Introduction

Neuromodulation achieves analgesia without producing destruction of nerves. Attempts to alter pain perception in the cerebral cortex by neuromodulation can take place via the peripheral nervous or the central nervous system. Neuromodulation within the central nervous system occurs at the spinal cord level or within the brain. All of these sites have been used to decrease the perception of pain and produce analgesia.

The recent clinical focus of neuromodulation techniques in chronic pain medicine center on two main areas: electrical (nondrug) stimulation and intrathecal drug delivery. This chapter looks at electrical stimulation as a neuromodulatory way of producing pain relief in patients suffering from chronic pain syndromes. Specifically, the chapter focuses on percutaneous and surgical techniques of spinal cord stimulation (SCS).

Patient Selection

Several factors are considered when selecting patients for SCS. The type (neuropathic vs. nociceptive) and location (radicular vs. axial) of pain are considered key components in the selection process. However, it is equally important to consider physical characteristics of the patient when deciding if he or she is an appropriate candidate for SCS. This consideration is also important for implanters who only perform percutaneous techniques to a decision about potentially referring the patient for surgical or paddle lead placement.

Patient selection for SCS is reviewed in detail elsewhere in this book. The ideal candidates for percutaneous leads are younger patients who do not have significant degenerative spine disease or pronounced scoliosis and/or kyphosis. The patients depicted in Figs. 6-1 and 6-2 have significant scoliosis and were considered potentially difficult percutaneous placements. Approaching from the convex side of the scoliotic curve, the trials proceeded uneventfully as did the subsequent permanent percutaneously placed epidural leads.

Fig. 6-1 Scoliotic patient undergoing trial spinal cord stimulator lead placement. Note position of needle on convex side of scoliosis. This approach facilitates successful lead placement.

Fig. 6-2 Scoliotic patient with hardware in place from previous surgery. Note approach is from convex side of scoliotic curve.

It can prove difficult to place and anchor percutaneous leads to appropriate fascial tissue in morbidly obese patients. Percutaneously placed trial leads can often be maintained in the obese patient for 1 to 2 weeks without incident.

Migration of percutaneously placed spinal cord stimulator leads has been reported in many studies.^1–3 The reported incidence ranges from 5% to 23% in different series.^1–4 Proper patient selection should help to minimize the likelihood of subsequent migration. As with morbidly obese patients, very thin patients may prove more technically challenging for the percutaneous implanter. This may include finding appropriate space for the generator and anchors and fixation of the leads.

Trialing

Trialing methods for SCS vary from one implanter to the next. No consensus exists for duration of trials, percutaneous or surgically implanted leads, or number of leads used for the trial (commonly, one or two leads).

Most implanters trial from 48 hours to 10 days. The overwhelmingly and single most important goal of a spinal cord stimulator trial attempts to determine the likelihood of a patient achieving clinically meaningful long-term pain relief from the device. No studies have determined the length of time needed for a trial to answer this question.

How much pain relief is necessary before considering a trial successful for subsequent permanent implant? The literature often reports 50% pain relief as an outcome for judging a successful trial.^5–6 No good studies have looked at whether a criterion of 50% pain relief during a trial period predicts long-term success with SCS. It is quite possible that some patients with less than 50% relief may find acceptable relief long term and/or significant improvement in activities of daily living and increased functional abilities. It is known that some patients who report 50% or greater pain relief during a trial do not sustain relief long term and eventually become therapy failures.

Pain relief estimated at 50% often does not correlate with 50% reduction as measured by a numeric pain rating scale (NPRS) (Rauck, unpublished data). It is unclear if implanters who use 50% as a cutoff for determining whether to implant a permanent system should use the patient’s verbal response or an NPRS.

Percutaneously implanted spinal cord stimulator leads can be inserted less invasively than surgically implanted trial leads. Surgically implanted trial leads are commonly sutured to spinal elements. An unsuccessful trial requires a second surgical procedure to remove the leads.

Percutaneously implanted spinal cord stimulator leads can be tunneled and exteriorized. These leads commonly are anchored in the deep fascia tissue of the back at the time of insertion. If the trial is unsuccessful, a second surgical procedure is required to remove the anchors and lead(s).

Many percutaneous implanters place leads through a special epidural needle with no intention of leaving the lead in place for long-term use. The lead or leads are exteriorized and sutured in place against the skin. At the end of the trial the lead(s) can be removed in the office without a surgical procedure or use of fluoroscopy.

The effort to place surgically implanted trial leads or percutaneous leads that are anchored, tunneled, and exteriorized may present a bias for second-stage completion or implantation of the generator. Patients with marginal results during the trial may opt for permanent implantation rather than agreeing to a second surgical procedure for removal. The downside to removing the trial leads at the end of the trial is the risk that subsequently placed permanent lead(s) may not recreate paresthesias exactly as the trial. Patients may report that the permanent system does not perform as well as the trial.

The implanting physician must decide whether to use one or two leads during a trial. Cost, increased procedural time, and increased risk of complications mitigate against placing two leads for all trial purposes. However, placing only one spinal cord stimulator lead for the trial may prove inadequate for producing sufficient paresthesias and analgesia for the patient. Many implanters use two leads for all permanent implants; thus one can argue that two leads should be used at trial to mimic the long-term paresthesias one expects over time.

Risk of infection during spinal cord stimulator trial has been infrequent.^7–8 Meticulous sterile technique should be followed during the trial placement. Literature from other implantable trial catheters suggest that the risk of infection increases with the duration of the trial.⁹ A recent study with intrathecal catheters (in which the risk of serious neuraxial infection would be expected to be greater than epidurally placed spinal cord stimulator leads) reported no infections until week 3 and thereafter an incidence of 16% for catheters placed longer than 2 weeks.⁹

Our policy (RR and JN) has been to implant one percutaneously placed lead for most of our spinal cord stimulator trials. During paresthesia mapping in the fluoroscopy suite, we place a second lead if we are unable to get adequate paresthesias in the area of pain as reported by the patient. This occurs occasionally in our practice.

We inform patients to expect the trial to last 1 week. Patients are followed closely by telephone to make sure that the trial is progressing smoothly. If paresthesias are inadequate, too light, or too strong, patients are brought back to the clinic for reprogramming. Occasionally trials are extended beyond 1 week if patients cannot adequately assess the results of the trial.

If the lead migrates or pulls out (very rare) before 1 week, the results of the trial are assessed and reviewed with the patient. If sufficient pain relief (>50%) is reported and the patient and implanter agree, the decision to go forward with a permanent implantation is often made. Similarly, if no or minimal pain relief has been obtained, no permanent implantation is performed. Occasionally the trial lead is replaced 1 to 2 weeks later if the patient cannot assess the effects of the trial at the time of lead migration.

Finally, after removal of the percutaneous trial lead, we wait 1 to 2 weeks before permanent implantation. This allows for any indolent infection from the trial to manifest before moving forward with a permanent implantation. It is always preferred to find a subcutaneous or other skin infection from the trial and eradicate it before placing the permanent implant.

Positioning the Spinal Cord Stimulator Lead

Ultimately the location of any spinal cord stimulator lead depends on individual patient characteristics. No single location produces the same paresthesias in all patients. The optimal time for finding the location that produces paresthesias in the patient’s pain distribution is during trial lead placement. The implanter and the programmer coordinate efforts along with feedback from the patient to map paresthesias that cover the patient’s complaints of pain. Nevertheless, it is important to mention that it is not always possible to reproduce the paresthesias reported during the trial by implanting the permanent leads in the same radiologically defined location. Often, the permanent lead is implanted in the vicinity of the area where the trial lead was placed.

Patients are most commonly positioned prone for both percutaneous trial and permanent lead placement. If necessary, clippers are preferred over shaving and should be used the day of surgery. Prophylactic intravenous antibiotics should be administered within 1 hour of the permanent implant unless there is a strong contraindication present. We also routinely administer intravenous antibiotics before the percutaneous trial implant, although all implanters do not agree on its efficacy.

After sterile surgical preparation (e.g., chlorhexidine), many implanters use an Ioban drape over the surgical site. Standard sterile surgical techniques are used. Needle entry for percutaneous placement depends in part on anticipated final placement of the lead(s). Common needle entry for the lower extremity and/or axial low back pain is the midlumbar region. Skin entry commonly is marked at L2-3, L3-4, or L4-5. Entry into the epidural space should be as flat as possible, dependent in part on the body habitus of the patient. Entry into the epidural space is either one or often two levels above skin insertion. A paramedian approach should be used to avoid both the forces of the supraspinous and interspinous ligaments and the tendency of the spinous process to fracture a lead placed through a midline approach. The percutaneous implanter should not hesitate to use a longer-than-standard epidural needle to ensure that the angle of approach to the epidural space is shallow (less than 45 degrees whenever possible). A lateral view should be taken to ensure that the lead has not migrated anteriorly in the epidural space or into the dura (Fig. 6-3).

Fig. 6-3 Lateral view of spinal cord stimulator lead positioned erroneously in the anterior epidural space. Note that it is not possible to ascertain that a lead is anterior in the epidural space with a standard anteroposterior view.

Common lead placement for lower-extremity paresthesias vary from T9 to T12 (Figs. 6-4 and 6-5). Lead placement below T12 will not consistently stimulate posterior columns since the spinal cord often terminates at L1 or L2. Stimulation for axial back paresthesias commonly requires placement of the leads at T7 and/or T8. As stated previously, final lead placement should always be individualized to the patient response during intraoperative mapping.

Fig. 6-4 Lead placement for typical spinal cord stimulation of lower extremities and axial back pain.

Fig. 6-5 Spinal cord lead placement for axial low back pain with leads positioned across vertebral body levels of T8-T9.

For upper-extremity paresthesias, leads are commonly placed from C2 to C7. Needle insertion should be in the thoracic region. A choice exists between the upper and lower thoracic region. The upper thoracic area is a stable and good site for needle placement. Skin location of T4 or T5 allows the needle to advance easily in a shallow orientation to the epidural space. The normal kyphotic nature of the upper thoracic spine can make insertion in this area technically difficult or nonintuitive to the novice implanter. On the skin the needle often appears perpendicular, whereas the fluoroscopic image (particularly the lateral image) verifies that the approach to the epidural space is shallow (Fig. 6-6). Although insertion at T3 or T2 is certainly achievable, this kyphotic tendency enhances the direction of the needle. Skill and practice in learning this technique are necessary before implementation. The lateral view on fluoroscopic imaging can be difficult to interpret (see Fig. 6-6). However, as described in the following paragraphs, there are significant advantages to using the lateral view as one advances the needle in this location. With practice, the implanter can learn to interpret and understand this approach in the lateral view. The major advantage of needle placement in this area is less threading or passing of the electrode lead in the epidural space.

Fig. 6-6 Approach to epidural space in the upper thoracic region. Paramedian approach is used on anteroposterior view. Lateral view can be difficult to understand but becomes easier with repeated use.

Proponents of a lower thoracic needle placement with subsequent passage of the lead to cervical sites counter that the lower skin insertion site for the needle and lead obviates the need to tunnel the lead(s) from the upper thoracic space to the generator site (usually a flank or buttock location).

In our practice we use a lower thoracic approach when the patient is young or we expect threading or passing of the lead in the epidural space to proceed without difficulty. In the older patient or one with significant spine disease we commonly choose the upper thoracic space for needle entry. Passing the lead(s) long distances through the thoracic spine region in the presence of significant spondylolysis, facet arthropathy, spurs, thickened ligamentum flavum, or other degenerative diseases can be technically challenging and fraught with difficulty. The lead can bow when significant obstructions are encountered, with painful paresthesias reported by the patient if the lead contacts a corresponding nerve root (Fig. 6-7).

Fig. 6-7 Spinal cord stimulator that has bowed laterally. This lead can easily produce paresthesias on the nerve root and can migrate anteriorly in the epidural space.

In our experience it can be very difficult to obtain adequate paresthesias with SCS in patients who have significant neck pain. The best attempt at achieving paresthesia relies on placement of the electrode at C2 with an exaggerated lateral location as seen on the anteroposterior (AP) fluoroscopic image (Fig. 6-8). One alternative is to attempt a trial with a paddle lead placed retrograde over C1-C2, in a midline position. In the experience of the authors, this often allows for paresthesia coverage of the neck.

Fig. 6-8 Cervical trial lead positioned at C2 with lead extending laterally. This placement provides optimal chance of producing pain-relieving paresthesias in the neck.

There have been case reports demonstrating four-limb paresthesias with leads placed in the lower cervical region.¹⁰ Most patients considered candidates for SCS do not have pain in three to four extremities. However, this observation of four-limb stimulation has been useful in the rare patient and may prevent placement of a second generator and two additional leads.

Lead placement for angina pectoris patients is commonly performed at C7 to T3 (Fig. 6-9). Needle insertion is usually in the lower thoracic level (T9 to 12). Two leads are commonly used and are positioned over both posterior columns. The leads should be placed close to the midline to avoid uncomfortable paresthesias of the chest wall.

Fig. 6-9 Spinal cord stimulator leads placed in patient with refractory angina pectoris. Note leads are positioned between C7 and T2.

A note should be made about the placement of two percutaneous leads. For trial purposes, the second needle is sometimes placed on the opposite side of the spine from the first needle or an interspace one level below the first needle. One should avoid placing the second needle above the first needle on the same side of the spine because the needle can potentially damage the lead coming from the first needle (Fig. 6-10). It is possible to use the same interspace as the first needle, use a skin insertion site 1 to 2 cm below the first needle, and approach the space using the first needle as a guide.

Fig. 6-10 Lateral view demonstrating the position of two 17-gauge Tuohy needles at the same vertebral level. Note that second needle is appropriately positioned beneath the first needle/lead to prevent possibility of the needle shearing or damaging the first lead.

For permanent implantation, two leads are commonly used. This often produces optimal paresthesia possibilities and also assists the programmer in maintaining adequate coverage if there is minor migration following surgery. Both leads can be inserted through the same intervertebral space using the same paramedian approach from either the right or left side of the spine. In fact, it is our personal preference to insert both epidural needles at the same level and on the same side of the spine. The second epidural needle should always be inferior to the first needle. The second needle is positioned after the first percutaneous lead has been positioned. It is acceptable for the second needle to slide slightly medial or lateral relative to the first needle, depending on the location of the first needle and the desired location of the second lead. Lateral fluoroscopy greatly helps the implanter watch the second needle as it approaches the epidural space. The implanter can easily discern that the needle is staying inferior relative to the first needle as it approaches the epidural space. The first needle also serves as an excellent marker for where the posterior epidural space lies.

If there is difficulty placing the second needle, the opposite paramedian side or a lower epidural space can be used. It is very unusual to have to use an additional space; once the implanter becomes comfortable using the same interspace, the initial needle serves as an excellent landmark for depth and location of the posterior epidural space.

Subcutaneous Cut-Down and Lead Anchoring Techniques

Implanters vary during permanent implantation on when to perform the subcutaneous cut-down for anchoring the leads. Some implanters perform the cut-down before either needle placement. Others prefer to perform it after the initial needle is placed but before the second needle. Either method is acceptable and can result in satisfactory outcomes.

If the cut-down is performed after needle and lead placement, the needle should be left in place to protect the lead during cut-down. A self-retaining retractor aids exposure. Monopolar cautery should be avoided in the area of the epidural needles because the electrical current and heat from contact with the Bovie can be carried down the needle to the epidural space.

Once site preparation is complete, stylets and needles should be removed. Fluoroscopic guidance is used to ensure no change in position of the leads. Final lead arrangement can have several patterns. In addition, some electrode asymmetry can be considered acceptable in the majority of cases.

Meticulous anchoring of the leads is essential to prevent subsequent migration of the lead array. Anchoring devices are preferred by the authors as sutures placed directly around the leads may facilitate lead deterioration, failure, and possibly fracture. Nonabsorbable sutures placed into deep fascial planes help hold the anchor in place. Multiple sutures around some of the anchoring devices help to minimize subsequent migration (Fig. 6-11). When using an anchoring sleeve, it should be positioned forward on the lead such that the distal end is abutting or directed into the deep fascial tissue. Placement of the anchor in this fashion minimizes the chance of the lead migrating out of the epidural space and into the space between the fascial tissue and the anchoring sleeve. It is a good idea to take frequent fluoroscopic images during the anchoring process to ensure that no lead migration occurs. A final fluoroscopic image in both the AP and lateral views at the end of this process ensures that there has been no migration.

Fig. 6-11 Suturing technique of anchor. Note constriction of anchor around the lead using a sturdy suture.

Lateral Fluoroscopic Imaging

Fluoroscopic imaging is used to aid in placement of the epidural needle and subsequent passage of the spinal cord stimulator lead into the posterior epidural space. Standard fluoroscopic machines are limited to two-dimensional views. AP views inform the implanter if the needle or lead is too medial or lateral. AP views cannot tell the depth of the needle. Lateral fluoroscopic views cannot help with left/right orientation of the needle but can help visualize the depth of the needle.

Some implanters use AP fluoroscopic guidance exclusively during placement of the epidural needle, relying on the feel of the lamina and a loss of resistance technique to guide them into the epidural space. In the authors’ technique, the needle is placed firmly into fascial tissue using the AP view. The AP view is used initially to ascertain left/right direction and orientation. Once it is evident that the needle is headed in the correct direction, the fluoroscopic machine is rotated to the lateral view for final approach to the epidural space. Careful attention is made not to change the left/right orientation during lateral imaging. Whether the lead is in the lumbar, lower thoracic, or upper thoracic space, our preference is to position the fluoroscopic machine in the lateral view as the needle approaches the epidural space. (This is also our approach for cervical epidurals without SCS and all other approaches to the epidural space.) It takes skill and practice to interpret the location of the posterior epidural space, particularly in the upper thoracic area, but with experience it can be done consistently and reliably.

Further, the lateral fluoroscopic view shows where the lumbar lamina is and where access to the epidural space can be gained (Fig. 6-12). If the implanter learns the anatomy and understands the images, it becomes possible to predict if the approach of the needle will gain access to the epidural space or require redirection to avoid the protective lamina.

Fig. 6-12 Lateral fluoroscopic imaging showing anatomic relationships to entry site for epidural space. Red lines (shadows of pedicles and lamina) are analogous to stairs with entry on “risers” of steps. Blue line is superior endplate of vertebral body.

Generator Site Location and Preparation

The generators, particularly for rechargeable systems, have greatly diminished in size over the past several years. This allows additional options for placement of the generator. Generators can be placed in the posterior hip area/buttock area, the flank (single incision), midaxillary line over the inferior ribs, or subclavian area (for upper-extremity leads and upper-thoracic needle placement). Abdominal placement of the pulse generators is also possible and it has been suggested that it may reduce the risk of lead migration, although this is not a consensus among implanters.

The most common site for generator placement has been in the posterior hip area. The goal is to place the generator above where the patient sits but below the belt line. This can sometimes be problematic and the generators may coincide with the belt line. Incisions are usually made horizontally. The depth of the pocket is commonly 1 to 3 cm, and recommendations vary among the suppliers for actual depth of generator placement. Ideally the generator is placed deep enough to avoid subsequent discomfort or erosion but superficial enough for recharging and interrogation of the system with subsequent programming.

Adequate hemostasis in the pocket is important before placement of the generator to prevent blood products from entering the portals between the generator and the leads. The leads or extension wires are tunneled between the electrode and anchor site and the pocket site using a tunneler. Some implanters prefer to leave a tensioning loop in the back incision to minimize the potential for the leads migrating when the patient moves.

The electrical contact sites on the lead should be inspected for blood or tissue products and cleaned with a wet, then dry, sponge before insertion into the generator. After connecting to the generator, excess lead is coiled and maintained under or around the generator. Connecting cables may be needed to reach the IPG site. This should be considered against the possibility that each additional connector adds risk for disconnecting or for electrical interface problems.

The integrity of the system should be checked before closure. The generator is sutured into the pocket to prevent its migration, flipping over, or twiddling by the patient. A final AP and lateral fluoroscopic image is obtained to demonstrate no migration during the process.

An alternative location for the generator is in the flank. This location allows for placement of the generator lateral to where the leads are anchored in lumbar epidural insertions. Large generators are not well tolerated in this area. However, small (rechargeable) generators work very well in a lateral flank location. Fig. 6-13 shows final lead placement with generator positioned in the flank. Flank locations make it unnecessary to have two incisions. This should reduce the risk of skin or superficial infections. In addition, no tunneling is necessary between leads and pocket site. The implanter must make the pocket large enough so the generator stays off the midline area. Patients do not tolerate pockets migrating over the spinal processes. Besides anchoring the generator to fascial tissue, an additional two or three retaining sutures to close the pocket site help keep the generator from migrating.

Fig. 6-13 Final placement of bilateral spinal cord stimulator leads with pocket and generator placed in right flank/paramedian locations.

A few implanters and patients prefer the pocket in the abdominal region, in the midaxillary line, or more anteriorly over the inferior ribs. These are acceptable sites but may require the use of longer leads or connected extensions. It also can be problematic to perform the entire procedure in the prone position and may require re-preparation and re-draping of the patient for pocket incision and generator placement.

Revisions

Complications of SCS placement are reviewed elsewhere in this book. However, it should be noted that all implanters will experience a need to revise SCS systems. This may be replacing a generator, moving a generator to a different site, replacing or repositioning a migrated lead, or removing an infected system.

The steps to perform a permanent, percutaneously placed SCS system can be relatively straightforward. Managing SCS systems after placement can prove challenging in some cases. Anyone who chooses to do permanent SCS implantations should be prepared to deal with the complications and perform the necessary maintenance and revisions to provide long-term satisfaction and favorable outcomes.

Summary

Percutaneously placed SCS leads can be performed for both trial conditions and permanent implants. The procedure for permanent implantations is less invasive than surgically placed leads. Long-term favorable outcomes can be expected with many patients and physicians who choose this route of implantation.

Paddle/Plate Leads

Cylindrical leads, inserted percutaneously through a Touhy needle or through superficial incisions as discussed in previous paragraphs, are a less invasive option. Incisions are small and typically do not breach the fascia. The advantages of this technical option include limited postoperative pain, discomfort, and length of stay. Most patients can be discharged on the same day as the procedure. Implantation of percutaneous leads is routinely done under local anesthesia with or without intravenous sedation. This minimizes anesthesia-related risk, particularly in the elderly and in those with multiple co-morbidities. However, the design limitations inherent in a cylindrical lead may limit stimulation efficacy. The tips of percutaneous leads are frequently placed several segments rostral to the spinal entry level and then anchored at the point of entry. Although technique modifications¹¹ and improved anchoring devices may limit longitudinal lead migration, lateral lead migration, not always easily identifiable in x-ray films, remains a concern. Furthermore, cylindrical leads are prone to positional effects with changes in activity or posture. This often limits efficacy and practical clinical value in patients who are (or want to be) more active and in those who continue to work.

Plate leads are a valuable alternative to cylindrical leads. Unlike cylindrical leads, plate leads have contacts that are insulated on one side. The contacts are flat and therefore cover a greater surface area. The expanded electrode-tissue interface maximizes efficiency and requires lower stimulation amplitude to generate a similar therapeutic response.¹² In 2005, North and associates¹² reported a prospective study comparing 12 patients implanted with cylindrical leads to 12 patients implanted with plate leads for the management of failed back surgery syndrome. Patients with plate leads had significantly better clinical results than those with cylindrical leads at 1.9 months follow-up.² Similarly in a retrospective study Villavicencio and colleagues¹³ reported that patients implanted with paddle leads also had improved pain control. Lead migration and positional effects observed with cylindrical leads are minimized with the flat or semicurved lead body design of the plate lead. Depending on the model, the lead body can be anchored to spinal elements, or extensions from the lead body can be anchored to the muscle or fascia during wound closure. These anchoring options provide stabilization points that are closer to the electrical contacts than with cylindrical leads. The risk for lead migration is reduced, although migration can still occur in some cases, particularly at the nonanchored distal tip of the lead.

Implantation Technique

The technique for implantation of a paddle lead is sometimes straightforward for the experienced surgeon. However, revision surgery is often challenging. Imaging of the spine can be helpful to assess the size of the canal and evaluate the degree of degenerative disease or post-surgical changes that may influence the type of lead selected and implant strategy. As for other posterior thoracic or cervical spinal procedures, there is always a concern for spinal cord injury, dural tears, or other complications.^3,14–16 In this chapter we focus on the thoracic implantation of plate leads in two common scenarios: (1) implantation of the plate lead following a temporary test with a percutaneous cylindrical lead, and (2) implantation of a plate lead after removing chronically implanted SCS cylindrical leads.

Implantation of a New Plate Spinal Cord Stimulator Lead

Plate leads can be selected as the first choice for chronic implantation of a SCS system. This decision often follows a test period of SCS with externalized cylindrical leads. If the placement of the test leads is without complication, placement of the permanent system can follow shortly after. Adhesions in the epidural space tend to be minimal after a percutaneous trial and usually do not pose a significant challenge for permanent implantation.

The procedure can be performed under general anesthesia or sedation, with or without regional anesthesia.^17–20 If the implant is done under sedation, intraoperative testing of the pattern of paresthesia coverage can be verified during the procedure, in a fashion similar to that of implantation of percutaneous leads.^18,20 The plate leads can be adjusted until the induced paresthesias cover all or the majority of the area of chronic pain. Implantation under general anesthesia is often preferred. Reasons include patient choice and concerns for deep sedation in the prone position. Although this option precludes intraoperative testing, implantation of the plate lead can often be successfully accomplished with intraoperative radiographic verification by placing the lead in the topographical region that was adequate for stimulation during the test period. Intraoperative neurophysiological monitoring is often used as for other spine procedures. If, after surgery, paresthesias cannot be created in the areas of chronic pain, revision of the lead may be required to accomplish this goal. The posterior elements of the spine are exposed with standard surgical exposure techniques. The length of the incision may vary with the patient’s body habitus, and usually it is not necessary to extend the dissection as laterally as for spinal instrumentation. Depending on the lead type and planned level of implantation, a unilateral partial laminectomy or laminotomy of the inferior margin of a lamina, sparing the spinous process and supraspinal ligament, is performed. For some lead models and for midline implantation, the exposure may be extended to include a midline partial laminectomy or laminotomy to clear an opening that allows for safe passage. Under most circumstances the tip of the paddle lead is advanced into the canal until the length of the lead is just beyond the laminotomy (Fig. 6-14).

Fig. 6-14 Small midline laminotomy for placement of a paddle lead. Note that only the bottom of the lead is seen while most of the body is sublaminar. The bottom of the lead was sutured to the ligamentum flavum for anchoring

If the opening of bone and ligamentum flavum is insufficient, resistance at the edges can direct the lead along unintended trajectories in the spinal canal or against the dural sac and cord. To protect the thecal sac and spinal cord from the paddle lead tip, the inferior spinous process may be removed to accommodate a shallower approach. As for cylindrical leads, implantation of the plate lead is usually planned for a median or paramedian position to optimize stimulation of the spinal cord while limiting spread of stimulation to the nerve roots laterally. The partial laminectomy or laminotomy is planned below the level intended for the tip of the lead. If a relatively short lead is used and the plan is to place the tip at the level of the top of a thoracic vertebral body, the opening can be created on the inferior margin of the corresponding lamina. However, depending on the length of the lead and the site for planned implantation, the laminectomy or laminotomy is created one or more levels below the intended vertebral body level for tip placement. The lead is inserted and aimed at the midline or paramedian position that favors the side of the pain. The location can be verified with intraoperative x-ray films and/or fluoroscopy.

Epidural adhesions, osteophytes, or thickening of the ligamentum flavum can cause resistance when the lead is passed. The nerve roots or the dural sac may also block lead advancement. In these situations it is often necessary to remove the lead and attempt reimplantation until a satisfactory trajectory into the canal and final location are seen. The dural guide is a useful instrument to facilitate dissection. Some plate lead kits include a tool for dissection of epidural adhesions. When resistance is identified, fluoroscopy can be helpful to identify what level and structure may be causing the resistance. For example, if the lead is directed laterally, resistance may be related to a nerve root, and repeated attempts may result in damage to the root or dural tear. In addition, forcing the lead beyond resistance could push it toward the sac and result in tear or cord injury. Complications from dissection of the sublaminar epidural space and lead placement are not limited to direct injury to the cord and roots during dissection alone. Hematoma formation,²¹ inadvertent dural tear with leakage of cerebrospinal fluid, and pseudomeningocele formation are infrequent but recognized complications for this type of operation.

The lead can be anchored directly to the ligamentum flavum. Anchoring to the dura can be attempted, but the suture through the dura may create fluid leakage. Some leads have a portion of the body configured for suture anchoring. If the lead is not anchored directly to the spinal elements, the extensions from the lead body can be anchored to the soft tissues during closure. Some anchors provided with the SCS leads or in separate packages are designed for this purpose. Different anchoring techniques and anchoring devices can be used. Bench data have suggested that the position and type of anchor may influence the stress on the lead and promote hardware failure,¹⁴ although these findings have not yet been validated clinically. Implantation of the implantable pulse generator (IPG) and connection to the lead directly or via extension wires is performed in a similar fashion as described for cylindrical leads.

Implantation of a Spinal Cord Stimulator Plate Lead After Removal of Cylindrical Leads

Patients experiencing limited benefits or complications associated with cylindrical lead implantation such as migration, positional effects, or hardware failure are often referred for revision of the SCS system and implantation of a paddle lead replacing the cylindrical leads. If cylindrical lead migration or limited benefits are observed early, before extensive scar formation, replacement with a plate lead is less complicated. The complexity and risk of the operation increases over time as scar formation around the cylindrical leads thickens. Computed tomography myelograms may be of value to assess the lead location and canal dimensions when revision surgery is planned. Magnetic resonance imaging is not possible in patients implanted with current SCS systems. Surgical exposure is essentially the same as for a new lead; however, caution (or avoidance) must be used with Bovie electrocautery in a patient already implanted with a neurostimulation system. The cylindrical leads are removed from their entry and anchoring site. The laminectomy is planned in accordance with the intended level of implantation of the lead. The scar buildup around the previously implanted leads may be obvious in the epidural space (Fig. 6-15). Dural guides and implantation tools for clearing the sublaminar space can be used to dissect epidural scar. If significant resistance is noted with passing the lead and the epidural scar is resistant to dissection, another laminotomy may be opened one or two levels cranially to facilitate dissection of the epidural scar and adhesions. In some cases a full laminectomy is needed. This should be weighed against the risk of a late deformity. However, if at least a “bridge” of lamina can be preserved, it can assist in stabilizing the lead and preventing dorsal migration away from the dura (Fig. 6-16).

Fig. 6-15 Laminectomy in a patient who had previous spinal cord stimulation implanted with percutaneous leads. Note the thick cylindrical scar bands in the epidural space (arrow) corresponding to the tracts of encapsulation around the percutaneous leads.

Fig. 6-16 Implantation of a paddle lead via a small midline laminotomy. Note that a second midline laminotomy was opened cranially to the initial (bottom) laminotomy to allow for dissecting epidural scar and adhesions that prevented adequate placement of the lead through a single opening.

Summary

Percutaneously placed cylindrical leads are used for trialing and permanent implantation. The procedure for permanent implantation of a cylindrical lead is less invasive than required to implant a plate lead and good outcomes can be accomplished. However, surgically implanted plate/paddle leads may be less likely to migrate, less susceptible to positional effects, and may result in better clinical outcomes in some patients when compared to percutaneous leads. In candidates for SCS, these potential benefits should be weighed against the more invasive surgery required for implantation.