C. Moreau, MD L. Defebvre, MD, PhD A. Destée, MD, PhD S. Bleuse, PhD F. Clement, MD J.L. Blatt, MD, PhD P. Krystkowiak, MD, PhD D.

Size: px

Start display at page:

Download "C. Moreau, MD L. Defebvre, MD, PhD A. Destée, MD, PhD S. Bleuse, PhD F. Clement, MD J.L. Blatt, MD, PhD P. Krystkowiak, MD, PhD D."

Laura Young
5 years ago
Views:

1 ARTICLES STN-DBS frequency effects on freezing of gait in advanced Parkinson disease C. Moreau, MD L. Defebvre, MD, PhD A. Destée, MD, PhD S. Bleuse, PhD F. Clement, MD J.L. Blatt, MD, PhD P. Krystkowiak, MD, PhD D. Devos, MD, PhD Address correspondence and reprint requests to Dr. David Devos, Clinique Neurologique, Hôpital R. Salengro, CHU, F Lille cedex, France ABSTRACT Background: Severe gait disturbances and freezing episodes (frequently resistant to optimal dopaminergic treatment) often appear in advanced Parkinson disease (PD). Even several years after initiation, high-frequency subthalamic nucleus deep brain stimulation (STN-DBS) is still very effective for controlling segmental symptoms. However, there are no long-term data on the management of gait disorders and freezing in STN-DBS. Objectives: To compare the effects of various STN-DBS parameters on freezing of gait and to determine whether such effects are more related to stimulation energy (usual voltages vs high voltages at 130 Hz) or frequency (130 Hz vs approximately half this frequency: 60 Hz). Methods: We blindly assessed STN-DBS parameters in 13 PD patients reporting severe gait disorders. We compared the effects on gait of two different voltages (the patient s usual voltage [median 3 volts] and a high voltage [median 3.7 volts]) and two different frequencies (60 and 130 Hz, while maintaining the same total energy delivered) vs off-stimulation conditions. Results: The number of freezing episodes was significantly lower at the 60-Hz high voltage/ equivalent energy and higher at the 130-Hz/high voltage than for off stimulation. The slight improvement in the Unified Parkinson s Disease Rating Scale motor score observed (at 130 Hz) did not achieve statistical significance. Conclusions: Our results prompt consideration of a new strategy for two-stage subthalamic nucleus deep brain stimulation (STN-DBS) frequency optimization, with stimulation at 130 Hz and the usual voltage during the initial years of STN-DBS and then at 60 Hz at a high voltage in Parkinson disease patients who develop severe gait disorders. Neurology 2008;71:80 84 GLOSSARY DBS deep brain stimulation; MLA mesencephalic locomotor area; PD Parkinson disease; PPN pedunculopontine nucleus; Q1 first quartile; Q3 third quartile; STN subthalamic nucleus; SWS Stand Walk Sit; TEED total electrical energy delivered; UPDRS Unified Parkinson s Disease Rating Scale. Supplemental data at In long-term follow-up studies of subthalamic nucleus deep brain stimulation (STN-DBS), the maintained control of many Parkinson disease (PD) symptoms contrasts with a progressive increase in gait dysfunction (and notably the appearance of freezing episodes) in approximately 5% to 20% of the patient population. This becomes a major cause of disability and dependence 1-7 (table e-1 on the Neurology Web site at However, we do not know whether the axial symptoms that seem to predominate after several years of successful DBS are signs of the progression of nondopaminergic degenerative PD lesions 8 or whether they emerge simply because they are less sensitive to the commonly used stimulation settings. Optimal control of segmental parkinsonian signs is initially obtained at an STN stimulation frequency of 130 Hz (and potentially up to 185 Hz) in patients who are free of severe gait disorders, with a typical worsening of motor symptoms with STN-DBS 30 Hz. 9,10 These observations Editorial, page 76 e-pub ahead of print on April 16, 2008, at From the Department of Neurology and Movement Disorders (C.M., L.D., A.D., P.K., D.D.) and Department of Clinical Neurophysiology (S.B., J.L.B.), EA 2683, IFR 114, University of Lille 2, Lille, France; and Department of Neurology (F.C.), Roeselare, H.-Hartziekenhuis Roeselaere-Menen, Belgium. Disclosure: The authors report no disclosures. 80 Copyright 2008 by AAN Enterprises, Inc.

2 prompted us to wonder whether gait disorders could be managed by using other combinations of STN-DBS parameters. First, the stimulation voltage seems to be the most critical parameter in STN-DBS, followed by the frequency 9 ; one logical option would be to modify these settings to control parkinsonian symptoms. Second, recent data on the stimulation of structures near the STN (part of the mesencephalic locomotor area [MLA] and the pedunculopontine nucleus [PPN]) evidenced an improvement in gait disorders in PD The stimulation frequency range used was lower than for the STN area (from Hz 11,12 up to 70 Hz 14 ). Given the anatomic proximity of the STN and the PPN (just 5 mm), one could imagine that high STN-DBS frequencies might exert unfavorable effects on the MLA. We investigated PD patients who developed severe gait disorders and freezing episodes after several years of STN-DBS by looking at their response to 1) an energy effect (i.e., high voltages vs the patient s usual voltage) at a frequency of 130 Hz and 2) a frequency effect (i.e., 130 Hz vs approximately half this frequency [60 Hz] but at equivalent delivered energy levels). 15 METHODS Thirteen PD patients who developed severe gait disorders in the 5 years after initiation of STN-DBS gave their informed consent to participate. The patients median age was 70 [first quartile third quartile, Q1 Q3: 66 72] years, and the median disease duration was 18 [13 22] years. None of the patients had exhibited dopa-resistant axial signs before surgery. The median time since STN-DBS initiation was 5 [4 5] years. The preoperative L-dopa motor response on the Unified Parkinson s Disease Rating Scale (UPDRS) was 55% [45 60%]. After 1 year of STN-DBS, all patients displayed a good outcome, with a median reduction in the motor UPDRS score of 52% [48 60%] and the disappearance of motor fluctuations. By the time of the study, the average on vs off STN-DBS reduction in UPDRS was 45% [43 55%] compared with 52% in the early postoperative period. The median L-dopa sensitivity (% UPDRS III score improvement) was 45% at the time of the study. The L-dopa median daily dose equivalent was 675 mg. All patients displayed the same profile: the progressive appearance of gait disorders (notably including increased gait hypokinesia and postural instability), which were partially improved by L-dopa (median time of appearance after surgery: 2.5 [2 3] years). Furthermore, all patients presented freezing episodes in the off-medication state. None of the patients presented a directly L-dopa induced ( on-freezing ) episode. Before the study, the STN-DBS and dopa-therapy parameters had been optimized for each patient (the median stimulation settings being 130 Hz, 3 volts [V], and 60 s). We used a standardized, clinical, timed test for gait evaluation: the Stand Walk Sit (SWS) Test over a distance of 7 meters. 16 The test results the number of freezing episodes (the primary endpoint), completion time, and the number of steps were always assessed by the same, blinded investigator, whereas the stimulation parameters were modified by a second neurologist. Patients were evaluated under off-drug conditions after 12 hours of therapy withdrawal. Monopolar stimulation was performed with the contact that had given the best clinical results over the previous years: ventral contacts (0 or 1) were used in 10 patients, and dorsal contacts (2 or 3) were used in 3 patients. For each randomized condition (i.e., 60 Hz, 130 Hz, or off), the starting voltage was also randomized (either high or low). The applied voltage was then increased (or decreased) in 0.5-V steps. First part: A possible reduction in gait disorders in response to a voltage effect was assessed by blindly increasing the voltage in 0.5-V steps from 2 to 5.5 V at a constant frequency of 130 Hz. Second part: A possible frequency effect on gait disorders was blindly assessed by comparing 60-Hz and 130-Hz stimulation, with a voltage increase in 0.5-V steps from 2 to 5.5 V. To limit the number of statistical comparisons, three conditions were chosen: 1) stimulation switched off (off-dbs), 2) STN-DBS at 130 Hz (130-DBS) at the usual voltage (median 3 [Q1 Q3: 2 3.4] V), and 3) 130-DBS at a high voltage (3.7 [ ] V). The usual voltage corresponded to the preexperiment STN-DBS voltage used at 130 Hz over the preceding years and for which optimal clinical results had been observed in terms of UPDRS motor scores. The high voltage was the highest voltage in each patient that was well tolerated without triggering intolerable and persistent adverse effects, i.e., 5.5 V. To specifically study the frequency effect (and not a general energy effect), we maintained the same energy level for stimulation at both 60 and 130 Hz by varying the voltage. For each patient, the 60-DBS conditions were equivalent to the 130-DBS conditions in terms of energy delivered (calculated according to the total electrical energy delivered (TEED) formula: TEED (1 second) voltage 2 frequency pulse width/impedance). 12 The pulse width was constant (60 s). To limit the number of statistical comparisons, five conditions were chosen: 1) off-dbs, 2) 130-DBS at the usual voltage (3 [2 3.4] V), 3) 130-DBS at a high voltage (3.7 [ ] V), 4) STN-DBS at 60 Hz (60-DBS) at the usual voltage/equivalent energy (4.4 [3 5] V), and 5) 60- DBS at the high voltage/equivalent energy (5.5 [ ] V). The UPDRS Part III score was recorded under three conditions: during the off-dbs condition and during both 130-DBS and 60-DBS conditions at the voltage yielding the best results on the SWS Test. Subscores for rest tremor (Item 20), rigidity (22), and akinesia (23 26) and components of the axial subscore [including gait (30), stance (27), posture (28), postural stability (29), and speech (18)] were also compared. The assessment lasted approximately 4 hours in a single, morning session, to minimize circadian fluctuation. To limit the impact of stimulation after-effects, we strictly maintained an interval of 1 hour between each frequency change and 10 minutes between each voltage change. 17 The Conover nonparametric test was used to compare conditions. 18 Principal effects seen in an analysis of variance on ranks were further analyzed in contrast studies using the Bonferroni post hoc test. A significance level of 0.05 was chosen. Neurology 71 July 8,

3 Table Data concerning the 13 Parkinson disease patients Off-DBS 130-DBS, usual voltage, 3 [2 3.4] V 60-DBS, usual voltage/ equivalent energy, 4.4 [3 5] V 130-DBS, high voltage, 3.7 [ ] V 60-DBS, high voltage/ equivalent energy, 5.5 [ ] V SWS: completion time, s 50* [30 75] 41* [26 54] 30* [23 52] 90* [27 120] 24* [20 38] SWS: number of steps 48* [37 80] 41* [31 67] 32* [26 48] 60* [34 90] 25* [21 40] SWS: number of freezing episodes 3* [ ] 2* [1 2.5] 0.75* [0 1.25] 4* [3.5 10] 0* [0 0] UPDRS III (maximum: /108) 49 [43 51] 26 [21 30] 30 [25 34] UPDRS subscore: axial (/16) 5 [3 5] 5 [3 5] 4 [3 4] UPDRS subscore: gait (/4) 2.5 [2 3] 1 [1 2] 1 [1 2] UPDRS subscore: rest tremor (/20) 2 [2 3] 1 [0 2] 1 [1 2] UPDRS subscore: rigidity (/20) 8 [6 9] 5 [3 6] 6 [3 9] UPDRS subscore: akinesia (/32) 16 [ ] 16 [ ] 11 [9 11] Results are expressed as median [first quartile third quartile]. For the Stand Walk Sit (SWS) Test, all of the conditions were significantly different from each other (*p 0.05), except that 130 deep brain stimulation (DBS) in the high- and usual-voltage conditions did not differ significantly from the off-dbs condition in terms of the completion time and the number of steps. For the Unified Parkinson s Disease Rating Scale (UPDRS), only the best condition for each frequency was tested: significant difference between the off-dbs condition on one hand and the 60-DBS and the 130-DBS conditions on the other, with no difference between the latter two. RESULTS First part. In the 130-Hz/high-voltage condition, the SWS completion time and the number of steps and freezing episodes increased. The results were similar to those seen in the off DBS condition (table). In the 130-Hz condition, all patients displayed the same profile of worsening gait and more freezing as the voltage increased. There was an effect of condition on the SWS completion time [F(2,11) 4.7, p 0.024] and the number of steps [F(2,11) 7, p 0.01] and freezing episodes [F(2,11) 11.2, p 0.002]. Post hoc comparisons showed a slight, nonsignificant reduction in the number of the freezing episodes for the usual voltage when compared with off-dbs, which contrasted with the significant increase in freezing episodes observed at high-voltage values. In terms of the completion time and the number of steps, we observed a worsening in the 130-DBS high-voltage condition, compared with 130-DBS at the usual voltage. There were no significant differences between 130-DBS at the high voltage and the off-dbs condition (figure 1). Second part. The best results in terms of completion time and the number of steps and freezing episodes were obtained with 60-Hz high voltage/equivalent energy. There was an effect of condition on the completion time [F(4,9) 13.8, p 0.001] and the number of steps [F(4,9) 10.5, p 0.002] and freezing episodes [F(4,9) 11.6, p 0.001]. Post hoc comparisons revealed that 60-Hz DBS at the high energy/equivalent voltage gave the best results for each parameter. Both 60-DBS conditions were better than the two 130-DBS conditions and the off-dbs condition. In terms of the UPDRS III scores, there was no difference between 60 and 130 Hz, and all on conditions were better than off DBS [F(3,14) 8.8, p 0.002]. Compared with the off-dbs condition, there was an improvement in gait, tremor, rigidity, and akinesia in the 60-DBS and 130-DBS conditions, with no difference between the latter. Follow-up data. After 8 months, the clinical gait benefit with the 60-Hz high energy/equivalent voltage was still satisfactory in 85% (n 11) of the patients, although it had been necessary to slightly increase the daily L-dopa dose (by approximately 200 mg/day) to achieve the same clinical benefits for segmental parkinsonian symptoms. Two patients went back to 130 Hz, respectively 2 and 3 months later, because of a worsening of segmental symptoms, especially tremor, which was not well controlled by the slight L-dopa increase. DISCUSSION Our results demonstrated 1) a negative effect on gait of increasing voltages at a fixed frequency of 130 Hz in PD patients with severe gait disorders after several years of STN-DBS and 2) a positive effect of reduced frequency at equivalent delivered energy levels, with the greatest benefit on freezing episodes appearing in the 60-Hz highvoltage/equivalent-energy condition. A small, nonsignificant increase in the UPDRS motor score (which never exceeded 4 points) was observed with 60 Hz at the high energy/equivalent voltage, when compared with 130 Hz at the usual 82 Neurology 71 July 8, 2008

4 Figure 1 Median number of freezing, time, and steps in the three conditions Figure 2 Median values of the UPDRS scores in the three conditions *Significant difference (p 0.05) between the off deep brain stimulation (DBS) condition and the 60-DBS and the 130-DBS conditions, with no difference between the latter two. UPDRS Unified Parkinson s Disease Rating Scale. The numbers 1 4 represent the following stimulation settings: 1) 130 Hz, usual voltage; 2) 130 Hz, high voltage; 3) 60 Hz, usual voltage/equivalent energy; 4) 60 Hz, high voltage/equivalent energy. Median values and quartiles are given for the population as a whole. For each frequency, the voltage was increased in 0.5-volt (V) steps from 2 to 5.5 V. The threshold of adverse events for the 130 deep brain stimulation (DBS) condition varied from one patient to another but did not exceed 5 V. * Significant difference (p 0.05). The solid black line represents the off-dbs condition; first and third quartile [Q1 Q3] values correspond to the black dotted line. The black line with triangles represents the 130-Hz condition, with 1) the usual voltage value [Q1 Q3] and 2) the high voltage value [Q1 Q3]. The dark gray line with squares represents the 60-Hz equivalent voltage values, with 3) the usual energy/equivalent voltage [Q1 Q3] and 4) the high energy/equivalent voltage [Q1 Q3]. voltage. However, we saw no differences between the 60-DBS and 130-DBS conditions in terms of axial and gait subscores, probably because of the moderate clinical improvements in terms of speech, standing, and postural stability and the subscores lower sensitivity to change. The slight reduction in akinesia for 60 Hz at the high voltage/equivalent energy (which nevertheless did not reach significance) may have contributed to the reduction in the number of freezing episodes during gait. 19 The 130- and 60-Hz conditions did not differ significantly in terms of the tremor and rigidity subscores (figure 2). The underlying pathophysiologic mechanisms and structures involved in these effects remain to be determined. In very advanced PD, a change in STN function caused by direct high-frequency stimulation of this area (with a negative impact on gait) is one obvious possibility. Furthermore, the greater clinical impact of high voltages (more than 3 V) could also suggest current diffusion to structures around the STN electrode in general and to fibers that have lower thresholds than somata in particular. 20 Because the PPN is just 5 mm away from the STN (and its projections are even closer), high frequency STN- DBS could exert unfavorable effects on these areas. Direct, low-frequency (10- to 25-Hz) stimulation of the human PPN produces a gait improvement, whereas high frequencies ( 80 Hz) produce a worsening Furthermore, in animals, 60-Hz stimulation of the PPN s descending projections may exert a push toward locomotion. 21 Further confirmatory studies are required, but our results prompt consideration of a new strategy for twostage STN-DBS frequency optimization, with 130 Hz at the usual voltage during the initial years of STN-DBS and then 60 Hz at a high energy/equivalent voltage in patients developing severe gait disorders after several years of STN-DBS. ACKNOWLEDGMENT The authors thank Dr. David Fraser (Biotech Communication, Damery, France) for proofreading the manuscript. Received June 1, Accepted in final form November 7, Neurology 71 July 8,

5 REFERENCES 1. Krack P, Batir A, Van Blercom N, et al. Five-year follow-up of bilateral stimulation of the subthalamic nucleus in Parkinson s disease. N Engl J Med 2003;349: Schüpbach WM, Chastan N, Welter ML, et al. Stimulation of the subthalamic nucleus in Parkinson s disease: a 5 year follow up. J Neurol Neurosurg Psychiatry 2005;16: Rodriguez-Oroz MC, Obeso JA, Lang AE, et al. Bilateral deep brain stimulation in Parkinson s disease: a multicentre study with 4 years follow-up. Brain 2005;128: Liang G, Chou K, Baltuch G, et al. Long term outcomes of bilateral subthalamic nucleus stimulation in patients with advanced Parkinson s disease. Stereotact Funct Neurosurg 2006;84: Ostergaard K, Aa Sunde N. Evolution of Parkinson s disease during 4 years of bilateral deep brain stimulation of the subthalamic nucleus. Mov Disord 2006;21: Piboolnurak P, Lang AE, Lozano AM, Miyasaki JM. Levodopa response in long-term bilateral subthalamic stimulation for Parkinson s disease. Mov Disord 2007;22: Derost P, Ouchchane L, Morand D, et al. Is STN DBS appropriate to treat severe Parkinson s disease in an elderly population? Neurology 2007;68: Marckham CH, Diamond SG. Long-term follow-up of early dopa treatment in Parkinson s disease. Arch Neurol 1986;19: Moro E, Esselink A, Xie J, et al. The impact of PD of electrical parameters settings in STN stimulation. Neurology 2002;59: Timmerman L, Wotjecki L, Gross J, et al. Ten hertz stimulation of subthalamic nucleus deteriorates motor symptoms in Parkinson s disease. Mov Disord 2004;19: Plaha P, Gill S. Bilateral deep brain stimulation of the pedunculopontine nucleus for PD. Neuroreport 2005;16: Mazzone P, Lozano A, Stanzione P, et al. Implantation of human pedunculopontine nucleus: a safe and clinically relevant target in PD. Neuroreport 2006;16: Stefani A, Lozano A, Peppe A, et al. Bilateral deep brain stimulation of the pedunculo-pontine and subthalamic nuclei in severe Parkinson s disease. Brain 2007;130: Moro E, Zadikoff C, Alkhairallah T, et al. Unilateral stimulation in progressive supranuclear palsy and Parkinson s disease: preliminary data. American Academy of Neurology 2007;S13: Koss AM, Alterman RL, Tagliati M, Shils JL. Calculating total electrical energy delivered by deep brain stimulation systems. Ann Neurol 2004;56: Langston JW, Widner H, Goetz CG, Brooks D, Fahn S, Freeman T, Watts R. Core assessment program for intracerebral transplantations (CAPIT). Mov Disord 1992;7: Temperli P, Ghika J, Villemure JG, et al. How do Parkinsonian signs return after discontinuation of STN DBS? Neurology 2003;60: Conover WJ, Iman RL. Rank transformation as a bridge between parametric and nonparametric statistics. Am Stat Assoc 1982;35: Nieuwboer A, Dom R, De Weert W, et al. Abnormalities of the spatiotemporal characteristics of gait at the onset of freezing in PD. Mov Disord 2001;16: Ranck JB. Which elements are excited in electrical stimulation of mammalian central nervous system: a review. Brain Res 1998;1975: Garcia-Rill E, Homma Y, Skinner R. Arousal mechanisms related to posture and locomotion, 1: descending modulation. Prog Brain Res 2004;143: Experience the 2008 AAN Annual Meeting... from your home or office. Whether you were able to attend the 60 th AAN Annual Meeting in Chicago or not, the 2008 Virtual Annual Meeting suite of portable products offers an excellent opportunity to experience what you may have missed all from the comfort and convenience of your home, office, or car. Audio MP3 Files starting at $10 Webcasts-on-Demand $199 members / $299 non-members 2008 Syllabi on CD-ROM $199 members / $299 non-members NEW! Practice CD-ROM $39 members Visit today! 84 Neurology 71 July 8, 2008

Tao Xie 1*, Mahesh Padmanaban 1, Lisa Bloom 2, Ellen MacCracken 2, Breanna Bertacchi 1, Abraham Dachman 3 and Peter Warnke 4

Tao Xie 1*, Mahesh Padmanaban 1, Lisa Bloom 2, Ellen MacCracken 2, Breanna Bertacchi 1, Abraham Dachman 3 and Peter Warnke 4 Xie et al. Translational Neurodegeneration (2017) 6:13 DOI 10.1186/s40035-017-0083-7 REVIEW Effect of low versus high frequency stimulation on freezing of gait and other axial symptoms in Parkinson patients