Acknowledgments: This study was supported in part by theNational Health Research Institutes, Taiwan (NHRI-EX93-9133NI). We thank Dr. L.M. Liu for assistance in neurologicalexamination.

APPENDIX

Single Nucleotide Polymorphism

Sequence variations exist at defined positions withingenomes and are responsible for individual phenotypiccharacteristics, including a person’s propensity towardcomplex disorders such as heart disease and cancer. Astools for understanding human variation and moleculargenetics, sequence variations can be used for gene map-ping, definition of population structure, and performanceof functional studies.

Hairpin-Mediated Strand Slippage

During the replication of DNA, repeat motifs can buildslipped–stranded structures or hairpins during movement ofpolymerase. In the replicating DNA, thermodynamicallystable hairpins may build in trinucleotide repeat regions,which bypass the polymerase. Conversely, during the syn-thesis of lagging strand, formation of hairpins in Okazakistrands, slippage and realignment of the fragments, maycause large expansions.

Cis or Trans Genetic Effects

Cis implies that the genetic regions in questionreside on the same chromosome as a part of a contig-uous region of DNA, such as promoter, enhancer, orstructures within the gene. On the other hand, transgenetic factors are factors such as transcription factorsthat are not on the same chromosomes and are prod-ucts of the genes located remotely, but may influencethe function of the gene.

REFERENCES1. Pulst S-M, Nechiporuk A, Nechiporuk T, et al. Moderate expansion

of a normally biallelic trinucleotide repeat in spinocerebellar ataxiatype 2. Nat Genet 1996;14:269–276.

2. Lu C-S, Chou Y-HW, Kuo P-C, Chang H-C, Weng Y-H. Theparkinsonian phenotype of spinocerebellar ataxia type 2. Arch Neu-rol 2004;61:35–38.

3. Shan DE, Soong BW, Sun CM, Lee SJ, Liao KK, Liu RS. Spino-cerebellar ataxia type 2 presenting as familial levodopa-responsiveparkinsonism. Ann Neurol 2001;50:812–815.

4. Gwinn-Hardy K. When is ataxia not ataxia? Arch Neurol 2004;61:25–26.5. Shan DE, Liu RS, Sun CM, Lee SJ, Liao KK, Soong BW. Presence

of spinocerebellar ataxia type 2 gene mutation in a patient withapparently sporadic Parkinson’s disease: clinical implications. MovDisord 2004;19:1357–1360.

6. Kock N, Muller B, Vieregge P, et al. Role of SCA2 mutations inearly- and late-onset dopa-responsive parkinsonism. Ann Neurol2002;52:257–258.

7. Subramony SH, Hernandez D, Adam A, et al. Ethnic differences in theexpression of neurodegenerative disease: Machado-Joseph disease inAfricans and Caucasians. Mov Disord 2002;17:1068–1071.

Osteoporosis in Parkinson’sDisease

Brian Wood, MD, FRCP,* andRichard Walker, MD, FRCP

Northumbria Healthcare NHS Trust, Wansbeck GeneralHospital, Woodhorn Lane, Ashington, Northumberland,

United Kingdom

Abstract: There are few studies of osteoporosis in Parkin-son’s disease (PD). We assessed the prevalence of osteopo-rosis in a PD clinic cohort. All subjects with a confirmeddiagnosis of PD attending a clinic were invited to partici-pate. All consenting subjects had bone density measured bydual energy X-ray absorptiometry scanning. Further data,including demography, disease duration, and disease sever-ity, were collected. One hundred five subjects participated;median age was 75 (54–92) years. Fifty-one (49%) patientswere men. Of the men: median T score, �1.3 (range, �4.7to 3.8); median Z score, 0.0 (�3.2 to 4.7); diagnostic cate-gories: osteoporosis, 20%; osteopenia, 41%; normal, 39%.Of the women: median T score �2.7 (�4.7 to 1.4); medianZ score, �0.25 (�2.6 to 4.2); diagnostic categories: osteo-porosis, 63%; osteopenia, 28%; and normal, 9%. Wholesample: osteoporosis, 42%; osteopenia, 34%; and normal,24%. There were associations between age, depression, dis-ease duration, and osteoporosis but not with disease sever-ity. Female gender was an independent predictor of osteo-porosis. The prevalence of osteoporosis/osteopenia isconsiderable in PD patients but does not exceed that ofother people of similar age. Osteoporosis/osteopenia waspresent in almost all women of this age group with PD.© 2005 Movement Disorder Society

Key words: osteoporosis, Parkinson’s disease, bonedensitometry

We assessed the prevalence of osteoporosis in a cohortof patients attending a Parkinson’s disease (PD) clinic ata District General Hospital and assessed factors associ-ated with the presence of osteoporosis. Osteoporosis is asystemic skeletal disease characterized by low bone massand micro-architectural deterioration, with a consequentincrease in bone fragility and fracture susceptibility.1

Deterioration of bone mass accompanies the normal age-ing process in both sexes but tends to begin earlier andmore aggressively in postmenopausal women.

*Correspondence to: Dr. Brian Wood, Northumbria Healthcare NHSTrust, Wansbeck General Hospital, Woodhorn Lane, Ashington,Northumberland NE63 9JJ, United Kingdom.E-mail: brian.wood@northumbria-healthcare.nhs.uk

Received 24 January 2005; Revised 14 March 2005; Accepted 28March 2005

Published online 17 August 2005 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/mds.20643

1636 B. WOOD AND R. WALKER

Movement Disorders, Vol. 20, No. 12, 2005

The current “gold standard” method for diagnosingand assessing osteoporosis is bone mineral density mea-surement with dual energy X-ray absorptiometry(DEXA). Results are reported as statistical comparisonswith mean bone density; Z scores compare bone densityto an age-matched population, whereas T scores comparebone density to the young adult population. According tothe World Health Organization (WHO) criteria, a T scoreof � �2.5 (i.e., 2.5 standard deviations below the normalmean for the adult population) is defined as osteoporosisand between �2.5 and �1 is defined as osteopenia.Therefore, T scores are used in clinical practice. A de-crease of one standard deviation of bone density has beenshown to increase the risk of femoral neck fracture by afactor of 2.7 (95% confidence intervals 1.6–4.6).2

The earlier studies investigated whether levodopa af-fected bone mass (through postulated effects on growthhormone secretion) but had very small numbers and didnot use bone densitometry.3,4 The first study to usedensitometry in PD was carried out in Japan.5 Metacarpalbone density was assessed rather than measurements atthe hip or spine, which have greater clinical relevance.Only Z scores were reported making comparisons withmore recent studies difficult, but the results demonstratedthat 54% of women and 26% of men were osteopenic. AChinese study used DEXA to assess osteoporosis in 22(mostly men) PD patients and demonstrated significantlevels of osteoporosis in the spine.6 In 1995, Taggart andCrawford compared bone mineral density in 55 PD pa-tients with age- and sex-matched healthy controls.7 Therewas a small but significant difference in hip bone densitybetween the two groups but no difference in spine den-sity. The presence of osteoporosis, which is arguablymore clinically relevant, was not reported. A study fromSpain8 using three different methods to assess bone massin PD showed a tendency toward reduced bone mass inPD patients compared to controls. A more recent Japa-

nese study9 investigated both metacarpal bone mineraldensity and indices of bone turnover in 71 patients withPD and 33 controls. There was a significant progressivereduction of Z score and decreased serum levels ofvitamin D in more severe disease. The authors believedthat hypovitaminosis D was more important in causingdecreased bone mineral density than immobilization.Therefore, the evidence is inconclusive as to exactlywhat the prevalence of osteoporosis is in PD and whetherosteoporosis is more common in PD patients than thenormal population.

PATIENTS AND METHODS

As part of a study investigating falls in idiopathicPD,10 patients were recruited from a database of patientsattending a specialist PD clinic at a District GeneralHospital. All patients on the database were approachedunless they were permanently bed fast or had potentiallylife-threatening major medical conditions. Diagnosis wasconfirmed using United Kingdom Parkinson’s DiseaseSociety Brain Bank Criteria.11 As part of the falls study,subjects underwent intensive multidisciplinary assess-ment, including Unified Parkinson’s Disease RatingScale (UPDRS),12 Hoehn and Yahr staging,13 Mini-Men-tal State Examination (MMSE),14 and Tinetti scores.15

Local ethical committee approval was obtained. Within 2weeks of this initial assessment, bone densitometry wasperformed. A single trained nurse undertook all assess-ments using a HOLOGIC QDR-4500A scanner. The

FIG. 1. T scores at the hip. N � 105.00; mean � �1.5; SD � 1.32.

TABLE 1. Z and T scores for the hip and spine

T score(hip)

T score(spine)

Z score(hip)

Z score(spine)

Whole sample 105 105 97 97Median �1.4 �1.7 �0.1 �0.1Minimum �5.3 �4.7 �3.3 �3.2Maximum 2.1 3.8 2.8 4.7

Men 51 51 49 49Median �0.8 �1.3 0.0 0.0Minimum �5.3 �4.7 �3.3 �3.2Maximum 2.1 3.8 2.8 4.7

Women 54 54 48 48Median �2.1 �2.7 �0.5 �0.25Minimum �5.3 �4.7 �3.3 �2.6Maximum 0.5 1.4 2.8 4.2

OSTEOPOROSIS IN PARKINSON’S DISEASE 1637

Movement Disorders, Vol. 20, No. 12, 2005

bone mineral content and area were measured at thefemoral neck, trochanteric and intertrochanteric regions,and in the first four lumbar vertebrae. The three scoresfor the hip and four scores for the spine were addedtogether, and the bone mineral content was then dividedby the area to produce separate bone mineral densityvalues for the hip and spine. From these values, T and Zscores were derived from normative data. Associationsof osteoporosis with the other baseline data collectedwere performed using standard statistical tests. Indepen-dent sample t tests were carried out for normally distrib-uted variables. Skewed baseline data were comparedwith the presence of osteoporosis using Mann–WhitneyU tests. Categorical variables were compared with thepresence of osteoporosis by using Chi-squared test or,where appropriate, Fisher’s exact test.

RESULTS

From 141 patients attending the PD Service (exclud-ing the bedridden/medically unstable patients), 105 con-sented to take part in the study. The median age was 75(range, 54–92) years, 47.8% were men, and 91% wereliving independently. The sample had a median of 3years since diagnosis of PD (range, 0–31), with a medianUPDRS score of 33 (range, 8–64) and a median modi-fied Hoehn and Yahr rating of 2.0 (range, 1.0–4.0). Eightof the subjects were under the age for a Z score to bequoted. The median scores along with their ranges areoutlined in Table 1. The results shown are for both thehip and spine.

Concentrating on the T scores, Figures 1 and 2 dem-onstrate the absolute values of the T scores at both thehip and the spine in the entire sample. The scores for thehip are normally distributed, with both the mean andmedian score being well into the osteopenic range. Thescores for the spine are slightly skewed toward the os-teoporotic levels, and the mean and median T scores arelower for the spine than the hip.

Forty-four subjects (41.9%) met the WHO criteria forthe diagnosis of osteoporosis. A further 36 (34.3%) metthe criteria for the diagnosis of osteopenia.

Ten men (19.6%) had osteoporosis, and a further 21(41%) had osteopenia. In the female group, there were 34cases of osteoporosis (63%) and a further 15 cases ofosteopenia (28%). Only five women (9%) had normalbone density.

For the entire group, there was a statistically signifi-cant association between increasing age, elevated geriat-ric depression score, higher Hoehn and Yahr score, andtotal Tinetti gait and balance score with osteoporosis.There was a positive association with years since diag-nosis and a strongly positive association with being fe-

male. These data are displayed in Tables 2 and 3. Logis-tic regression was carried out to investigate predictors ofosteoporosis. All predictors were entered into a singlemodel. The only independently predicting variable wasbeing female. Women with PD were seven times morelikely to have osteoporosis than men (95% confidenceinterval, 2.9–16.9).

DISCUSSION

As far as the authors are aware, this report is thelargest series of PD patients who have undergone bonedensitometry and is, as such, an important addition to thelimited evidence base of osteoporosis in PD. The resultsof previous studies of the prevalence of osteoporosis inPD are inconsistent. Not all studies have used the samemethod of assessment or the WHO definitions.

The authors acknowledge that one limitation of thestudy was the lack of a control group, but it was believedthat bone densitometry inherently has its own controlsowing to the method of the statistical measurement of Tand Z scores. As such, a mean Z score of 0 would equateto an equivalent age-matched population.

The overwhelming finding in this study, and an im-portant clinical message, is that over three quarters of allthe subjects had abnormal bone density (osteopenia 34%,osteoporosis 42%). Over 90% of the women in this studyhad osteoporosis or osteopenia. The frailest individualsattending the PD service were excluded or refused toparticipate, and it can be reasonably extrapolated thatthey would have an even greater prevalence of osteopo-

FIG. 2. T scores at the spine. N � 105.00; mean � �1.7; SD � 1.78.

1638 B. WOOD AND R. WALKER

Movement Disorders, Vol. 20, No. 12, 2005

rosis. Although it is not surprising that abnormal bonedensity is more common in women (owing to the post-menopausal decrease in estrogen), such a high preva-lence has not been demonstrated previously in studies ofosteoporosis in PD.

It is interesting to note that T scores are lower in thespine than the hip for this group of PD patients. It wouldbe expected for T scores to be normally distributed atboth the hip and spine. This finding could perhaps beassociated with decreased axial mobility in PD and mer-its further study.

There are univariate associations in this study betweenthe presence of osteoporosis and increased age, depres-sion, impairment of gait and balance, and disease dura-tion (but not severity). There did not seem to be anassociation between the severity of the disease and thepresence of osteoporosis. The association with diseaseduration, but not severity, is particularly interesting. Al-though typically disease severity and duration will bewell correlated, some aspects of these two variables maydiffer. Perhaps a longer period of mild immobility(which could occur with a longer duration of PD that is

not too severe) adversely affects bone mass more thansevere disease of short duration. This is an area thatmerits further study. It is also interesting that there wasno association with total L-dopa dose.

It is very important to note that the median Z scores forthe entire sample for both the hip and spine were veryclose to zero. Therefore, the prevalence of osteoporosisis similar to other people of a similar age. As such,osteoporosis in PD may not be more common than in thegeneral population. The absolute prevalence of osteopo-rosis, however, is undoubtedly very high in this sample.Osteoporosis is often described as a “silent epidemic”and is underdiagnosed in older people. An assessment ofa general elderly faller should include assessment ofosteoporosis, and it could be argued that the same wouldhold true for a group of patients with PD, since this groupis particularly prone to falling.10 Appropriate assessmentand/or treatment of osteoporosis should be considered inall patients with idiopathic PD. In the future, it would behelpful to see further studies specifically looking at ap-propriate treatment strategies for osteoporosis in individ-uals with PD and decreased bone density.

Key Learning Points

Osteoporosis is common in idiopathic PD. Bone den-sitometry should be considered in all patients with PD.Appropriate treatment or referral to a specialist serviceshould be considered in PD patients suffering fromosteoporosis.

Acknowledgments: We thank Janet Hunter and Dr. ShelaghBaillie for performing and analyzing the bone densitometry,and Dr. Fraser Birrell for his review of the final draft of thepaper.

TABLE 2. Univariate analyses of association with osteoporosis

Continuous variables Osteoporosis Number Mean SD Mean difference Sig.

Age (yr) N 61 73.02 8.50 �2.89 0.043Y 44 75.91 5.99

Total UPDRS N 61 33.51 11.70 �1.58 0.483Y 44 35.09 10.89

Total GDS N 61 5.2 3.10 �1.39 0.034Y 44 6.59 3.51

Total Tinetti N 59 35.15 7.75 �3.31 0.043Y 43 38.47 8.46

Levodopa dose N 61 434.4 267.8 15.1 0.764Y 44 419.32 231.3

Years since diagnosis N 61 Mean rank � 53.26 0.043Y 44 Mean rank � 52.64

MMSE N 61 Mean rank � 54.02 0.483Y 44 Mean rank � 51.58

UPDRS, Unified Parkinson’s Disease Rating Scale; GDS, Hoehn and Yahr General Disability Scale; MMSE,Mini-Mental State Examination.

TABLE 3. Analysis of categorical variables associated withosteoporosis

Categorical variablesOsteoporosis

absentOsteoporosis

present �2 Significance

Sex 20.25 �0.001Male 41 10Female 20 34

Dopamine agonist 3.46 0.072N 47 40Y 14 4

Hoehn and Yahr �3 0.03 0.863N 48 34Y 13 10

OSTEOPOROSIS IN PARKINSON’S DISEASE 1639

Movement Disorders, Vol. 20, No. 12, 2005

REFERENCES

1. Anonymous. Diagnosis, prophylaxis and treatment of osteoporosis.Am J Med 1991;90:170–210.

2. Greenspan SL, Myers ER, Maitland LA, Resnick NM, Hayes WC.Fall severity and bone mineral density as risk factors for hipfracture in ambulatory elderly. JAMA 1994;271:128–133.

3. Rico H, Vazquez A, Cabranes JA, et al. Long-term influence oflevodopa on bone mass and growth hormone in postmenopausalwomen with Parkinson’s disease. Clin Neuropharmacol 1987;10:87–91.

4. Rubinacci A, Scotti A, Tessari L. The effect of long-term treatmentwith L-dopa on bone mass of postmenopausal women. CalcifTissue Int 1983;35(Suppl.):A4.

5. Ishizaki F, Harada T, Katayama S, Abe H, Nakamura S. Relation-ship between osteopenia and clinical characteristics of Parkinson’sdisease. Mov Disord 1993;8:507–511.

6. Kao CH, Chen CC, Wang SJ, Chia LG, Yeh SH. Bone mineraldensity in patients with Parkinson’s disease measured by dualphoton absorptiometry. Nucl Med Commun 1994;15:173–177.

7. Taggart H, Crawford V. Reduced bone density of the hip in elderlypatients with Parkinson’s disease. Age Ageing 1995;24:326–328.

8. Revilla M, de la Sierra G, Aguado F, et al. Bone mass in Parkin-son’s disease: a study with three methods. Calcif Tissue Int 1996;58:311–315.

9. Sato Y, Kikuyama M, Oizumi K. High prevalence of vitamin Ddeficiency and reduced bone mass in Parkinson’s disease. Neurol-ogy 1997;49:1273–1278.

10. Wood BH, Bilclough JA, Bowron A, Walker RW. Incidence andprediction of falls in Parkinson’s disease: a prospective multidis-ciplinary study. J Neurol Neurosurg Psychiatry 2002;72:721–725.

11. Hughes AJ, Daniel SE, Kilford L, Lees AJ. Accuracy of clinicaldiagnosis of idiopathic Parkinson’s disease: a clinico-pathologicalstudy of 100 cases. J Neurol Neurosurg Psychiatry 1992;55:181–184.

12. Fahn S, Elton RL. Unified Parkinson’s Disease Rating Scale. In:Fahn S, Marsden CD, Calne D, Goldstein M, editors. Recentdevelopments in Parkinson’s disease. New York: MacmillanHealth Care Information; 2000. p 153–163.

13. Hoehn MM, Yahr MD. Parkinsonism: onset, progression and mor-tality. Neurology 1967;17:427–442.

14. Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”. Apractical method for grading the cognitive state of patients for theclinician. J Psychiatr Res 1975;12:189–198.

15. Tinetti ME. Performance-oriented assessment of mobility prob-lems in elderly patients. J Am Geriatr Soc 1986;34:119–126.

Dopaminergic Influence onDisturbed Spatial Discrimination

in Parkinson’s Disease

Hae-Won Shin, MD,1 Suk Y. Kang, MD,1

and Young H. Sohn, MD1,2*

1Department of Neurology and Brain Research Institute,Yonsei University College of Medicine, Seoul, Korea; 2Brain

Korea 21 Project for Medical Science, Yonsei UniversityCollege of Medicine, Seoul, Korea

Abstract: Various sensory symptoms and disturbed sensoryperception are often observed in patients with idiopathicParkinson’s disease (PD). The basis of sensory disturbancein PD is unknown but possibly reflects a role for the basalganglia in sensory processing. To investigate the relation-ship between the sensory dysfunction and dopaminergicdeficiency in PD, we measured spatial discrimination usingthe Grating Orientation Task in 21 drug-naive patientswith PD, before and after long-term antiparkinson therapy,and 25 age-matched healthy controls. The grating orienta-tion threshold was significantly higher in patients (3.03 �0.84) than controls (2.03 � 0.79). After 3 to 10 months ofantiparkinson therapy, the grating orientation thresholdwas significantly lowered (2.66 � 0.84), although it was stillhigher than that in controls. Improvement in the patients’sensory function was significantly correlated with motorimprovement (r � 0.44). These results suggest that sensorydysfunction in Parkinson’s disease is related at least in partto the dopaminergic deficit. © 2005 Movement DisorderSociety

Key words: Parkinson’s disease; sensation; dopamine; grat-ing orientation threshold

Various sensory symptoms often accompany idio-pathic Parkinson’s disease (PD), and they are usuallyproportional to the degree of motor dysfunction.1 Inaddition, disturbed sensory perception has been observedin patients with PD.2,3 The basis of sensory disturbancein PD is unknown but possibly reflects a role for the basalganglia in sensory processing.2,3 To investigate the rela-tionship between the sensory dysfunction and dopami-nergic deficiency in PD, we measured spatial discrimi-nation using the Grating Orientation Task in drug-naive

*Correspondence to: Dr. Young H. Sohn, Department of Neurology,Yonsei University Medical Center, 134 Shinchon-dong, Seodaemun-gu, Seoul 120-752, South Korea.E-mail: yhsohn62@yumc.yonsei.ac.kr

Received 18 December 2004; Revised 23 March 2005; Accepted 30March 2005

Published online 9 August 2005 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/mds.20642

1640 H.-W. SHIN ET AL.

Movement Disorders, Vol. 20, No. 12, 2005