- Email: [email protected]
Therapy Habituation at 12 Months: Spinal Cord Stimulation Versus Dorsal Root Ganglion Stimulation for Complex Regional Pain Syndrome Type I and II
ARTICLE IN PRESS The Journal of Pain, Vol 00, No 00 (), 2019: pp 1−10 Available online at www.jpain.org and www.sciencedirect.com
Original Reports Therapy Habituation at 12 Months: Spinal Cord Stimulation Versus Dorsal Root Ganglion Stimulation for Complex Regional Pain Syndrome Type I and II Robert M. Levy,* Nagy Mekhail,y Jeffrey Kramer,z Lawrence Poree,x Kasra Amirdelfan,{ Eric Grigsby,k Peter Staats,** Allen W. Burton,yy Abram H. Burgher,zz James Scowcroft,xx Stan Golovac,{{ Leonardo Kapural,kk Richard Paicius,*** Jason Pope,yyy Sam Samuel,y William Porter McRoberts,zzz Michael Schaufele,xxx Alexander R. Kent,yy Adil Raza,yy and Timothy R. Deer{{{ Marcus Neuroscience Institute, Boca Raton, Florida, yCleveland Clinic, Cleveland, Ohio, zVolta Research and University of Illinois College of Medicine, Chicago, Illinois, xUniversity of California at San Francisco, California, {IPM Medical Group, Inc., Walnut Creek, California, kNeurovations, Napa, California, **Premier Pain Center, Shrewsbury Township, New Jersey, yyAbbott Neuromodulation, Plano, Texas, zzHOPE Research − TPC, Phoenix, Arizona, xxPain Management Associates, Independence, Missouri, {{Florida Pain, Merritt Island, Florida, kkCarolinas Pain Institute, Winston-Salem, North Carolina, ***Newport Beach Headache and Pain, Newport Beach, California, yyyEvolve Restorative Center, Santa Rosa, California, zzzHoly Cross Hospital, Ft. Lauderdale, Florida, xxxDrug Studies America, Marietta, Georgia, {{{Spine and Nerve Center of the Virginias, Charleston, West Virginia *
Abstract: The ACCURATE randomized, controlled trial compared outcomes of dorsal root ganglion (DRG) stimulation versus tonic spinal cord stimulation (SCS) in 152 subjects with chronic lower extremity pain due to complex regional pain syndrome (CRPS) type I or II. This ACCURATE substudy was designed to evaluate whether therapy habituation occurs with DRG stimulation as compared to SCS through 12-months. A modified intention-to-treat analysis was performed to assess percentage pain relief (PPR) and responder rates at follow-up visits (end-of-trial, 1, 3, 6, 9, 12-months postpermanent implant) for all subjects that completed trial stimulation (DRG:N = 73, SCS:N = 72). For both groups, mean PPR was significantly greater at end-of-trial (DRG = 82.2%, SCS =0 77.0%) than all other follow-ups. Following permanent DRG system implantation, none of the time points were significantly different from one another in PPR (range = 69.3−73.9%). For the SCS group, PPR at 9-months (58.3%) and 12-months (57.9%) was significantly less than at 1-month (66.9%). The responder rate also decreased for the SCS group from 1-month (68.1%) to 12-months (61.1%). After stratifying by diagnosis, it was found that only the CRPS-I population had diminishing pain relief with SCS. DRG stimulation resulted in more stable pain relief through 12-months, while tonic SCS demonstrated therapy habituation at 9- and 12-months. Trial Registration: The ACCURATE study was registered at ClinicalTrials.gov with Identifier NCT01923285. Received November 1, 2018; Revised May 3, 2019; Accepted August 27, 2019. This work was supported by Abbott” and not funded. This study was a sub-analysis from the ACCURATE Trial (NCT01923285). Disclosures: Drs. Mekhail, Deer, Poree, Amirdelfan, Grigsby, Staats, Burton, Burgher, Scowcroft, Golovac, Kapural, Paicius, Pope, Samuel, McRoberts, Schaufele, and Levy were paid by Spinal Modulation & St. Jude Medical as investigators for the ACCURATE study. Dr. Levy has served as a consultant to Mainstay Medical, Nalu, Saluda Medical, and Abbott. Dr. Levy is or has been a minority shareholder in Saluda Medical, Spinal Modulation, Bioness, Vertos, and Nevro. Dr. Mekhail formerly had a consultation agreement with Spinal Modulation to serve as medical monitor of the ACCURATE study. Currently he is a consultant for Abbott, Saluda Medical, StimWave, Medtronic Neurological and Flowonix. Dr. Amirdelfan is a consultant for Nevro, Saluda Medical, Nalu Medical, Biotronik and ENSO. Dr. Scowcroft has served as a consultant for Abbott and Boston Scientific. Dr. Kapural is a consultant for Abbott, Nevro, Neuros, SPR Therapeutics, and Saluda Medical. Dr. Paicius is a consultant for Abbott, Nuvectra, Mesoblast, Vertiflex, Vertos, and Pacific Stem Cells. Dr. Pope is a consultant for Abbott, Flowonix, Saluda Medical, Jazz Pharmaceuticals,
SPR Therapeutics, and Vertiflex. Dr. McRoberts serves or has served as a consultant for Abbott, Medtronic, Nevro, Boston Scientific, Bioness, Vertiflex, and SPR Therapeutics. Dr. Deer is a consultant for Axonics, Bioness, Flowonix, Medtronic, Jazz Pharmaceuticals, Nevro, Abbott, and Saluda Medical and has consulting/equity for Axonics and Bioness. Dr. Deer formerly had equity in Spinal Modulation and Nevro. Dr. Kramer was a former consultant for Abbott, Nalu Medical, Autonomic Technologies, Thimble Bioelectronics, Circuit Therapeutics, and CereVu. Dr. Kramer has equity in Abbott as a former employee, and formerly had equity in Spinal Modulation. Dr. Kramer is now an employee of Medtronic. Dr. Burton, Dr. Kent, and Mr. Raza receive salary and have equity as employees of Abbott. The remaining authors report no conflicts of interest. Address reprint requests to Alexander Kent, PhD, Abbott Neuromodulation, 645 Almanor Ave, Sunnyvale, CA 94085. E-mail: [email protected] 1526-5900/$36.00 © 2019 by the American Pain Society https://doi.org/10.1016/j.jpain.2019.08.005
1
ARTICLE IN PRESS 2 The Journal of Pain
Therapy Habituation for SCS Versus DRG Stimulation
Perspective: This article reports on an ACCURATE substudy, which found that long-term therapy habituation occurred at 12-months with SCS, but not DRG stimulation, in patients with CRPS. The underlying mechanisms of action for these results remain unclear, although several lines of inquiry are proposed. © 2019 by the American Pain Society Key Words: Spinal cord stimulation, dorsal root ganglion stimulation, habituation, tolerance, complex regional pain syndrome.
Introduction Complex regional pain syndrome (CRPS) type I and CRPS type II diminish quality of life and can lead to significant disability.15 Both conditions have varied and complex clinical features, with chronic debilitating pain being the primary symptom, usually involving the lower or upper extremities.16,24 The conditions are differentiated by the absence (CRPS-I) or presence (CRPS-II) of demonstrable nerve damage as an underlying etiology. Spinal cord stimulation (SCS) is an accepted treatment for CRPS-I and CRPS-II when conventional treatments have failed.10,24 SCS systems consist of stimulation lead (s) placed near the spinal cord, and a pulse generator which generates electrical stimulation pulses. Tonic SCS has been used for decades and is considered the standard stimulation mode. In this paradigm, electrical pulses are delivered in a continuous manner at a low frequency (<100 Hz), and generally cause a paresthesia sensation often described as buzzing or tingling.5 Dorsal root ganglion (DRG) stimulation uses lead(s) placed near the target DRG, which are connected to a pulse generator to deliver stimulation at approximately 20 Hz.5 The ACCURATE study was a randomized controlled multicenter trial (RCT) designed to evaluate the safety and effectiveness of DRG stimulation as compared to tonic SCS therapy.5 This study was performed in 152 subjects with lower extremity pain due to CRPS-I or CRPS-II, and who were enrolled at 22 clinical sites in the United States. The results of the ACCURATE study demonstrated that DRG stimulation therapy was associated with a significantly greater level of treatment success (≥50% pain relief at trial and 3 month follow-up compared to baseline, and freedom from stimulationrelated neurological deficits) than SCS (81.2% versus 55.7%; noninferiority: P < .001, superiority: P < .001). The literature provides reports of habituation with low-frequency, tonic SCS therapy for a variety of pain conditions, including CRPS. Habituation (or “tolerance”) has been defined as the “development of inadequate pain relief [after acute success], despite appropriate stimulation coverage that cannot be explained by a hardware-related issue.”9,14 Kumar reported that habituation occurred in 25.9% of 328 patients at long-term follow-up (range = 1−22 years), including 17.9% of the 28 CRPS-I and CRPS-II subject population.14 Hayek found that 13.7% of 235 patients experienced habituation (mean follow-up = 45 § 25 months).9 Moreover, habituation was the most common reason for device explant, with a median time to explant of 19.6 months. Similarly, North found a 13% habituation rate for 205 patients at long-term follow-up (mean = 7.1 years, range = 1.5−20.4
found that 18% of 80 patients were years).18 Alo explanted between 24 and 30 months because of insufficient pain relief despite good paresthesia coverage.1 Burchiel documented 14.5% of 145 patients who discontinued use of SCS or had their device explanted due to ineffective stimulation at one year.4 Lastly, Kemler reported that the analgesic effects of SCS with physical therapy were not significantly different from physical therapy alone starting at the 3-year follow-up.11 SCS therapy habituation is not well understood, but multiple complex factors have been suggested to explain this phenomenon. Habituation may be due to disease progression in CRPS-I and CRPS-II, plasticity of central pain pathways, fibrosis around the stimulating electrodes, or other physiological and psychological factors (eg, short-term placebo effect, Hawthorne effect).3,14 Moreover, patients may exaggerate analgesia during the trial period to receive a permanent SCS implant, leading to artificially enhanced pain relief scores in the short-term. The current paper reports on a subanalysis of the ACCURATE RCT data that evaluated habituation with DRG stimulation and SCS therapies. The primary hypothesis was that SCS would show a significant decline in pain relief at 12 months compared to 1 month, whereas no significant change would be observed between these timepoints for DRG stimulation.
Methods Study Design The ACCURATE RCT was a randomized, controlled multicenter study conducted at 22 investigative centers in the United States from 2013 through 2016. This study evaluated the safety and effectiveness of DRG stimulation compared to tonic SCS for the treatment of chronic lower extremity pain due to CRPS-I or CRPS-II for at least 6 months.5 All sites obtained Institutional Review Board approval, and subjects signed informed consent prior to enrollment.
Subject Selection All subjects were screened per the study’s inclusion/ exclusion criteria. Briefly, enrolled subjects (nay¨ve to neurostimulation) had pain in the chief area of complaint of ≥60 mm on a visual analog scale (VAS), failed at least 2 pharmacologic treatments from 2 drug classes, had stable neurological function, and no psychological pathology that contraindicated an implantable device. All subjects’ medical, psychological, and imaging records
ARTICLE IN PRESS Levy et al were evaluated by a medical monitor, independent of the study sponsor and study investigators, to ensure appropriate subject selection. The diagnosis of lower extremity CRPS-I or CRPS-II was confirmed by an experienced medical monitor (N.M.) for strict adherence to the following criteria during the ACCURATE study.5 Diagnosis of CRPS-I was based on the Budapest criteria.8 CRPS-II (referred to as causalgia in ACCURATE) was defined in the study as a painful condition arising from damage to a nerve resulting in chronic pain, which may or may not have secondary symptoms.21
Devices and Procedure The DRG stimulation system (Axium Neurostimulation System, Abbott, IL) components were percutaneous leads designed to stimulate the DRG, an external trial pulse generator, and an implantable pulse generator. The SCS devices were commercially-available pulse generators (RestoreUltra and RestoreSensor, Medtronic, MN). The postural sensing feature of the RestoreSensor pulse generator was not activated for this study, which allowed comparison of DRG stimulation to the traditional SCS paradigm without postural adjustment that was most commonly used at the time of the study. A maximum of 16 contacts were placed per patient in the DRG stimulation and SCS groups.
Data Collection and Statistical Analysis Outcome data was collected by study staff on case report forms in a nonblinded manner. A description of all scales used to collect primary and second endpoints is provided in the ACCURATE study publication.5 The primary outcome measure for this substudy was change in percentage pain relief (PPR) over time, as measured by the percentage change from baseline for in-clinic VAS scores in the primary area of pain to each study follow-up visit: End of Trial neurostimulation, 1 month, 3 months, 6 months, 9 months, and 12 months following permanent implant. As a secondary outcome measure, the responder rate at each study follow-up was calculated as the percentage of subjects with at least 50% pain relief. This modified intention-to-treat analysis (MITT) included all subjects who completed the trial stimulation phase, with missing data from any subsequent time points imputed by last observation carried forward (LOCF) so that patients were not excluded for missing a single study visit. Of the 320 patients consented in the study, a total of 152 subjects were enrolled and randomized 1:1 in the ACCURATE study (76 DRG stimulation, 76 SCS). Six subjects (3 DRG stimulation, 3 SCS) withdrew prior to undergoing the trial stimulation phase, and 1 additional SCS subject never completed the trial stimulation phase. Subjects that failed at the end of the trial visit (<50% pain relief), deemed treatment failures, were not eligible for a permanent implant and were exited from the ACCURATE study per protocol. The MITT population for this analysis completed the end of
The Journal of Pain 3 trial stimulation visit, and consisted of 145 subjects (73 DRG stimulation, 72 SCS). This MITT population was based on standard intention-to-treat principles, in which subjects were analyzed based on their initial randomized treatments. Subsequently, any subjects that exited the study for any reason at any time were included in this analysis. LOCF was applied at follow-up visits for several subjects in the DRG group (1 month follow-up: 12 subjects; 3 months: 14 subjects; 6 months: 14 subjects; 9 months: 18 subjects; 12 months: 18 subjects) and SCS group (1 month follow-up: 18 subjects; 3 months: 18 subjects; 6 months: 20 subjects; 9 months: 23 subjects; 12 months: 22 subjects). Statistical analyses were performed in MATLAB R2016B (MathWorks, Natick, MA). Prior to assessing therapy habituation for each stimulation type (DRG or SCS), a 4-way repeated-measures analysis of variance (ANOVA) was performed, in which follow-up visit, diagnosis, gender and stimulation type were the independents variable and PPR was the dependent variable (a = .05). If the 4-way interaction term was significant, 1-way repeated-measures ANOVA tests were performed, in which follow-up visit was the independent variable and PPR was the dependent variable (a = .05). If the ANOVA was significant, post-hoc Fisher’s least significant different tests were performed to identify which follow-up visits were statistically different. Performing the Fisher’s least significant different post-hoc tests only after finding that the global ANOVA was significant protected against type I error from multiple comparisons. This analysis was performed for all subjects, and then repeated after stratifying by diagnosis (CRPS-I or CRPS-II) or gender to determine if any observed habituation effects were dependent on the pain condition or patient demographics. Finally, to assess significant changes in responder rates, logistic regression was performed for each stimulation type, in which the follow-up visit was the predictor variables and responder (≥50% pain relief) was the response variable.
Results Subjects’ Baseline Characteristics An overview of subject’s baseline characteristics is provided in Table 1. Briefly, subjects’ diagnoses were CRPS-I (DRG stimulation 57.5% and SCS 56.9%) and CRPS-II (DRG stimulation 42.5% and SCS 43.1%). The average age of subjects was 52.5 § 12.7 years in the DRG stimulation group and 52.3 § 11.7 years in the SCS group. Females comprised 53.4% (DRG stimulation) and 51.4% (SCS) of the study population. Average duration of chronic lower limb pain was 7.5 § 7.6 years in the DRG stimulation group, and 7.0 § 7.7 years in the SCS group. The primary regions of lower limb pain treated in the study included groin pain (DRG stimulation 9.6%, SCS 12.5%), buttock pain (DRG stimulation 2.7%, SCS 5.6%), leg pain (DRG stimulation 30.1%, SCS 33.3%), and foot pain (DRG stimulation 57.5%, SCS 48.6%).
ARTICLE IN PRESS Therapy Habituation for SCS Versus DRG Stimulation
4 The Journal of Pain
Baseline Characteristics of Subjects in the DRG and SCS Groups
Table 1.
BASELINE CHARACTERISTICS N Age Mean (SD) Gender % Male (N) % Female (N) Race % White (N) % Other (N) Primary diagnosis % CRPS-I (N) % CRPS-II (N) Duration of lower limb pain (y) Mean (SD) Region of lower limb Pain % Groin (N) % Buttocks (N) % Leg (N) % Foot (N) Baseline VAS score (mm) Mean (SD)
DRG STIMULATION
SCS
P VALUE*
73
72
N/A
52.5 (12.7)
52.3 (11.7)
0.93
46.6% (34) 53.4% (39)
48.6% (35) 51.4% (37)
0.87
94.5% (69) 5.5% (4)
91.7% (66) 8.3% (6)
0.53
57.5% (42) 42.5% (31)
56.9% (41) 43.1% (31)
1.00
7.5 (7.6)
7.0 (7.7)
0.68
9.6% (7) 2.7% (2) 30.1% (22) 57.5% (42)
12.5% (9) 5.6% (4) 33.3% (24) 48.6% (35)
0.65
81.2 (10.5)
81.3 (10.6)
0.96
Table 2. Baseline Characteristics of Subjects With CRPS-I and CRPS-II BASELINE CHARACTERISTICS N Age Mean (SD) Gender % Male (N) % Female (N) Race % White (N) % Other (N) Duration of lower limb pain (y) Mean (SD) Region of lower limb Pain % Groin (N) % Buttocks (N) % Leg (N) % Foot (N) Baseline VAS score (mm) Mean (SD)
CRPS-I
CRPS-II
P VALUE*
83
62
N/A
50.8 (12.7)
54.5 (11.2)
0.08
56.6% (47) 53.2% (33) 43.4% (36) 46.8% (29)
0.74
92.8% (77) 93.5% (58) 7.2% (6) 6.5% (4)
1.00
6.7 (7.3)
8.1 (8.1)
0.28
6.0% (5) 17.7% (11) 6.0% (5) 1.6% (1) 31.3% (26) 32.3% (20) 56.6% (47) 48.4% (30)
0.09
82.2 (11.0)
0.21
79.9 (9.9)
*Statistics: t-test (compare means), Fisher’s Exact test (compare proportions with 2 outcomes), Chi-Squared test (compare proportions with >2 outcomes).
*Statistics: t-test (compare means), Fisher’s Exact Test (compare proportions with 2 outcomes), Chi-Squared Test (compare proportions with >2 outcomes).
Subjects’ average baseline VAS scores were 81.2 § 10.5 mm in the DRG stimulation group, and 81.3 § 10.6 mm in the SCS group. Pain medications and comorbidities were similar in both groups. As shown in Table 1, there were no significant differences between the groups in baseline characteristics (P ≥ .65). Finally, there was no significant difference in baseline characteristics between subjects diagnosed with CRPS-I or CRPS-II (Table 2). The 4-way repeated-measures ANOVA yielded significant results for the interaction term (P < .0465), thus allowing us to perform 1-way repeated-measures ANOVA for all subjects and stratified by diagnosis and gender.
Pain Relief for All Subjects The average PPR as measured by VAS is displayed in Fig 1 for the DRG stimulation and SCS groups. Subjects in both groups experienced significantly greater pain relief at the end of the trial stimulation period (when subjects were assessed for eligibility for permanent implant) than at any other time point (P < .01). At the end of Trial neurostimulation, DRG stimulation and SCS subjects experienced average PPR of 82.2% and 77.0%, respectively. During ongoing follow-up visits, the DRG stimulation group’s level of pain relief was 69.8% at 1 month and remained stable from 1 month through 12 months, when pain relief was 69.3%. The SCS group’s pain relief at 1 month was 66.9%, but this declined significantly to 58.3% at 9 months and 57.9% at 12 months (P < .01). A full summary of the statistical results is provided in Appendix A.
The responder rate at each study follow-up visit for the DRG stimulation and SCS groups is shown in Fig 2. The responder rate was highest at the end of the trial stimulation period for both groups, in which 89.0% of DRG stimulation subjects and 86.1% of SCS subjects had at least 50% pain relief from baseline. For the DRG stimulation group, this responder rate declined to 74.0% at 1 month, but remained relatively stable at 75.3% by 12 months. In contrast, for the SCS group, the responder rate declined to 68.1% at 1 month, and continued to fall to 61.1% at 12 months.
Pain Relief by Diagnosis The average pain relief at each study follow-up visit for subjects diagnosed with CRPS-I in the DRG stimulation and SCS groups is presented in Fig 3. CRPS-I subjects in the DRG stimulation group had significantly greater pain relief at the end of the trial stimulation period (79.1%) than at any other time point except at 3 months (74.3%) (P < .05). Pain relief was 65.4% at 1 month, increased significantly to 74.3% at 3 months (P < .05), and remained relatively stable through 12 months (67.7%). CRPS-I subjects in the SCS group also had significantly greater pain relief at the end of the trial stimulation period than at any other time point (77.6%) (P < .005). Pain relief at 1 month was 68.6%, but continued to decline significantly to 56.0% and 54.6% at 9 months and 12 months, respectively (P < .01). The average pain relief at each study follow-up visit for subjects diagnosed with CRPS-II in the DRG stimulation and SCS groups is presented in Fig 4. CRPS-II subjects in both groups had significantly greater pain relief at the end of the trial stimulation period than at any other time point, 86.5% and 76.3%, respectively (P < .5). For the DRG group, pain relief stabilized at 1 month
ARTICLE IN PRESS Levy et al
The Journal of Pain 5
Figure 1. Percentage pain relief over time in the DRG stimulation (n = 73) and SCS (n = 72) groups. Error bars are standard error of
the mean. One-way ANOVA: P < .0001 (DRG stimulation and SCS groups). # indicates significant differences between End of TNS and all other follow-up visits within the same group (P < .01). * indicates significant differences between 1 month and the indicated follow-up visits for the SCS group (P < .01). The numbers near the bottom of each bar provide the percentage pain relief value corresponding to that bar. Abbreviations: DRG, dorsal root ganglion; SCS, spinal cord stimulation; TNS, Trial neurostimulation.
Figure 2. Responder rates over time in the DRG stimulation (n = 73) and SCS (n = 72) groups. * indicates significant differences between End of TNS and the indicated follow-up visits within the same group (logistic regression, P < .05).
(75.8%) and there were no significant changes in pain relief through 12 months (71.4%). Pain relief for CRPS-II subjects in the SCS group at 1 month was 64.7% and remained stable to 12 months (62.2%).
Pain Relief by Gender The mean pain relief for male and female subjects in the DRG stimulation and SCS groups is shown in Fig. 5 and 6, respectively. For males in the DRG group, pain
ARTICLE IN PRESS 6 The Journal of Pain
Therapy Habituation for SCS Versus DRG Stimulation
Figure 3. Percentage pain relief over time for CRPS-I subjects in the DRG stimulation (n = 42) and SCS (n = 41) groups. Error bars are standard error of the mean. One-way ANOVA: p<0.01 (DRG stimulation group) and P < .0001 (SCS group). # indicates significant differences between end of TNS and all other follow-ups within the same group, except for the 3-month follow-up for the DRG stimulation group and the 1-month follow-up for the SCS group (P < .05). * indicates significant differences between 1 month and the indicated follow-up visits for the SCS group (P < .01). ^ indicates a significant difference between 1 month and the indicated follow-up visit for the DRG stimulation group (P < .05).
Figure 4. Percentage pain relief over time for CRPS-II subjects in the DRG stimulation (n = 31) and SCS (n = 31) groups. Error bars are standard error of the mean. One-way ANOVA: P < .01 (DRG stimulation group) and P < .05 (SCS group). # indicates significant differences between end of TNS and all other follow-ups within the same group (P < .05).
relief was significantly higher at the end of trial period (78.8%) than any other follow-up visit (P < .001). This was also true for male subjects in the SCS group, for which pain relief was highest at the end of trial (72.6%)
compared to all other follow-ups (P < .05). These male subjects in the SCS group saw a significant decline in pain relief from 1 month (65.3%) to 9 months (56.0%) and 12 months (54.6%) (P < .05). Female subjects in the
ARTICLE IN PRESS Levy et al
The Journal of Pain 7
Figure 5. Percentage pain relief over time for male subjects in the DRG stimulation (n = 34) and SCS (n = 35) groups. Error bars are standard error of the mean. One-way ANOVA: P < .001 (DRG stimulation group) and P < .0005 (SCS group). # indicates significant differences between end of TNS and all other follow-ups within the same group, except for the 1-month follow-up for the SCS stimulation group (P < .05). * indicates significant differences between 1 month and the indicated follow-up visits for the SCS group (P < .05).
Figure 6. Percentage pain relief over time for female subjects in the DRG stimulation (n = 39) and SCS (n = 37) groups. Error bars are standard error of the mean. One-way ANOVA: P < .05 (DRG stimulation group) and P < .0005 (SCS group). # indicates significant differences between end of TNS and all other follow-ups within the same group, except for the 3- and 6-month follow-ups for the DRG stimulation group (P < .05). ^ indicates a significant difference between 3 months and both the 1- and 12-month follow-ups for the DRG stimulation group (P < .05). DRG stimulation and SCS groups had significantly greater pain relief at the end of trial (85.2% and 81.2%, respectively) compared to other follow-ups (P < .05), except for the 3 and 6 months visits for the
DRG group. Finally, the pain relief obtained by females in the DRG group at 3 months (84.1%) was significantly greater than at 1 month (75.3%) and 12 months (74.2%) (P < .05).
ARTICLE IN PRESS 8 The Journal of Pain
Discussion This ACCURATE study subanalysis sought to compare therapy habituation for SCS and DRG stimulation in patients with CRPS-I or CRPS-II. Previous studies have demonstrated that long-term habituation to SCS occurs in 13 to 25.9% of patients across different pain diagnoses and follow-up durations,1,4,9,11,14,18 but it was unknown whether habituation also occurred with DRG stimulation. Here, it was found that subjects’ pain intensity was reduced compared to baseline following treatment with either the DRG or SCS system. Moreover, subjects with DRG stimulation maintained the analgesic effects of stimulation over time from 1 month through 12 months, while the SCS group experienced significant diminishing pain relief at 9- and 12-months postpermanent implant. This was also observed as a more substantial decline in the responder rate from 1 month to 12 months for the SCS group than the DRG group. These results suggest that DRG stimulation is less susceptible to therapeutic habituation than SCS. The reason that DRG stimulation may be less susceptible to habituation than SCS needs further exploration. However, a variety of lines of inquiry may help elucidate this finding. One hypothesis is that DRG stimulation may affect the underlying pathology that results in pain symptoms differently than SCS. Previous research has suggested that pathophysiological changes of primary sensory neurons within the DRG are critically involved in the development and maintenance of chronic pain in CRPS-I and CRPS-II.12,24 Injured primary sensory neurons can become hyper-excitable and generate ectopic firing from the cell soma, which are located within the DRG.6,26 This ectopic activity propagates downstream and can produce afferent pain signaling in the spinal cord. Stimulation of neurons, directly at the relevant DRG anatomy where the pathology originates, can reduce the neuronal hyperexcitability that can otherwise generate pain signals.12,13 Since the analgesic effects of DRG stimulation may in part be generated locally within the DRG, rather than via downstream synaptic networks, this may limit the extent or impact of maladaptive plasticity changes at downstream synaptic networks that could otherwise produce habituation.14 In contrast, SCS is thought to generate therapeutic effects by altering activity in the dorsal horn synaptic network, and so may be more susceptible to maladaptive plasticity. A second possible explanation for the differences in habituation is that there are multiple, synergistic effects resulting from DRG stimulation that produce a more robust change in the chronic pain state, such as the impact of DRG stimulation on the inflammatory reflex.7 Third, it was previously shown that DRG stimulation provided greater improvement in psychological disposition versus tonic SCS,5 possibly through modulation by DRG stimulation of the medial pain pathway associated with the emotional aspects of pain.20 Alleviating the adverse psychological factors associated with pain could enhance long-term outcomes.22 Finally, some patients with CRPS-I or CRPS-II may find that the paresthesia-sensation associated with SCS becomes less
Therapy Habituation for SCS Versus DRG Stimulation pleasant over time and consequently reduce their use of SCS, whereas utilization of paresthesia-free DRG stimulation in some patients25 may limit the occurrence of this trend. However, these theories warrant further examination and research. Different trends in pain relief over time were observed in the SCS group for CRPS-I versus CRPS-II subjects, and male versus female subjects. CRPS-I subjects in the SCS group demonstrated significantly reduced pain relief at 9- and 12-months postpermanent implant, while CRPS-II subjects’ level of pain relief was stable from the 1-month follow-up visit through the 12-month visit. The reasons for the different trends observed between CRPS-I and CRPS-II subjects in pain relief are uncertain. A similar trend was observed with male subjects in the SCS group demonstrating significantly reduced pain relief at 9- and 12-months, whereas female subjects showed stable pain relief. While differences in pain perception and treatment effects have been noted between males and females,17,19 it is unknown why male subjects appeared to be more susceptible to habituation of SCS therapy, and a more general analysis of gender differences would be worthwhile. There were several limitations to this study. First, this was an underpowered, retrospective analysis of data from the ACCURATE RCT study, which was not performed for the specific purpose of this analysis. Because there were no specific methodological considerations in the ACCURATE RCT to ensure equitable treatments through the 12-month visit, the results observed in this analysis could be attributed to some unknown or unmeasured influence. Second, the ACCURATE RCT was a nonblinded study, which may have introduced expectation bias by the patients and bias in outcomes assessment by the investigators or study staff. Third, only 12 months of data was available for our retrospective analysis. It is unclear from these data what the effect of habituation is over longer time horizons. Fourth, the use of LOCF to impute missing data requires the assumption that subjects’ pain intensity remained unchanged following their last VAS measurement at the time of exit. However, since all subjects had chronic pain for an average of greater than 7 years, and had failed prior treatments, it is unlikely that subjects would experience a clinically meaningful change in pain intensity following exit. Fifth, the study conclusions are limited to CRPSI and CRPS-II patients who had chronic severe lower limb pain, baseline VAS scores of at least 60 mm, and pain lasting on average 7.5 § 7.6 years in the DRG stimulation group, and 7.0 § 7.7 years in the SCS group. It is unknown if the same patterns of habituation would be observed in a population of CRPS-I and CRPS-II patients treated at an earlier stage in their disease progression, with pain in other anatomical locations such as the upper limbs, or different disease indications. Sixth, the diagnosis of CRPS-II used in the ACCURATE study may not have followed Budapest diagnosis criteria. Finally, the assessment of habituation for SCS was specific to the low-frequency, tonic waveform, and newer SCS waveforms, such as high-frequency
ARTICLE IN PRESS Levy et al and BurstDR stimulation, may have different rates of habituation.2,23
Conclusions
The Journal of Pain 9
Acknowledgments The authors would like to acknowledge the investigators of the ACCURATE study, including Drs. Jon Obray, Christopher Kim, Thomas Yearwood, Hazmer Cassim, Mark Netherton, Nathan Miller, Edward Tavel, and Timothy Davis. Linda Johnson made writing contributions to the manuscript.
Tonic SCS is an accepted treatment for CRPS-I and CRPSII, but patients treated with DRG stimulation appeared to have more durable pain relief over time as compared to patients treated with SCS. More specifically, DRG stimulation showed stable pain relief from 1 month through 12 months postimplant, while tonic SCS demonstrated diminishing pain relief to therapy at 9- and 12-months. Long-term therapy habituation occurred at 12 months in the SCS group but not the DRG group, despite equally attentive follow-up and reprogramming.
Supplementary Data
References
10. Kemler MA, Barendse GA, van Kleef M, de Vet HC, Rijks e CA, van den Wildenberg FA: Spinal cord stimulaCP, Furne tion in patients with chronic reflex sympathetic dystrophy. N Engl J Med 343:618-624, 2000
KM, Yland MJ, Charnov JH, Redko V: Multiple pro1. Alo gram spinal cord stimulation in the treatment of chronic pain: Follow-up of multiple program SCS. Neuromodulation 2:266-272, 1999 KM, Oakley J, Feler CA: Spinal cord stim2. Bennett DS, Alo ulation for complex regional pain syndrome I [RSD]: A retrospective multicenter experience from 1995 to 1998 of 101 patients. Neuromodulation 2:202-210, 1999 3. Bennett DS, Brookoff D: Complex regional pain syndromes (reflex sympathetic dystrophy and causalgia) and spinal cord stimulation. Pain Med 7:S64-S96, 2006 4. Burchiel KJ, Anderson VC, Brown FD, Fessler RG, Friedman WA, Pelofsky S, Weiner RL, Oakley J, Shatin D: Prospective, multicenter study of spinal cord stimulation for relief of chronic back and extremity pain. Spine (Phila Pa 1976) 21:2786-2794, 1996 5. Deer TR, Levy RM, Kramer J, Poree L, Amirdelfan K, Grigsby E, Staats P, Burton AW, Burgher AH, Obray J, Scowcroft J, Golovac S, Kapural L, Paicius R, Kim C, Pope J, Yearwood T, Samuel S, McRoberts PW, Cassim H, Netherton M, Miller N, Schaufele M, Tavel E, Davis T, Davis K, Johnson L, Mekhail N: Dorsal root ganglion stimulation yielded higher treatment success rate for complex regional pain syndrome and causalgia at 3 and 12 months: a randomized comparative trial. Pain 158:669-681, 2017 6. Devor M: Ectopic discharge in Abeta afferents as a source of neuropathic pain. Exp Brain Res 196:115-128, 2009 7. Gravius N, Chaudhry SR, Muhammad S, Bostroem A, Gravius S, Randau T, Scheele D, Westhofen P, Kruppenbacher J, Stoffel-Wagner B, Maier C, Weidlich A, Yearwood TL, Chakravarthy K, Kramer J, Hurlemaan R, Kinfe T: Selective L4 dorsal root ganglion stimulation evokes pain relief and changes of inflammatory markers: Part I profiling of saliva and serum molecular patterns. Neuromodulation 22:44-52, 2019 8. Harden RN, Bruehl S, Stanton-Hicks M, Wilson PR: Proposed new diagnostic criteria for complex regional pain syndrome. Pain Med 8:326-331, 2007 9. Hayek SM, Veizi E, Hanes M: Treatment-limiting complications of percutaneous spinal cord stimulator implants: A review of eight years of experience from an academic center database. Neuromodulation 18:603-608, 2015
Supplementary data related to this article can be found at https://doi.org/10.1016/j.jpain.2019.08.005.
11. Kemler MA, de Vet HC, Barendse GA, van den Wildenberg FA, van Kleef M: Effect of spinal cord stimulation for chronic complex regional pain syndrome Type I: Five-year final follow-up of patients in a randomized controlled trial. J Neurosurg 108:292-298, 2008 12. Koopmeiners AS, Mueller S, Kramer J, Hogan QH: Effect of electrical field stimulation on dorsal root ganglion neuronal function. Neuromodulation 16:304-311, 2013. discussion 10-1 13. Krames ES: The role of the dorsal root ganglion in the development of neuropathic pain. Pain Med 15:1669-1685, 2014 14. Kumar K, Hunter G, Demeria D: Spinal cord stimulation in treatment of chronic benign pain: Challenges in treatment planning and present status, a 22-year experience. Neurosurgery 58:481-496, 2006. discussion 481-96 15. Lohnberg JA, Altmaier EM: A review of psychosocial factors in complex regional pain syndrome. J Clin Psychol Med Settings 20:247-254, 2013 16. Merskey H, Bogduk N: Classification of Chronic Pain: Descriptions of Chronic Pain Syndromes and Definitions of Pain Terms, 2nd ed. Seattle, IASP Press, 1994 17. Murphy KR, Han JL, Hussaini SM, Yang S, Parente B, Xie J, Lad SP: The volume-outcome effect: impact on trial-topermanent conversion rates in spinal cord stimulation. Neuromodulation 20:256-262, 2017 18. North RB, Kidd DH, Zahurak M, James CS, Long DM: Spinal cord stimulation for chronic, intractable pain: Experience over two decades. Neurosurgery 32:384-394, 1993. discussion 94-5 19. Paller CJ, Campbell CM, Edwards RR, Dobs AS: Sexbased differences in pain perception and treatment. Pain Med 10:289-299, 2009 20. Pawela CP, Kramer JM, Hogan QH: Dorsal root ganglion stimulation attenuates the BOLD signal response to noxious sensory input in specific brain regions: Insights into a possible mechanism for analgesia. Neuroimage 147:10-18, 2017 21. Schott GD: Mechanisms of causalgia and related clinical conditions. The role of the central and sympathetic nervous systems. Brain 109:717-738, 1986
ARTICLE IN PRESS 10
The Journal of Pain
22. Sparkes E, Duarte RV, Mann S, Lawrence TR, Raphael JH: Analysis of psychological characteristics impacting spinal cord stimulation treatment outcomes: A prospective assessment. Pain Physician 18:E369-E377, 2015 23. Van Buyten JP, Wille F, Smet I, Wensing C, Breel J, Karst € ggel-Kra €mer K, Vesper J: Therapy-related E, Devos M, Po explants after spinal cord stimulation: Results of an international retrospective chart review study. Neuromodulation 20:642-649, 2017 24. van Eijs F, Stanton-Hicks M, Van Zundert J, Faber CG, Lubenow TR, Mekhail N, van Kleep M, Huygen F: Evidencebased interventional pain medicine according to clinical
Therapy Habituation for SCS Versus DRG Stimulation diagnoses. 16. Complex regional pain syndrome. Pain Pract 11:70-87, 2011 25. Subthreshold Stimulation of the Dorsal Root Ganglia: Retrospective Study, in Verrills P, Mitchell B, Vivian D, Cusack W, Kramer J, (eds): Subthreshold Stimulation of the Dorsal Root Ganglia: Retrospective Study, Las Vegas, North American Neuromodulation Society, 2017 26. Wu G, Ringkamp M, Murinson BB, Pogatzki EM, Hartke TV, Weerahandi HM, Campbell JN, Griffin JW, Meyer RA: Degeneration of myelinated efferent fibers induces spontaneous activity in uninjured C-fiber afferents. J Neurosci 22:7746-7753, 2002
Original Reports Therapy Habituation at 12 Months: Spinal Cord Stimulation Versus Dorsal Root Ganglion Stimulation for Complex Regional Pain Syndrome Type I and II Robert M. Levy,* Nagy Mekhail,y Jeffrey Kramer,z Lawrence Poree,x Kasra Amirdelfan,{ Eric Grigsby,k Peter Staats,** Allen W. Burton,yy Abram H. Burgher,zz James Scowcroft,xx Stan Golovac,{{ Leonardo Kapural,kk Richard Paicius,*** Jason Pope,yyy Sam Samuel,y William Porter McRoberts,zzz Michael Schaufele,xxx Alexander R. Kent,yy Adil Raza,yy and Timothy R. Deer{{{ Marcus Neuroscience Institute, Boca Raton, Florida, yCleveland Clinic, Cleveland, Ohio, zVolta Research and University of Illinois College of Medicine, Chicago, Illinois, xUniversity of California at San Francisco, California, {IPM Medical Group, Inc., Walnut Creek, California, kNeurovations, Napa, California, **Premier Pain Center, Shrewsbury Township, New Jersey, yyAbbott Neuromodulation, Plano, Texas, zzHOPE Research − TPC, Phoenix, Arizona, xxPain Management Associates, Independence, Missouri, {{Florida Pain, Merritt Island, Florida, kkCarolinas Pain Institute, Winston-Salem, North Carolina, ***Newport Beach Headache and Pain, Newport Beach, California, yyyEvolve Restorative Center, Santa Rosa, California, zzzHoly Cross Hospital, Ft. Lauderdale, Florida, xxxDrug Studies America, Marietta, Georgia, {{{Spine and Nerve Center of the Virginias, Charleston, West Virginia *
Abstract: The ACCURATE randomized, controlled trial compared outcomes of dorsal root ganglion (DRG) stimulation versus tonic spinal cord stimulation (SCS) in 152 subjects with chronic lower extremity pain due to complex regional pain syndrome (CRPS) type I or II. This ACCURATE substudy was designed to evaluate whether therapy habituation occurs with DRG stimulation as compared to SCS through 12-months. A modified intention-to-treat analysis was performed to assess percentage pain relief (PPR) and responder rates at follow-up visits (end-of-trial, 1, 3, 6, 9, 12-months postpermanent implant) for all subjects that completed trial stimulation (DRG:N = 73, SCS:N = 72). For both groups, mean PPR was significantly greater at end-of-trial (DRG = 82.2%, SCS =0 77.0%) than all other follow-ups. Following permanent DRG system implantation, none of the time points were significantly different from one another in PPR (range = 69.3−73.9%). For the SCS group, PPR at 9-months (58.3%) and 12-months (57.9%) was significantly less than at 1-month (66.9%). The responder rate also decreased for the SCS group from 1-month (68.1%) to 12-months (61.1%). After stratifying by diagnosis, it was found that only the CRPS-I population had diminishing pain relief with SCS. DRG stimulation resulted in more stable pain relief through 12-months, while tonic SCS demonstrated therapy habituation at 9- and 12-months. Trial Registration: The ACCURATE study was registered at ClinicalTrials.gov with Identifier NCT01923285. Received November 1, 2018; Revised May 3, 2019; Accepted August 27, 2019. This work was supported by Abbott” and not funded. This study was a sub-analysis from the ACCURATE Trial (NCT01923285). Disclosures: Drs. Mekhail, Deer, Poree, Amirdelfan, Grigsby, Staats, Burton, Burgher, Scowcroft, Golovac, Kapural, Paicius, Pope, Samuel, McRoberts, Schaufele, and Levy were paid by Spinal Modulation & St. Jude Medical as investigators for the ACCURATE study. Dr. Levy has served as a consultant to Mainstay Medical, Nalu, Saluda Medical, and Abbott. Dr. Levy is or has been a minority shareholder in Saluda Medical, Spinal Modulation, Bioness, Vertos, and Nevro. Dr. Mekhail formerly had a consultation agreement with Spinal Modulation to serve as medical monitor of the ACCURATE study. Currently he is a consultant for Abbott, Saluda Medical, StimWave, Medtronic Neurological and Flowonix. Dr. Amirdelfan is a consultant for Nevro, Saluda Medical, Nalu Medical, Biotronik and ENSO. Dr. Scowcroft has served as a consultant for Abbott and Boston Scientific. Dr. Kapural is a consultant for Abbott, Nevro, Neuros, SPR Therapeutics, and Saluda Medical. Dr. Paicius is a consultant for Abbott, Nuvectra, Mesoblast, Vertiflex, Vertos, and Pacific Stem Cells. Dr. Pope is a consultant for Abbott, Flowonix, Saluda Medical, Jazz Pharmaceuticals,
SPR Therapeutics, and Vertiflex. Dr. McRoberts serves or has served as a consultant for Abbott, Medtronic, Nevro, Boston Scientific, Bioness, Vertiflex, and SPR Therapeutics. Dr. Deer is a consultant for Axonics, Bioness, Flowonix, Medtronic, Jazz Pharmaceuticals, Nevro, Abbott, and Saluda Medical and has consulting/equity for Axonics and Bioness. Dr. Deer formerly had equity in Spinal Modulation and Nevro. Dr. Kramer was a former consultant for Abbott, Nalu Medical, Autonomic Technologies, Thimble Bioelectronics, Circuit Therapeutics, and CereVu. Dr. Kramer has equity in Abbott as a former employee, and formerly had equity in Spinal Modulation. Dr. Kramer is now an employee of Medtronic. Dr. Burton, Dr. Kent, and Mr. Raza receive salary and have equity as employees of Abbott. The remaining authors report no conflicts of interest. Address reprint requests to Alexander Kent, PhD, Abbott Neuromodulation, 645 Almanor Ave, Sunnyvale, CA 94085. E-mail: [email protected] 1526-5900/$36.00 © 2019 by the American Pain Society https://doi.org/10.1016/j.jpain.2019.08.005
1
ARTICLE IN PRESS 2 The Journal of Pain
Therapy Habituation for SCS Versus DRG Stimulation
Perspective: This article reports on an ACCURATE substudy, which found that long-term therapy habituation occurred at 12-months with SCS, but not DRG stimulation, in patients with CRPS. The underlying mechanisms of action for these results remain unclear, although several lines of inquiry are proposed. © 2019 by the American Pain Society Key Words: Spinal cord stimulation, dorsal root ganglion stimulation, habituation, tolerance, complex regional pain syndrome.
Introduction Complex regional pain syndrome (CRPS) type I and CRPS type II diminish quality of life and can lead to significant disability.15 Both conditions have varied and complex clinical features, with chronic debilitating pain being the primary symptom, usually involving the lower or upper extremities.16,24 The conditions are differentiated by the absence (CRPS-I) or presence (CRPS-II) of demonstrable nerve damage as an underlying etiology. Spinal cord stimulation (SCS) is an accepted treatment for CRPS-I and CRPS-II when conventional treatments have failed.10,24 SCS systems consist of stimulation lead (s) placed near the spinal cord, and a pulse generator which generates electrical stimulation pulses. Tonic SCS has been used for decades and is considered the standard stimulation mode. In this paradigm, electrical pulses are delivered in a continuous manner at a low frequency (<100 Hz), and generally cause a paresthesia sensation often described as buzzing or tingling.5 Dorsal root ganglion (DRG) stimulation uses lead(s) placed near the target DRG, which are connected to a pulse generator to deliver stimulation at approximately 20 Hz.5 The ACCURATE study was a randomized controlled multicenter trial (RCT) designed to evaluate the safety and effectiveness of DRG stimulation as compared to tonic SCS therapy.5 This study was performed in 152 subjects with lower extremity pain due to CRPS-I or CRPS-II, and who were enrolled at 22 clinical sites in the United States. The results of the ACCURATE study demonstrated that DRG stimulation therapy was associated with a significantly greater level of treatment success (≥50% pain relief at trial and 3 month follow-up compared to baseline, and freedom from stimulationrelated neurological deficits) than SCS (81.2% versus 55.7%; noninferiority: P < .001, superiority: P < .001). The literature provides reports of habituation with low-frequency, tonic SCS therapy for a variety of pain conditions, including CRPS. Habituation (or “tolerance”) has been defined as the “development of inadequate pain relief [after acute success], despite appropriate stimulation coverage that cannot be explained by a hardware-related issue.”9,14 Kumar reported that habituation occurred in 25.9% of 328 patients at long-term follow-up (range = 1−22 years), including 17.9% of the 28 CRPS-I and CRPS-II subject population.14 Hayek found that 13.7% of 235 patients experienced habituation (mean follow-up = 45 § 25 months).9 Moreover, habituation was the most common reason for device explant, with a median time to explant of 19.6 months. Similarly, North found a 13% habituation rate for 205 patients at long-term follow-up (mean = 7.1 years, range = 1.5−20.4
found that 18% of 80 patients were years).18 Alo explanted between 24 and 30 months because of insufficient pain relief despite good paresthesia coverage.1 Burchiel documented 14.5% of 145 patients who discontinued use of SCS or had their device explanted due to ineffective stimulation at one year.4 Lastly, Kemler reported that the analgesic effects of SCS with physical therapy were not significantly different from physical therapy alone starting at the 3-year follow-up.11 SCS therapy habituation is not well understood, but multiple complex factors have been suggested to explain this phenomenon. Habituation may be due to disease progression in CRPS-I and CRPS-II, plasticity of central pain pathways, fibrosis around the stimulating electrodes, or other physiological and psychological factors (eg, short-term placebo effect, Hawthorne effect).3,14 Moreover, patients may exaggerate analgesia during the trial period to receive a permanent SCS implant, leading to artificially enhanced pain relief scores in the short-term. The current paper reports on a subanalysis of the ACCURATE RCT data that evaluated habituation with DRG stimulation and SCS therapies. The primary hypothesis was that SCS would show a significant decline in pain relief at 12 months compared to 1 month, whereas no significant change would be observed between these timepoints for DRG stimulation.
Methods Study Design The ACCURATE RCT was a randomized, controlled multicenter study conducted at 22 investigative centers in the United States from 2013 through 2016. This study evaluated the safety and effectiveness of DRG stimulation compared to tonic SCS for the treatment of chronic lower extremity pain due to CRPS-I or CRPS-II for at least 6 months.5 All sites obtained Institutional Review Board approval, and subjects signed informed consent prior to enrollment.
Subject Selection All subjects were screened per the study’s inclusion/ exclusion criteria. Briefly, enrolled subjects (nay¨ve to neurostimulation) had pain in the chief area of complaint of ≥60 mm on a visual analog scale (VAS), failed at least 2 pharmacologic treatments from 2 drug classes, had stable neurological function, and no psychological pathology that contraindicated an implantable device. All subjects’ medical, psychological, and imaging records
ARTICLE IN PRESS Levy et al were evaluated by a medical monitor, independent of the study sponsor and study investigators, to ensure appropriate subject selection. The diagnosis of lower extremity CRPS-I or CRPS-II was confirmed by an experienced medical monitor (N.M.) for strict adherence to the following criteria during the ACCURATE study.5 Diagnosis of CRPS-I was based on the Budapest criteria.8 CRPS-II (referred to as causalgia in ACCURATE) was defined in the study as a painful condition arising from damage to a nerve resulting in chronic pain, which may or may not have secondary symptoms.21
Devices and Procedure The DRG stimulation system (Axium Neurostimulation System, Abbott, IL) components were percutaneous leads designed to stimulate the DRG, an external trial pulse generator, and an implantable pulse generator. The SCS devices were commercially-available pulse generators (RestoreUltra and RestoreSensor, Medtronic, MN). The postural sensing feature of the RestoreSensor pulse generator was not activated for this study, which allowed comparison of DRG stimulation to the traditional SCS paradigm without postural adjustment that was most commonly used at the time of the study. A maximum of 16 contacts were placed per patient in the DRG stimulation and SCS groups.
Data Collection and Statistical Analysis Outcome data was collected by study staff on case report forms in a nonblinded manner. A description of all scales used to collect primary and second endpoints is provided in the ACCURATE study publication.5 The primary outcome measure for this substudy was change in percentage pain relief (PPR) over time, as measured by the percentage change from baseline for in-clinic VAS scores in the primary area of pain to each study follow-up visit: End of Trial neurostimulation, 1 month, 3 months, 6 months, 9 months, and 12 months following permanent implant. As a secondary outcome measure, the responder rate at each study follow-up was calculated as the percentage of subjects with at least 50% pain relief. This modified intention-to-treat analysis (MITT) included all subjects who completed the trial stimulation phase, with missing data from any subsequent time points imputed by last observation carried forward (LOCF) so that patients were not excluded for missing a single study visit. Of the 320 patients consented in the study, a total of 152 subjects were enrolled and randomized 1:1 in the ACCURATE study (76 DRG stimulation, 76 SCS). Six subjects (3 DRG stimulation, 3 SCS) withdrew prior to undergoing the trial stimulation phase, and 1 additional SCS subject never completed the trial stimulation phase. Subjects that failed at the end of the trial visit (<50% pain relief), deemed treatment failures, were not eligible for a permanent implant and were exited from the ACCURATE study per protocol. The MITT population for this analysis completed the end of
The Journal of Pain 3 trial stimulation visit, and consisted of 145 subjects (73 DRG stimulation, 72 SCS). This MITT population was based on standard intention-to-treat principles, in which subjects were analyzed based on their initial randomized treatments. Subsequently, any subjects that exited the study for any reason at any time were included in this analysis. LOCF was applied at follow-up visits for several subjects in the DRG group (1 month follow-up: 12 subjects; 3 months: 14 subjects; 6 months: 14 subjects; 9 months: 18 subjects; 12 months: 18 subjects) and SCS group (1 month follow-up: 18 subjects; 3 months: 18 subjects; 6 months: 20 subjects; 9 months: 23 subjects; 12 months: 22 subjects). Statistical analyses were performed in MATLAB R2016B (MathWorks, Natick, MA). Prior to assessing therapy habituation for each stimulation type (DRG or SCS), a 4-way repeated-measures analysis of variance (ANOVA) was performed, in which follow-up visit, diagnosis, gender and stimulation type were the independents variable and PPR was the dependent variable (a = .05). If the 4-way interaction term was significant, 1-way repeated-measures ANOVA tests were performed, in which follow-up visit was the independent variable and PPR was the dependent variable (a = .05). If the ANOVA was significant, post-hoc Fisher’s least significant different tests were performed to identify which follow-up visits were statistically different. Performing the Fisher’s least significant different post-hoc tests only after finding that the global ANOVA was significant protected against type I error from multiple comparisons. This analysis was performed for all subjects, and then repeated after stratifying by diagnosis (CRPS-I or CRPS-II) or gender to determine if any observed habituation effects were dependent on the pain condition or patient demographics. Finally, to assess significant changes in responder rates, logistic regression was performed for each stimulation type, in which the follow-up visit was the predictor variables and responder (≥50% pain relief) was the response variable.
Results Subjects’ Baseline Characteristics An overview of subject’s baseline characteristics is provided in Table 1. Briefly, subjects’ diagnoses were CRPS-I (DRG stimulation 57.5% and SCS 56.9%) and CRPS-II (DRG stimulation 42.5% and SCS 43.1%). The average age of subjects was 52.5 § 12.7 years in the DRG stimulation group and 52.3 § 11.7 years in the SCS group. Females comprised 53.4% (DRG stimulation) and 51.4% (SCS) of the study population. Average duration of chronic lower limb pain was 7.5 § 7.6 years in the DRG stimulation group, and 7.0 § 7.7 years in the SCS group. The primary regions of lower limb pain treated in the study included groin pain (DRG stimulation 9.6%, SCS 12.5%), buttock pain (DRG stimulation 2.7%, SCS 5.6%), leg pain (DRG stimulation 30.1%, SCS 33.3%), and foot pain (DRG stimulation 57.5%, SCS 48.6%).
ARTICLE IN PRESS Therapy Habituation for SCS Versus DRG Stimulation
4 The Journal of Pain
Baseline Characteristics of Subjects in the DRG and SCS Groups
Table 1.
BASELINE CHARACTERISTICS N Age Mean (SD) Gender % Male (N) % Female (N) Race % White (N) % Other (N) Primary diagnosis % CRPS-I (N) % CRPS-II (N) Duration of lower limb pain (y) Mean (SD) Region of lower limb Pain % Groin (N) % Buttocks (N) % Leg (N) % Foot (N) Baseline VAS score (mm) Mean (SD)
DRG STIMULATION
SCS
P VALUE*
73
72
N/A
52.5 (12.7)
52.3 (11.7)
0.93
46.6% (34) 53.4% (39)
48.6% (35) 51.4% (37)
0.87
94.5% (69) 5.5% (4)
91.7% (66) 8.3% (6)
0.53
57.5% (42) 42.5% (31)
56.9% (41) 43.1% (31)
1.00
7.5 (7.6)
7.0 (7.7)
0.68
9.6% (7) 2.7% (2) 30.1% (22) 57.5% (42)
12.5% (9) 5.6% (4) 33.3% (24) 48.6% (35)
0.65
81.2 (10.5)
81.3 (10.6)
0.96
Table 2. Baseline Characteristics of Subjects With CRPS-I and CRPS-II BASELINE CHARACTERISTICS N Age Mean (SD) Gender % Male (N) % Female (N) Race % White (N) % Other (N) Duration of lower limb pain (y) Mean (SD) Region of lower limb Pain % Groin (N) % Buttocks (N) % Leg (N) % Foot (N) Baseline VAS score (mm) Mean (SD)
CRPS-I
CRPS-II
P VALUE*
83
62
N/A
50.8 (12.7)
54.5 (11.2)
0.08
56.6% (47) 53.2% (33) 43.4% (36) 46.8% (29)
0.74
92.8% (77) 93.5% (58) 7.2% (6) 6.5% (4)
1.00
6.7 (7.3)
8.1 (8.1)
0.28
6.0% (5) 17.7% (11) 6.0% (5) 1.6% (1) 31.3% (26) 32.3% (20) 56.6% (47) 48.4% (30)
0.09
82.2 (11.0)
0.21
79.9 (9.9)
*Statistics: t-test (compare means), Fisher’s Exact test (compare proportions with 2 outcomes), Chi-Squared test (compare proportions with >2 outcomes).
*Statistics: t-test (compare means), Fisher’s Exact Test (compare proportions with 2 outcomes), Chi-Squared Test (compare proportions with >2 outcomes).
Subjects’ average baseline VAS scores were 81.2 § 10.5 mm in the DRG stimulation group, and 81.3 § 10.6 mm in the SCS group. Pain medications and comorbidities were similar in both groups. As shown in Table 1, there were no significant differences between the groups in baseline characteristics (P ≥ .65). Finally, there was no significant difference in baseline characteristics between subjects diagnosed with CRPS-I or CRPS-II (Table 2). The 4-way repeated-measures ANOVA yielded significant results for the interaction term (P < .0465), thus allowing us to perform 1-way repeated-measures ANOVA for all subjects and stratified by diagnosis and gender.
Pain Relief for All Subjects The average PPR as measured by VAS is displayed in Fig 1 for the DRG stimulation and SCS groups. Subjects in both groups experienced significantly greater pain relief at the end of the trial stimulation period (when subjects were assessed for eligibility for permanent implant) than at any other time point (P < .01). At the end of Trial neurostimulation, DRG stimulation and SCS subjects experienced average PPR of 82.2% and 77.0%, respectively. During ongoing follow-up visits, the DRG stimulation group’s level of pain relief was 69.8% at 1 month and remained stable from 1 month through 12 months, when pain relief was 69.3%. The SCS group’s pain relief at 1 month was 66.9%, but this declined significantly to 58.3% at 9 months and 57.9% at 12 months (P < .01). A full summary of the statistical results is provided in Appendix A.
The responder rate at each study follow-up visit for the DRG stimulation and SCS groups is shown in Fig 2. The responder rate was highest at the end of the trial stimulation period for both groups, in which 89.0% of DRG stimulation subjects and 86.1% of SCS subjects had at least 50% pain relief from baseline. For the DRG stimulation group, this responder rate declined to 74.0% at 1 month, but remained relatively stable at 75.3% by 12 months. In contrast, for the SCS group, the responder rate declined to 68.1% at 1 month, and continued to fall to 61.1% at 12 months.
Pain Relief by Diagnosis The average pain relief at each study follow-up visit for subjects diagnosed with CRPS-I in the DRG stimulation and SCS groups is presented in Fig 3. CRPS-I subjects in the DRG stimulation group had significantly greater pain relief at the end of the trial stimulation period (79.1%) than at any other time point except at 3 months (74.3%) (P < .05). Pain relief was 65.4% at 1 month, increased significantly to 74.3% at 3 months (P < .05), and remained relatively stable through 12 months (67.7%). CRPS-I subjects in the SCS group also had significantly greater pain relief at the end of the trial stimulation period than at any other time point (77.6%) (P < .005). Pain relief at 1 month was 68.6%, but continued to decline significantly to 56.0% and 54.6% at 9 months and 12 months, respectively (P < .01). The average pain relief at each study follow-up visit for subjects diagnosed with CRPS-II in the DRG stimulation and SCS groups is presented in Fig 4. CRPS-II subjects in both groups had significantly greater pain relief at the end of the trial stimulation period than at any other time point, 86.5% and 76.3%, respectively (P < .5). For the DRG group, pain relief stabilized at 1 month
ARTICLE IN PRESS Levy et al
The Journal of Pain 5
Figure 1. Percentage pain relief over time in the DRG stimulation (n = 73) and SCS (n = 72) groups. Error bars are standard error of
the mean. One-way ANOVA: P < .0001 (DRG stimulation and SCS groups). # indicates significant differences between End of TNS and all other follow-up visits within the same group (P < .01). * indicates significant differences between 1 month and the indicated follow-up visits for the SCS group (P < .01). The numbers near the bottom of each bar provide the percentage pain relief value corresponding to that bar. Abbreviations: DRG, dorsal root ganglion; SCS, spinal cord stimulation; TNS, Trial neurostimulation.
Figure 2. Responder rates over time in the DRG stimulation (n = 73) and SCS (n = 72) groups. * indicates significant differences between End of TNS and the indicated follow-up visits within the same group (logistic regression, P < .05).
(75.8%) and there were no significant changes in pain relief through 12 months (71.4%). Pain relief for CRPS-II subjects in the SCS group at 1 month was 64.7% and remained stable to 12 months (62.2%).
Pain Relief by Gender The mean pain relief for male and female subjects in the DRG stimulation and SCS groups is shown in Fig. 5 and 6, respectively. For males in the DRG group, pain
ARTICLE IN PRESS 6 The Journal of Pain
Therapy Habituation for SCS Versus DRG Stimulation
Figure 3. Percentage pain relief over time for CRPS-I subjects in the DRG stimulation (n = 42) and SCS (n = 41) groups. Error bars are standard error of the mean. One-way ANOVA: p<0.01 (DRG stimulation group) and P < .0001 (SCS group). # indicates significant differences between end of TNS and all other follow-ups within the same group, except for the 3-month follow-up for the DRG stimulation group and the 1-month follow-up for the SCS group (P < .05). * indicates significant differences between 1 month and the indicated follow-up visits for the SCS group (P < .01). ^ indicates a significant difference between 1 month and the indicated follow-up visit for the DRG stimulation group (P < .05).
Figure 4. Percentage pain relief over time for CRPS-II subjects in the DRG stimulation (n = 31) and SCS (n = 31) groups. Error bars are standard error of the mean. One-way ANOVA: P < .01 (DRG stimulation group) and P < .05 (SCS group). # indicates significant differences between end of TNS and all other follow-ups within the same group (P < .05).
relief was significantly higher at the end of trial period (78.8%) than any other follow-up visit (P < .001). This was also true for male subjects in the SCS group, for which pain relief was highest at the end of trial (72.6%)
compared to all other follow-ups (P < .05). These male subjects in the SCS group saw a significant decline in pain relief from 1 month (65.3%) to 9 months (56.0%) and 12 months (54.6%) (P < .05). Female subjects in the
ARTICLE IN PRESS Levy et al
The Journal of Pain 7
Figure 5. Percentage pain relief over time for male subjects in the DRG stimulation (n = 34) and SCS (n = 35) groups. Error bars are standard error of the mean. One-way ANOVA: P < .001 (DRG stimulation group) and P < .0005 (SCS group). # indicates significant differences between end of TNS and all other follow-ups within the same group, except for the 1-month follow-up for the SCS stimulation group (P < .05). * indicates significant differences between 1 month and the indicated follow-up visits for the SCS group (P < .05).
Figure 6. Percentage pain relief over time for female subjects in the DRG stimulation (n = 39) and SCS (n = 37) groups. Error bars are standard error of the mean. One-way ANOVA: P < .05 (DRG stimulation group) and P < .0005 (SCS group). # indicates significant differences between end of TNS and all other follow-ups within the same group, except for the 3- and 6-month follow-ups for the DRG stimulation group (P < .05). ^ indicates a significant difference between 3 months and both the 1- and 12-month follow-ups for the DRG stimulation group (P < .05). DRG stimulation and SCS groups had significantly greater pain relief at the end of trial (85.2% and 81.2%, respectively) compared to other follow-ups (P < .05), except for the 3 and 6 months visits for the
DRG group. Finally, the pain relief obtained by females in the DRG group at 3 months (84.1%) was significantly greater than at 1 month (75.3%) and 12 months (74.2%) (P < .05).
ARTICLE IN PRESS 8 The Journal of Pain
Discussion This ACCURATE study subanalysis sought to compare therapy habituation for SCS and DRG stimulation in patients with CRPS-I or CRPS-II. Previous studies have demonstrated that long-term habituation to SCS occurs in 13 to 25.9% of patients across different pain diagnoses and follow-up durations,1,4,9,11,14,18 but it was unknown whether habituation also occurred with DRG stimulation. Here, it was found that subjects’ pain intensity was reduced compared to baseline following treatment with either the DRG or SCS system. Moreover, subjects with DRG stimulation maintained the analgesic effects of stimulation over time from 1 month through 12 months, while the SCS group experienced significant diminishing pain relief at 9- and 12-months postpermanent implant. This was also observed as a more substantial decline in the responder rate from 1 month to 12 months for the SCS group than the DRG group. These results suggest that DRG stimulation is less susceptible to therapeutic habituation than SCS. The reason that DRG stimulation may be less susceptible to habituation than SCS needs further exploration. However, a variety of lines of inquiry may help elucidate this finding. One hypothesis is that DRG stimulation may affect the underlying pathology that results in pain symptoms differently than SCS. Previous research has suggested that pathophysiological changes of primary sensory neurons within the DRG are critically involved in the development and maintenance of chronic pain in CRPS-I and CRPS-II.12,24 Injured primary sensory neurons can become hyper-excitable and generate ectopic firing from the cell soma, which are located within the DRG.6,26 This ectopic activity propagates downstream and can produce afferent pain signaling in the spinal cord. Stimulation of neurons, directly at the relevant DRG anatomy where the pathology originates, can reduce the neuronal hyperexcitability that can otherwise generate pain signals.12,13 Since the analgesic effects of DRG stimulation may in part be generated locally within the DRG, rather than via downstream synaptic networks, this may limit the extent or impact of maladaptive plasticity changes at downstream synaptic networks that could otherwise produce habituation.14 In contrast, SCS is thought to generate therapeutic effects by altering activity in the dorsal horn synaptic network, and so may be more susceptible to maladaptive plasticity. A second possible explanation for the differences in habituation is that there are multiple, synergistic effects resulting from DRG stimulation that produce a more robust change in the chronic pain state, such as the impact of DRG stimulation on the inflammatory reflex.7 Third, it was previously shown that DRG stimulation provided greater improvement in psychological disposition versus tonic SCS,5 possibly through modulation by DRG stimulation of the medial pain pathway associated with the emotional aspects of pain.20 Alleviating the adverse psychological factors associated with pain could enhance long-term outcomes.22 Finally, some patients with CRPS-I or CRPS-II may find that the paresthesia-sensation associated with SCS becomes less
Therapy Habituation for SCS Versus DRG Stimulation pleasant over time and consequently reduce their use of SCS, whereas utilization of paresthesia-free DRG stimulation in some patients25 may limit the occurrence of this trend. However, these theories warrant further examination and research. Different trends in pain relief over time were observed in the SCS group for CRPS-I versus CRPS-II subjects, and male versus female subjects. CRPS-I subjects in the SCS group demonstrated significantly reduced pain relief at 9- and 12-months postpermanent implant, while CRPS-II subjects’ level of pain relief was stable from the 1-month follow-up visit through the 12-month visit. The reasons for the different trends observed between CRPS-I and CRPS-II subjects in pain relief are uncertain. A similar trend was observed with male subjects in the SCS group demonstrating significantly reduced pain relief at 9- and 12-months, whereas female subjects showed stable pain relief. While differences in pain perception and treatment effects have been noted between males and females,17,19 it is unknown why male subjects appeared to be more susceptible to habituation of SCS therapy, and a more general analysis of gender differences would be worthwhile. There were several limitations to this study. First, this was an underpowered, retrospective analysis of data from the ACCURATE RCT study, which was not performed for the specific purpose of this analysis. Because there were no specific methodological considerations in the ACCURATE RCT to ensure equitable treatments through the 12-month visit, the results observed in this analysis could be attributed to some unknown or unmeasured influence. Second, the ACCURATE RCT was a nonblinded study, which may have introduced expectation bias by the patients and bias in outcomes assessment by the investigators or study staff. Third, only 12 months of data was available for our retrospective analysis. It is unclear from these data what the effect of habituation is over longer time horizons. Fourth, the use of LOCF to impute missing data requires the assumption that subjects’ pain intensity remained unchanged following their last VAS measurement at the time of exit. However, since all subjects had chronic pain for an average of greater than 7 years, and had failed prior treatments, it is unlikely that subjects would experience a clinically meaningful change in pain intensity following exit. Fifth, the study conclusions are limited to CRPSI and CRPS-II patients who had chronic severe lower limb pain, baseline VAS scores of at least 60 mm, and pain lasting on average 7.5 § 7.6 years in the DRG stimulation group, and 7.0 § 7.7 years in the SCS group. It is unknown if the same patterns of habituation would be observed in a population of CRPS-I and CRPS-II patients treated at an earlier stage in their disease progression, with pain in other anatomical locations such as the upper limbs, or different disease indications. Sixth, the diagnosis of CRPS-II used in the ACCURATE study may not have followed Budapest diagnosis criteria. Finally, the assessment of habituation for SCS was specific to the low-frequency, tonic waveform, and newer SCS waveforms, such as high-frequency
ARTICLE IN PRESS Levy et al and BurstDR stimulation, may have different rates of habituation.2,23
Conclusions
The Journal of Pain 9
Acknowledgments The authors would like to acknowledge the investigators of the ACCURATE study, including Drs. Jon Obray, Christopher Kim, Thomas Yearwood, Hazmer Cassim, Mark Netherton, Nathan Miller, Edward Tavel, and Timothy Davis. Linda Johnson made writing contributions to the manuscript.
Tonic SCS is an accepted treatment for CRPS-I and CRPSII, but patients treated with DRG stimulation appeared to have more durable pain relief over time as compared to patients treated with SCS. More specifically, DRG stimulation showed stable pain relief from 1 month through 12 months postimplant, while tonic SCS demonstrated diminishing pain relief to therapy at 9- and 12-months. Long-term therapy habituation occurred at 12 months in the SCS group but not the DRG group, despite equally attentive follow-up and reprogramming.
Supplementary Data
References
10. Kemler MA, Barendse GA, van Kleef M, de Vet HC, Rijks e CA, van den Wildenberg FA: Spinal cord stimulaCP, Furne tion in patients with chronic reflex sympathetic dystrophy. N Engl J Med 343:618-624, 2000
KM, Yland MJ, Charnov JH, Redko V: Multiple pro1. Alo gram spinal cord stimulation in the treatment of chronic pain: Follow-up of multiple program SCS. Neuromodulation 2:266-272, 1999 KM, Oakley J, Feler CA: Spinal cord stim2. Bennett DS, Alo ulation for complex regional pain syndrome I [RSD]: A retrospective multicenter experience from 1995 to 1998 of 101 patients. Neuromodulation 2:202-210, 1999 3. Bennett DS, Brookoff D: Complex regional pain syndromes (reflex sympathetic dystrophy and causalgia) and spinal cord stimulation. Pain Med 7:S64-S96, 2006 4. Burchiel KJ, Anderson VC, Brown FD, Fessler RG, Friedman WA, Pelofsky S, Weiner RL, Oakley J, Shatin D: Prospective, multicenter study of spinal cord stimulation for relief of chronic back and extremity pain. Spine (Phila Pa 1976) 21:2786-2794, 1996 5. Deer TR, Levy RM, Kramer J, Poree L, Amirdelfan K, Grigsby E, Staats P, Burton AW, Burgher AH, Obray J, Scowcroft J, Golovac S, Kapural L, Paicius R, Kim C, Pope J, Yearwood T, Samuel S, McRoberts PW, Cassim H, Netherton M, Miller N, Schaufele M, Tavel E, Davis T, Davis K, Johnson L, Mekhail N: Dorsal root ganglion stimulation yielded higher treatment success rate for complex regional pain syndrome and causalgia at 3 and 12 months: a randomized comparative trial. Pain 158:669-681, 2017 6. Devor M: Ectopic discharge in Abeta afferents as a source of neuropathic pain. Exp Brain Res 196:115-128, 2009 7. Gravius N, Chaudhry SR, Muhammad S, Bostroem A, Gravius S, Randau T, Scheele D, Westhofen P, Kruppenbacher J, Stoffel-Wagner B, Maier C, Weidlich A, Yearwood TL, Chakravarthy K, Kramer J, Hurlemaan R, Kinfe T: Selective L4 dorsal root ganglion stimulation evokes pain relief and changes of inflammatory markers: Part I profiling of saliva and serum molecular patterns. Neuromodulation 22:44-52, 2019 8. Harden RN, Bruehl S, Stanton-Hicks M, Wilson PR: Proposed new diagnostic criteria for complex regional pain syndrome. Pain Med 8:326-331, 2007 9. Hayek SM, Veizi E, Hanes M: Treatment-limiting complications of percutaneous spinal cord stimulator implants: A review of eight years of experience from an academic center database. Neuromodulation 18:603-608, 2015
Supplementary data related to this article can be found at https://doi.org/10.1016/j.jpain.2019.08.005.
11. Kemler MA, de Vet HC, Barendse GA, van den Wildenberg FA, van Kleef M: Effect of spinal cord stimulation for chronic complex regional pain syndrome Type I: Five-year final follow-up of patients in a randomized controlled trial. J Neurosurg 108:292-298, 2008 12. Koopmeiners AS, Mueller S, Kramer J, Hogan QH: Effect of electrical field stimulation on dorsal root ganglion neuronal function. Neuromodulation 16:304-311, 2013. discussion 10-1 13. Krames ES: The role of the dorsal root ganglion in the development of neuropathic pain. Pain Med 15:1669-1685, 2014 14. Kumar K, Hunter G, Demeria D: Spinal cord stimulation in treatment of chronic benign pain: Challenges in treatment planning and present status, a 22-year experience. Neurosurgery 58:481-496, 2006. discussion 481-96 15. Lohnberg JA, Altmaier EM: A review of psychosocial factors in complex regional pain syndrome. J Clin Psychol Med Settings 20:247-254, 2013 16. Merskey H, Bogduk N: Classification of Chronic Pain: Descriptions of Chronic Pain Syndromes and Definitions of Pain Terms, 2nd ed. Seattle, IASP Press, 1994 17. Murphy KR, Han JL, Hussaini SM, Yang S, Parente B, Xie J, Lad SP: The volume-outcome effect: impact on trial-topermanent conversion rates in spinal cord stimulation. Neuromodulation 20:256-262, 2017 18. North RB, Kidd DH, Zahurak M, James CS, Long DM: Spinal cord stimulation for chronic, intractable pain: Experience over two decades. Neurosurgery 32:384-394, 1993. discussion 94-5 19. Paller CJ, Campbell CM, Edwards RR, Dobs AS: Sexbased differences in pain perception and treatment. Pain Med 10:289-299, 2009 20. Pawela CP, Kramer JM, Hogan QH: Dorsal root ganglion stimulation attenuates the BOLD signal response to noxious sensory input in specific brain regions: Insights into a possible mechanism for analgesia. Neuroimage 147:10-18, 2017 21. Schott GD: Mechanisms of causalgia and related clinical conditions. The role of the central and sympathetic nervous systems. Brain 109:717-738, 1986
ARTICLE IN PRESS 10
The Journal of Pain
22. Sparkes E, Duarte RV, Mann S, Lawrence TR, Raphael JH: Analysis of psychological characteristics impacting spinal cord stimulation treatment outcomes: A prospective assessment. Pain Physician 18:E369-E377, 2015 23. Van Buyten JP, Wille F, Smet I, Wensing C, Breel J, Karst € ggel-Kra €mer K, Vesper J: Therapy-related E, Devos M, Po explants after spinal cord stimulation: Results of an international retrospective chart review study. Neuromodulation 20:642-649, 2017 24. van Eijs F, Stanton-Hicks M, Van Zundert J, Faber CG, Lubenow TR, Mekhail N, van Kleep M, Huygen F: Evidencebased interventional pain medicine according to clinical
Therapy Habituation for SCS Versus DRG Stimulation diagnoses. 16. Complex regional pain syndrome. Pain Pract 11:70-87, 2011 25. Subthreshold Stimulation of the Dorsal Root Ganglia: Retrospective Study, in Verrills P, Mitchell B, Vivian D, Cusack W, Kramer J, (eds): Subthreshold Stimulation of the Dorsal Root Ganglia: Retrospective Study, Las Vegas, North American Neuromodulation Society, 2017 26. Wu G, Ringkamp M, Murinson BB, Pogatzki EM, Hartke TV, Weerahandi HM, Campbell JN, Griffin JW, Meyer RA: Degeneration of myelinated efferent fibers induces spontaneous activity in uninjured C-fiber afferents. J Neurosci 22:7746-7753, 2002