Spinal Cord Stimulation

Published on 26/03/2015 by admin

Last modified 22/04/2025

Print this page

This article have been viewed 2264 times

CHAPTER 167 Spinal Cord Stimulation

Since its development 40 years ago, spinal cord stimulation (SCS) has become an increasingly used treatment for chronic pain. In one of the early examples of “bench-to-bedside” research translation, the gate theory of pain transmission was first put to use by Shealy’s development of SCS in 1967,¹^,² a mere 2 years after it was proposed. However, pain relief through the use of electrical stimulation had not been an entirely novel idea. For centuries, eels, as biologic sources of electrical current, have been used as analgesic aids. This effect was largely written off as magical until the work of Melzack and Wall offered a reasonable explanation. The ensuing clinical experience and accumulated body of knowledge comprises one of the most successful examples of useful, translational research in the history of neurosurgery. This chapter outlines the proposed mechanisms by which SCS works and discusses contemporary issues and clinical uses of this technology.

Mechanism of Action

Gate Theory

The gate-control theory of pain transmission proposed by Melzack and Wall ³ in 1965 describes a balance between small and large sensory fibers, with positive and negative feedback loops controlling the activity of the dorsal horn cells of the spinal cord. The gate theory postulates that a predominance of small-diameter sensory fiber activity opens “gates” within the dorsal horn of the spinal cord, whereas large-fiber dominance closes them. Shealy reasoned that because large fibers have a lower threshold for depolarization by an electrical field applied to a peripheral nerve, they may be recruited selectively by an externally applied field. Moreover, because large-diameter sensory fibers within peripheral nerves are segregated into the dorsal columns, they may be more selectively activated by electrical stimulation of the dorsal aspect of the spinal cord. It is for this reason that the primary electrical effect of SCS has been assumed to be mediated by the dorsal columns.

As elegant as it is, the gate theory alone does not fully explain the clinical effect seen with SCS. For example, there are many pain conditions that are not effectively treated by SCS, including the pain of acute injury.⁴ In contrast, SCS is effective at treating hyperalgesia, which is signaled by large fibers. This may indicate that relief of pain by electrical stimulation is due to frequency-related conduction block, acting at primary afferent branch points where dorsal column fibers and dorsal horn collaterals diverge.⁵ This suggests that other mechanisms involving interneurons in the dorsal horn or involving descending fibers or sympathetic mechanisms may exist.⁶

One of the major limitations in our understanding of the mechanisms underlying SCS has been the paucity of animal models that reproduce the human chronic pain condition. Initial animal models using acute tissue injury have given way to those employing peripheral nerve injury, in an effort to create a reliable chronic pain model. However, it has been difficult to develop a rat model of chronic pain because rats generally recover from painful insults without long-term pain.⁵ Even in humans, peripheral nerve injury does not reliably result in neuropathic pain, and objective assessment of pain in the rat relies on behavioral changes. We therefore rely largely on the data accumulated over the past three decades of human use to further elucidate a possible mechanism.

Neurotransmitters

To better clarify the mechanism of action of SCS, investigators have also examined the role of local neurotransmitters within the spinal cord on pain conduction. Early studies explored opioid receptors as the conventional model of pharmacologic analgesia. It was found that administration of the narcotic antagonist naloxone had no effect on the relief of pain by SCS.⁷ Other studies revealed that substance P is increased in the cerebrospinal fluid (CSF) following SCS. It has also been noted that SCS causes a decrease in the release of excitatory amino acids, such as glutamate and aspartate, while at the same time increasing the release of γ-aminobutyric acid (GABA).⁶ This evidence suggests that neuropathic pain may be a state of imbalance between excitatory and inhibitory neurotransmitters and that SCS may restore that balance (Fig. 167-1).⁵ This is further corroborated by studies that show that the GABA agonist baclofen improves the effect of SCS, whereas GABA antagonists inhibit its effect.⁸ These lines of research leave open the door for the possibility of adjunctive pharmacotherapy.

FIGURE 167-1 Schematic illustration of the postulated effect of spinal cord stimulation (SCS). A, Dorsal horn neurons under normal conditions. Afferent inputs release excitatory neurotransmitters into the synaptic cleft. Simultaneously, γ-aminobutyric acid (GABA) neurons are activated exerting both presynaptic and postsynaptic inhibition, balancing the state of excitation. B, After peripheral nerve injury, GABA-ergic activity is reduced, and excitatory neurotransmission is enhanced. These changes can give rise to allodynia and hyperalgesia. C, SCS activates dorsal column fibers, and the resulting nerve activity is antidromically mediated to neurons in the dorsal horn. The paresthesias experienced during SCS are due to orthodromic activation of the dorsal columns. D, SCS-evoked nerve activity induces the release of GABA from dorsal horn interneurons, reducing hyperexcitability. AMPA, α-amino-3-hydroxyl-5-methyl-4-isoxazole-propionate; NMDA, N-methyl-D-aspartate.

(Adapted from Cui JG, O’Connor WT, Ungerstedt U, et al. Spinal cord stimulation attenuates augmented dorsal horn release of excitatory amino acids in mononeuropathy via a GABAergic mechanism. Pain. 1997;73:87.)

Summary of Mechanisms

The mechanisms underlying SCS-induced analgesia are still open to question. However, it is clearly related to modulation of signals mediated by dorsally located fibers within the spinal cord. Neurohumoral changes in the spinal cord may be secondary to such activity modulation and may explain poststimulation analgesia, which is quite prominent in some patients.⁹

Indications and Outcomes

Early uses of SCS spanned the gamut of chronic pain etiologies, and outcomes were decidedly mixed. Clinicians soon learned that electrical stimulation was well suited for the treatment of certain diagnoses, whereas its use in others yielded disappointing results. With this in mind, before implantation of a spinal cord stimulator, the cause of the patient’s pain should be established. There should be appropriate, objective evidence of a pain disorder for which SCS has been shown to have efficacy. In general terms, pain conditions marked by nerve injury (so-called neuropathic pain) have had the most favorable track record following SCS. For an exhaustive description of indications and outcomes in SCS, the reader is encouraged to refer to Krames’ 1999 paper.¹⁰ In the sections that follow, we highlight a few of the most common diagnoses treated with SCS.

Failed Back Surgery Syndrome

Chronic neuropathic pain is most commonly located in the back and legs. Of patients undergoing lumbosacral spine surgery for treatment of this pain, 10% to 40% eventually develop persistent or recurrent pain.¹¹ This postsurgery pain is often referred to as failed back surgery syndrome (FBSS). Patients commonly have complaints of axial low back pain and lower extremity pain. In general, SCS is indicated in patients with FBSS whose radicular leg symptoms are more severe than the component of axial back pain. Objective evidence for a source of their neuropathic pain should be sought, including radiographic evidence of an anatomically successful surgery for herniated disk with an appropriate radiculopathy, matching the pain complaint.

FBSS is the most common indication for SCS and has the most evidence supporting its use. The Prospective, Randomized, Controlled, Multicenter Study of Patients with Failed Back Surgery Syndrome (PROCESS) trial, a large, multicenter, randomized controlled trial of SCS, has provided some of the best evidence for the use of SCS in failed back surgery syndrome.¹²^–¹⁴ The data emerging from this ongoing trial, first presented in 2005, followed pain relief and cost-effectiveness when compared with conventional medical management. Pain relief was significantly better in patients treated with spinal cord stimulators than in those managed conventionally.¹³ Patients also reported significant improvement in health care–related quality of life. However, the most recent data from this study show that, at 6 months, health care costs were significantly higher in the SCS group, likely related to the costs of the device and implantation.¹⁴ Longer follow-up of these patients is required to assess the true cost of this therapy.

The role of SCS in the management of recurrent back and leg pain following spine surgery has been examined in a randomized controlled trial comparing this therapy with repeat lumbar spine surgery.¹⁵ In this series of 42 patients with operable spinal pathology, SCS proved more effective than redo spine surgery. Moreover, a significant proportion of patients who had been randomized to spine surgery and failed to benefit from it subsequently achieved good results with SCS. Finally, the cost of therapy with SCS proved lower than repeat lumbar spine surgery, despite the significant cost of the hardware used.¹⁶

Complex Regional Pain Syndrome

SCS has been in use for the treatment of complex regional pain syndrome (CRPS). Also known as reflex sympathetic dystrophy, this is a condition for which there are few effective treatment options. The pathophysiology is unclear, with pain, dysfunction, and trophic changes occurring in an affected limb following trauma or surgery to that limb.¹⁷ Kemler and colleagues¹⁸ have recently released the 5-year follow-up results of a randomized, controlled trial comparing SCS with physical therapy to physical therapy alone. They found significantly better relief of symptoms in SCS patients than in those conservatively managed during the first 3 years after implantation. However, a spontaneous improvement in the control group appears to gradually diminish the effect of SCS over longer periods, with no statistical difference between the groups at 5 years. Despite these results, 95% of patients with SCS reported that they would undergo implantation again.

Ischemic Pain

It has been noted that SCS may alter vascular tone and improve tissue perfusion. It is thought that, through a combination of sympathetic outflow modulation and antidromic activation of sensory fibers, SCS acts to dilate peripheral vasculature.¹⁹ The use of SCS in the treatment of peripheral vascular disease was first proposed in the 1970s.²⁰ Subsequently, stimulator placement in the low thoracic or lumbar region of the spine has been shown to be effective in the treatment of pain caused by lower extremity peripheral artery occlusive disease.²¹^–²³ Placement of the stimulator electrode at higher thoracic levels has been used in the treatment of intractable angina, with studies showing improved exercise tolerance and less ST-segment changes with SCS treatment.¹⁹ Recently, the effect of cervical SCS on the cerebral vasculature has been explored, with focus on cerebral vasospasm following subarachnoid hemorrhage.²⁴

Psychological Screening

One of the practical issues in SCS is the screening of patients. Clearly, not every pain syndrome responds equally well to SCS. In the same vein, not every patient with chronic pain responds well to this therapy. In an effort to sort out who responds well to SCS, many practitioners have turned to psychological screening. Aside from its ability to identify patients with major psychiatric morbidity, psychological screening has been reported to have some predictive value in selecting patients who would benefit from the therapy.²⁵ Some studies have found that it is important to assess the psychological profile of the patient to determine the suitability of SCS for treatment of the patient’s pain. In particular, preimplantation screening protocols have examined overall stability of the patient and the patient’s defensiveness, self-confidence, and self-efficacy. Other aspects examined are the patient’s response to and concern about the illness and the patient’s ability to cope with setbacks. Family support and a willingness to be an active participant in the patient’s care are also important.²⁶ However, the utility of psychological screening is not uniformly accepted; there are also data that suggest that the predictive value of psychological testing is low.²⁷ In summary, data regarding the usefulness of psychological screening of patients before SCS are mixed. However, this type of screening continues to be widely practiced.

Early Lead Configurations

The initial foray into SCS used a small laminectomy for placement of the stimulating electrode into the epidural, endodural, or subarachnoid space. The placement of the electrode with a laminectomy under local anesthesia provided limited access to different levels of the spinal cord, yielding little intraoperative flexibility in tailoring the stimulation field to the patient’s pain. The advent in the 1970s of slender electrodes that may be inserted percutaneously through a Tuohy needle increased the maneuverability of the electrode longitudinally along the spinal canal. This, in turn, has allowed for improved ability to determine the optimal level for electrode placement, maximizing stimulation in affected areas.

Modern-Day Lead Types

Contemporary stimulator lead types consist of either insulated wires containing anywhere from one to eight contacts at their tips in a linear array or paddle-shaped electrodes with two or three columns of disk-shaped electrodes (Fig. 167-2). The contacts are composed of a nonferromagnetic alloy such as platinum-iridium. In general, the insulating material is silastic based to maintain flexibility and durability. The percutaneous, wire-shaped electrodes can be placed through a needle under fluoroscopic guidance, allowing the surgeon broad access to multiple levels of the spinal cord in a minimally invasive fashion. However, by virtue of their cylindrical shape, these contacts radiate current in all directions, requiring more energy and potentially stimulating dorsally located nerve fibers in addition to the ventrally located spinal cord. The paddle-shaped leads allow for more focused dispersion of current, but require a more invasive implantation procedure.²⁸

FIGURE 167-2 Spinal cord stimulation electrode arrays and generators in contemporary use. The linear arrays on the left may be placed through a Tuohy needle, whereas the flat, paddle-shaped electrodes (center) require open surgery for placement. Several implantable generators made by the various manufacturers are pictured on the right.

The percutaneous technique was initially used for temporary trials of stimulation before laminectomy and placement of a permanent lead. However, the percutaneous leads have since become popular for long-term use, owing to ease of placement. The earliest percutaneous electrodes had a single contact and required the placement of multiple leads to achieve bipolar stimulation. In the 1980s, leads were developed with increasing numbers of contacts, providing more maneuverability in stimulation. Earlier lead designs incorporated large contacts with long intercontact spacing because these were thought to broaden the effect of stimulation. However, more contemporaneous lead designs use smaller contacts with narrow intercontact spacing. Work with finite element modeling of the spinal canal has demonstrated that this narrow intercontact spacing provides superior targeting of the dorsal columns of the spinal cord (Fig. 167-3).²⁹ In addition, it appears that a transverse orientation of anode and cathode minimizes uncomfortable radicular stimulation and maximizes stimulation of the dorsal columns. The “transverse tripole,” which involves a negative terminal flanked transversely by two positive terminals, is thought to provide for optimal stimulation characteristics.³⁰ Accomplishing a transverse orientation requires that multiple percutaneous leads be placed parallel to each other or, alternatively, requires the placement of a paddle electrode with multiple linear electrode arrays.

FIGURE 167-3 A, Transverse section of cervical three-dimensional model of the spinal canal and surrounding tissues; cathode is centered at this section, and dorsal roots are not included in the model. B, Isopotential lines in unipolar stimulation. C, Isocurrent density lines in transverse tripolar stimulation. CSF, cerebrospinal fluid.

(Adapted from Holsheimer J. Computer modelling of spinal cord stimulation and its contribution to therapeutic efficacy. Spinal Cord. 1998;36:53.)

Laminectomy Compared with Percutaneous Placement

The ability to implant leads for SCS by a minimally invasive, percutaneous technique has resulted in a profound increase in the popularity of this therapy during the past 20 years. During the past 5 years, however, a number of reports have surfaced suggesting that the leads implanted through laminectomy may deliver better results than those placed percutaneously.²⁸^,³¹ In the early 2000s, a randomized prospective controlled trial was conducted to compare both the technical and clinical results associated with the percutaneous and laminectomy electrodes. Initial technical data, published in 2002, touted the increased insulation of the paddle-style electrodes for longer battery life and the ability to use lower stimulation amplitudes.²⁸ All patients participating in the trial initially underwent temporary percutaneous implantation, followed by randomization to either permanent implantation of percutaneous or laminectomy-placed leads. Patients who were able to compare the two types of leads reported better ratings of paresthesia coverage of pain, and patients reported improved low back coverage. Clinical data, published in 2005, revealed that this advantage was durable to a follow-up point of 2 years, but the statistical significance disappeared at a mean of 2.9 years of follow-up. The evidence therefore does not clearly support the theoretical and anecdotal reports of laminotomy lead superiority.³¹ Still, laminotomy leads are clearly indicated in patients in whom percutaneous implantation cannot be accomplished because of epidural adhesions and in patients who have experienced repeated percutaneous lead migration.

Generators

Advances in the miniaturization of electronics and battery technology have provided significant improvement in the flexibility of SCS systems. The earliest generators employed radiofrequency transmitters worn externally to energize the stimulator electrode. These have no implanted battery that would require replacement, and revisions were rarely required. The major disadvantage of such systems is the need for the patient to wear an external power-transmitting device, which may have variable communication with the internal component, altering the perceived stimulation amplitude. Radiofrequency-coupled devices are therefore used infrequently but are still used in selected patients with very high energy requirements.

Implantable pulse generators became available in the 1980s with the advent of lithium ion battery technology (see Fig. 167-2). These have the obvious advantage of providing more convenience and a better cosmetic result. Battery life in older models is limited, however, and requires balancing stimulation parameters with battery life to achieve the best result. Patients must undergo surgical replacement of their implanted generators when the battery has been exhausted.

Although fully implanted pulse generators now make up the lion’s share of systems in clinical use, a new generation of rechargeable generators has entered the marketplace. These devices allow patients to enjoy the convenience of an internalized power source, without the need for frequent surgical procedures to replace the generator. To recharge these generators, the patient wears a radiofrequency-coupled external device periodically for several hours. Unfortunately, compliance with recharging these devices is very important; completely depleted, the rechargeable generator may be irreversibly damaged and require surgical replacement. The price of rechargeability therefore appears to be more significant patient compliance in a regular maintenance schedule, which interferes with the convenience of this device.³² Therefore, the new, rechargeable generators do not necessarily replace the primary cell, implantable generators that have dominated this clinical arena.

Patient Programmers

One of the advantages enjoyed by SCS units is their ability to allow the patients to interact with their therapy. Patient programmers allow the user to modify their stimulation characteristics to optimize their pain management. Over the years, the programmers have become more sophisticated. Initially, patients were provided a magnet that simply turned the generator on or off. Contemporary programmers are considerably more sophisticated, permitting the patient to adjust amplitude, frequency, pulse width, and contact polarities. Some of these programmers allow patients to experience multiple stimulation programs, which alternate with each other on a millisecond basis, essentially becoming simultaneous in their effect. Such advanced programming options have not been shown to be superior to traditional stimulation programs but allow a glimpse into the possibilities offered by advances in electrical engineering. Time will tell whether these devices will improve the efficacy of spinal cord stimulators.

Surgical Technique

Electrode Placement

Placement of epidural stimulation electrodes may be accomplished in a number of ways. In the simplest technique, the electrode may be placed percutaneously through a Tuohy needle, with the leads connected to a generator that may be either external (temporary) or implanted. At the other end of the range, a single-level laminectomy may be performed for placement of the paddle-style electrode. These procedures are usually performed in a conscious patient, using local anesthetics and mild sedation to allow the patient to assist in optimal placement of the stimulating electrode.

A trial of stimulation before permanent implantation is usually performed to ensure that the effects of SCS are therapeutic. This involves an initial lead placement with connection to an external generator, allowing for evaluation of the treatment for several days. The electrodes may then be used permanently or changed at that time.

Determination of the spinal level for electrode insertion and placement is key to successful stimulation. The location of pain treated and the type of electrode being placed guide the level of insertion. Needle-type electrodes have more flexibility in that they may be advanced a number of levels cranially through the epidural space to ensure the best stimulation coverage. Conversely, paddle-type electrodes may generally be placed only one to two levels cranial to the location of the laminectomy. For upper extremity symptoms, cervical placement is required. One should be mindful of the cervical cord enlargement, and electrode insertion should ideally be performed below the T1-2 level. A practical rule of thumb is to insert the electrode at T2-3 or T3-4 for upper extremity pathology and at L1-2 or L2-3 for lower extremity targets. The target level for the electrical contacts varies by patient and pain region being treated. Common target levels for lower extremity coverage range from T8-T12, spanning the lumbar enlargement of the cord. Similarly, common levels for upper extremity symptoms range from C3 to C6, covering the cervical enlargement. Placement where the spinal cord is of small caliber may result in unpleasant local segmental effects. Barolat ³³^,³⁴ has published an exhaustive review of stimulation levels and their physiologic correlates.

Percutaneous placement of spinal cord stimulator electrodes is an attractive option, especially for temporary trial of stimulation. The patient is placed in the prone position with biplanar fluoroscopy. Several centimeters of the lead should lie in the epidural space to stabilize the electrode and minimize migration. This is best accomplished by entering the spine at least two segments below the target stimulation level. The technique for insertion uses a Tuohy needle to gain access to the epidural space.

The epidural space may be identified through the loss-of-resistance method. It is important to identify subarachnoid placement of the needle as opposed to epidural placement, and this may be accomplished in a number of ways. Of course, the appearance of CSF will indicate violation of the dura. In the absence of CSF flow, a guidewire or the electrode inserted into the subarachnoid space will glide more easily, and the wire may be seen to “float” in the CSF on fluoroscopy. Finally, electronic stimulation within the subarachnoid space will elicit stimulation response at extremely low thresholds.

When the epidural space has been identified, the electrode may be advanced through the Tuohy needle to the appropriate stimulation position. Care should be taken to recognize whether the electrode has migrated ventrally, which will be indicated on anteroposterior fluoroscopy by slight lateral deviation followed by medial curvature. Lateral imaging may be useful in determining this, as well.

The electrode should be secured with multiple points of fixation to reduce the chance of dislodgement. Strain relief loops may be used around the insertion site to deflect tension away from the trajectory of the electrode. Anchors and nonabsorbable suture are used to fix the electrode to the interspinous ligaments and to the fascia before tunneling toward the generator.

Generator Placement

Internal pulse generators are implanted in a subcutaneous pocket in a position where they will not interfere with bony prominences. Two common locations are the lower quadrant of the abdomen or the buttock. It is important to consider the ease of the patient to access the generator for routine programming. Also, care should be taken to avoid placing the generator where undue pressure will be placed on it, causing skin breakdown from waistbands or sitting. Some practitioners advocate placing a generator in the region of the buttock because it is more convenient surgically and is cosmetically acceptable to most patients. However, there are data indicating that such generator placement increases the strain on the leads, increasing the propensity for lead migration or fracture.³⁵

In preparation for tunneling and generator placement, it has been our practice to close the midline spinal incision and place the patient in the lateral position under general anesthesia. Any trial leads exiting the skin are disconnected from the generator and are cut close to the skin. The patient is prepared and draped, and the subcutaneous pocket is created. The back incision is then reopened, and an extension lead is tunneled subcutaneously between the two incisions.

Complications

SCS has proved to be a safe technique for the management of chronic neuropathic pain. The incidence of life-threatening infection is low; the most severe complications require reoperation, whereas others may simply affect pain relief.³⁵^,³⁶ Complications may be divided into two categories: technical and biologic. Technical complications relate to both the implantation technique and long-term durability of the hardware. A recent review of the literature found that one of the most frequent complications is electrode migration. Percutaneously placed electrodes are more susceptible to this issue than are paddle electrodes. Paralysis has been reported rarely, usually related to development of epidural abscess.³⁷