Authors: Faisal Meshael AlAbbas
Categories: Systematic Review, Brachial plexus neuropathy, brachial plexus avulsion, chronic regional pain syndrome, neuromodulation, neurostimulation
Source: Medical Archives
Traumatic brachial plexus injuries are common among young adults, with a majority of patients succumbing to chronic pain syndromes. Conservative management is usually not satisfactory in these cases and surgical interventions are often required. We have conducted a systematic review and meta-analysis examining one of the neurosurgical techniques, spinal cord stimulation (SCS), in chronic pain neuromodulation in cases of chronic pain syndrome after traumatic brachial plexus injuries.
This systematic review aims to explore the reported use of cervical spinal cord stimulation as a neuromodulator in patients with chronic pain syndromes following traumatic brachial plexus injury.
A systematic literature search was conducted using MEDLINE through the OVID interface, ProQuest, Web of Science, The Cochrane Library, and Scopus. Our own files and reference lists of identified key articles were also searched.
A total of 13 studies (8 case reports and 5 case series), comprising 29 patients were included. Most brachial plexus injuries were sustained in motor vehicle accidents. 86% (25/29) of patients showed a good initial response to SCS, however, the response decreased over time, and 69% (20/29) of the patients reported a good response at the end of follow-up. Lead migration was the only complication reported in two studies.
SCS is a less invasive procedure with significantly fewer neurological side effects. A trial period of SCS is suggested in patients who have failed conservative treatment modalities before other neurosurgical interventions are considered.
Keywords: Brachial plexus neuropathy, brachial plexus avulsion, chronic regional pain syndrome, neurostimulation, neuromodulation
Brachial plexus injuries are devastating and disabling conditions commonly affecting young adults after motor vehicle accidents (1, 2). The injury may vary considerably from the involvement of the roots (either complete or partial) to individual peripheral nerves arising from the cords. Between 30% to 80% of patients with traumatic brachial plexus injuries develop chronic pain syndromes. (1, 2) In approximately 40% of cases, this pain is severe. (3) These syndromes include complex regional pain syndrome, brachial plexopathies, and secondary neuropathic pain due to direct nerve injury. For the most part, these syndromes are not responsive to medical therapy, and invasive interventions are common (4). Many patients are injured in motor vehicle accidents or at work, are quite young, and require effective long-term pain management. Surgical interventions to manage these conditions include nerve reconstruction procedures, such as brachial plexus nerve root transfer or sural nerve grafts. However, these techniques report inconsistent success and have variable incidences of persistent postoperative pain. (3,4) Dorsal root entry zone (DREZ) lesioning has been successfully used to manage deafferentation pain after brachial plexus avulsion. However, its utility in those patients reporting ongoing background pain has not been good (6).
Cervical spinal cord stimulation (SCS) is an intervention that has been successfully used in some cases of traumatic brachial plexus injury (2, 7, 8).
This systematic review aims to explore the reported use of cervical spinal cord stimulation as a neuromodulator in patients with chronic pain syndromes following traumatic brachial plexus injury
The review was conducted according to the Cochrane Handbook and reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis guidelines (PRISMA) (9,10). A PRISMA checklist was completed and can be viewed as Appendix A.
** Eligibility criteria **
All letters to the editor, case reports, case series, observational studies, or randomized controlled trials (RCTs) examining SCS aimed at chronic pain neuromodulation in patients with traumatic brachial plexus injuries were considered eligible for inclusion in this systematic review. We excluded studies a) no chronic pain condition was present (> 6 months of pain), b) only drug therapy was used for pain neuromodulation, c) radiofrequency ablation, deep-brain stimulation, or peripheral muscle or nerve stimulation was used, or d) Nerve transfer was performed.
** Information sources and search strategy **
The following electronic databases were searched for articles published up to December 23, PubMed, EMBASE, CT.gov, ICTRP, CINAHL, and The Cochrane Library. We also searched our files, consulted with experts, reviewed reference lists from identified articles, and searched for cited references of key publications. The following combination of keywords and Medical Subject Headings (MeSH) terms were “Brachial Plexus Neuropathies” AND “Treatment”.
** Eligibility assessment **
We screened the titles and abstracts of each identified citation. Those reports possibly meeting the eligibility criteria were extracted for full-text review.
** Outcomes of interest **
The primary outcomes of interest were the success of SCS in modulating pain following traumatic brachial plexus injury. We further sought to identify the criteria used to select patients for SCS and the rate of postoperative complications
We assessed the quality and risk of bias of each study using the following 1) Is the study’s objective clearly stated? 2) Is this a prospective study? 3) Is this a multicenter study? 4) Is the mechanism of injury stated? 5) Are adequate patient characteristics provided? 6) Is the stimulation intervention clearly described? 7) Are the outcomes measures defined? 8) Are the outcomes measures appropriate for the study aims? 9) Is the length of follow-up reported? 10) Is the number of patients lost to follow-up reported? 11) Are adverse events reported? 12) Are the study conclusions supported by the results? 13) Have competing interests and sources of support been reported?
** Statistical analysis **
A-priori, it was decided that no formal data meta-analysis would be attempted if less than three RCTs were included in the review.
Table I details the design of the included studies representing 29 patients. The most were case reports (8,11,13-15,17,18,20) (n=8) with the remainder being case series (n=19).2,7,12,16,19 Most injuries were sustained in motor vehicle accidents, (2, 8,11-13,16,17,19,20) industrial injuries contributed significantly (2,16) and two cases were due to iatrogenic intraoperative injuries. (14,15) Before undergoing SCS, most patients had been treated with combinations of antidepressants, anticonvulsants, opioids, stellate ganglion blocks, and in some cases, epidural corticosteroid injections. One of the patients had previously failed low-frequency SCS treatment 15 and another had two failed DREZ surgeries. (17) All studies used cervical dorsal column spinal cord stimulation, except for one that extended electrode coverage to include T1-T2.(7) and a second that included ventral stimulation. (20) Follow-ups ranged from 1 month (11) to 58 months (16) with the most common follow-up period being 12 months.
Table II details the characteristics of the spinal cord stimulators used, their settings, and the field of electrode coverage. Manufactures included Medtronic (Resume TL, Pisces- Quad, 8-contact electrodes) (8,12,13,19,20) Advanced Neuromodulation Systems (7,12), and St Jude’s Medical (13, 18) The remainder of the studies did not report the type of stimulator used. (2,11,14,15) Low-frequency settings (20 to 130 Hz) were used in all but one study where a high-frequency stimulator at 10 kHz was used.15Pulse width ranged from 60 to 500 microseconds and pulse intensity from 1.2 to 13 mA. In most studies electrode position was determined by initial imaging of the injury and then fine-tuned with direct patient feedback. A temporary pulse generator was used for a trial period in nine of the studies. (2,7,8,11-13,15,16,19)
A summary of each risk of bias, presented as a percentage across all included studies, can be found in Figure 2. Figure 3 reports the risk of bias for the individual bias items for all included studies. All included studies were case reports or case series – no RCTs were identified or included. None of the studies were prospectively designed or were multicenter in nature. Generally, patient details, mechanism of injury, outcome measures, length of follow-up, and loss of follow-up were well reported. Many studies failed to report on the presence of adverse events specifically, and most did not report competing interests or sources of financial support.
Table III describes the patient pain scoring systems, postintervention pain score changes, and reported adverse events after implementing SCS. All studies showed a good (>40% reduction in pain scores, or a cessation or reduction in oral analgesia use) immediately after initiation of SCS, except for Garci-March et al., where two of the six patients were reported to have a “fair” response, (16) and Teixeira et al. where two of the four patients did not have any response to SCS. (19) Thus, a total of 86% (25/29) patients showed a good initial response to SCS. In addition, some studies further reported associated improvements in quality of life after SCS (2,13,15) Importantly, three of the studies. (4,16,19) representing nine patients (31%), reported deterioration in pain control over the follow-up period, irrespective of their initial response to SCS. Choi et al. reported worse pain scores at six months, which required the addition of peripheral nerve stimulation. (14) In the series reported by Garci-March et al., only one of the six patients was pain-free at 28 months, and three were considered to have failed treatment. (16) All three of these patients progressed to DREZ treatment. Teixeira et al. found that two of the four patients having SCS had no sustained reduction in pain scores, and one patient who had initially shown “considerable” pain reduction experienced a significant pain relapse at 14 months. (19) Thus, by the end of follow-up, 69% (20/29) of patients retained a good response to SCS**.**
** SCS complications **
Two studies reported lead migration at the last follow-up. (12,13) No studies reported any postoperative infections.
Brachial plexus avulsion is a complex and devastating injury. The injury includes direct nerve root injury and may often cause preganglionic injury. (21) The avulsion may be partial or complete, including all five roots. In cases of partial brachial plexus avulsion, it is unclear whether the avulsed roots play a role in the chronic pain complexes or whether the preserved roots are more important in the pathophysiology (1, 6, 22).
The spinal ganglia are the interface between the central and peripheral nervous systems. This provides the mechanism behind the complex chronic pain phenomena described by patients, including burning, shooting, crushing, and phantom limb pain (22). Complex pain syndromes are managed using a broad and diverse range of interventions. Non- surgical interventions include non-steroidal anti-inflammatory drugs, antidepressants, antiepileptics, local anesthetic blocks, infusions of lidocaine or ketamine, and experimental drugs such as cannabinoids. Gebreyohanes et al. classifies surgical interventions for managing brachial plexus avulsions as ablative, modulatory, or reconstructive (22). Ablative interventions include thalamotomy stereotactic mesencephalotomy and spinal cord anterolateral cordotomy, and DREZ lesioning. Modulatory interventions include electrical motor cortex stimulation, thalamic deep brain stimulation, and spinal cord stimulation. Most reconstructive techniques have focused on nerve transfer.
The ideal management approach for patients with chronic pain after traumatic brachial plexus injuries remains unclear. Current practices focus on the use of SCS or DREZ lesioning, with a tendency to favor DREZ.
SCS stimulation is thought to have its effect through A-fiber modulation. Low-frequency SCS has sought to induce paresthesia in the target area, described by patients as a tingling or uncomfortable sensation of vibration. Anterograde stimulation of the large-diameter fibers belonging to the dorsal column is thought to mediate this sensation. In addition, pain relief may further be mediated by the inhibition of wide-dynamic-range neurons in the lamina V – a critical factor in driving neuropathic pain (23, 24) Chien et al. provide an excellent discussion of the possible physiological mechanism underlying pain modulation in SCS (13).
This analysis has shown that the long-term response to traditional low-frequency SCS is unpredictable. While most patients experience early benefits, approximately 30% see a reduction in efficacy after 6 to 12 months. Still, other studies have reported a loss of efficacy up to 2 years after surgery. Therefore, it is essential to appreciate that most of the studies reported in this analysis had follow-up periods shorter than two years.
Recently, studies have reported success with high-frequency SCS in patients with chronic back and limb pain. (25-27). High-frequency SCS uses frequencies of 10,000 Hz instead of traditional low-frequency stimulation with frequencies of between 30 and 150 Hz. High-frequency stimulation uses short-duration pulses, approximately 30 microseconds, with an amplitude of 2 to 5 mA. This stimulation does not seem to cause paresthesia by initiating an action potential in the lemniscal pathway. Instead, it is postulated that high-frequency stimulation may entrain the small and medium-diameter dorsal column fibers, thereby causing pain modulation (15).
The most common alternative to SCS is DREZ lesioning surgery, and studies have reported good results in more than 75% of these patients. (22) However, DREZ lesioning may not reach lamina IV to VI, thereby causing treatment failure. Further, due to atrophy and distortion of the cervical cord, DREZ procedures may not reach the targeted sites. Finally, DREZ’s semi-blind landmark technique may not be precise enough to target the substantia gelatinosa accurately. As a result of these limitations, DREZ studies report pain recurrence within 6 to 12 months, good pain relief in approximately 75% of patients five years after treatment, and a high incidence of paralysis and proprioceptive disorders (7).
From our analysis of these papers, it was instructive to see the progressive development of standardized approaches to screening and implementation of SCS (25) Appropriate patient selection lays the foundation of clinical success when using SCS. The primary eligibility criterion for SCS is the presence of chronic pain that has been refractory to at least three months of conservative management. Conservative management should include pain medication, pharmacological and behavioral interventions, physical therapy, and possibly epidural injections or nerve blocks. Pain should have a VAS intensity of 5 or more and be accompanied by high disability index scores (e.g., Oswestry Disability Index 40 to 80 out of 100). (25) Care should be taken to meticulously document all conservative management modalities, as well as baseline pain and disability scores. Critical exclusion criteria should be sought. These include the presence of active disruptive psychiatric or psychological disorders. Other conditions that may affect pain perception or the inability to comply with postoperative follow-up plans may need to be identified as well.
We identified two new aspects related to SCS. The first was the combination of both dorsal and ventral leads as described by Watanabe et al. (20). This method of stimulation may induce transverse spinal cord stimulation in a dorsoventral direction, thereby potentially generating a wider range of paresthesia and greater neuromodulation. However, it is likely that standard SCS limitations and possible loss of neuromodulation over time would still apply to this technique. Further, there is insufficient data to understand if routinely placing ventral leads would provide improved neuromodulation or if it would improve the duration of pain modulation. The second aspect is the use of high-frequency SCS (15, 25). Its utility for chronic back and limb pain continues to be demonstrated and we believe this holds exceptional promise in patients with traumatic brachial plexus injuries. Based on this review, we wish to offer the following suggestions for managing patients with traumatic brachial plexus injuries. First, we suggest that SCS be considered the first-line treatment in patients with chronic pain after brachial plexus avulsion who have failed conservative treatment modalities. SCS is less invasive than DREZ, allows for a trial period of temporary stimulation, and has significantly less permeant neurological side effects. Its two major risk factors are a failure to provide adequate pain modulation and catheter migration, both of which are not permanent. Second, the benefits of SCS may be further enhanced if high-frequency SCS is used as a first-line intervention. Third, we suggest that a period of trial stimulation be used in all patients before generator implantation. Fourth, in patients who fail high-frequency SCS, DREZ
lesioning should be considered the next step. Finally, we suggest the establishment of an international register to track the efficacy of patients receiving high-frequency SCS for brachial plexus avulsion. Again, SCS is less invasive than DREZ, allows for a trial period of temporary stimulation, and has significantly less permeant neurological side effects.
Our analysis carries inherent limitations. First, all studies reported here are retrospective and therefore suffer from inherent bias. This is essentially the result of the small number of SCS studies that have been published. Second, the follow-up period of many of these studies is less than two years. This is important as many of the studies with longer-term follow-up have shown that the efficacy of SCS may wane dramatically over time. Thus, these results must be understood to reflect a cohort of patients with a follow-up period of less than two years.
Third, there is wide heterogeneity across time and methodology between these studies. There is a significant risk in comparing studies done decades apart that have used very different equipment and methodologies. (16, 20) This difference will become more important as high-frequency SCS stimulation begins to enter clinical use. Fourth, the pain scale measurements vary considerably across studies and cannot be readily compared. Fifth, in contrast with the broader literature, very few studies in this review have used high-frequency SCS in brachial plexus injuries.
SCS has been reported to successfully and safely neuromodulate the chronic pain experienced by patients with traumatic brachial plexus injuries. The reported response is very good (86%) initially but the response decreases (69%) over the follow-up period. High-frequency SCS may hold significant potential in treating patients with traumatic brachial plexus injury. Because of a better safety profile, we suggest giving an early trial of SCS as a first-line neuro intervention in patients who have failed to respond to conservative management
The author was involved on all steps of preparation this article including final proofreading.
The author have no conflicts of interest to disclose.
No specific funding was received in relation to this study.