Deep Brain Stimulation May Improve Quality of Life in People With Parkinson's Disease Without Affecting Caregiver Burden

Deep Brain Stimulation May Improve Quality ofLife in People With Parkinson’s Disease WithoutAffecting Caregiver BurdenGenko Oyama, MD, PhD*; Michael S. Okun, MD*†; Peter Schmidt, PhD‡;Alexander I. Tröster, PhD§; John Nutt, MD¶; Criscely L. Go, MD*;Kelly D. Foote, MD*†; Irene A. Malaty, MD*; on behalf of the NPF-QIIInvestigators

Objective: This study aims to investigate the influence of deep brain stimulation (DBS) on caregiver burden and quality of life inParkinson’s disease.

Methods: A cross-sectional retrospective study utilizing the National Parkinson Foundation Quality Improvement Initiative clini-cal study was conducted. A group of 275 patients who had undergone DBS for Parkinson’s disease were extracted from 2916subjects who were included in this data base. The data were compared to an age, sex, and disease severity matched control group.A secondary analysis was then performed on two more control groups that were matched to account for presence or absence ofmotor fluctuations. The multidimensional caregiver strain index and Parkinson’s disease quality-of-life questionnaire 39 summaryindex were compared.

Results: The multidimensional caregiver strain index did not differ between the DBS group (16.9 � 11.8) and a matched non-DBSgroup (16.1 � 17.6, p = 0.618). The quality-of-life index was, however, significantly better in the DBS group (28.9 � 15.6) than in thenon-DBS group (32.3 � 17.6, p = 0.034). A secondary analysis revealed that the total caregiver strain score was lower in the nomotor fluctuation control group than the other two groups (p < 0.05). Regression analysis revealed significant relationshipsbetween the quality-of-life index and caregiver strain index total scores (p < 0.001), between caregiver strain index total score andage at surgery (p = 0.027), and also between the interval since surgery (p = 0.048).

Conclusions: Although there were several limitations to this study, DBS seems to improve quality of life without significantlyincreasing caregiver burden.

Keywords: Caregiver burden, caregiver strain, DBS, Parkinson’s disease, QOL

Conflict of Interests: The authors report no conflict of interest.

Address correspondence to: Michael S. Okun, MD, Department of Neurology, McKnight Brain Institute, 100 S Newell Dr, Room L3-101, Gainesville, FL 32610, USA.Email: [email protected]

* Department of Neurology, University of Florida Center for Movement Disorders and Neurorestoration, Gainesville, FL, USA;† Department of Neurosurgery, University of Florida Center for Movement Disorders and Neurorestoration, Gainesville, FL, USA;‡ National Parkinson Foundation, Miami, FL, USA;§ Department of Neurology, University of North Carolina School of Medicine, Chapel Hill, NC, USA; and¶ Department of Neurology, Oregon Health & Science University, Portland, OR, USA

For more information on author guidelines, an explanation of our peer review process, and conflict of interest informed consent policies, please go to http://www.wiley.com/bw/submit.asp?ref=1094-7159&site=1Study funding: Supported by National Parkinson FoundationFinancial disclosures: Dr. Oyama had received grant support for fellowship from Medtronic. Dr. Okun serves as a consultant for the National Parkinson Foundation and hasreceived research grants from National Institutes of Health (NIH), NPF, the Michael J. Fox Foundation, the Parkinson Alliance, Smallwood Foundation, and the UFFoundation. Dr. Okun has in the past >36 months received no support from industry including travel. Dr. Okun has received royalties for publications with Demos, Manson,and Cambridge (movement disorders books). Dr. Okun has participated in Continuing Medical Education (CME) activities on movement disorders sponsored by theUniversity of South Florida CME office, PeerView, and by Vanderbilt University. The institution and not Dr. Okun receives grants from Medtronic and AdvancedNeuromodulation Systems (ANS)/St. Jude, and the PI has no financial interest in these grants. Dr. Okun has participated as a site PI and/or co-PI for several NIH, foundation,and industry sponsored trials over the years but has not received honoraria. Dr. Foote has participated in Medtronic-sponsored DBS teaching courses (prior to June 2009)and Medtronic-sponsored DBS fellowship training programs through grants to the institution. Dr. Tröster has served as consultant to Medtronic, St. Jude Medical andBoston Scientific. Dr. Tröster has served grants from Medtronic Speaker fees from Medtronic and St Jude. Dr. Malaty has received grant support from the NationalParkinson Foundation (NPF). Dr. Nutt has received grant support from Schering-Plough (now Merck) and has served as consultant to XenoPort Inc., Impax Laboratories,Neurogen Inc., Synosia, Neuroderm Ltd, Merck, Lilly/Medtronics, Elan Pharmaceuticals, Addex Pharma SA, Lundbeck Inc., Merz Pharmaceuticals, and SynAgile Corp. Drs.Go and Schmidt report no disclosures.

Neuromodulation: Technology at the Neural Interface

Received: November 18, 2012 Revised: April 30, 2013 Accepted: May 22, 2013

(onlinelibrary.wiley.com) DOI: 10.1111/ner.12097

126

www.neuromodulationjournal.com Neuromodulation 2014; 17: 126–132© 2013 International Neuromodulation Society

mailto:[email protected]

http://www.wiley.com/bw/submit.asp?ref=1094-7159&site=1

http://www.wiley.com/bw/submit.asp?ref=1094-7159&site=1

INTRODUCTION

Deep brain stimulation (DBS) is becoming a common option for care-fully selected patients with advanced Parkinson’s disease (PD) (1,2). DBShas been widely considered as an option for patients with intractabletremor, for those patients who are affected by long-term complicationsof levodopa therapy (e.g., motor fluctuations and dyskinesias), or forthose who cannot tolerate medications due to side effects (1). Evidenceof DBS efficacy, tolerability, and its positive impact on quality of life (QOL)has been reported by several studies (1–15). The impact of DBS on thecaregiver has not, however, been well studied. Recent studies reveal thatgreater motor symptom severity and presence of neuropsychiatricsymptoms both increase caregiver burden in PD (4,16,17). Thus, onemight predict caregiver strain to decline after a partner’s successful DBS.Contrary to expectation, there has been an expanding concern that DBSmay amplify caregiver strain due to the increased time demands (e.g.,medical appointments for device programming and maintenance) andthe increased financial burden (e.g., need for full-time care and insurancecopayments) (18). We aimed to more carefully investigate the influenceof DBS on caregiver burden in PD by utilizing a large dataset acquiredthrough the National Parkinson Foundation (NPF) Quality ImprovementInitiative (QII) Study (19). The Multidimensional Caregiver Strain Index(MCSI) (20) and Parkinson’s disease quality of life questionnaire (PDQ-39)(21) were routinely administered in the QII study dataset, and thereforeoffered a unique opportunity to more carefully examine the issue ofcaregiver strain and its relationship to DBS. We tested two primaryhypotheses in this study. First, the MCSI score would be higher (worse) inpatients treated with DBS than in those not treated with DBS and,second, that the PDQ-39 score would be lower (better) in patients treatedwith DBS than those not treated with DBS.

METHODSDesign

This cross-sectional study drew data from the NPF QII ClinicalStudy (19). The data were obtained from 16 participating NPFCenters of Excellence within the United States.

SubjectsAll patients visited participating NPF Centers of Excellence and

were asked to participate to NPF QII Clinical Study. If they indicatedan interest in participating, informed consent was obtained.Included were all patients registered in the NPF QII (19) between2009 and 2010, who had a regular caregiver, and who were living athome (not in nursing home). A caregiver was defined as any personwho lived with the patient and held direct responsibility for ensur-ing care delivery to the patient. All patients included were diag-nosed with PD by a movement disorders expert. The data basequery yielded a total of 2916 of total records (these included 330patients with DBS). From these 2916 cases, 35 were excluded for lackof crucial information (age, diagnosis, the presence of DBS, informa-tion of caregiver or living place), 123 were excluded because thepatients did not dwell at home (were in a nursing home), and 429were excluded because the patients did not have a caregiver or hada paid caregiver. A total of 2329 subjects had potentially viablerecords to be included in this study.

MeasurementsThevariablesincludedgender,age,ageatonset,ageatsurgery,disease

duration, Hoehn andYahr stage, interval from the surgery, weight, height,

timedup&go, immediate5wordrecall,delayed5wordrecall,presenceofmotor fluctuation, and the PDQ-39 scores and the MCSI scores, and allwere extracted from the QII data base. The MCSI is a validated scalemeasuring six dimensions of subjective caregiver strain, including physi-calstrain,socialconstraints,financialstrain,timeconstraints, interpersonalstrain, and elder demanding/manipulative issues (20). It consists of 18items, each rated on a five-point scale, with higher numbers reflectingmore frequent occurrence of strain.The PDQ-39 is comprised of 39 itemseach of which indicates the extent to which a certain activity has beenimpededbyPDutilizingafive-pointscale(never,occasionally,sometimes,often, or always). The PDQ-39 measures “quality of life” in eight discretedomains (mobility, activity of daily livings (ADLs), emotional well-being,stigma, social support, cognition, communication, and pain). Scores foreach domain are expressed as a percentage (100 indicating greater dis-ruption and dissatisfaction with a domain). The PDQ-39 summary indexscore (PDQ39-SI) can be computed by summing the eight domain scoresand standardizing the score on a 0–100 scale (21).

Primary AnalysisFrom this cohort (a total of 2329 subjects), 275 patients were

identified as having DBS. Once the DBS group (N = 275) was formed,then an age, sex, and disease severity (Hoehn and Yahr stage)matched PD control group (age � two years and stage � 1) wasextracted from the data base. The total scores of the MCSI andPDQ-39 summary index were compared between the two groups.Thereafter, secondary comparative analysis of the different domainsof the MCSI and PDQ-39 were performed. A subanalysis on MCSIand PDQ39-SI between the unilateral DBS group and bilateral DBSgroup was conducted. To investigate potential confounds, physicalstatus (weight and height), motor function (timed up & go test,without pushing off from chair or using cane/walker), memory(immediate 5 word recall and delayed 5 word recall) also were com-pared between the two groups.

Secondary Analysis to Examine the Impact ofMotor Fluctuations

Because motor fluctuation is considered as a major troublesomesymptom that may worsen patient’s QOL and caregiver strain, weassessed the association of motor fluctuations, and then extractedtwo other age- and sex-matched control groups either with orwithout the presence of motor fluctuations. We compared a groupwith DBS (N = 275) with a group without DBS and without motorfluctuations (N = 275), and we also compared the DBS group with agroup without DBS but with motor fluctuations (N = 275). The MCSIand PDQ-39 were compared among the three groups. Finally, thepredictors of MCSI were analyzed for the DBS cases only.

Statistical AnalysisTwo primary hypotheses were tested (first, the MCSI score would

be higher in patients treated with DBS than those not treated withDBS and, second, the PDQ-39 score would be lower in patientstreated with DBS than those not treated with DBS). A Shapiro–Wilkstest was used to test normality. If the data were normally distrib-uted, then an unpaired t-tests (two-tailed) was used. Otherwise, thetwo groups’ means were compared using nonparametric Mann–Whitney U-tests. Chi-square tests were used for group heterogene-ity of sex, Hoehn and Yahr staging, and presence of motorfluctuation. For the secondary analysis, Kruskal–Wallis tests wereused to compare the groups. The Steel–Dwass test was performed

127DBS AND CAREGIVER BURDEN AND QOL IN PD


for multiple comparisons among individual groups. Pearson’s linearregression analyses were performed to evaluate the prediction ofcaregiver burden from age, age at onset, age at surgery, diseaseduration, interval from the surgery, and PDQ39-SI. The minimal levelof significance was fixed at p < 0.05. Considering a mean differenceamong groups of <5 points and a standard deviation of 15, theminimum sample size required to have an 80% probability to detecta statistically significant differences was 142 patients (alpha: 0.05).

RESULTSPrimary Analysis

Table 1 shows characteristics of the two groups from the primaryanalysis. The groups did not differ in presence of motor fluctuations.In the DBS group, disease duration was longer than in the non-DBSgroup. There were no differences between the two groups in physi-cal status (weight and height) and memory (immediate 5-wordrecall and delayed 5-word recall), but motor function (timed up andgo test) was better in the non-DBS group than in the DBS group. TheDBS targets included subthalamic nucleus (STN) (71 unilateral and75 bilateral), globus pallidus internus (GPi) (12 unilateral and 10bilateral), others (two unilateral and two bilateral), and unknown (87unilateral and 16 bilateral).

Figure 1a shows the results of the MCSI total score and thePDQ39-SI in for each group. There was no significant difference intotal MCSI score between the DBS group (16.9 � 11.8, 95% confi-dence interval [CI] 15.1, 18.4) and the non-DBS group (16.1 � 17.6,95% CI 30.2, 34.4, p = 0.618), whereas the PDQ39-SI was significantly

lower (indicating better QOL) in the DBS group (28.9 � 15.6, 95% CI27.1, 30.8) than in the non-DBS group (32.3 � 17.6, 95% CI 30.2, 34.4,p = 0.034).

The profile of specific domains of the MCSI and PDQ-39 for eachgroup was shown in Figure 1b. There were no differences betweenthe two groups in MCSI domains. There were differences in thePDQ-39 domains of emotional well-being (p < 0.001), stigma (p <0.001), cognition (p < 0.001), and pain (p = 0.031). There were nodifferences between the unilateral and bilateral DBS groups in MCSI(p = 0.956) and PDQ-39 SI (p = 0.485).

Secondary AnalysisTable 2 shows characteristics of the two additional control

groups. The non-DBS group without motor fluctuations revealed anearlier disease stage, a shorter disease duration than the other twogroups, better motor function (timed up and go test), and bettermemory function. The non-DBS group with motor fluctuationsrevealed a shorter disease duration than the DBS group but a longerdisease duration than the no motor fluctuation group. The DBSgroup revealed lower (worse) verbal fluency than the others.Figure 2a shows the results of MCSI total score and PDQ39-SI foreach group. There was a significant difference in total MCSI scorebetween the non-DBS without motor fluctuation group and theother groups (p < 0.05). The PDQ39-SI was significantly differentamong the groups (p < 0.05). Figure 2b shows the profile of specificdomains of the MCSI and PDQ-39. There were significant differencesin the MCSI domains of physical strain, social constraints, and timeconstraints between the no-motor fluctuation control and theothers (p < 0.05). Financial strain and presence of a demanding elderwere lower in the no-motor fluctuation control group than themotor fluctuation control group (p < 0.05). The PDQ-39 domains ofemotional well-being, stigma, social support, cognition, and painwere significantly higher in the motor fluctuation group than theothers (p < 0.05). Mobility and communication were significantlylower in the no-motor fluctuation group (p < 0.05). The ADLs werethe highest (worst) for the motor fluctuation control group, fol-lowed in order by the DBS group and then the no motor fluctuationcontrol group (p < 0.05).

Regression AnalysisThe results of Pearson’s linear regression analysis are summarized

in Table 3. There was significant positive relationship between the

Table 1. Primary analysis of DBS vs. non-DBS group

DBS group Non-DBS group P value

N 275 275Age (year) 62.6 � 8.8 62.6 � 8.8 0.961Sex (M/F) 2.8 2.8 1.000Hoehn & Yahr stage 2.8 � 0.7 2.8 � 0.8 0.542Disease duration (year) 15.0 � 6.3 12.3 � 6.2 <0.001Motor fluctuation + (No) 183 181 0.928Weight (kg) 183.8 � 39.5 178.6 � 39.3 0.090Height (cm) 67.8 � 4.2 67.8 � 4.1 0.647Timed up & go test 16.5 � 8.9 14.4 � 8.1 0.003Immediate 5-word recall 4.3 � 0.9 4.3 � 1.1 0.387Delayed 5 word recall 2.8 � 1.3 2.8 � 1.4 0.937

Table 2. Secondary analysis of DBS vs. a motor fluctuator and nonfluctuator group

DBS group Non-DBS group Non-DBS group P valuewithout motor fluctuation with motor fluctuation

N 275 275 275Age (year) 62.6 � 8.8 62.6 � 8.8 62.6 � 8.7 0.999Sex (M/F) 2.8 2.8 2.8 1.000Hoehn & Yahr stage 2.8 � 0.7 2.8 � 0.8* 2.9 � 0.7 <0.001Disease duration (year) 15.0 � 6.3* 9.0 � 6.2* 12.9 � 5.8* <0.001Motor fluctuation + (No) 183* 0* 275* <0.001Weight (kg) 183.8 � 39.5 184.8 � 42.1 178.1 � 40.9 0.097Height (cm) 67.8 � 4.2 68.1 � 3.9 67.7 � 4.5 0.631Timed up & go test (sec) 16.5 � 8.9 12.9 � 7.1* 15.4 � 8.6 <0.001Immediate 5 word recall 4.3 � 0.9 4.4 � 1.0* 4.1 � 1.1 0.008Delayed 5 word recall 2.8 � 1.3 2.9 � 1.4 2.7 � 1.4 0.036

128

OYAMA ET AL.


PDQ39-SI and MCSI total scores (p < 0.001). The PDQ39-SI showed apositive association with the disease duration (p = 0.018); however,the MCSI did not show a significant relationship with the diseaseduration (p = 0.135). The MCSI total score also was negatively pre-dicted by age at surgery (p = 0.027) and positively predicted byinterval since surgery (p = 0.048).

DISCUSSION

Contrary to our initial hypothesis, the main finding of this cross-sectional study was that caregiver strain did not differ between theDBS and non-DBS groups. On the other hand, consistent with ourhypothesis, QOL was better in the DBS than the non-DBS group.Although this could imply a causative relationship, particularly con-sidering the literature on effects of DBS on QOL (1–15), this can only

Figure 1. Comparisons DBS vs. non-DBS on caregiver strain and quality of life.(a) Total scores and (b) domains. DBS: deep brain stimulation. MCSI: The Multi-dimensional Caregiver Strain Index, PDQ-39: Parkinson’s disease quality of lifequestionnaire. *p < 0.05.

Figure 2. Fluctuator vs. nonfluctuator comparison on caregiver strain andquality of life. (a) Total scores and (b) domains. DBS: deep brain stimulation.MCSI: The Multidimensional Caregiver Strain Index, PDQ-39: Parkinson’s diseasequality of life questionnaire. *p < 0.05.



be speculative in the absence of presurgical data for comparison.The results do however suggest that DBS therapy may improve QOLof patients without increasing caregiver burden. This was reinforcedby the secondary analysis that accounted for the presence of motorfluctuations. The addition of a motor fluctuation control group,although not the ideal group, did offer a more comparable group ofpatients for analysis. Caregiver strain did not differ between the DBSgroup and the motor fluctuation group; however, the QOL in theDBS group was better than the motor fluctuation group.

Although we suspected that DBS would worsen caregiver strainprimarily due to increasing time and financial constraints, this sus-picion was not supported by the data. It is true however that toachieve excellent care of patients, DBS therapy requires frequentfollow-up visits for programming, hardware evaluation, and batterychecks, as well as attention to medications. Caregivers must under-stand how to check devices and how to avoid potentially compli-cating scenarios (e.g., avoiding a non head receive coil brain MRI andavoiding diathermy) (22). The data from this study suggest thatcaregiver strain was similar whether there was a DBS present or not;however, QOL was superior in the DBS group. The results of theregression analysis revealed that there was a significant relationshipbetween caregiver burden and QOL, though we were not able totease out this relationship with the current data set which revealedsimilar QOL but differences on MCSI.

Nonmotor symptoms in general have been correlated to car-egiver strain (17,23–28). Caregiver burden was previously associ-ated with disability, the presence of depression, hallucinations,and/or the presence of confusion (29). Caregiver strain has beenthought to be related to overall disease progression (28). Anotherfactor that could have affected our findings was the disease stage.D’Amelio et al. reported that PD severity (Hoehn and Yahr stage)also was associated with caregiver burden as well as mentalsymptoms (16).

Among the PDQ-39 domains, emotional well-being, stigma, andpain were better in the DBS group, which was consistent with pre-vious studies of STN or GPi-DBS (1,2,4–6,8–10,13), whereas mobilityand ADLs did not differ between the DBS group and non-DBS group.This latter finding contrasts to most existing studies that havereported that the PDQ-39 subscale data improve following STN orGPi-DBS especially in mobility and ADLs (1–6,8–10,13). The improve-ment of mobility and ADLs is thought to be potentially due to anoverall improvement in post-DBS motor function.

The main limitation of this study was its cross-sectional designand the limitations of the matched control groups (the variableshave not fleshed out disease duration). As the disease duration wassignificantly longer in the DBS group, one might consider that car-egiver strain would be more and that QOL would be worse in theDBS group, but the opposite was found. If the DBS group had actu-

ally been compared with a group with similar disease duration, thedifference may have been even greater favoring DBS. Our datarevealed that the MCSI score did not correlate to disease durationthough the PDQ39-SI was associated. Thus, the complexity of theconcept of QOL and of caregiver burden may also have impactedthe results, as many variables could have potentially contributed tothese measures. Although cognitive problems are one of the pre-dictors of caregiver burden, patients with cognitive problems areless likely to undergo DBS implantation. Additionally, as this studywas based on a patient-based questionnaire and as not all subjectswere actually implanted at the NPF center collecting the data, infor-mation on the DBS target was incomplete and could have beeninaccurate. Our original study design did not depend on the DBStarget; however, the relatively large number of unknown DBS targetsites might have impacted the results. Despite these limitations, thesample size was relatively large, and the study drew from 16 centersand was conducted in a real-world setting and not a clinical trialsetting. Although there is a small possibility of a type II error and wewill have to cautiously interpret these data, the overall resultssuggest that DBS therapy may improve QOL without increasingcaregiver burden. Factors that may contribute to caregiver strain inDBS patients should be clarified in a prospective longitudinalfollow-up study.

Disclosure/Acknowledgments

We would like to acknowledge the support of the NPF Center ofExcellence and the data derived from the NPF QII Clinical Study.

Authorship Statement

Dr. Okun helped to design and conduct the study and coordinateresults with all sides and helped in patient recruitment, data collec-tion, and analysis. All authors approve the final manuscript file (UFand NPF). Statistical support analyzing the data from Dr. Wu andfunding from the NPF. All authors had complete access to studydata.

How to Cite This ArticleOyama C., Okun M.S., Schmidt P., Tröster A.I., Nutt J., GoC.L., Foote K.D., Malaty I.A. 2014. Deep Brain StimulationMay Improve Quality of Life in People with Parkinson’sDisease without Affecting Caregiver Burden.Neuromodulation 2014; 17: 126–132

REFERENCES

1. Deuschl G, Schade-Brittinger C, Krack P et al. A randomized trial of deep-brainstimulation for Parkinson’s disease. N Engl J Med 2006;355:896–908.

2. Follett KA, Weaver FM, Stern M et al. Pallidal versus subthalamic deep-brain stimu-lation for Parkinson’s disease. N Engl J Med 2010;362:2077–2091.

3. Just H, Ostergaard K. Health-related quality of life in patients with advanced Parkin-son’s disease treated with deep brain stimulation of the subthalamic nuclei. MovDisord 2002;17:539–545.

4. Martinez-Martin P, Valldeoriola F, Tolosa E et al. Bilateral subthalamic nucleus stimu-lation and quality of life in advanced Parkinson’s disease. Mov Disord 2002;17:372–377.

5. Patel NK, Plaha P, O’Sullivan K, McCarter R, Heywood P, Gill SS. MRI directedbilateral stimulation of the subthalamic nucleus in patients with Parkinson’sdisease. J Neurol Neurosurg Psychiatry 2003;74:1631–1637.

Table 3. Pearson’s linear regression analysis.

The PDQ39-SI(P value)

The MCSI total score(P value)

The PDQ39-SI – R2 = 0.220 (<0.001)The MCSI total scores R2 = 0.220 (<0.001) –Age R2 = -0.004 (p = 0.990) R2 = 0.009 (p = 0.179)Age at onset R2 = 0.004 (p = 0.275) R2 = 0.016 (p = 0.080)Age at surgery R2 = 0.005 (0.262) R2 = 0.027 (0.027)Disease duration R2 = 0.017 (0.018) R2 = 0.012 (0.135)Interval since surgery R2 = 0.022 (0.048) R2 = 0.22 (p = 0.048)

130

OYAMA ET AL.


6. Lezcano E, Gomez-Esteban JC, Zarranz JJ et al. Improvement in quality of life inpatients with advanced Parkinson’s disease following bilateral deep-brain stimula-tion in subthalamic nucleus. Eur J Neurol 2004;11:451–454.

7. Troster AI, Fields JA, Wilkinson S, Pahwa R, Koller WC, Lyons KE. Effect of motorimprovement on quality of life following subthalamic stimulation is mediated bychanges in depressive symptomatology. Stereotact Funct Neurosurg 2003;80:43–47.

8. Drapier S, Raoul S, Drapier D et al. Only physical aspects of quality of life are signifi-cantly improved by bilateral subthalamic stimulation in Parkinson’s disease.J Neurol 2005;252:583–588.

9. Siderowf A, Jaggi JL, Xie SX et al. Long-term effects of bilateral subthalamic nucleusstimulation on health-related quality of life in advanced Parkinson’s disease. MovDisord 2006;21:746–753.

10. Zahodne LB, Okun MS, Foote KD et al. Greater improvement in quality of life follow-ing unilateral deep brain stimulation surgery in the globus pallidus as compared tothe subthalamic nucleus. J Neurol 2009;256:1321–1329.

11. Slowinski JL, Putzke JD, Uitti RJ et al. Unilateral deep brain stimulation of the sub-thalamic nucleus for Parkinson disease. J Neurosurg 2007;106:626–632.

12. Woods SP, Fields JA, Lyons KE et al. Neuropsychological and quality of lifechanges following unilateral thalamic deep brain stimulation in Parkinson’sdisease: a one-year follow-up. Acta Neurochir (Wien) 2001;143:1273–1277. discus-sion 1278.

13. Rodrigues JP, Walters SE, Watson P, Stell R, Mastaglia FL. Globus pallidus stimulationimproves both motor and nonmotor aspects of quality of life in advanced Parkin-son’s disease. Mov Disord 2007;22:1866–1870.

14. Okun MS, Fernandez HH, Wu SS et al. Cognition and mood in Parkinson’s disease insubthalamic nucleus versus globus pallidus interna deep brain stimulation: theCOMPARE trial. Ann Neurol 2009;65:586–595.

15. Williams A, Gill S, Varma T et al. Deep brain stimulation plus best medical therapyversus best medical therapy alone for advanced Parkinson’s disease (PD SURG trial):a randomised, open-label trial. Lancet Neurol 2010;9:581–591.

16. D’Amelio M, Terruso V, Palmeri B et al. Predictors of caregiver burden in partners ofpatients with Parkinson’s disease. Neurol Sci 2009;30:171–174.

17. Martinez-Martin P, Forjaz MJ, Frades-Payo B et al. Caregiver burden in Parkinson’sdisease. Mov Disord 2007;22:924–931. quiz 1060.

18. Okun MS, Rodriguez RL, Mikos A et al. Deep brain stimulation and the role of theneuropsychologist. Clin Neuropsychol 2007;21:162–189.

19. Okun MS, Siderowf A, Nutt JG et al. Piloting the NPF data-driven quality improve-ment initiative. Parkinsonism Relat Disord 2009;16:517–521.

20. Stull DE. The multidimensional caregiver strain index (MCSI): its measurement andstructure. J Clin Geropsychol 1996;2:175–196.

21. Jenkinson C, Fitzpatrick R, Peto V, Dummett S, Morley D, Saunders P. The Parkinson’sdisease questionnaires. User manual. (PDQ-39, PDQ-8, PDQ summary index & PDQ-carer), 3rd ed. Oxford: Isis Outcomes, 2012.

22. Tagliati M, Jankovic J, Pagan F, Susatia F, Isaias IU, Okun MS. Safety of MRI in patientswith implanted deep brain stimulation devices. Neuroimage 2009;47 (Suppl. 2):T53–T57.

23. Aarsland D, Larsen JP, Karlsen K, Lim NG, Tandberg E. Mental symptoms in Parkin-son’s disease are important contributors to caregiver distress. Int J Geriatr Psychiatry1999;14:866–874.

24. Aarsland D, Bronnick K, Ehrt U et al. Neuropsychiatric symptoms in patients withParkinson’s disease and dementia: frequency, profile and associated care giverstress. J Neurol Neurosurg Psychiatry 2007;78:36–42.

25. Meara J, Mitchelmore E, Hobson P. Use of the GDS-15 geriatric depression scale as ascreening instrument for depressive symptomatology in patients with Parkinson’sdisease and their carers in the community. Age Ageing 1999;28:35–38.

26. Happe S, Berger K. The association between caregiver burden and sleep distur-bances in partners of patients with Parkinson’s disease. Age Ageing 2002;31:349–354.

27. Thommessen B, Aarsland D, Braekhus A, Oksengaard AR, Engedal K, Laake K. Thepsychosocial burden on spouses of the elderly with stroke, dementia and Parkin-son’s disease. Int J Geriatr Psychiatry 2002;17:78–84.

28. Carter JH, Stewart BJ, Archbold PG et al. Living with a person who has Parkinson’sdisease: the spouse’s perspective by stage of disease. Parkinson’s Study Group. MovDisord 1998;13:20–28.

29. Schrag A, Jahanshahi M, Quinn N. What contributes to quality of life in patients withParkinson’s disease? J Neurol Neurosurg Psychiatry 2000;69:308–312.

COMMENTS

The authors are to be commended on an outstanding analysis uti-lizing data from the National Parkinson Foundation Quality Improve-ment Initiative.

Charles P. David, M.D.Nashville, TN, USA

***This is a cross-sectional, retrospective review study using the NPFQuality Improvement Initiative data on 275 patients who had under-gone DBS, comparing them with PD controls. This is an importantstudy as shows that DBS may improve quality of life without affectingcaregiver burden. One expected limitation on studies like this is to beable to produce a good control group that shares the same criticalcharacteristics (those that affect the primary and secondary endpoints) with the DBS group.

Diego Torres, M.D.Omaha, NE, USA

***Caregiver burden is an important issue in advancing Parkinson’sdisease and may be related to the complexity of the disease as well asthe complexity of Parkinson’s therapeutics. The impact on caregiverburden with Deep Brain Stimulation (DBS) has not been well addressedand it is this knowledge gap that this large, cross-sectional study aimsto address.

The authors report the comparison of 275 patients who underwentDBS against an equal sized control group who did not have DBS.Secondary analysis were made with 2 additional non-DBS controlgroups of 275 patients—one with fluctuations and one without. Theywere all extracted from a database of 2,329 patients and matched byage +/-2 years, Hoehn and Yahr stage +/-1, year and gender. UPDRSscores and disease duration were not matched.

The primary analysis was the Multidimensional Caregiver StrainIndex (MCSI) and quality of life with the PDQ-39. While they hypoth-esized that DBS would increase caregiver strain (due to more complexcare and appointments) and improve quality of life, they found thatcaregiver strain was not significantly different between groups and,quality of life was indeed improved with DBS.

Confounding the interpretation of this was that the study was cross-sectional and there was possible bias introduced by who was given ornot given DBS. Possibly reflecting this was the finding that the DBSgroup had significantly longer disease duration (15.0 yrs. vs. 12.3 yrs.).However, this did not result in any significant difference in rates ofmotor fluctuations between the groups. Further, this does not detractfrom the significance of better quality of life in the DBS group. It’s alsonotable that caregiver strain was equivalent between groups despitethe DBS group having longer disease duration. At a minimum, it seemsbelievable the DBS doesn’t worsen caregiver strain.

In order to further explore the confounds in this dataset due to itscross-sectional design they further compared the DBS group againstthe 2 additional non-DBS controls, with and without motor fluctua-tions. This analysis revealed mostly an effect of motor fluctuations: Thenon-fluctuating/no DBS group showed improved caregiver strain overthe DBS group and fluctuating/no DBS group. Quality of life was sig-nificantly better in the non-fluctuating/no DBS group, than the DBSgroup which was better than the fluctuating/non-DBS group.

Within the limits of interpretation of a cross-sectional study, we canstill conclude that DBS improves quality of life and does not likelyworsen caregiver strain. Caregiver strain was the same leaving open thepossibility that it would be improved with DBS when properly com-pared to a pre-surgical comparison. No such study has evaluated this.However, the recent EARLYSTIM trial does support the above findingswith regards to quality of life.1 This large, prospective trial randomizingpatients to DBS or best medical therapy found a 7.8 point improve-ment in the PDQ-39 while the medical group showed a 0.2 pointworsening. The disease burden on the caregiver is an important one,and this work makes important strides in understanding variable that



affect this burden. Surely, more prospective studies need to be done tofurther substantiate these relationships.

Benjamin L. Walter, M.D.Cleveland, OH, USA

REFERENCE1. Schuepbach WM, Rau J, Knudsen K, et al. Neurostimulation for Parkinson’s disease with

early motor complications. N Engl J Med. Feb 14 2013;368(7):610–622.

Comments not included in the Early View version of this paper.

132

OYAMA ET AL.


Documents

Deep Brain Stimulation May Improve Quality of Life in People With Parkinson's Disease Without Affecting Caregiver Burden