- Email: [email protected]
Loss of benefit in VIM thalamic deep brain stimulation (DBS) for essential tremor (ET): How prevalent is it?
Parkinsonism and Related Disorders 19 (2013) 676 679
Contents lists available at SciVerse ScienceDirect
Parkinsonism and Related Disorders journal homepage: www.elsevier.com/locate/parkreldis
Loss of benefit in VIM thalamic deep brain stimulation (DBS) for essential tremor (ET): How prevalent is it? Ludy C. Shih a, *, Kathrin LaFaver b, Chen Lim c, Efstathios Papavassiliou d, Daniel Tarsy a a
Department of Neurology, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, USA Human Motor Control Section, National Institute of Neurological Disease and Stroke, National Institutes of Health, Bethesda, MD, USA c New York Medical College, Valhalla, NY, USA d Division of Neurosurgery, Department of Surgery, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, USA b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 25 September 2012 Received in revised form 6 March 2013 Accepted 22 March 2013
Ventralis intermedius (Vim) thalamic deep brain stimulation for medication refractory essential tremor (ET) has been shown to significantly improve severity of limb tremor in several large case series with significant reduction in objective motor scores. A variable proportion of patients experience decline in benefit over time, however, most studies have not been designed to describe the phenomenon of waning benefit in terms that are helpful for patient counseling. In this retrospective single center study, we define waning benefit as a phenomenon that occurs after patients begin to require reprogramming visits to optimize DBS benefit on tremor. We employed a survival analysis with time to escape (TTE) as a quantitative measure of time elapsed between implantation and the need for subsequent reprogram ming. In our cohort of ET patients operated on with Vim DBS from 1994 to 2009, among 45 subjects who met inclusion criteria, 73% reported waning benefit at some point during a mean follow up period of 56 months (range 12 152 months). The mean TTE from implantation date was 18 months (range 3 75 months). We conclude that loss of benefit over time from Vim DBS for ET is more prevalent than pre viously published estimates have indicated and should be discussed during patient counseling regarding durability of expected benefit. In addition, this data suggests that a disease based explanation rather than technical factors are more likely to explain the decline in benefit. Ó 2013 Elsevier Ltd. All rights reserved.
Keywords: Essential tremor Deep brain stimulation Long-term outcome
1. Introduction Medication refractory essential tremor (ET) is a disabling con dition which often shows considerable improvement following ventralis intermedius (Vim) thalamic deep brain stimulation (DBS). Despite several large case series showing long term efficacy with tremor score reduction between 41 and 78% at long term follow up [1e3], an estimated 35% of patients observe a decline in benefit over time [4] although other groups have reported varying preva lence of decline of 10e40% depending upon the definition of worsening [5e7]. There is considerable variation in how this worsening is defined as well as debate regarding the mechanism for this decline. Suboptimal lead localization may account for declining benefit, especially in the case of early failures [5]. Other suggested mechanisms include development of tolerance to stim ulation and progression of the underlying disease [6]. Several
* Corresponding author. Kirstein 228, Department of Neurology, Beth Israel Deaconess Medical Center, 330 Brookline Avenue, Boston, MA 02215, USA. E-mail address: [email protected] (L.C. Shih). 1353-8020/$ see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.parkreldis.2013.03.006
prospective long term studies with three year follow up or longer have examined the degree of improvement over time with number of subjects ranging between 19 and 37 [1e3,6,8e13]. While most of these studies showed no statistically significant decline in benefit as measured by mean tremor rating scores, the studies with long term follow up of six years or more noted a significant increase in on stimulation mean tremor scores across the cohort compared to the initial on stimulation tremor scores at 1e2 years, with durable reduction in action tremor possibly being more problematic than for postural tremor [3,8]. There has been recent interest in zona incerta DBS for ET to avoid waning benefit [14,15], yet few studies agree on the preva lence of waning benefit with Vim DBS and whether there are specific clinical features which predict waning of benefit. Previous literature has consistently employed a four point tremor scale to assess changes in tremor which may be insensitive to changes that are clinically meaningful for the patient. In addition, other studies have analyzed possible causes of complete stimulation failure, but this occurs infrequently and does not reflect the much more com mon occurrence of declining durability of benefit over time despite reprogramming. As a result it has been difficult to estimate the
L.C. Shih et al. / Parkinsonism and Related Disorders 19 (2013) 676 679
likelihood that a patient might experience declining benefit over time, and previous estimates of tolerance incidence may not be an accurate reflection of this likelihood. In order to provide effective counseling to the patient weighing the decision of whether to undergo Vim DBS for tremor, information on possible clinical pre dictors for waning benefit is needed. Our goal for this study was to provide a practical clinical defi nition of waning benefit, which we defined as the need to adjust stimulation settings in order to maintain tremor control on at least two consecutive follow up visits and to report its prevalence in a large case series of ET patients undergoing Vim DBS. 2. Methods We conducted a retrospective chart review approved by the Beth Israel Deaconess Medical Center Institutional Review Board. Subjects were included if they carried a diagnosis of probable ET as made by a movement disorder specialist following Movement Disorders Society consensus guidelines [16] and who underwent lead implantation and programming with either unilateral or bilateral Vim thalamic DBS at our center between January 1994 and September 2009 performed by two neurosurgeons. Exclusion criteria were: 1) development of parkinsonism, cerebellar ataxia, or possible drug-induced worsening of tremor following DBS, 2) less than 12 months of follow-up at our center, or 3) no immediate benefit following initial programming or loss of benefit within first 3 months. Clinical information regarding severity of limb tremor (rated as mild, moderate, or severe), gender, presence of family history, voice tremor or head tremor was collected for all subjects. Time to escape (TTE) was defined as the time which elapsed between date of initial lead implantation and the date of the second consecutive visit during which the following occurred: 1) patient reported subjective waning benefit, 2) worsening tremor was confirmed and documented on examination by a movement disorders neurologist or movement disorders nurse practitioner by comparison of handwriting or tremor rating from the current visit to the prior visit, and 3) subsequent reprogramming during that visit provided clear objective improvement of tremor. Reprogramming of DBS parameters typically included increases in stimulation amplitude by 0.2 1 V, PW increases of 30 90 ms or rate increases maximally to 185 Hz. In bilaterally implanted patients, TTE was calculated with respect to the side that was first implanted, if carried out in a staged manner. Similarly, waning benefit was measured in the limb contralateral to the side that was initially operated. Details of the surgical procedure have been described in a previous publication [5]. Postoperative imaging was performed using either computerized tomography or magnetic resonance imaging. In cases of computerized tomography, a scan was obtained with the gantry angled either in or parallel to the AC PC plane. In the subgroup of patients whose postoperative imaging was suitable for lead location analysis, cuts were made at 1.25-mm intervals through the region of the thalamus. Implantable pulse generators were turned on approximately 4 weeks after implantation of the electrodes. During the first three months postoperatively, stimulation parameters were adjusted to achieve optimal control of tremor without side effects. Follow-up was performed at routinely scheduled visits, usually every 3 6 months, unless the patient reported an emergent complication or other issue. Efficacy, detailed stimulation parameters, and impedance analyses for possible open or short circuits were recorded at each follow-up visit. The goal of the study was to view loss of benefit as a binary outcome. Therefore, statistical analysis consisted of calculating odds ratios for developing waning benefit, which were used for those independent variables that were of a binary nature (i.e. sex, presence or absence of family history, voice tremor or head tremor). For the group analysis examining clinical factors relating to loss of benefit, the “time to escape” was employed as a continuous variable, from which smaller values indicated a worse phenomenon, while larger values indicated a more desirable phenomenon, i.e. a longer duration of time before loss of benefit was noticed by the patient and the investigator. Simple linear regression analysis was then used to analyze the relationship between age or disease duration and TTE, and Wilcoxon rank sum test to examine lead location and stimulation parameter differences between groups. Post hoc determination of power, based on the ability to detect an odds ratio of 5 or greater using presence of family history as a binary variable, a sample size of 45 subjects, with 33 cases and 0.3 controls per case, reveals a power of 49% using alpha 0.05.
3. Results Sixty three patients with a diagnosis of ET underwent either unilateral or bilateral Vim thalamic DBS lead implantation at our center between 1994 and 2009. Of the 63 patients, 18 were excluded from analysis because of subsequent development of parkinsonism (n 3), cerebellar ataxia (n 3), possible
677
neuroleptic induced worsening of tremor (n 1), no significant benefit within 3 months of programming (n 2) and less than 12 months follow up at our center (n 9, two of whom died within 12 months following surgery from complications not due to tremor or DBS surgery). After application of inclusion and exclusion criteria, 45 subjects were included for analysis, consisting of 21 men and 24 women. Average patient age at time of surgery was 64.8 13.9 years. Average duration of ET prior to surgery was 28.3 17.6 (range 1e 60) years. The sides of implantation and the locations of symptoms are summarized in Table 1. Mean follow up period was 55.9 34.4 (range 12e152) months. Of the nine subjects who had less than 12 months follow up, there were four females and five males. The average age was 68.2 13.0 years. Average duration of ET prior to surgery was 22.6 19.4 years. Six of the subjects had head tremor, three subjects had voice tremor. Of the 45 patients analyzed, 33 (73.3%) reported waning benefit at a mean time of 18.8 15.1 months (range 3e75 months) following lead implantation. Positive family history of tremor (n 30) and presence of head (n 28) or voice tremor (n 16) were not associated with a statistically significantly higher risk of developing waning benefit. Those clinically rated as “severe” limb tremor (n 19) prior to surgery were not significantly more likely to experience waning benefit. However, female subjects were significantly more likely to experience waning benefit (OR 5.25, 95% CI 1.18e23.2). Additionally, there was no significant correlation between TTE and age or disease duration at time of surgery (r 0.149 and r 0.027, respectively). After application of in clusion criteria, all subjects included in the analysis initially responded to reprogramming efforts to improve tremor within the office visit. Therefore, stimulation parameters were often changed to try to compensate for partial loss of tremor control. Compared to initial stimulation parameters, pulse width and frequency differed significantly from stimulation parameters recorded at the most recent follow up (p 0.009 and <0.001, respectively), while lead location for a subgroup of patients with optimal post operative imaging for lead localization (n 12) showed no significant dif ference between those with (n 6) or without waning benefit (n 6, Table 2). 4. Discussion Vim DBS remains the most well accepted treatment option for patients with disabling medication refractory tremor, but waning benefit over time appears to be common, present in 73% of our cohort of 45 subjects followed longitudinally over a mean of 55.9 months. To our knowledge, this is the largest long term follow up
Table 1 Patient characteristics. Patient characteristics Sex (M:F) 21:24 Age at surgery (yrs) 64.8 (SD 13.9) Disease duration (yrs) 28.3 (SD 17.6) Tremor severity of targeted limb before surgery Mild 1 Moderate 25 Severe 19 Positive family history 30 Head tremor 28 Voice tremor 16 Unilateral:bilateral surgery 24:21 Mean follow up (months) 55.9 (SD 34.4) Waning benefit (n) 33 Time to escape (months) 18.8 (SD 15.1)
46.7:53.3% 27 89 (range) 1 60 (range)
66.7% 62.3% 35.6% 53.3:46.7% 12 152 (range) 73.3% 3 75 (range)
678
L.C. Shih et al. / Parkinsonism and Related Disorders 19 (2013) 676 679
Table 2 Risk of waning benefit and correlation with clinical/surgical factors. Clinical factor
Odds ratio (95% CI)
Female sex “Severe” tremor Positive FH Head tremor Voice tremor
5.25 (1.18 23.21) 0.88 (0.23 3.49) 1.0 (0.24 4.06) 2.0 (0.52 7.66) 1.95 (0.44 8.58)
Lead location (n
12)
Mean x coordinate from mid-AC-PC Mean x coordinate from wall of 3rd ventricle Mean y coordinate from PC Mean z coordinate from mid-AC-PC Stimulation parameters (n Amplitude (V) Pulse width (ms) Frequency (Hz)
45)
Maintained response (n
6)
Waning benefit (n
6)
12.24 (SD 1.7)
13.7 (SD 2.3)
9.88 (SD 1.3)
10.12 (SD 2.6)
5.85 (SD 1.1)
5.9 (SD 1.4)
0.06 (SD 3.7)
0.08 (SD 2.9)
Maintained response (n
12)
2.7 (SD 0.8) 78 (SD 28.9) 155.5 (SD 26.9)
Waning benefit (n
33)
3.1 (SD 0.7) 112 (SD 47.8) 184 (SD 1.9)
single center case series reported. Our estimate exceeds previously published estimates of waning benefit [1,4e7]. There may be several explanations for why our findings differ from previous estimates. First, our definition of waning benefit, as the need for repeated attempts at reprogramming after an initial period of sustained benefit of at least 3 months or more following surgery, may be more sensitive to the clinical experience of patients who have the need for reprogramming and less durable benefit on tremor between reprogramming sessions. This experience may not be adequately captured in measures of 4 point tremor rating scores in which a single point change may reflect a logarithmic change in severity for each point [17] and may not reflect smaller but clini cally meaningful changes in tremor severity. Secondly, it is unclear whether some of the long term follow up studies published permitted programmed changes prior to recording of the clinical rating score. In at least one published study, the protocol allowed patients to be reprogrammed in order to capture optimal tremor scores at the study visit, which may have influenced the appearance of durability of benefit in that study [1]. A third possibility is that these long term follow up studies were not specifically designed to be sensitive to subtle changes in tremor benefit over time, though several studies have noted slight in creases in various stimulation parameters over time, implying a fairly reproducible need for reprogramming [1e3,8]. Finally, our follow up time is one of the longest reported dura tion follow up among those published in the literature. As noted above, studies with longer follow up time were more likely to report greater degrees of waning benefit that were detectable on four point clinical scales. Tolerance to stimulation, implied as an adaptation by the brain to high frequency electrical stimulation [7] and disease progression have been proposed as potential causes of waning benefit, however, previously lower estimates of 10e40% [1,4e7] do not support the existence of a robust physiologically based phenomenon that occurs in a consistent fashion across this patient population. We believe our finding of higher prevalence is likely to reflect the common worsening in action tremor most ET patients experience over time. Surprisingly, our analyses did not yield a significant correlation between any of the clinical features we examined such as age at time of surgery, disease duration, gender, severity of tremor, presence of midline tremor or a positive family history. Being female was
associated with waning benefit but it is unclear whether this repre sents patient care preference or a distinct association between female sex and rate of progression of disease. Additionally, to our knowledge there is no data to support a sex based difference in response to DBS. We also confirmed a statistically significant difference in stim ulation parameters between groups with and without waning benefit which suggests that waning benefit drives the statistically significant increases in stimulation parameters seen by other groups [1e3,8,11e13]. Taken together with the high prevalence of waning benefit, we believe these data to be more consistent with a natural history of progressively worsening tremor of the limbs in ET, which may become less dependent upon excessive cer ebellothalamocortical synchronization and possibly more depen dent upon other factors not ameliorated by Vim DBS. This study has several limitations that might limit our in terpretations. First, this was a retrospective study, subject to se lection bias and underpowering inherent in this type of study. It is possible that the nine subjects who had less than 12 months follow up actually had absence of loss of benefit, thereby falsely increasing the overall proportion of those with loss of benefit. While this is possible, we examined the reasons for lack of follow up, when available. Three of the subjects died following DBS surgery, two of which were unrelated to tremor or DBS surgery itself. We conclude that it is unlikely that patients who did not follow up did excep tionally well following surgery. We also acknowledge that the sample size enabled only modest power to detect an odds ratio of 5 or greater. Therefore, the association of gender and the lack thereof with family history, presence of head or voice tremor should be taken as exploratory analyses requiring greater numbers in future studies to draw conclusions about these associations. Second, formalized FahneTolosaeMarin tremor rating scale scores and quantitative tremor recordings were not conducted in these sub jects and are worth further study as clinical features that may predict worsening over time. Other clinical features were not examined such as structural or functional MRI of relevant cerebellar and thalamic anatomy. Recent studies indicate impaired white matter integrity in patients with ET and possible loss of Purkinje cells in cerebellar cortex [18e21]. Given recent data that suggest DBS exerts its clinical effect partly through axonal stimulation, the integrity of the cerebellothalamocortical tracts or spinocerebellar afferent tracts could play a role in both long term durability of stimulation benefit and severity of progression of ET. We were not able to verify the stereotactic coordinates of all leads implanted in our case series. However, the lack of a statistical association between waning benefit and lead location suggests lack of significant influence of lead locations on this phenomenon, although lead location may still influence efficacy of initial tremor suppression. Stereotactic localization that deviates from well accepted statistically averaged x,y,z coordinates for effective VIM location cannot be relied upon as the only marker of satisfactory surgical lead placement. Data concerning other factors such as the presence of somatosensory or tremor cells in microelectrode re cordings and relief of tremor with intraoperative macrostimulation should be included in future studies in order to examine the cor relation between lead localization and waning benefit over time. In summary, waning benefit and a need for repeated reprog ramming is common in Vim DBS for medication refractory ET. As defined here, waning benefit and time elapsed before developing the need for repeated programming reflects the experience of pa tients who undergo Vim DBS more accurately than objective clin ical scores. These results suggest that partial loss of tremor control and progressive worsening of tremor while receiving VIM DBS is more common than identified in previous estimates and patients considering this therapy should be counseled accordingly. Given how common the phenomenon is, both in our study but as
L.C. Shih et al. / Parkinsonism and Related Disorders 19 (2013) 676 679
demonstrated by significant changes in long term programming in other studies as well, factors inherent to the disease process itself seem more likely than surgical technical factors to be the cause of this phenomenon. While post mortem analysis of brains affected by essential tremor after years of DBS may determine whether pathological changes in cerebellum can be identified in those who lose benefit, the degree to which these pathological changes in the cerebellum occur in ET are just beginning to be analyzed [22]. Additional studies are needed in which potential biomarkers, including LINGO1 polymorphisms, fMRI, PET, perfusion or diffusion tensor imaging changes may help distinguish those who have a more severe and progressive course and whether these markers can also identify those individuals who go on to have durable benefit with VIM DBS. Acknowledgments L.C. Shih was supported in part by the Clinical Investigator Training Program: Beth Israel Deaconess Medical Center and Har vard/MIT Health Science and Technology, in collaboration with Pfizer, Inc. and Merck & Co. The authors have no relevant financial disclosures to make regarding relevant companies involved in this work over the past three years. References [1] Pahwa R, Lyons KE, Wilkinson SB, Simpson Jr RK, Ondo WG, Tarsy D, et al. Long-term evaluation of deep brain stimulation of the thalamus. J Neurosurg 2006;104:506 12. [2] Koller WC, Lyons KE, Wilkinson SB, Troster AI, Pahwa R. Long-term safety and efficacy of unilateral deep brain stimulation of the thalamus in essential tremor. Mov Disord 2001;16:464 8. [3] Sydow O, Thobois S, Alesch F, Speelman JD. Multicentre European study of thalamic stimulation in essential tremor: a six year follow up. J Neurol Neurosurg Psychiatr 2003;74:1387 91. [4] Benabid AL, Benazzouz A, Hoffmann D, Limousin P, Krack P, Pollak P. Longterm electrical inhibition of deep brain targets in movement disorders. Mov Disord 1998;13(Suppl. 3):119 25. [5] Papavassiliou E, Rau G, Heath S, Abosch A, Barbaro NM, Larson PS, et al. Thalamic deep brain stimulation for essential tremor: relation of lead location to outcome. Neurosurgery 2004;54:1120 9 [discussion 9 30].
679
[6] Pilitsis JG, Metman LV, Toleikis JR, Hughes LE, Sani SB, Bakay RA. Factors involved in long-term efficacy of deep brain stimulation of the thalamus for essential tremor. J Neurosurg 2008;109:640 6. [7] Benabid AL, Pollak P, Gao D, Hoffmann D, Limousin P, Gay E, et al. Chronic electrical stimulation of the ventralis intermedius nucleus of the thalamus as a treatment of movement disorders. J Neurosurg 1996;84:203 14. [8] Rehncrona S, Johnels B, Widner H, Tornqvist AL, Hariz M, Sydow O. Long-term efficacy of thalamic deep brain stimulation for tremor: double-blind assessments. Mov Disord 2003;18:163 70. [9] Favilla CG, Ullman D, Wagle Shukla A, Foote KD, Jacobson CEt, Okun MS. Worsening essential tremor following deep brain stimulation: disease progression versus tolerance. Brain 2012;135:1455 62. [10] Putzke JD, Uitti RJ, Obwegeser AA, Wszolek ZK, Wharen RE. Bilateral thalamic deep brain stimulation: midline tremor control. J Neurol Neurosurg Psychiatr 2005;76:684 90. [11] Blomstedt P, Hariz GM, Hariz MI, Koskinen LO. Thalamic deep brain stimulation in the treatment of essential tremor: a long-term follow-up. Br J Neurosurg 2007;21:504 9. [12] Hariz MI, Shamsgovara P, Johansson F, Hariz G, Fodstad H. Tolerance and tremor rebound following long-term chronic thalamic stimulation for Parkinsonian and essential tremor. Stereotact Funct Neurosurg 1999;72: 208 18. [13] Zhang K, Bhatia S, Oh MY, Cohen D, Angle C, Whiting D. Long-term results of thalamic deep brain stimulation for essential tremor. J Neurosurg 2010;112: 1271 6. [14] Plaha P, Khan S, Gill SS. Bilateral stimulation of the caudal zona incerta nucleus for tremor control. J Neurol Neurosurg Psychiatr 2008;79:504 13. [15] Plaha P, Patel NK, Gill SS. Stimulation of the subthalamic region for essential tremor. J Neurosurg 2004;101:48 54. [16] Deuschl G, Bain P, Brin M. Consensus statement of the Movement Disorder Society on Tremor. Ad Hoc Scientific Committee. Mov Disord 1998;13(Suppl. 3): 2 23. [17] Elble RJ, Pullman SL, Matsumoto JY, Raethjen J, Deuschl G, Tintner R, et al. Tremor amplitude is logarithmically related to 4- and 5-point tremor rating scales. Brain 2006;129:2660 6. [18] Klein JC, Lorenz B, Kang JS, Baudrexel S, Seifried C, van de Loo S, et al. Diffusion tensor imaging of white matter involvement in essential tremor. Hum Brain Mapp;32:896 904. [19] Nicoletti G, Manners D, Novellino F, Condino F, Malucelli E, Barbiroli B, et al. Diffusion tensor MRI changes in cerebellar structures of patients with familial essential tremor. Neurology;74:988 94. [20] Shin DH, Han BS, Kim HS, Lee PH. Diffusion tensor imaging in patients with essential tremor. AJNR Am J Neuroradiol 2008;29:151 3. [21] Louis ED, Vonsattel JP. The emerging neuropathology of essential tremor. Mov Disord 2008;23:174 82. [22] Louis ED, Faust PL, Vonsattel JP. Purkinje cell loss is a characteristic of essential tremor: towards a more mature understanding of pathogenesis. Parkinsonism Relat Disord 2012 Sep;18(8):1003 4. http://dx.doi.org/10.1016/j.parkreldis.2012.06.017. [Epub 2012 Jul 11].
Contents lists available at SciVerse ScienceDirect
Parkinsonism and Related Disorders journal homepage: www.elsevier.com/locate/parkreldis
Loss of benefit in VIM thalamic deep brain stimulation (DBS) for essential tremor (ET): How prevalent is it? Ludy C. Shih a, *, Kathrin LaFaver b, Chen Lim c, Efstathios Papavassiliou d, Daniel Tarsy a a
Department of Neurology, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, USA Human Motor Control Section, National Institute of Neurological Disease and Stroke, National Institutes of Health, Bethesda, MD, USA c New York Medical College, Valhalla, NY, USA d Division of Neurosurgery, Department of Surgery, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, USA b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 25 September 2012 Received in revised form 6 March 2013 Accepted 22 March 2013
Ventralis intermedius (Vim) thalamic deep brain stimulation for medication refractory essential tremor (ET) has been shown to significantly improve severity of limb tremor in several large case series with significant reduction in objective motor scores. A variable proportion of patients experience decline in benefit over time, however, most studies have not been designed to describe the phenomenon of waning benefit in terms that are helpful for patient counseling. In this retrospective single center study, we define waning benefit as a phenomenon that occurs after patients begin to require reprogramming visits to optimize DBS benefit on tremor. We employed a survival analysis with time to escape (TTE) as a quantitative measure of time elapsed between implantation and the need for subsequent reprogram ming. In our cohort of ET patients operated on with Vim DBS from 1994 to 2009, among 45 subjects who met inclusion criteria, 73% reported waning benefit at some point during a mean follow up period of 56 months (range 12 152 months). The mean TTE from implantation date was 18 months (range 3 75 months). We conclude that loss of benefit over time from Vim DBS for ET is more prevalent than pre viously published estimates have indicated and should be discussed during patient counseling regarding durability of expected benefit. In addition, this data suggests that a disease based explanation rather than technical factors are more likely to explain the decline in benefit. Ó 2013 Elsevier Ltd. All rights reserved.
Keywords: Essential tremor Deep brain stimulation Long-term outcome
1. Introduction Medication refractory essential tremor (ET) is a disabling con dition which often shows considerable improvement following ventralis intermedius (Vim) thalamic deep brain stimulation (DBS). Despite several large case series showing long term efficacy with tremor score reduction between 41 and 78% at long term follow up [1e3], an estimated 35% of patients observe a decline in benefit over time [4] although other groups have reported varying preva lence of decline of 10e40% depending upon the definition of worsening [5e7]. There is considerable variation in how this worsening is defined as well as debate regarding the mechanism for this decline. Suboptimal lead localization may account for declining benefit, especially in the case of early failures [5]. Other suggested mechanisms include development of tolerance to stim ulation and progression of the underlying disease [6]. Several
* Corresponding author. Kirstein 228, Department of Neurology, Beth Israel Deaconess Medical Center, 330 Brookline Avenue, Boston, MA 02215, USA. E-mail address: [email protected] (L.C. Shih). 1353-8020/$ see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.parkreldis.2013.03.006
prospective long term studies with three year follow up or longer have examined the degree of improvement over time with number of subjects ranging between 19 and 37 [1e3,6,8e13]. While most of these studies showed no statistically significant decline in benefit as measured by mean tremor rating scores, the studies with long term follow up of six years or more noted a significant increase in on stimulation mean tremor scores across the cohort compared to the initial on stimulation tremor scores at 1e2 years, with durable reduction in action tremor possibly being more problematic than for postural tremor [3,8]. There has been recent interest in zona incerta DBS for ET to avoid waning benefit [14,15], yet few studies agree on the preva lence of waning benefit with Vim DBS and whether there are specific clinical features which predict waning of benefit. Previous literature has consistently employed a four point tremor scale to assess changes in tremor which may be insensitive to changes that are clinically meaningful for the patient. In addition, other studies have analyzed possible causes of complete stimulation failure, but this occurs infrequently and does not reflect the much more com mon occurrence of declining durability of benefit over time despite reprogramming. As a result it has been difficult to estimate the
L.C. Shih et al. / Parkinsonism and Related Disorders 19 (2013) 676 679
likelihood that a patient might experience declining benefit over time, and previous estimates of tolerance incidence may not be an accurate reflection of this likelihood. In order to provide effective counseling to the patient weighing the decision of whether to undergo Vim DBS for tremor, information on possible clinical pre dictors for waning benefit is needed. Our goal for this study was to provide a practical clinical defi nition of waning benefit, which we defined as the need to adjust stimulation settings in order to maintain tremor control on at least two consecutive follow up visits and to report its prevalence in a large case series of ET patients undergoing Vim DBS. 2. Methods We conducted a retrospective chart review approved by the Beth Israel Deaconess Medical Center Institutional Review Board. Subjects were included if they carried a diagnosis of probable ET as made by a movement disorder specialist following Movement Disorders Society consensus guidelines [16] and who underwent lead implantation and programming with either unilateral or bilateral Vim thalamic DBS at our center between January 1994 and September 2009 performed by two neurosurgeons. Exclusion criteria were: 1) development of parkinsonism, cerebellar ataxia, or possible drug-induced worsening of tremor following DBS, 2) less than 12 months of follow-up at our center, or 3) no immediate benefit following initial programming or loss of benefit within first 3 months. Clinical information regarding severity of limb tremor (rated as mild, moderate, or severe), gender, presence of family history, voice tremor or head tremor was collected for all subjects. Time to escape (TTE) was defined as the time which elapsed between date of initial lead implantation and the date of the second consecutive visit during which the following occurred: 1) patient reported subjective waning benefit, 2) worsening tremor was confirmed and documented on examination by a movement disorders neurologist or movement disorders nurse practitioner by comparison of handwriting or tremor rating from the current visit to the prior visit, and 3) subsequent reprogramming during that visit provided clear objective improvement of tremor. Reprogramming of DBS parameters typically included increases in stimulation amplitude by 0.2 1 V, PW increases of 30 90 ms or rate increases maximally to 185 Hz. In bilaterally implanted patients, TTE was calculated with respect to the side that was first implanted, if carried out in a staged manner. Similarly, waning benefit was measured in the limb contralateral to the side that was initially operated. Details of the surgical procedure have been described in a previous publication [5]. Postoperative imaging was performed using either computerized tomography or magnetic resonance imaging. In cases of computerized tomography, a scan was obtained with the gantry angled either in or parallel to the AC PC plane. In the subgroup of patients whose postoperative imaging was suitable for lead location analysis, cuts were made at 1.25-mm intervals through the region of the thalamus. Implantable pulse generators were turned on approximately 4 weeks after implantation of the electrodes. During the first three months postoperatively, stimulation parameters were adjusted to achieve optimal control of tremor without side effects. Follow-up was performed at routinely scheduled visits, usually every 3 6 months, unless the patient reported an emergent complication or other issue. Efficacy, detailed stimulation parameters, and impedance analyses for possible open or short circuits were recorded at each follow-up visit. The goal of the study was to view loss of benefit as a binary outcome. Therefore, statistical analysis consisted of calculating odds ratios for developing waning benefit, which were used for those independent variables that were of a binary nature (i.e. sex, presence or absence of family history, voice tremor or head tremor). For the group analysis examining clinical factors relating to loss of benefit, the “time to escape” was employed as a continuous variable, from which smaller values indicated a worse phenomenon, while larger values indicated a more desirable phenomenon, i.e. a longer duration of time before loss of benefit was noticed by the patient and the investigator. Simple linear regression analysis was then used to analyze the relationship between age or disease duration and TTE, and Wilcoxon rank sum test to examine lead location and stimulation parameter differences between groups. Post hoc determination of power, based on the ability to detect an odds ratio of 5 or greater using presence of family history as a binary variable, a sample size of 45 subjects, with 33 cases and 0.3 controls per case, reveals a power of 49% using alpha 0.05.
3. Results Sixty three patients with a diagnosis of ET underwent either unilateral or bilateral Vim thalamic DBS lead implantation at our center between 1994 and 2009. Of the 63 patients, 18 were excluded from analysis because of subsequent development of parkinsonism (n 3), cerebellar ataxia (n 3), possible
677
neuroleptic induced worsening of tremor (n 1), no significant benefit within 3 months of programming (n 2) and less than 12 months follow up at our center (n 9, two of whom died within 12 months following surgery from complications not due to tremor or DBS surgery). After application of inclusion and exclusion criteria, 45 subjects were included for analysis, consisting of 21 men and 24 women. Average patient age at time of surgery was 64.8 13.9 years. Average duration of ET prior to surgery was 28.3 17.6 (range 1e 60) years. The sides of implantation and the locations of symptoms are summarized in Table 1. Mean follow up period was 55.9 34.4 (range 12e152) months. Of the nine subjects who had less than 12 months follow up, there were four females and five males. The average age was 68.2 13.0 years. Average duration of ET prior to surgery was 22.6 19.4 years. Six of the subjects had head tremor, three subjects had voice tremor. Of the 45 patients analyzed, 33 (73.3%) reported waning benefit at a mean time of 18.8 15.1 months (range 3e75 months) following lead implantation. Positive family history of tremor (n 30) and presence of head (n 28) or voice tremor (n 16) were not associated with a statistically significantly higher risk of developing waning benefit. Those clinically rated as “severe” limb tremor (n 19) prior to surgery were not significantly more likely to experience waning benefit. However, female subjects were significantly more likely to experience waning benefit (OR 5.25, 95% CI 1.18e23.2). Additionally, there was no significant correlation between TTE and age or disease duration at time of surgery (r 0.149 and r 0.027, respectively). After application of in clusion criteria, all subjects included in the analysis initially responded to reprogramming efforts to improve tremor within the office visit. Therefore, stimulation parameters were often changed to try to compensate for partial loss of tremor control. Compared to initial stimulation parameters, pulse width and frequency differed significantly from stimulation parameters recorded at the most recent follow up (p 0.009 and <0.001, respectively), while lead location for a subgroup of patients with optimal post operative imaging for lead localization (n 12) showed no significant dif ference between those with (n 6) or without waning benefit (n 6, Table 2). 4. Discussion Vim DBS remains the most well accepted treatment option for patients with disabling medication refractory tremor, but waning benefit over time appears to be common, present in 73% of our cohort of 45 subjects followed longitudinally over a mean of 55.9 months. To our knowledge, this is the largest long term follow up
Table 1 Patient characteristics. Patient characteristics Sex (M:F) 21:24 Age at surgery (yrs) 64.8 (SD 13.9) Disease duration (yrs) 28.3 (SD 17.6) Tremor severity of targeted limb before surgery Mild 1 Moderate 25 Severe 19 Positive family history 30 Head tremor 28 Voice tremor 16 Unilateral:bilateral surgery 24:21 Mean follow up (months) 55.9 (SD 34.4) Waning benefit (n) 33 Time to escape (months) 18.8 (SD 15.1)
46.7:53.3% 27 89 (range) 1 60 (range)
66.7% 62.3% 35.6% 53.3:46.7% 12 152 (range) 73.3% 3 75 (range)
678
L.C. Shih et al. / Parkinsonism and Related Disorders 19 (2013) 676 679
Table 2 Risk of waning benefit and correlation with clinical/surgical factors. Clinical factor
Odds ratio (95% CI)
Female sex “Severe” tremor Positive FH Head tremor Voice tremor
5.25 (1.18 23.21) 0.88 (0.23 3.49) 1.0 (0.24 4.06) 2.0 (0.52 7.66) 1.95 (0.44 8.58)
Lead location (n
12)
Mean x coordinate from mid-AC-PC Mean x coordinate from wall of 3rd ventricle Mean y coordinate from PC Mean z coordinate from mid-AC-PC Stimulation parameters (n Amplitude (V) Pulse width (ms) Frequency (Hz)
45)
Maintained response (n
6)
Waning benefit (n
6)
12.24 (SD 1.7)
13.7 (SD 2.3)
9.88 (SD 1.3)
10.12 (SD 2.6)
5.85 (SD 1.1)
5.9 (SD 1.4)
0.06 (SD 3.7)
0.08 (SD 2.9)
Maintained response (n
12)
2.7 (SD 0.8) 78 (SD 28.9) 155.5 (SD 26.9)
Waning benefit (n
33)
3.1 (SD 0.7) 112 (SD 47.8) 184 (SD 1.9)
single center case series reported. Our estimate exceeds previously published estimates of waning benefit [1,4e7]. There may be several explanations for why our findings differ from previous estimates. First, our definition of waning benefit, as the need for repeated attempts at reprogramming after an initial period of sustained benefit of at least 3 months or more following surgery, may be more sensitive to the clinical experience of patients who have the need for reprogramming and less durable benefit on tremor between reprogramming sessions. This experience may not be adequately captured in measures of 4 point tremor rating scores in which a single point change may reflect a logarithmic change in severity for each point [17] and may not reflect smaller but clini cally meaningful changes in tremor severity. Secondly, it is unclear whether some of the long term follow up studies published permitted programmed changes prior to recording of the clinical rating score. In at least one published study, the protocol allowed patients to be reprogrammed in order to capture optimal tremor scores at the study visit, which may have influenced the appearance of durability of benefit in that study [1]. A third possibility is that these long term follow up studies were not specifically designed to be sensitive to subtle changes in tremor benefit over time, though several studies have noted slight in creases in various stimulation parameters over time, implying a fairly reproducible need for reprogramming [1e3,8]. Finally, our follow up time is one of the longest reported dura tion follow up among those published in the literature. As noted above, studies with longer follow up time were more likely to report greater degrees of waning benefit that were detectable on four point clinical scales. Tolerance to stimulation, implied as an adaptation by the brain to high frequency electrical stimulation [7] and disease progression have been proposed as potential causes of waning benefit, however, previously lower estimates of 10e40% [1,4e7] do not support the existence of a robust physiologically based phenomenon that occurs in a consistent fashion across this patient population. We believe our finding of higher prevalence is likely to reflect the common worsening in action tremor most ET patients experience over time. Surprisingly, our analyses did not yield a significant correlation between any of the clinical features we examined such as age at time of surgery, disease duration, gender, severity of tremor, presence of midline tremor or a positive family history. Being female was
associated with waning benefit but it is unclear whether this repre sents patient care preference or a distinct association between female sex and rate of progression of disease. Additionally, to our knowledge there is no data to support a sex based difference in response to DBS. We also confirmed a statistically significant difference in stim ulation parameters between groups with and without waning benefit which suggests that waning benefit drives the statistically significant increases in stimulation parameters seen by other groups [1e3,8,11e13]. Taken together with the high prevalence of waning benefit, we believe these data to be more consistent with a natural history of progressively worsening tremor of the limbs in ET, which may become less dependent upon excessive cer ebellothalamocortical synchronization and possibly more depen dent upon other factors not ameliorated by Vim DBS. This study has several limitations that might limit our in terpretations. First, this was a retrospective study, subject to se lection bias and underpowering inherent in this type of study. It is possible that the nine subjects who had less than 12 months follow up actually had absence of loss of benefit, thereby falsely increasing the overall proportion of those with loss of benefit. While this is possible, we examined the reasons for lack of follow up, when available. Three of the subjects died following DBS surgery, two of which were unrelated to tremor or DBS surgery itself. We conclude that it is unlikely that patients who did not follow up did excep tionally well following surgery. We also acknowledge that the sample size enabled only modest power to detect an odds ratio of 5 or greater. Therefore, the association of gender and the lack thereof with family history, presence of head or voice tremor should be taken as exploratory analyses requiring greater numbers in future studies to draw conclusions about these associations. Second, formalized FahneTolosaeMarin tremor rating scale scores and quantitative tremor recordings were not conducted in these sub jects and are worth further study as clinical features that may predict worsening over time. Other clinical features were not examined such as structural or functional MRI of relevant cerebellar and thalamic anatomy. Recent studies indicate impaired white matter integrity in patients with ET and possible loss of Purkinje cells in cerebellar cortex [18e21]. Given recent data that suggest DBS exerts its clinical effect partly through axonal stimulation, the integrity of the cerebellothalamocortical tracts or spinocerebellar afferent tracts could play a role in both long term durability of stimulation benefit and severity of progression of ET. We were not able to verify the stereotactic coordinates of all leads implanted in our case series. However, the lack of a statistical association between waning benefit and lead location suggests lack of significant influence of lead locations on this phenomenon, although lead location may still influence efficacy of initial tremor suppression. Stereotactic localization that deviates from well accepted statistically averaged x,y,z coordinates for effective VIM location cannot be relied upon as the only marker of satisfactory surgical lead placement. Data concerning other factors such as the presence of somatosensory or tremor cells in microelectrode re cordings and relief of tremor with intraoperative macrostimulation should be included in future studies in order to examine the cor relation between lead localization and waning benefit over time. In summary, waning benefit and a need for repeated reprog ramming is common in Vim DBS for medication refractory ET. As defined here, waning benefit and time elapsed before developing the need for repeated programming reflects the experience of pa tients who undergo Vim DBS more accurately than objective clin ical scores. These results suggest that partial loss of tremor control and progressive worsening of tremor while receiving VIM DBS is more common than identified in previous estimates and patients considering this therapy should be counseled accordingly. Given how common the phenomenon is, both in our study but as
L.C. Shih et al. / Parkinsonism and Related Disorders 19 (2013) 676 679
demonstrated by significant changes in long term programming in other studies as well, factors inherent to the disease process itself seem more likely than surgical technical factors to be the cause of this phenomenon. While post mortem analysis of brains affected by essential tremor after years of DBS may determine whether pathological changes in cerebellum can be identified in those who lose benefit, the degree to which these pathological changes in the cerebellum occur in ET are just beginning to be analyzed [22]. Additional studies are needed in which potential biomarkers, including LINGO1 polymorphisms, fMRI, PET, perfusion or diffusion tensor imaging changes may help distinguish those who have a more severe and progressive course and whether these markers can also identify those individuals who go on to have durable benefit with VIM DBS. Acknowledgments L.C. Shih was supported in part by the Clinical Investigator Training Program: Beth Israel Deaconess Medical Center and Har vard/MIT Health Science and Technology, in collaboration with Pfizer, Inc. and Merck & Co. The authors have no relevant financial disclosures to make regarding relevant companies involved in this work over the past three years. References [1] Pahwa R, Lyons KE, Wilkinson SB, Simpson Jr RK, Ondo WG, Tarsy D, et al. Long-term evaluation of deep brain stimulation of the thalamus. J Neurosurg 2006;104:506 12. [2] Koller WC, Lyons KE, Wilkinson SB, Troster AI, Pahwa R. Long-term safety and efficacy of unilateral deep brain stimulation of the thalamus in essential tremor. Mov Disord 2001;16:464 8. [3] Sydow O, Thobois S, Alesch F, Speelman JD. Multicentre European study of thalamic stimulation in essential tremor: a six year follow up. J Neurol Neurosurg Psychiatr 2003;74:1387 91. [4] Benabid AL, Benazzouz A, Hoffmann D, Limousin P, Krack P, Pollak P. Longterm electrical inhibition of deep brain targets in movement disorders. Mov Disord 1998;13(Suppl. 3):119 25. [5] Papavassiliou E, Rau G, Heath S, Abosch A, Barbaro NM, Larson PS, et al. Thalamic deep brain stimulation for essential tremor: relation of lead location to outcome. Neurosurgery 2004;54:1120 9 [discussion 9 30].
679
[6] Pilitsis JG, Metman LV, Toleikis JR, Hughes LE, Sani SB, Bakay RA. Factors involved in long-term efficacy of deep brain stimulation of the thalamus for essential tremor. J Neurosurg 2008;109:640 6. [7] Benabid AL, Pollak P, Gao D, Hoffmann D, Limousin P, Gay E, et al. Chronic electrical stimulation of the ventralis intermedius nucleus of the thalamus as a treatment of movement disorders. J Neurosurg 1996;84:203 14. [8] Rehncrona S, Johnels B, Widner H, Tornqvist AL, Hariz M, Sydow O. Long-term efficacy of thalamic deep brain stimulation for tremor: double-blind assessments. Mov Disord 2003;18:163 70. [9] Favilla CG, Ullman D, Wagle Shukla A, Foote KD, Jacobson CEt, Okun MS. Worsening essential tremor following deep brain stimulation: disease progression versus tolerance. Brain 2012;135:1455 62. [10] Putzke JD, Uitti RJ, Obwegeser AA, Wszolek ZK, Wharen RE. Bilateral thalamic deep brain stimulation: midline tremor control. J Neurol Neurosurg Psychiatr 2005;76:684 90. [11] Blomstedt P, Hariz GM, Hariz MI, Koskinen LO. Thalamic deep brain stimulation in the treatment of essential tremor: a long-term follow-up. Br J Neurosurg 2007;21:504 9. [12] Hariz MI, Shamsgovara P, Johansson F, Hariz G, Fodstad H. Tolerance and tremor rebound following long-term chronic thalamic stimulation for Parkinsonian and essential tremor. Stereotact Funct Neurosurg 1999;72: 208 18. [13] Zhang K, Bhatia S, Oh MY, Cohen D, Angle C, Whiting D. Long-term results of thalamic deep brain stimulation for essential tremor. J Neurosurg 2010;112: 1271 6. [14] Plaha P, Khan S, Gill SS. Bilateral stimulation of the caudal zona incerta nucleus for tremor control. J Neurol Neurosurg Psychiatr 2008;79:504 13. [15] Plaha P, Patel NK, Gill SS. Stimulation of the subthalamic region for essential tremor. J Neurosurg 2004;101:48 54. [16] Deuschl G, Bain P, Brin M. Consensus statement of the Movement Disorder Society on Tremor. Ad Hoc Scientific Committee. Mov Disord 1998;13(Suppl. 3): 2 23. [17] Elble RJ, Pullman SL, Matsumoto JY, Raethjen J, Deuschl G, Tintner R, et al. Tremor amplitude is logarithmically related to 4- and 5-point tremor rating scales. Brain 2006;129:2660 6. [18] Klein JC, Lorenz B, Kang JS, Baudrexel S, Seifried C, van de Loo S, et al. Diffusion tensor imaging of white matter involvement in essential tremor. Hum Brain Mapp;32:896 904. [19] Nicoletti G, Manners D, Novellino F, Condino F, Malucelli E, Barbiroli B, et al. Diffusion tensor MRI changes in cerebellar structures of patients with familial essential tremor. Neurology;74:988 94. [20] Shin DH, Han BS, Kim HS, Lee PH. Diffusion tensor imaging in patients with essential tremor. AJNR Am J Neuroradiol 2008;29:151 3. [21] Louis ED, Vonsattel JP. The emerging neuropathology of essential tremor. Mov Disord 2008;23:174 82. [22] Louis ED, Faust PL, Vonsattel JP. Purkinje cell loss is a characteristic of essential tremor: towards a more mature understanding of pathogenesis. Parkinsonism Relat Disord 2012 Sep;18(8):1003 4. http://dx.doi.org/10.1016/j.parkreldis.2012.06.017. [Epub 2012 Jul 11].