinson's disease: clinical correlates and outcome

Parkinsonism and Related Disorders 10 (2003) 23–28 www.elsevier.com/locate/parkreldis

Depression in Parkinson’s disease: clinical correlates and outcome A. Rojo*, M. Aguilar, M.T. Garolera, E. Cubo, I. Navas, S. Quintana Neurology and Intensive Care Units, Servicio de Neurologia, Hospital Mutua de Terrassa, c/Castell, 25, Terrassa, Barcelona 08221, Spain Received 2 August 2002; revised 8 April 2003; accepted 24 April 2003

Abstract Depression has been shown to be more common in Parkinson’s disease (PD) than in other chronic and disabling disorders. Neurochemical and functional disturbances are important etiopathogenic factors. The prevalence and clinical features associated with depression in PD remain controversial. The purpose of this study is to estimate the prevalence of depressive symptoms in our patients, as related to other clinical data, and to assess clinical outcomes of these symptoms. A series of PD patients were evaluated over a 9-year period, using the Unified Parkinson’s Disease Rating Scale (UPDRS), Hoehn and Yahr stage (HY), Schwab and England Scale (SE), Mini-Mental State Examination (MMSE), and Yesavage Geriatric Depression Scale (GDS). Presence of depressive symptoms was considered if GDS score was higher than 10: mild – moderate (MD) for GDS scores between 11 and 20 and moderate – severe (SD) for GDS scores greater than 20. Three hundred and fifty-three patients were included in this study and additional follow up information was obtained for 184 patients. MD and SD were found in 40.2 and 16.7% of PD patients, respectively. Female gender, high HY, high UPDRS total and subtotal, and low MMSE and SE scores were significantly associated with depressive symptoms. According to changes in GDS score, 34% of patients remained stable, 35% showed an improvement, and 30.9% worsened in the follow up study. Gender, age, age of onset, HY, UPDRS, and PD duration are not related to depression outcome. q 2003 Elsevier Ltd. All rights reserved. Keywords: Depression; Geriatric; Geriatric depression scale; Parkinson’s disease; Yesavage

1. Introduction Depression is the most common neuropsychiatric disturbance in Parkinson’s disease (PD) and has been shown to be more common in PD than in other chronic and disabling disorders. Previous studies have concluded that between 4 and 60% of patients show depressive symptoms [1 – 3] and up to 26% of these patients were on antidepressant treatment [4]. Depression is one of the most important factors impairing quality of life in this disease [5,6]. Due to methodological issues such as selection bias, inaccurate definitions of depression (major vs. minor depressive episode, dysthymic disorder…), and the use of different scales and diagnostic criteria, the risk factors and findings associated with depression in PD have remained controversial. A previous study has shown that a prior depressive episode is a risk factor for depression development in PD [7]. Other related factors were cognitive * Corresponding author. Tel.: þ 34-93-736-50-38; fax: þ 34-93-736-5050. E-mail address: [email protected] (A. Rojo). 1353-8020/03/$ - see front matter q 2003 Elsevier Ltd. All rights reserved. doi:10.1016/S1353-8020(03)00067-1

impairment, higher Unified Parkinson’s Disease Rating Scale (UPDRS) score, higher requirements of L -DOPA, and the presence of motor fluctuations [8]. Other studies have related depression in PD to right side parkinsonism, akinesia, severe disability, anxiety and psychosis [9 – 11], and to female gender, early onset of PD, longer disease duration, severe motor deficit, axial bradykinesia, and gait and balance impairment [12 – 14]. In a previous study with 87 patients, our group showed an association of depression with female gender, Hoehn and Yahr (HY) stage, impairment of daily living activities, and cognitive deficits [1]. Although functional impairment, as well as sociological and psychological factors, have an influence on mood state in PD, there is some evidence concerning organic factors involved in the etiopatogeny of depression in PD. Dopamine, norepinephrine and serotonin deficits have been implicated in the neurochemical basis of this disorder. Degeneration in the mesolimbic system, tegmental ventral area, locus coeruleus, and serotoninergic nucleus has been reported. Depression in PD has been associated with lower levels of 5-HIIA (5-hidroxiindolacetic acid) in cerebrospinal fluid [15], mesencephalic hypoecogenicity [16], and

24

A. Rojo et al. / Parkinsonism and Related Disorders 10 (2003) 23–28

frontomedial hypometabolism in 18-F-glucose PET scan [17 –19]. Previous studies of depression in PD have used different depression scales: Hamilton Depression Scale [12,16,20], Zung Self-Rating Scale [21], Beck Depression Inventory ˚ sberg Depression Rating Scale [8,19], [8], Montgomery-A Visual Analog Mood Scales [22], Von Zerssen’s depression scale [23], and short Geriatric Depression Scale (GDS-15) [24]. Other studies have used DSM III-R diagnostic criteria [13,17,25,26]. The GDS of Yesavage, including 30 questions with a dichotomic answer (yes/no) was chosen because it is appropriate for the geriatric and PD population (due to the absence of somatic questions, which could be related to motor impairment and could be difficult to interpret (e.g. constipation)). Normal value in geriatric population is 5.75 ^ 4.34 (vs. 15 ^ 6.5 in mild depression, and 22.8 ^ 5.07 in severe depression) [27]. Using a score of 11 as a cut off value for depression diagnosis in geriatric population, sensitivity and specificity of this scale are 84 and 95%, respectively, [28]. It is also useful in detecting depression in demented patients [27] and has been used previously in Parkinson’s disease [29,30]. This study was carried out using GDS to estimate the prevalence of depressive symptoms in our PD outpatients and to evaluate the relationship between depressive scores and clinical features of PD (single evaluation study). In some patients a second evaluation was performed in order to assess the outcome of depressive symptoms in PD and to identify factors associated with changes in mood state in PD.

Table 1 Clinical and demographic data in patients with Parkinson’s disease included in the single evaluation study Single evaluation study Prospective study Number Sex Age at examination (years) Age of onset Disease duration Hoehn andYahr stage: 1–1.5 2 3 4–5 Education (years) GDS score MMSE score UPDRS score: Sub I-mental score (0–16) Sub II-DLA score (0–52) Sub III-motor score (0–56) Total UPDRS score (0–124) Schwab England score:

353 145 M; 208 F 68.8 ^ 96.6 63.5 ^ 11.1 4.6 ^ 5.3

184 74 M, 110 F 67.2 ^ 8.9 62.1 ^ 10.5 4.9 ^ 5.2

44 (12.5) 134 (38.0) 144 (40.8) 30 (8.5) 6.3 ^ 4.3 12.7 ^ 7.1 25.2 ^ 4.4

27 (14.6) 73 (39.7) 70 (38.1) 14 (7.6) 6.2 ^ 4.6 13 ^ 5.3 26.1 ^ 3.6

3.9 ^ 2.9 14.2 ^ 8.9 30.1 ^ 16.3 48.6 ^ 26.1 78.6 ^ 21.3

3.9 ^ 2.6 14.2 ^ 8.2 28.6 ^ 10.6 45.8 ^ 23.9 77.2 ^ 18.2

Values given as mean ^ SD when applicable. Numbers in parentheses are percentages. GDS: Geriatric Depression Scale; MMSE: Mini Mental State examination; DLA: daily living activities; UPDRS: Unified Parkinson’s Disease Rating Scale.

Prospective follow up data were also obtained for a subgroup of patients. Demographic and clinical features for these patients were similar to the entire group (Table 1). Protocolarized follow up evaluations were done every 6 months during a mean time of 2.7 ^ 1.6 years. 2.3. Statistical analysis

2. Patients and methods 2.1. Study population A consecutive series of PD patients referred to Neurology Service, Hospital Mutua de Terrassa, Barcelona (Spain) over a 9-year period (1990 – 1999) was studied. 2.2. Procedure PD was diagnosed using the Brain Bank Criteria [31]. A standard neurological evaluation to assess Parkinson’s syndrome was performed, including a complete neurological examination using the UPDRS, Hoehn and Yahr stage (HY), and Schwab and England scale (SE). A mental evaluation was made using the Mini-Mental State Examination (MMSE) and the GDS as screening test for depressive symptoms. In patients showing motor fluctuations, the evaluations were made during an ‘on’ phase. Demographic and clinical features of our study population are summarized in Table 1. Antiparkinsonian therapy was modified and antidepressant drugs were added if necessary and were adjusted individually.

Quantitative data are shown as mean ^ standard deviation and qualitative data are shown as percentages. Normal distribution was tested with Kolmogorov Smirnov test. We used parametric tests (t de Student, one way ANOVA (post hoc Scheffe test)) and non-parametric tests (Mann –Whitney U test, Spearman Correlation or Willconxon test for paired values), according to its normality. A regression analysis was used to test independence of variables in the depressive score. The variables with a p value , 0.10 on a bivariate analysis were included in stepwise multiple regression analysis. Furthermore, according to the GDS score, patients were classified as: no depressive symptoms (ND, GDS , 11), mild –moderate depressive symptoms (MD, GDS 11 –20), and patients with moderate – severe depressive symptoms (SD, GDS . 20) to analyze clinical characteristics in these groups. Finally, in patients with more than one evaluation, we categorized changes in GDS score, as follows: stable (same initial and final GDS results (^ 1 point difference)), improvement (GDS final score two or more points lower than initial score), or worsening (GDS final score two or

A. Rojo et al. / Parkinsonism and Related Disorders 10 (2003) 23–28

more points higher than initial score). The Wilcoxon test was used to analyze data from these patients. Statistical analyses were done with the SPSSw 7.0 software (SPSS Institute Inc, USA).

3. Results 3.1. Single evaluation study Three hundred and fifty-three patients (145 men, 208 women) were included in the study. MD and SD were found in 40.2% ðn ¼ 142Þ and 16.7% ðn ¼ 59Þ of the PD patients, respectively. Clinical and demographic data for patients with Parkinson’s disease in three subgroups according to the GDS score (ND, MD, SD) are summarized in Table 2. Prevalence of depressive scores was 68.6% in women and 40.64% in men ðp ¼ 0:0001Þ: GDS score in women was significantly higher than in men (14.3 ^ 6.8 vs. 10.3 ^ 6.7; p ¼ 0:0001) and SD was particularly more frequent in women. The presence of depressive symptoms increased with higher HY stage ðp ¼ 0:003Þ; higher scores in UPDRS subscales, total score ðp ¼ 0:0001Þ; and was associated with lower scores in SE ðp ¼ 0:0001Þ: Tremor scores (but not rigidity scores) were higher in patients with depression than

25

in non-depressed patients ðp ¼ 0:03Þ: Items for akinesia in UPDRS were higher in severely depressed patients ðp ¼ 0:0002Þ: Patients in SD group showed a lower score in MMSE ðp ¼ 0:004Þ: These findings were confirmed by correlation analyses. However, some variables are not independent of each other. In a stepwise multiple regression analysis, the UPDRS II score (daily living activities), UPDRS III score (motor), and female gender were the independent variables that significantly predicted the GDS score. These variables predicted 30.8% of the variability in the GDS score ðR ¼ 0:55Þ: 3.2. Prospective study (two evaluations) Data from 184 patients (110 women, 74 men) were available. There were no significant differences in GDS scores between the first visit and the last one (13.2 ^ 7.3 vs. 12.7 ^ 6.6). On the second visit, 44% of patients were ranked as MD and 14.7% as SD (Table 3). According to changes in GDS score, 34% of patients remained stable, 35% showed an improvement, and 30.9% worsened (Table 4). In initial visits, 75 patients (41%) were ranked as normal. In the follow up visit, although the majority of these patients remained non-depressant, 25% were scored in the depressive range (24% MD, 1% SD). Of the 74 patients classified

Table 2 Clinical and demographic data in patients with Parkinson’s disease in three subgroups: no depressive symptoms patients (ND, Geriatric Depression Scale , 11), patients with mild–moderate depressive symptoms (MD, GDS 11–20), and patients with moderate–severe depressive symptoms (SD, GDS .20)

Number Sex Male Female H and Y 1–1.5 2 2.5–3 4–5 Sub I UPDRS (mental) Sub II UPDRS (DLA) Sub III UPDRS (motor) Total UPDRS Tremor Rigidity Akynesia Age of onset Age PD course (y) Schwab England Initial MMSE Fluctuations

ND

MD

SD

p

152 (43.1)

142 (40.2)

59 (16.7)

86 (56) 66 (43)

44 (31) 98 (69)

15 (25.5) 44 (74.5)

0.0001#

21 (13.8) 70 (46) 58 (38.2) 3 (2) 1.9 ^ 1.9** 11.6 ^ 7.8 26.1 ^ 14.9 39.6 ^ 22.3** 3.3 ^ 3.5 7.1 ^ 3.7 9.0 ^ 6.5 63.5 ^ 11.1 68.4 ^ 9.4 4.6 ^ 2.0 84.1 ^ 18.3** 25.9 ^ 3.9 39 (25.6)

19 (13.5) 44 (31.2) 63 (44.7) 15 (10.6) 4.8 ^ 2.2** 15.4 ^ 8.5* 31.6 ^ 16.8* 51.8 ^ 24.7** 4.8 ^ 4.6* 7.8 ^ 4.2 10.5 ^ 6.4 63.7 ^ 11.5 69.0 ^ 10.1 4.4 ^ 1.7 75.5 ^ 22.2 25.2 ^ 4.1 31(21.8)

4 (6.78) 20 (33.8) 23 (39) 12 (20.3) 8.8 ^ 12.0** 18.4 ^ 10.5* 37.1 ^ 16.0* 64.2 ^ 29.6** 5.2 ^ 4.4* 8.6 ^ 3.7 12.9 ^ 6.6* 63.03 ^ 10.2 69.1 ^ 9.2 4.3 ^ 1.6 71.8 ^ 23.1 23.4 ^ 5.2** 15 (25.4)

0.0003#

0.0001 0.0001 0.0001 0.0001 0.03 0.05 0.0002 0.76 0.78 0.90 0.0001 0.004 0.38

Values given as mean ^ SD where applicable. Numbers in parentheses are percentages; #Chi square test. One way ANOVA (post hoc test: Scheffe)* statistically significant vs. non-depressed, ** statistically significant vs. two other groups; DLA: daily living activities; H and Y: Hoehn and Yahr stage; UPDRS: Unified Parkinson’s Disease Rating Scale.

26

A. Rojo et al. / Parkinsonism and Related Disorders 10 (2003) 23–28

Table 3 Distribution ðnÞ in subgroups: no depressive symptoms (ND), mild– moderate depressive symptoms (MD), and moderate–severe depressive symptoms (SD) on initial and final visit for a population of 184 patients Second visit

Initial visit

ND ðn ¼ 75Þ MD ðn ¼ 74Þ SD ðn ¼ 35Þ

ND ðn ¼ 76Þ

MD ðn ¼ 81Þ

SD ðn ¼ 27Þ

56 15 5

18 52 11

1 7 19

after the initial visit as MD, seven patients (9%) worsened, scoring in the SD range. We found improvement to the normal range in 15 (20%) patients initially ranked in the mild depression group and in 5 (14%) patients initially ranked in the severe depression group. Eleven patients (31%) of the SD group improved to MD (Table 3). No statistically significant differences were found in gender, HY, UPDRS, PD onset and PD course, when comparing patients who improved, worsened or remained stable (Table 4). GDS score improvement was associated with lower scores in initial MMSE and higher scores in UPDRS subscale I. Worsening in GDS score was associated with higher initial GDS score ðp , 0:05Þ: Table 4 Distribution of the variables in the first evaluation of the prospective study of 184 patients divided in three subgroups: stable group (Geriatric Depression Scale—GDS—same initial and final GDS results (^1 point difference), improvement group (GDS final score two or more points lower than initial score) and worsening group (GDS final score two or more points higher than initial score)

Gender ðnÞ Male Female Hoehn Yahr ðnÞ 1 –1.5 2 2.5 –3 4 –5 Age (x ^ SD) PD course (y) (x ^ SD) Age of onset (y) (x ^ SD) UPDRS initial Sub I (mental) Sub II (DLA) Sub III (motor) UPDRS total Tremor items Rigidity items MMSE initial MMSE final GDS initial

Stable ðn ¼ 63Þ

Improvement ðn ¼ 64Þ

Worsening ðn ¼ 57Þ

26 37

20 44

28 29

10 20 27 5 67.8 ^ 9.2 4.8 ^ 4.2 62.6 ^ 10.7

9 27 23 5 67.1 ^ 9.5 4.8 ^ 6.0 61.4 ^ 11.7

8 26 20 3 66.8 ^ 7.9 5.1 ^ 4.7 62.4 ^ 9.2

3.5 ^ 3.1 13.6 ^ 8.2 27.8 ^ 16.6 44.9 ^ 25.2 3.9 ^ 4.2 6.9 ^ 3.6 26.6 ^ 3.5 26.4 ^ 4.2 9.2 ^ 5.3

5.1 ^ 2.4* 13.7 ^ 8.9 29.9 ^ 15.2 48.8 ^ 23.8 4.1 ^ 3.7 7.7 ^ 4.0 24.9 ^ 4.2* 26.3 ^ 3.7 12.5 ^ 6.2

3.05 ^ 2.5 12.6 ^ 7.4 28.1 ^ 15.7 43.9 ^ 22.8 4.0 ^ 4.6 7.3 ^ 3.7 26.6 ^ 3.2 26.0 ^ 3.1 17.4 ^ 4.3*

p , 0:05: One way ANOVA (post hoc test: Scheffe). GDS: Geriatric Depression Scale; MMSE: Mini-Mental State examination; DLA: daily living activities; PD: Parkinson’s disease; UPDRS: Unified Parkinson’s Disease Rating Scale.

4. Discussion 4.1. Clinical correlates In our study, 56.9% of PD patients showed GDS scores higher than 10, suggesting depression (40.2% MD; 16.7% SD). Previous studies have shown that the frequency of depression in PD is about 40% [3]. This figure includes patients that meet the criteria for a major depressive episode and patients with dysthymia or minor depression. Our figures are similar to those previously reported [1,20,24,32], although lower values have been reported in other studies [13,19,25,26], perhaps due to differences in definitions (e.g. based on DSM III criteria). Our study did not intend to diagnose the depressive disorder; our only aim was to quantify the presence of depressive symptoms and their severity. As in previously published studies, most of our depressed patients showed MD; SD was less common [19]. It is important to understand whether SD is a specific problem, distinct from dysthymia in PD, or whether both disorders are in fact two degrees of the same condition. We know that GDS score does not allow to establish the diagnosis of dysthymia or major depression but it allows us to make some speculation. Some studies have suggested that severe depression is caused by an organic factor related to neurochemical disorders in PD and has an early onset, while the dysthymic condition is a patient’s reaction to suffering a progressive and disabling disease and therefore has a later onset. We have not found significant differences in disease duration or in age between both groups. A significant difference in MMSE score was found between patients with MD and SD; SD patients showing a lower MMSE score. However, the design of this study did not allow us to determine whether this was a cause or a consequence. A depression-female gender relationship has been previously reported [33,34]. Other studies have shown that depression is also more common in women in the general population [35]. This could be explained, at least in part, by a higher sensibility of scales to detect depression in women than in men [36]. Most of the previous studies were unable to show a significant relationship between severity of motor symptoms in PD and depression [37,39]. Tandberg et al. [19] showed a significant correlation of depression with UPDRS subscales I and II, but not with the motor subscale. Other studies, however, have shown an association with predominant akineto-rigid PD and with bradykinesia [38,40]. In order to avoid confounding factors, since motor symptoms associated with depression can impair motor and cognitive PD status [41], we chose to analyze tremor predominance. In the present study, depression was associated with scores in all UPDRS domains, including tremor and akinesia. As with most of the previously published studies [3], we could not find a relationship between depression and advanced age, age at PD onset, or PD duration. A relationship has been observed, however, by some authors [19] and

A. Rojo et al. / Parkinsonism and Related Disorders 10 (2003) 23–28

a higher prevalence of depression in younger patients has been reported [42]. Contrary to other studies, fluctuations were not more common in depressed patients in the present serie [43]. Several studies have shown a relationship between cognitive dysfunction and depression in PD [7,8,44]. Tandberg et al. [19] in a study of 245 patients, found cognitive dysfunction (MMSE , 24) to be a risk factor for depression occurrence in PD. The relationship between depression and cognition in PD is complex. Depression can cause and increase cognitive disorders in PD [7,40,43]; depressive symptoms are common in initial phases of dementia; and both syndromes could be accounted for the same neurochemical and neurodegenerative disorder. Previous studies have shown that depression in demented patients with PD is similar to depression in non-demented individuals [45,46]. No time course data for occurrence of depressive and cognitive disorders were available in our series. We are unable to determine whether cognitive changes precede or follow mood changes, due to a lack of information.

27

A higher initial GDS score suggests a worse outcome of depressive symptoms. These data do not support previous hypotheses suggesting that patients with major depression tend to improve while patients with mild depressive symptoms tend to remain stable [42]. Although we have analyzed multiple variables, no straightforward conclusions can be drawn, and the complex nature of mood should not be simplified. Life experiences during the disease, biographic personal factors, previous character, and psychological, social and family support could be the most important factors for depression development. However, these factors are more difficult to interpret and quantify. In conclusion, depression is common in PD and was associated with female gender and with the presence of cognitive, functional, and motor impairment. Gender, age, PD course, PD onset, HY, and motor and functional impairment are not related to the outcome of depressive symptoms.

Acknowledgements 4.2. Prospective study Overall prevalence of depressive symptoms in PD was similar in first and second evaluations. In spite of antidepressant therapy, which was added individually (and probably resulted in a reduced GDS score), the global GDS score did not show significant changes. We justified this because other patients developed depressive symptoms during the same time period (Table 3). This study has several limitations. We did not assess drug treatments and their efficacy. Furthermore, at the moment, prospective follow up data are only available for a subgroup of patients. This subgroup does not differ, however, from the whole series, in any demographic or clinical variable. About one third of the patients remained stable; one-third showed improvement and one third worsened. No statistically significant differences were found in gender, HY, UPDRS, PD onset, and PD course when comparing patients, who improved, worsened or remained stable. In the only previous prospective analysis, Brown et al. [47] repeated a Beck Depression Inventory assessment 1-year after the first evaluation. Most patients (61.4%) showed no depression in either their first or second visit. Although the authors did not report whether or not patients received treatment, they found that the initial percentage of depressed patients remained stable. The GDS score improvement was associated with lower scores in the initial MMSE and higher scores in the UPDRS subscale I. This finding was unexpected and it has proved difficult to interpret. However, we believe this cognitive impairment was related to depression because, in their final evaluation, patients showing an improved GDS score also showed an improved MMSE, with a score similar to the other patients (26.3 ^ 3.7) (Table 4).

Preliminary results of this study were presented at the XIV International Congress on Parkinson’s disease. July 27th – August 1st 2001. Helsinki, Finland.

References [1] Garolera M, Aguilar M. In: Korczyn AD, editor. Depression in patients with Parkinson’s disease, as measured on the Geriatric Depression Scale (GDS). International Symposium: Dementia in Parkinson’s Disease, Bologna: Monduzzi Editore; 1994. p. 12. [2] Zesiewicz TA, Gold M, Chari G, Hauser RA. Current issues in depression in Parkinson’s disease. Am J Geriatr Psychiatry 1999;7: 110–8. [3] Cummings JL. Depression and Parkinson’s disease: a review. Am J Psychiatry 1992;149:443 –57. [4] Richard IH, Kurlan R. A survey of antidepressant drug use in Parkinson’s disease. Parkinson Study Group. Neurology 1997;49: 1168– 70. [5] Schrag A, Jahanshahi M, Quinn N. What contributes to quality of life in patients with Parkinson’s disease? J Neurol Neurosurg Psychiatry 2000;69:308 –12. [6] Kuopio AM, Marttila RJ, Helenius H, Toivonen M, Rinne UK. The quality of life in Parkinson’s disease. Mov Disord 2000;15:216– 23. [7] Mayeux R, Stern Y, Rosen J, Leventhal J. Depression, intellectual impairment and Parkinson’s disease. Neurology 1981;31:645–50. [8] Tandberg E, Larsen JP, Aarsland D, Cummings JL. The occurrence of depression in Parkinson’s disease. A community-based study. Arch Neurol 1996;53:175–9. [9] Barber J, Tomer R, Sroka H, Myslobodsky MS. Does unilateral dopamine deficit contribute to depression? Psychiatry Res 1985;15: 17–24. [10] Cole SA, Woodard JL, Juncos JL, Kogos JL, Youngstrom EA, Watts RL. Depression and disability in Parkinson’s disease. J Neuropsychiatry Clin Neurosci 1996;8:20–5. [11] Tom T, Cummings JL. Depression in Parkinson’s disease. Pharmacological characteristics and treatment. Drugs Aging 1998; 12:55–74.

28

A. Rojo et al. / Parkinsonism and Related Disorders 10 (2003) 23–28

[12] Chia LG, Cheng LJ, Chuo LJ, Cheng FC, Cu JS. Studies of dementia, depression, electrophysiology and cerebrospinal fluid monoamine metabolites in patients with Parkinson’s disease. J Neurol Sci 1995; 133:73–8. [13] Cubo E, Bernard B, Leurgans S, Raman R. Cognitive and motor function in patients with Parkinson’s disease with and without depression. Clin Neuropharmacol 2000;23:331 –4. [14] Oertel WH, Hoglinger GU, Caraceni T, et al. Depression in Parkinson’s disease. An update. Adv Neurol 2001;86:373–83. [15] Asberg M, Bertilsson L, Martensson B, Scalia-Tomba GP, Thoren P, Transkman-Bendz L. CSF monoamine metabolites in melancholia. Acta Psychiatr Scand 1984;69:201–19. [16] Becker T, Becker G, Seufert J, et al. Parkinson’s disease and depression: evidence for an alteration of the basal limbic system detected by transcranial sonography. J Neurol Neurosurg Psychiatry 1997;63:590 –6. [17] Ring HA, Bench CJ, Trimble MR, Brooks DJ, Frackowiak RSJ, Dolan RJ. Depression in Parkinson disease. A positron emission study. Br J Psychiatry 1994;165:333 –9. [18] Mayeux R, Stern Y, Williams JBW, Cote L, Frantz A, Dyrenfurth Y. Clinical and biochemical features depression in Parkinson’s disease. Am J Psychiatry 1986;143:756– 9. [19] Tandberg E, Larsen JP, Aarsland D, Laake K, Cummings JL. Risk factors for depression in Parkinson’s disease. Arch Neurol 1997;54: 625–30. [20] Liu CY, Wang SJ, Fuh JL, Lin CH, Yang YY, Liu HC. The correlation of depression with functional activity in Parkinson’s disease. J Neurol 1997;244:493 –8. [21] Kuhn W, Heye N, Mu¨ller T, Kraus P, Klotz P, Friedrich B, Welter FL, Przuntek H. The motor performance test series in Parkinson’s disease is influenced by depression. J Neural Transm 1996;103:349–54. [22] Stern RA. Assessment of mood states in neurodegenerative disease: methodological issues and diagnostic recommendations. Semin Clin Neuropsychiatry 1996;1:315–24. [23] Heberlein I, Ludin HP, Scholz J, Vieregge P. Personality, depression, and premorbid lifestyle in twin pairs discordant for Parkinson’s disease. J Neurol Neurosurg Psychiatry 1998;64:262 –6. [24] Meara J, Mitchelmore E, Hobson P. Use of the GDS-15 geriatric depression scale as a screening instrument for depressive symptomatology in patients with Parkinson’s disease and their carers in the community. Age Ageing 1999;28:35–8. [25] Hantz P, Caradoc-Davies G, Cardoc-Davies T, Weatherall M, Dixon G. Depression in Parkinson’s disease. Am J Psychiatry 1994;151: 1010–4. [26] de Rijk C, Bijl RV. Prevalence of mental disorders in persons with Parkinson’s disease. Ned Tijdschr Geneeskd 1998;142:27–31. [27] Yesavage JA, Brink TL. Development and validation of a geriatric depression screening scale: a preliminary report. J Psychiat Res 1983; 17:37–49. [28] Brink TA, Yesavage JA, Lum O, Heersema P, Adey M, Rose TL. Screening test for geriatric depression. Clin Gerontol 1982;1:37–44. [29] Errea JM. Ara JR.Depression and Parkinson’s disease. Rev Neurol 1999;28:694 –8.

[30] Caap-Ahlgren M, Dehlin O. Insomnia and depressive symptoms in patients with Parkinson’s disease. Relationship to health-related quality of life. An interview study of patients living at home. Arch Gerontol Geriatr 2001;32:23–33. [31] Hugues A, Ben-Shlomo Y, Daniel SE. What features improve the accuracy of clinical diagnosis in Parkinson’s disease: a clinico pathologic study. Neurology 1992;42:1142 –6. [32] Zesiewicz TA, Gold M, Chari G, Hauser RA. Current issues in depression in Parkinson’s disease. Am J Geriatr Psychiatry 1999;7: 110 –8. [33] Gotham AM, Brown RG, Marsden CD. Depression in Parkinson’s disease: a quantitative and qualitative analysis. J Neurol Neurosurg Psychiatry 1986;31:645–56. [34] Brown RG, Mac Carthy B. Psychiatric morbidity in patients with Parkinson’s disease. Psychol Med 1990;20:77–87. [35] Dooneief G, Mirabello E, Bell K, Marder K, Stern Y, Mayeux R. An estimate of the incidence of depression in idiopathic Parkinson’s disease. Arch Neurol 1992;49:305 –7. [36] Allen-Burge R, Storandt M, Kinscherf DA, Rubin EH. Sex differences in the sensitivity of two self-report depression scales in older depressed inpatients. Psychol Aging 1994;9:443–5. [37] Mayeux R, Williams JBW, Stern Y, Cote L. Depression in Parkinson’s disease. Adv Neurol 1984;40:241–50. [38] Starkstein SE, Petracca G, Chemerinski E, Teson A, Sabe L, Merello M, Leiguarda R. Depression in classic versus akinetic-rigid Parkinson’s disease. Mov Disord 1998;13:29– 33. [39] Huber SJ, Freidenberg DL, Paulson GW, Shutlworth JA. The pattern of depressive symptoms varies with progression of Parkinson’s disease. J Neurol Neurosurg Psychiatry 1990;53(275):278. [40] Brown GL, Wilson WP. Parkinsonism and depression. South Med J 1972;65:540–5. [41] Kuhn W, Heye N, Mu¨ller T, Kraus P, Klotz P, Friedrich B, Welter FL, Przuntek H. The motor performance test series in Parkinson’s disease is influenced by depression. J Neural Transm 1996;103:349 –54. [42] Starkstein SE, Brethier ML, Bolduc PL, et al. Depression in patients with early versus late onset of Parkinson’s disease. Neurology 1989; 39:1441–5. [43] Larsen JP, Karlsen K, Tandberg E. Clinical problems in nonfluctuating patients with Parkinson’s disease: a community-based study. Mov Disord 2000;15:826 –9. [44] Starkstein SE, Preziosi TJ, Bolduc PL, Robison RG. Depression in Parkinson’s disease. J Nerv Ment Dis 1990;178:27– 31. [45] Troster AI, Stalp LD, Paolo AM, Fields JA, Koller WC. Neuropsychological impairment in Parkinson’s disease with and without depression. Arch Neurol 1995;52:1164–9. [46] Rogers D, Lees AJ, Smith E, Trimble M, Stern GM. Bradyphrenia in Parkinson’s disease and psychomotor retardation in depressive illness. Brain 1987;110:761 –6. [47] Brown RG, Mac Carthy B, Gotham Am, Der GJ, Marsden CD. Depression and disability in Parkinson’s disease: a follow up of 132 cases. Psychol Med 1988;18:49–55.