Spinal cord stimulation: Placement and management

INTRODUCTION — Spinal cord stimulation (SCS) is a neuromodulation technique that is used to treat neuropathic and sympathetically mediated chronic pain. SCS involves percutaneous or surgical implantation of electrodes in the epidural space, with power supplied by an implanted battery.

This topic will discuss placement of spinal cord stimulators, management of anesthesia for spinal cord stimulator placement, and management of SCS for patients who undergo other surgical procedures. It does not address the technique for laminectomy and surgical lead placement, dorsal root ganglion, or peripheral nerve stimulation. Evaluation of chronic pain, alternatives for therapy of chronic pain, and SCS for the treatment of specific conditions are discussed separately.

●(See "Investigational therapies for treating symptoms of lower extremity peripheral artery disease", section on 'Spinal cord stimulation'.)

●(See "Complex regional pain syndrome in adults: Treatment, prognosis, and prevention", section on 'Spinal cord and peripheral nerve stimulation'.)

●(See "Approach to the management of chronic non-cancer pain in adults".)

MECHANISM OF ACTION — The complex mechanism of action of SCS has not been fully elucidated and is multifactorial, involving neuromodulation of dysregulated pain pathways [1]. The mechanisms described below pertain to conventional or standard SCS. Knowledge about the mechanisms underlying pain relief from the newer SCS technologies, such as high-frequency (HF), burst, and most recently, closed loop evoked compound action potential (ECAP) stimulation, is evolving. (See 'Implantable pulse generators' below.)

●Neuropathic pain relief – SCS was originally thought to work through a gate control mechanism in the dorsal horn of the spinal cord. Thus, spinal cord stimulators were called dorsal column stimulators. According to gate control theory, the substantia gelatinosa in the dorsal horn was a functional "gate system" where pain was modulated. Small (painful A-delta and C) nerve fibers and large (A-beta) nerve fibers synapse at the "gate." Large A-beta fiber activation inhibited the painful small fibers, closing the "gate" and relieving pain [2]. Paresthesias are generated via depolarization of A-beta fibers [3,4].

The mechanism for pain relief with conventional SCS is now known to be more complex, involving both spinal and supraspinal mechanisms. Much of what is known comes from laboratory and animal studies. Relief of neuropathic pain by paresthesia-based SCS is mediated in part by wide dynamic range neuron suppression in the dorsal horn of the spinal cord, through A-beta mediated inhibitory factors involving gamma-aminobutyric acid (GABA), cholinergic, and other transmitter systems [1,5-7].

Supraspinal mechanisms of SCS occur via descending inhibitory pathways and are also thought to contribute significantly to the effects of SCS [8,9]. Pain relief from SCS can often be realized for days or even weeks after the stimulation has been turned off, which underscores the importance of supraspinal mechanisms of action [10,11]. An electroencephalographic (EEG) study of patients with chronic lower extremity neuropathic pain found that long-term SCS influenced both pain thresholds and cortical signaling, further evidence of a more central mechanism of analgesia [12]. Cortical inhibition by paresthesia-based SCS has also been demonstrated via amplitude reduction in somatosensory evoked potentials (SSEP) monitoring [1].

The mechanism of action for paresthesia-free forms of SCS (ie, burst, HF SCS) may be different from conventional SCS, and is the subject of ongoing research. As examples, conventional and burst SCS have different effects on the dorsal horn, the gracile nucleus in the dorsal column, and the brain [13], and pain relief from burst SCS does not appear to involve GABAergic systems [5]. HF SCS does not generate action potentials to the dorsal horn as is seen in paresthesia-based SCS, and does not appear to involve supraspinal mechanisms [1]. In 2020, an animal study found that paresthesia-independent SCS with a novel waveform (differential target multiplexed programming) had the ability to modulate glial cells in the CNS, specifically microglia and astrocytes [14]. The activation of these cells has been implicated in the development chronic neuropathic pain. (See 'Implantable pulse generators' below.)

●Ischemic pain relief – SCS analgesia in ischemic pain is more likely related to restoring the oxygen supply and demand imbalance [10,11]. Reduction of tissue ischemia is thought to be the result of increasing and/or redistributing blood flow to the area of tissue ischemia and/or decreasing tissue oxygen demand [11]. Vasodilatory effects of SCS may include the release of vasodilatory substances by sensory fibers via antidromic activation or actions on the sympathetic nervous system [11,15].

INDICATIONS AND EFFICACY — In general, SCS should be considered for patients with chronic pain who have failed conservative approaches. A spinal cord stimulator is not effective for all types of chronic pain, and is not effective for every patient with categories of pain that may respond. In the United States, the most common indication for spinal cord stimulator placement is chronic pain from failed back surgery syndrome (FBSS), which is also referred to as persistent spinal pain syndrome [16-19], although SCS may be effective for pain from other etiologies. A 2019 meta-analysis of 12 randomized trials including approximately 1000 patients with intractable spine or limb pain related to various disorders (ie, FBSS; chronic back, leg or trunk pain; diabetic neuropathy; peripheral vascular disease; complex regional pain syndrome [CRPS]) found that SCS was associated with increased odds of >50 percent pain reduction, compared with continued medical therapy [20]. High frequency and burst stimulation were associated with greater benefit than conventional SCS. (See 'Implantable pulse generators' below.)

In Europe, spinal cord stimulators are commonly placed for painful peripheral vascular disease and for refractory angina.

Indications for SCS include the following:

●Intractable back and radicular pain

•Radicular pain – There is moderate-quality evidence supporting the use of SCS in patients with persistent radicular pain due to FBSS. The benefits of SCS for patients with primarily axial back pain with FBSS are less clear. (See "Subacute and chronic low back pain: Surgical treatment", section on 'Spinal cord stimulation'.)

•Axial back pain – The benefits of SCS for patients with primarily axial back pain with FBSS are less clear. Axial back pain is more difficult to treat than radicular pain, and technical modifications including both hardware refinement and computerized stimulus modification are being studied. Newer forms of stimulation show promise for treatment of axial low back pain due to FBSS and other causes [21-25]. (See 'Implantable pulse generators' below.)

●Complex regional pain syndrome – SCS may be helpful for complex regional pain syndrome (CRPS) refractory to conventional therapy, particularly in patients with disease limited to one extremity. (See "Complex regional pain syndrome in adults: Treatment, prognosis, and prevention", section on 'Implanted devices'.)

●Painful peripheral vascular disease – Painful, inoperable peripheral vascular disease is the most common indication for SCS in Europe. (See "Investigational therapies for treating symptoms of lower extremity peripheral artery disease", section on 'Spinal cord stimulation'.)

●Intractable angina – SCS may be beneficial for patients with refractory angina. The technique is safe, has anti-ischemic effects, may improve New York Heart Association (NYHA) functional class, and may increase quality of life. (See "New therapies for angina pectoris", section on 'Spinal cord stimulation'.)

●Other indications – Limited literature suggests that SCS may be beneficial for patients with visceral abdominal and perineal pain [26-28] and for painful diabetic neuropathy [29,30]. (See "Management of diabetic neuropathy", section on 'Topical therapies or neuromodulation'.)

Patients with these and other chronic pain syndromes should be evaluated individually for spinal cord stimulator placement.

PATIENT SCREENING — Patient selection is an important determinant of successful SCS. In addition to evaluation of the type of chronic pain and appropriateness for SCS therapy, patients should be screened for comorbidities, contraindications to the technique, and potential technical difficulty.

A history and physical examination should be performed prior to spinal cord stimulator placement to assess for medical conditions that increase risk of failure or complications.

●Coagulopathy – Patients who undergo neuraxial procedures, including spinal cord stimulator placement, are at risk for spinal epidural hematoma (SEH), which can result in permanent neurologic damage. The baseline risk is low but is increased for patients with coagulopathy related to medical conditions or administration of anticoagulant or antiplatelet medications. Uncontrolled coagulopathy and severe thrombocytopenia are contraindications to spinal cord stimulator trial or implant.

We agree with the Joint Guidelines for Interventional Spine and Pain Procedures in Patients on Antiplatelet and Anticoagulant Medications published by a consortium of international pain societies [31]. These guidelines classify spinal cord stimulator trial and implant as procedures with high potential risk of serious intraspinal bleeding and provide guidance for discontinuation and reinstitution of anticoagulant and antiplatelet medications. A summary of those guidelines and a checklist for anticoagulation management for interventional spine procedures are shown in tables (table 1 and table 2).

Peri-procedure management of anticoagulant and antiplatelet medications should be coordinated with the patient's medical care providers. The risk of complications associated with withholding these medications may affect the length of the spinal cord stimulator trial.

The incidence of SEH from an in-situ spinal cord stimulator in patients who are chronically taking anticoagulant or antiplatelet medications is unknown [32]. Decisions to place stimulators in these patients must be individualized.

●Systemic or local infection – Active infection is a contraindication to spinal cord stimulator trial and implantation. We agree with the recommendations from the Neurostimulation Appropriateness Consensus Committee for Infection Prevention and Management, which state that remote infections should be identified and treated prior to neuromodulation trials and implants [33]. Recommendations for preventing and treating device related infection are shown in a table (table 3). We do not consider human immunodeficiency virus (HIV) a contraindication for spinal cord stimulator placement; in case reports, SCS has been effective for HIV-related polyneuropathy [34].

●Pacemakers – For patients with cardiac pacemakers and internal cardiac defibrillators, compatibility with SCS should be established prior to spinal cord stimulator trial. Spinal cord stimulators have little interaction with the newer pacemakers, provided strict bipolar right ventricular sensing is used [35-38]. Patients with defibrillators require closer observation to ensure stimulation does not result in defibrillator discharge. For patients with pacemakers, we generally consult a cardiologist as part of preprocedure evaluation.

●Laboratory evaluation – We perform laboratory evaluation as dictated by the medical history. If the spinal cord stimulator trial is successful, further preoperative testing is performed as usual for anesthesia, prior to permanent spinal cord stimulator implantation.

The Neurostimulation Appropriateness Consensus Committee (NACC) recommends that patients undergo preoperative testing for Staphylococcus aureus (methicillin-susceptible S. aureus [MSSA] and methicillin-resistant S. aureus [MSRA]), as a measure to avoid SCS infection [33]. In S. aureus-colonized patients, decolonization with nasal mupirocin ointment and chlorhexidine baths is recommended. Practice with respect to screening and decolonization varies. One contributor to this topic routinely screens patients for MRSA, whereas another does not screen patients, but routinely decolonizes all patients with intranasal betadine, and uses chlorhexidine skin preparation.

●Imaging studies – Spine imaging studies should be reviewed to anticipate technical difficulty and to identify those patients for whom surgery may be more appropriate than SCS. In patients with severe degenerative disease, spinal stenosis, or scoliosis of the spine, thoracic imaging can be helpful in preprocedural planning. Percutaneous SCS placement is usually avoided in patients with severe central canal spinal stenosis, to reduce the risk of spinal cord compression with the leads [39]. In some institutions, thoracic spine MRI is performed in all patients prior to SCS placement. (See 'Complications' below.)

●Psychologic screening – Psychologic screening should be performed prior to spinal cord stimulator placement [38,40,41] and may be required by insurance companies for approval for payment. We send most patients for psychologic evaluation as part of the preprocedural process, both to assess for underlying mood disorders and to understand the expectations they may have regarding the procedure. A systematic review and subsequent study found that high levels of depression, anxiety, coping, somatization, and hypochondriasis are associated with worse outcome for lumbar surgery and SCS [41,42]. Having one of these diagnoses is not an absolute contraindication to SCS. However, such conditions should be treated prior to considering SCS or other similarly invasive procedures.

●Duration of pain – SCS may be less effective if performed many years after the onset of chronic pain. A retrospective review of 400 cases in one institution reported a success rate of approximately 85 percent for SCS placed less than two years after onset of pain, versus 9 percent with a delay of over 15 years [43].

EQUIPMENT — Spinal cord stimulators consist of a stimulation lead(s), usually an implantable pulse generator (IPG; battery), and an extension cable that connects the lead with the generator.

Implantable pulse generators — A number of companies manufacture spinal cord stimulator generators and leads. The IPG delivers electrical stimulation that can be modified by altering the pulse width, frequency, and amplitude to achieve maximal pain suppression. The distinguishing characteristics include the method used for controlling electrical energy (ie, constant current versus constant voltage), the frequency range, stimulation waveform, rechargeable versus non-rechargeable IPG, characteristics of the leads, and magnetic resonance imaging (MRI) compatibility/conditionality.

Paresthesia-based versus paresthesia-free stimulation — With a rapid evolution of new SCS waveforms, there has been a transition toward sub-threshold or paresthesia-free therapy. Waveforms including burst, differential target multiplex, and high-frequency (HF) stimulation are considered paresthesia-independent. Traditionally, SCS systems were based on paresthesia-based stimulation for accurate placement and effective analgesia. Paresthesias are not required for pain relief with burst and HF SCS. Not all patients report total lack of paresthesia with burst stimulation, though the majority do, and the area of paresthesias that do occur is less for most patients [44].

Because patient feedback is not always required for placement of paresthesia-free stimulators, the trial and implant procedure time is shortened by the time normally spent obtaining coverage of the painful areas with paresthesias. The trial and the implant can theoretically be performed under general anesthesia (GA), if necessary. Caution must be used with percutaneous placement under GA due to the risk of undetected nerve or spinal cord trauma. For this safety reason, GA is not commonly chosen for trial or final implantation. If placement under GA is preferred, strong consideration should be given to using neuromonitoring during placement for patient safety. (See 'Choice of anesthetic technique' below.)

The onset of pain relief may be delayed by up to several days after starting paresthesia-free stimulation, whereas analgesia usually occurs within minutes of starting paresthesia-based stimulation. This delay or “wash-in period” must be considered when programming the pulse generator and making adjustments.

The choice of pulse generator should be individualized based on the type of pain to be treated, and provider and patient preferences. Some patients prefer to feel paresthesias in the area of pain, while others find them unpleasant. Some newer generators offer options for the patient to change the type of stimulation to optimize pain control and comfort.

Generator characteristics — A wide variety of impulse generators are available and innovative features are being developed. As technology has improved, IPGs have become smaller. A typical IPG has a volume 14 to 40 cm². In general, rechargeable IPGs are smaller compared to non-rechargeable units (picture 1).

Some of the distinguishing characteristics include the following:

Electrical control — Standard pulse generators deliver tonic pulses that keep either the voltage or current constant to stimulate tissue with variable impedance related to scar tissue and other factors [45,46].

Frequency of stimulation — Standard (paresthesia-based) SCS generators deliver stimulation at relatively low frequencies (ie, 50-Hz range). However, animal and human studies suggest that higher-frequency stimulation (eg, 1 to 10 kHz) may be effective for some chronic pain conditions [47]. HF generators use a biphasic stimulator wave form with pulse widths of approximately 30 microseconds at a rate of 10,000 Hz [22,23,48]. Specific programming algorithms are used to achieve the best result. Differences between HF and standard generators include:

●Since HF spinal cord stimulators do not reliably cause paresthesias, HF stimulator leads are placed empirically based on the location of pain, usually with one lead-tip at the top of T8 and the other lead at T9 in order to consistently capture low back and leg pain, in accordance with an algorithm.

●Most studies comparing the efficacy of HF with standard paresthesia-based SCS have reported superior pain relief with HF stimulation for back and leg pain. A prospective trial that randomly assigned 171 patients with back and leg pain from several etiologies to HF SCS or standard SCS therapy reported better pain control with HF therapy [22]. Compared with standard SCS therapy, HF SCS resulted in a greater percentage of patients achieving ≥50 percent reduction in back pain (85 versus 44 percent) and leg pain (83 versus 55 percent) at three months and 12 months [22]. Twenty-four month follow-up continued to show >50 percent pain relief in low back and leg pain, with greater relief of low back pain (67 percent relief) versus standard SCS (41 percent relief) [49]. Another prospective trial of HF SCS in 83 patients with back pain from mixed etiologies reported that more than 70 percent had >50 percent relief of pain at six months [23]. In contrast, a randomized trial including 55 patients with refractory back pain due to failed back surgery syndrome (FBSS) reported an average of 20 to 25 percent reduction in pain scores with no difference in pain at one year between HF SCS (ranging from 2 Hz to 10,000 Hz) and low frequency SCS [50].

●HF SCS uses more energy and may require more frequent battery charging compared with standard SCS, though the battery life before replacement is comparable; both last approximately 10 years [51].

Burst versus tonic stimulation — Burst stimulation is another form of stimulation that is referred to as paresthesia-free. It consists of five HF (500-Hz) burst spikes delivered at a rate of 40 Hz, with a pulse width of 1 millisecond [45]. Effective burst stimulation is less likely to cause paresthesias and may achieve better reduction of neuropathic pain than standard tonic stimulation. In one trial, 15 patients with neuropathic back or neck pain were randomly assigned to burst SCS, tonic SCS, or placebo [46,52]. When compared with standard tonic stimulation, burst stimulation resulted in a higher percent of patients with back pain improvement (51 versus 30 percent). [13]. A randomized multicenter crossover trial including 100 patients with chronic trunk and/or limb pain found that burst SCS provided noninferior pain relief at one year compared with standard tonic SCS [44]. Approximately 60 percent of patients had no paresthesias during burst stimulation, and another 27 percent had a smaller area of paresthesia than during standard tonic SCS. At one year, approximately 70 percent of patients preferred burst SCS, whereas 24 percent preferred tonic SCS.

Some clinicians, including the author, use low frequency tonic stimulation with paresthesias to initially place the SCS leads, and change the stimulation mode to burst after initial placement. Others place the leads anatomically, similar to HF stimulation.

Much like HF SCS, burst systems are increasingly being implanted using anatomic placement. In a randomized trial including 270 patients who had an SCS placed with nonlinear burst stimulation, trial success was similar in patients whose leads were placed with anatomic placement versus paresthesia mapping [53].

Closed loop evoked compound action potential stimulation — Closed loop evoked compound action potential (ECAP) stimulation is a novel form of SCS that uses a feedback algorithm to determine the stimulus amplitude. Standard SCS is an open loop system that may result in alteration of stimulation with postural changes due to changes in the distance between the SCS lead and the spinal cord. For example, when a patient lies supine, the spinal cord moves posteriorly, toward the epidural space. Closed loop stimulation aims to keep the recruitment of neural fibers constant by continually adjusting the stimulation input in real time, based on real time evoked action potentials. ECAP stimulation received US Food and Drug Administration (FDA) approval in the United States in 2022.

●A 2018 Australian study of 51 subjects using closed loop ECAP technology showed >50 percent improvement in axial low back and/or leg pain at three and six months [24].

●Closed loop stimulation may be more effective than open loop systems [54]. A multicenter randomized trial compared ECAP-controlled closed loop stimulation with standard SCS in 134 patients with intractable chronic back and leg pain despite conservative therapy [55]. At both 3 and 12 months, a greater proportion of patients who received closed loop ECAP-controlled stimulation achieved >50 percent reduction in pain (82 versus 60 percent at 3 months, 83 versus 61 percent at 12 months). At 12 months 55 percent of patients who received closed loop SCS had reduced or eliminated opioid use, compared with 40 percent of patients who received open loop SCS. A follow-up study of data from this trial found improved sleep at 12 months for patients who had ECAP-controlled closed loop stimulation [56].

Power source — SCS IPGs are powered by lithium-ion batteries, which are recharged through the patient's skin with external chargers, typically held in place with adhesive patches or in pouches that are held in place with a strap. Some manufacturers make devices with primary cell (non-rechargeable) batteries, with battery life that depends on the settings. The manufacturer stated lifespan of the IPG is approximately four to seven years for those with non-rechargeable batteries and nine years for those with rechargeable batteries. The non-rechargeable batteries are larger than the rechargeable batteries and may not be suitable for patients with very little subcutaneous fat.

With evolving technology and improvements in waveform delivery, rechargeable IPGs have gotten increasingly smaller and less burdensome to charge. In some cases, patients may only have to recharge their IPG five times yearly [57].

IPGs are replaced when the battery runs down, or at an interval specified by the device manufacturer. Replacement involves a surgical procedure similar to IPG implantation.

Wireless SCS — Several wireless SCS systems have been developed [58]. It includes an electrode and built-in receiver or IPG that communicate wirelessly through the skin to a transmitter and battery worn by the patient. Studies are ongoing to determine the optimal use for this technology [59].

MRI compatibility — Over 80 percent of patients with SCS will need an MRI within five years of implantation [60]. Therefore, MRI compatibility is an important feature for all spinal cord stimulator components. Concerns include the possibility of lead movement or heating, hardware damage, and reprogramming as a result of radiofrequency (RF) energy. Most spinal cord stimulator components are designated as MRI conditional and can safely be used for MRI of the head and extremities under specific conditions. Some systems are MRI conditionally safe for whole-body MRI; manufacturers' specifications vary and should be followed.

Leads — The two categories of leads for spinal cord stimulators are cylindrical and paddle (surgical) leads. Cylindrical leads are most commonly used for percutaneous procedures, while paddle leads are typically placed surgically via a laminotomy.

●Cylindrical leads – Cylindrical leads can be placed through a 14 gauge epidural needle, and emit current circumferentially. Cylindrical leads are usually easily removed if necessary, even after prolonged periods. Lead migration may occur, especially in the early postprocedure period. (See 'Complications' below.)

●Surgical paddle leads – Surgically placed "paddle leads" are flat in shape and have two to five columns of electrodes (image 1). They provide unidirectional stimulation directed towards the spinal cord, which allows deeper penetration of the electrical stimulus. Paddle leads are bulkier than cylindrical leads and should be avoided in areas of spinal stenosis. If surgical paddle leads are sutured in place, lead migration does not occur, but surgical removal is required to avoid dural tear.

Programmers — The patient is given a handheld programmer with settings determined by the stimulation that best relieves the patient's pain. Some companies have shifted to smartphone technology, allowing patients to control their devices with a mobile device app. In some cases, SCS devices may be reprogrammed remotely, decreasing the burden of office visits and improving the user experience. The patient can choose among multiple stimulation programs, and the pain clinician can adjust the choice of settings. Wireless technology is now available for stimulator programming.

TECHNIQUE — Percutaneous spinal cord stimulators are placed in two stages that consist of a trial period to prove efficacy followed by final implantation. Both trial and implantation are performed under fluoroscopic guidance and are outpatient procedures. These procedures are performed with strict aseptic technique, with full surgical skin preparation and draping, and with pre-procedure administration of antibiotics.

For surgical lead placement, a percutaneous trial is performed as it would be for percutaneous final implant. The radiographic images showing optimal lead placement for the trial are used by the surgeon to guide final lead placement.

Spinal level for lead placement — The optimal level for SCS lead position depends on the location of pain. For low back and lower extremity pain, leads are placed at low thoracic/upper lumbar levels (T8 to L1). Cervical leads, which are placed for treatment of pain related to cervical radicular pain or upper extremity complex regional pain syndrome (CRPS), may be accessed via the mid- to upper-thoracic epidural space [35,61].

The usual levels for lead position for pain in anatomic regions are as follows:

●Neck – Above C3

●Shoulder – Above C5

●Hand – C5, C6

●Thigh – Anterior thigh T7, T8, T11/12, posterior thigh T11 to L1

●Foot – L1

●Low back – T9 to T10 [62]

Spinal cord stimulator trial — The SCS trial is performed to test the effect on pain control and tolerability before permanent implantation. For trial, temporary leads are placed percutaneously and connected via an extension cable to an external generator for the duration of the trial. For standard, paresthesia-based SCS, correct lead placement is confirmed by the paresthesias in the area of the patient's pain. For systems with paresthesia-independent stimulation, leads are frequently inserted using anatomic placement (AP), meaning placement based on the location of the pain [53], though technique varies.

●For high frequency (HF) and differential target multiplex SCS leads are placed via AP.

●For burst stimulation, some clinicians (including the author) use low frequency tonic stimulation with paresthesias for lead placement during the trial and final implantation, whereas others place the leads for burst stimulation anatomically, similar to HF stimulator leads.

After the trial period, the leads are removed, and the permanent implantation is performed at a later date. We wait approximately two weeks after trial completion to perform implantation to make sure there has been no infection. The utility of SCS trials has been called into question. In a European study, pain outcomes were similar in patients with failed back surgery syndrome (FBSS) who did not undergo a trial compared with patients who had a successful trial [63]. More data are needed to determine whether a "no-trial" pathway is a reasonable option for certain populations of patients.

Alternatively, a tunneled trial can be performed in high-risk patients or for patients with challenging epidural access. For a tunneled trial, the trial leads become the permanent leads and are sutured in place as part of the trial. The leads are then connected to extensions, which are tunneled to flank area and connected to an external generator. After the trial period, the permanent implantation (or explantation if the trial fails) can be performed under local anesthesia (LA) with sedation or general anesthesia (GA), since the lead location has already been established [64]. Whether there is a difference in infection rates between these two techniques is unclear, based on retrospective data [65-67].

●Trial duration – There is no consensus on appropriate trial length, although on average, it lasts three to seven days [35,64,68-71]. For patients who take anticoagulant or antiplatelet medications, the length of the trial may be shortened to minimize the risk of thrombotic complications off medication. (See 'Patient screening' above.)

●Assessment of efficacy – A trial is considered successful and implantation is performed when the trial results in ≥50 percent pain relief with improved activity [35]. For paresthesia-based lead placement we aim to have paresthesias overlap with at least 80 percent of the pain topography at the time of lead placement. Loss of stimulation or efficacy during follow-up should prompt an evaluation for lead migration. If it is unclear, a radiograph with anteroposterior and lateral views can be very useful (image 2).

Implantation — After successful SCS trial, permanent implantation is performed weeks later after the trial insertion site has healed. For paresthesia-based lead placement, which requires patient feedback, implantation is usually performed with LA and monitored anesthesia care (MAC). If permanent leads are placed for the trial, or if anatomic lead placement is planned, GA may be used as well. (See 'Choice of anesthetic technique' below.)

●Lead placement – A small incision is made to allow insertion of epidural needles, anchoring the leads after insertion, and tunneling the lead extension cable. The leads are placed using fluoroscopy (image 3 and image 4 and image 5 and image 6) and are tested for paresthesias, if using. After stimulation is programmed, the level of sedation can be deepened as appropriate.

●Generator placement – A second incision is made to create the pocket for the implantable pulse generator (IPG). The IPG is commonly implanted in the flank area above the iliac crest (image 7), but can also be placed in the buttocks or abdomen.

The leads must then be tunneled under the skin from the site where they exit the spine to the IPG. After the leads are connected and the generator is in place, the wounds are irrigated and closed.

POSTOPERATIVE MANAGEMENT — The patient is seen in the clinic a few days to a week after the implant to monitor for wound integrity and signs of infection and to review the spinal cord stimulator settings and use of the programmer with the patient. The initial spinal cord stimulator settings often require slight adjustments in the first few weeks after implant, and are often stable thereafter.

COMPLICATIONS — Overall complications of SCS range from 5.3 to 40 percent and are most commonly due to hardware-related issues. The majority of complications are not life-threatening (eg, lead migration), but they do often require lead revision or explantation [72,73]. Spinal epidural hematoma (SEH) is a rare, potentially catastrophic complication of SCS.

●Lead migration – Lead migration with loss of appropriate pain coverage is one of the most common complications of SCS and is a frequent reason for lead revision [64,74,75]. Anchoring devices have been developed to minimize lead migration [43,74,76]. Lead migration is usually detected when the patient's stimulation has changed, either by loss of stimulation or stimulation in undesired locations (eg, abdominal stimulation). It can be confirmed radiographically (image 6).

The first step in addressing lead migration is reprogramming the electrodes within the lead or adjusting the amplitude. If reprogramming does not fix the problem, then the leads are either revised or removed.

The greater movement and flexibility of the neck compared with the thoracic and lumbar spines may result in an increase in the risk of lead migration in the cervical area [61,77].

●Lead fracture – Lead fracture has been reported in 0 to 9.1 percent of placements, and is similar with percutaneous and paddle leads [73,78,79]. Factors that contribute to fracture include poor anchoring technique and vigorous physical activity. Some patients with physically active lifestyles may be denied SCS because of concern for lead fracture. However, SCS has been used in patients in active-duty war zones with good results [80].

●Seroma – The surgical wounds should be closed in multiple layers to avoid creating pockets of space where fluid can collect [81]. Even with meticulous attention to hemostasis, seromas can occur. Forceful, blunt dissection and excessive use of electrocautery can increase the risk of seroma [64,73]. Many seromas will resolve on their own. If not, surgical incision and drainage may be needed [64,73,81].

●Infection – Device-related infection is a potentially serious complication that may require explantation of the device.

The risk of infection after spinal cord stimulator placement is estimated at between 2.5 and 12 percent, with more recent studies reporting lower rates [65,66,70,82-84]. In a large multicenter retrospective analysis of over 2700 patients who had infections within one year of implantation, the most common organism was Staphylococcus aureus and the battery (IPG) pocket was the most common SCS-related infection site. The median time to infection was 27 days. Elevated inflammatory markers were only present in one-half of the cases with infection [84]. Epidural abscess is a rare and potentially devastating complication of SCS.

Recommendations for prevention of SCS-related infection are shown in a table (table 3).

Superficial infections such as stitch abscess and skin inflammation can be treated with antibiotics. However, there should be a low threshold for removing all of the hardware when deep infection is suspected. When a spinal cord stimulator is explanted for infection, reimplantation should be delayed until eradication of infection off antibiotic therapy. Reimplantation should be performed at a site that was not involved with the prior infection.

●Dural puncture – Cerebrospinal fluid (CSF) leak after spinal cord stimulator placement is rare [85], and includes leaks related to dural puncture with the epidural needle or with the lead. However, if dural puncture does occur with the 14-gauge epidural needle, the risk of post dural puncture headache (PDPH) may be over 70 percent [86,87].

The risk of dural puncture may be increased in patients with difficult anatomy, when accessing the epidural space at a level that has had previous surgery, and in patients who are not able to remain still during the procedure.

There is no consensus regarding how to handle dural puncture during SCS. The procedure may be repeated above the level of the dural puncture, or the procedure may be aborted.

Epidural blood patch (EBP) may be considered with intractable PDPH; however, blood may enter the intrathecal space due to the large size of the dural defect. Strict aseptic technique is always required when EBP is performed, and may be particularly important in patients with an SCS in place. The optimal protocol is unclear; in one small series of four patients with SCS hardware in place, the EBP was performed with preoperative antibiotics, full surgical skin preparation and draping, and surgical scrub and attire [85]. (See "Post dural puncture headache", section on 'Epidural blood patch'.)

●Spinal cord injury or neurologic complications – Spinal cord injury (SCI) is a broad term that can describe conditions including direct spinal cord trauma, central cord syndrome, anterior cord syndrome, Brown-Sequard syndrome, spinal hematoma, concussion/edema of spinal cord, or nerve root injury.

•Spinal cord trauma – Direct spinal trauma is extremely rare during SCS placement, but is potentially devastating; there is a case report of quadriparesis following percutaneous SCS revision under general anesthesia (GA) [88]. Therefore, we prefer local anesthesia with appropriate sedation rather than GA for most patients, to allow patient report of pain that could indicate nerve trauma. We use neuromonitoring for patients who require general anesthesia. (See 'Choice of anesthetic technique' below.)

The reported rates of spinal cord injury (SCI) are low, and vary between 0.42 and 2.13 percent in retrospective database studies [89-91]. In a 2022 retrospective study of >70,000 SCS implants performed between 2010 and 2019, the incidence of SCI within 45 days of implantation was 0.42 percent [90]. The rate of SCI was 0.45 percent with percutaneous lead placement versus 0.36 percent after paddle lead placement, a statistically nonsignificant difference. An earlier retrospective database study of >8300 SCS implants from 2000 to 2009 found SCI in 2.13 percent of patients [89]. One possible explanation for the higher SCI rates in the earlier study is that it predated most of the current SCS guidelines that focus on patient selection, risk assessment, and screening.

•Spinal epidural hematoma – SEH is a neurosurgical emergency; timely decompression of the hematoma is mandatory to minimize the chance of permanent neurologic damage.

SEH is very rare after SCS in patients with normal coagulation. A review of over 8300 patients who underwent spinal cord stimulator implantation found an incidence of spinal hematoma of 0.71 percent, with no difference between percutaneous and paddle leads [89]. Patients who are anticoagulated are at increased risk of SEH; thus, preoperative guidelines for management of anticoagulant and antiplatelet medications should be followed. SCS is contraindicated in patients with uncontrolled coagulopathy and severe thrombocytopenia [92]. (See 'Patient screening' above.)

SEH may occur immediately, or days to weeks after spinal cord stimulator placement. Spinal epidural hematoma should be considered in any patient who has an increase in back pain or radicular pain, and should be acted upon emergently if there is associated sensory deficit, lower extremity weakness, or loss of bowel or bladder control [93,94]. (See "Disorders affecting the spinal cord", section on 'Spinal epidural hematoma'.)

●Tolerance – Tolerance is one of the more common undesirable effects of long-term SCS. In 20 to 40 percent of patients, the effectiveness of the stimulation wanes over time [95]. Daily washout periods may prevent or diminish tolerance. The spinal cord stimulator may be reprogrammed to recapture stimulation in the painful area.

●Explantation for inadequate pain relief – A European retrospective multicenter chart review, including 955 SCS implants from 2010 to 2013 reported a 7.9 percent explant rate with 52 percent due to inadequate pain relief. Rechargeable systems had higher explant rates (conventional SCS 5.5 percent/year and high-frequency [HF] SCS 5 percent/year) compared with conventional non-rechargeable systems (2.8 percent per year) [96].

ANESTHESIA FOR SPINAL CORD STIMULATOR PLACEMENT — The spinal cord stimulator trial is usually performed under local anesthesia (LA), with sedation as deemed necessary by the pain physician. Spinal cord stimulator implant after a successful trial is performed in an operating room with monitored anesthesia care (MAC). With the advent of paresthesia-free stimulation, there are some who perform percutaneous SCS under general anesthesia (GA) with neuromonitoring. Both the trial lead insertion and implantation are performed with the patient in the prone position.

Preoperative evaluation — Preanesthesia evaluation should include the usual history and anesthesia-directed physical examination, airway assessment, and appropriate laboratory evaluation. (See "Preoperative medical evaluation of the healthy adult patient" and "Airway management for induction of general anesthesia", section on 'Airway assessment' and "Preoperative evaluation for anesthesia for noncardiac surgery".)

In particular, the patient's ability to remain still and tolerate the prone position, level of anxiety, and predicted difficulty with airway management should be assessed and may affect the choice of anesthetic and surgical technique.

Choice of anesthetic technique — Options for thoracic or lumbar spinal cord stimulator implant include MAC, spinal anesthesia, and GA. The plan for anesthesia should be coordinated with the pain clinician, depending on patient factors and procedural requirements. We prefer MAC rather than GA for most patients, to allow patient report of pain that could indicate nerve trauma. For patients who require general anesthesia, we use neuromonitoring during the procedure.

Since these procedures are performed in the prone position, options for airway management during the procedure are limited. For patients with predicted difficulty with airway management or high risk of airway obstruction with moderate or deep sedation (eg, severe obesity, obstructive sleep apnea) GA with control of the airway from the start may be preferred. (See "Management of the difficult airway for general anesthesia in adults", section on 'Recognition of the difficult airway'.)

Placement of cervical SCS under sedation, with the patient in the sitting position with the face supported on a horseshoe headrest, has been reported [97].

Monitored anesthesia care — The most common anesthetic for percutaneous spinal cord stimulator implant is MAC with sedation, with local anesthetic (LA) infiltration by the pain clinician. During needle insertion, lead placement, and testing for paresthesia, the level of sedation, if any, must allow the patient to report pain that might indicate nerve trauma, and once the lead is placed, paresthesia coverage of their chronic pain. Once paresthesia testing is complete, the level of sedation can be increased.

The tunneling track that is necessary for connecting the lead to the IPG must be anesthetized to avoid discomfort. Deeper sedation should be titrated carefully while assessing ventilation and airway patency. We often administer a low-dose, titrated propofol infusion (eg, 25 to 75 mcg/kg/minute) or dexmedetomidine (0.2 to 2 mcg/kg/hour, with or without loading infusion). The doses and medications used for MAC should be individualized. (See "Monitored anesthesia care in adults", section on 'Drugs used for sedation and analgesia for monitored anesthesia care'.)

General anesthesia — GA may be administered for surgical lead implant (ie, via laminectomy) and when MAC is either unsafe or unlikely to be successful for percutaneous spinal cord stimulator implant. Alternative techniques can be used to compensate for the lack of patient feedback during GA, including the following:

●Permanent lead trial – When the need for GA for implant is anticipated, permanent trial leads can be placed under LA, followed by GA for generator placement, after successful trial.

●Neurophysiologic monitoring – Neuromonitoring (ie, sensory and motor evoked potentials) can be used during lead placement under GA to assess for nerve damage and dermatome coverage during SCS [98]. GA with neuromonitoring may be used for surgical lead placement, especially for cervical SCS. Neuromonitoring is also used for percutaneous stimulators performed under GA for paresthesia-free, anatomically placed stimulator leads.

The drugs and doses administered for GA must be modified for optimal neuromonitoring. (See "Neuromonitoring in surgery and anesthesia".)

ANESTHESIA FOR PATIENTS WITH AN EXISTING SPINAL CORD STIMULATOR — Perioperative concerns for patients with implanted spinal cord stimulators include the following:

●Electrosurgery – Electrosurgical energy can damage any component of a spinal cord stimulator. Manufacturer recommendations for the use of electrosurgical instruments should be investigated preoperatively. The following approach should be used to minimize the chance of damage and patient injury:

•Use bipolar electrosurgical instruments whenever possible at the lowest effective voltage setting.

•If monopolar tools are required:

-Test the spinal cord stimulator preoperatively for impedance as an indication of leaks, and turn the spinal cord stimulator off.

-Place the dispersive electrode (grounding pad) as far from the implanted pulse generator (IPG) as possible.

-Test the spinal cord stimulator postoperatively.

●Neuraxial anesthesia after SCS – Neuraxial anesthesia (ie, spinal, epidural, and combined spinal-epidural) should not be performed at the site of lumbar or thoracic spinal cord stimulator implant, to avoid lead disruption or entanglement. In addition, spread of epidural local anesthetic (LA) solutions may be altered by fibrous tissue or scarring after SCS placement, and may result in patchy or incomplete analgesia or anesthesia. Spinal radiographs should be reviewed for lead placement before performing neuraxial techniques for anesthesia in order to avoid needle or catheter placement that could disrupt spinal cord stimulator leads.

The decision to perform neuraxial anesthesia and/or analgesia for patients with SCS should be individualized based on patient factors and other alternatives.

●Magnetic resonance imaging (MRI) compatibility – For patients who require perioperative MRI, spinal cord stimulator MRI compatibility should be confirmed, and the manufacturer's guidelines should be followed. (See 'Equipment' above.) Many SCS systems now allow for conditional full-body scanning.

PREGNANCY — There are limited data, in the form of a small number of case reports, on the use of SCS during pregnancy [99]. The effects of SCS on the developing fetus and the risks of miscarriage and preterm delivery are unknown. Spinal cord stimulators are not placed in patients known to be pregnant, partly because of the need for imaging studies and fluoroscopy to guide the placement of the trial and permanent leads. However, patients with existing SCS may become or plan to become pregnant. In those cases, the lack of information on the risks associated with SCS should be discussed with the patient. The decision to continue using stimulation should be individualized.

Technical concerns regarding SCS during pregnancy include the following:

●For internal pulse generators (IPGs) placed in the flank or abdomen, the leads may be stretched as the abdomen expands during pregnancy, which can result in breakage or pain. In one reported case, the spinal cord stimulator lead extender was divided surgically for pain during the second trimester in a patient with an IPG in the low abdomen [100].

●A spinal cord stimulator may affect the options for regional labor analgesia and for regional anesthesia for cesarean delivery. Prenatal anesthesia consultation should be performed early in pregnancy to plan for labor analgesia and potential instrumental or operative delivery. In particular, prepregnancy radiographs or reports should be reviewed to help guide placement of an epidural catheter or spinal anesthesia.

For patients with lumbar leads, in whom epidural analgesia may not be safe or appropriate, spinal analgesia may be an option. (See "Neuraxial analgesia for labor and delivery (including instrumental delivery)", section on 'Neuraxial techniques'.)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Neuropathic pain" and "Society guideline links: Local and regional anesthesia".)

SUMMARY AND RECOMMENDATIONS

●Indications for spinal cord stimulation (SCS)

•SCS is a neuromodulation technique that may be used to treat neuropathic and sympathetically mediated chronic pain. (See 'Indications and efficacy' above.)

•SCS is most effective for radicular pain associated with failed back surgery syndrome (FBSS), complex regional pain syndrome (CRPS), painful ischemic peripheral vascular disease, and ischemic cardiac disease. (See 'Indications and efficacy' above.)

•Newer forms of SCS (ie, burst, high-frequency [HF], differential target multiplex) do not require paresthesias to be effective, and may provide more effective pain relief for some conditions. (See 'Indications and efficacy' above and 'Implantable pulse generators' above.)

•Patients should be screened for comorbidities that may increase the risk of complications or technical difficulty, including coagulopathy (table 1 and table 2 and table 3). (See 'Patient screening' above.)

●SCS placement

•SCS systems consist of leads that are placed either percutaneously or surgically in the epidural space and battery-powered implanted pulse generators (IPGs) (image 6 and image 4 and picture 1). (See 'Equipment' above.)

•SCS placement is a two-stage process. A trial is performed, followed by permanent implantation for successful trials. (See 'Technique' above.)

•Most SCS trials are performed under local anesthesia (LA) with or without sedation. SCS final implantation may be performed with monitored anesthesia care (MAC) or general anesthesia (GA), depending on the need for patient feedback and patient factors. For most patients, we prefer MAC rather than GA, to allow patient report of pain that might indicate nerve damage, and use neuromonitoring if GA is required. (See 'Anesthesia for spinal cord stimulator placement' above.)

●Complications – Most complications of SCS are minor (eg, lead migration), though they may require lead modification or explantation. Spinal epidural hematoma (SEH), spinal cord trauma, and epidural abscess are rare, potentially devastating complications. (See 'Complications' above.)

●Perioperative concerns for patients with SCS – (see 'Anesthesia for patients with an existing spinal cord stimulator' above):

•Bipolar electrosurgical instruments should be used whenever possible. If monopolar instruments are necessary, the spinal cord stimulator should be tested for leaks and turned off, and the dispersive electrode should be placed as far from the pulse generator as possible.

•Neuraxial anesthesia may be difficult or impossible after SCS.

•Magnetic resonance imaging (MRI) compatibility should be confirmed and manufacturer's recommendations should be followed for patients who undergo perioperative MRI.

●Pregnant patients – Spinal cord stimulators are not placed in pregnant patients. For patients with stimulators who become pregnant, options for neuraxial analgesia and anesthesia may be affected by the device. (See 'Pregnancy' above.)