INTRODUCTION

Intradiscal electrothermal therapy (IDET) is a recently proposed and investigated treatment for intractable discogenic pain. There is good evidence that IDET denatures collagen and causes changes in disc protein.¹ Yet, these changes do not alter the fundamental biomechanics of the intact motion segment.² Relatively speaking, it is a minimally invasive technique when compared to fusion or total disc arthroplasty. An important attribute of this treatment is that it does not preclude the future application of these more invasive treatments while the corollary is not true. Once disc replacement or fusion is undertaken, IDET is no longer a viable treatment option at that disc level.

Although IDET is a ‘minimally invasive’ intervention, this must not be construed to mean that the procedure is simple to perform or without potential deleterious side effects or complications. Both of these issues can be mitigated and optimal outcomes can be realized if meticulous technique is used. It is important to keep in mind the definition of technique: the systematic procedure by which a complex or scientific task is accomplished. The presentation of such a systematic, step-by-step approach is the aim of this chapter. By adhering to this approach, technical facility in clinical application is attained with every expectation of maximizing successful outcomes.

There exists an interplay between technique and clinical experience whereby good technique impacts clinical experience and vice versa. Undeniably, experience cannot be adequately transmitted through the generation of even the best written manuscript. On the other hand, the hard-earned hindsight that experience provides and which alters the basic procedure is teachable as technique. The technique presented here is tempered with our idiosyncratic experience to which, it is hoped, the future practitioner will add their own.

RELEVANT ANATOMY AND THE IDET PROCEDURE

Intradiscal electrothermal therapy entails the threading of a blunt-tipped thermal catheter through the disc with the goal of traversing the entire posterior anulus midway between endplates. Various aspects of the anatomy of the normal and degenerative disc conspire against the physician attempting to do so. There are small regional variations in disc structure due to the different demands placed upon them in the cervical, thoracic, and lumbar spine. As the main target of the IDET procedure, the lumbar disc will be the focus of discussion.

The intervertebral discs are interposed between adjacent vertebral bodies accounting for approximately 25% of the height of the column. The three main functions of the disc are to allow movement between spinal segments, serve as shock absorbers, and transmit the axial load of the body. The disc is composed of a tough outer anulus and a central nucleus pulposus. The nucleus is a remnant of the embryologic notochord. It is composed of semifluid ground substance and irregularly placed collagen fibers. When compressed, the gelatinous nucleus deforms and distributes load forces in all directions. The anulus is designed to contain these forces. It is composed of 10–20 fibrous lamellae arranged in concentric rings around the nucleus. Coincidental with the eccentric location of the nucleus, which is closer to the posterior anulus, there is a reduction in lamellar thickness posteriorly when compared to its anterior and lateral portions. Lumbar flexion places a stress on this relatively vulnerable area.

Two design features of the disc offset this weakness. First, like the lumbar vertebral bodies, the lumbar disc is kidney shaped with a slight concavity to its posterior border. This concavity increases the surface area of the posterior anulus and serves to strengthen it against distractive and compressive forces. It also increases the length the catheter must navigate as well as the acute angle of the posterolateral corners that must be negotiated to cover the posterior anulus. The passage of the catheter is more difficult as a result of increased resistance from the greater length and sharper angles. Also, this concavity must be considered when attempting to judge the proximity of the catheter to the thecal sac on a lateral fluoroscopic view prior to beginning the heating protocol. It may appear from this lateral orientation that the catheter is well contained within the substance of the anulus if its position is compared to the most dorsal extent of the vertebral body when, in fact, it has escaped at the turn around the posterolateral corner. When this transpires, the catheter will bridge the concavity of the posterior anulus instead of following its contour and, yet, appear to be safely within its confines.

Secondly, each lamella consists of obliquely oriented parallel collagen fibers. The direction of this obliquity is rotated 90° in adjacent lamellae so that the fibers between any two adjacent lamellae form a rough ‘X’ shape. This configuration contributes to the integrity of the anulus under similar forces. While the majority of lamellae form complete rings around the circumference of the disc, as many as 50% of the lamellae at the posterolateral corners are incomplete.³ Where a lamella ends, the superficial and deep rings approximate or fuse together (Fig. 26.1). Advancement of the blunt catheter tip past these fusion points may account for some of the difficulty of passing around the posterolateral corner. The tip, which tracks easily between lamellae when traveling within the anulus, may be unable to penetrate the fused lamellae at the terminus of the layer it is following.

Fig. 26.1 Illustration of normal lumbar disc.

Covering the entire nucleus and a portion of the annular ring is a layer of hyaline and fibrocartilage known as the vertebral endplate. It is bordered by the slightly raised perimeter of the vertebral body called the ring apophysis. The endplate is securely attached to the disc with the fibers of the inner lamellae being continuous with the fibrocartilage of the endplate.

In contrast, the outer annular fibers attach directly to the bony ring apophysis of the vertebral body and serve as the main ligamentous connection between segments. This function is aided by the anterior and posterior longitudinal ligaments (PLL). The PLL is intimately connected to the posterior anulus and serves to separate it from the dural sac. Both PLL and outer anulus are richly supplied with nerve fibers and are common tissue sources of back pain.

With aging, the water content of the disc diminishes and the disc becomes more fibrous. Collagen content of both the nucleus and anulus increases⁴ and the disc becomes increasingly less pliant as a result. The inner anulus expands at the expense of the nucleus and the boundary between them becomes less distinct. With dehydration the nucleus becomes less able to transmit forces equally and certain sections of anulus are subjected to disproportionately greater axial loads. Defects occur at the transition zone between the nucleus and anulus that can develop into radial tears extending through the anulus to the epidural space of the spinal canal. Increasing fibrillation of the lamellae is associated with similar defects.⁵ Circumferential tears of the anulus are the result of splitting between lamellar layers (Fig. 26.2).

Fig. 26.2 Illustration of degenerative disc with disruption of annular fibers.

Fibrotic change within the disc resulting from simple aging or scar formation following injury increases the possibility of obstruction to catheter passage and makes proper placement more difficult. Disruption of the lamellae with attendant loss of smooth tracking of the catheter between layers also increases the resistance to catheter passage. Defects in the nucleus and anulus may capture the advancing catheter tip and redirect it along paths of reduced resistance. These often lead to the epidural space as in complete radial tears or into dead ends such as the PLL or vertebral endplate.

The physician performing the IDET procedure is, in large measure, dependent upon patient anatomy for proper catheter placement; never more so than in the degenerated disc. Strategies for dealing with these problems are discussed below.

PATIENT SELECTION

The suitable candidate for IDET is a patient with a confirmed discogenic source of back pain without predominant leg symptoms unresponsive to aggressive conservative care including medications, activity modification, injection therapy, and an exercise program (Table 26.1). The need to deliver these treatments thoroughly and with expertise prior to considering IDET cannot be overemphasized.

Table 26.1 IDET Criteria

If the judgment has been made that there has been failure to progress, standing X-ray and recent magnetic resonance imaging (MRI) of the lumbosacral spine is required. Acceptable abnormalities for the performance of IDET may include disc space narrowing, disc desiccation and degeneration, or a small, contained disc protrusion. The presence of a high-intensity zone lesion does not preclude employing IDET. These images should also be scrutinized for abnormalities that could lead to the prohibition of the performance of IDET. Higher grade (II–IV) spondylolisthesis, especially isthmic and traumatic cases, will require flexion and extension plain views to rule out instability in the motion segment. Spondylolisthesis also increases the importance of ruling out nondiscogenic causes of axial low back pain. Pain greater with extension versus flexion of the spine with findings on imaging of central canal encroachment or fluid-filled zygapophyseal joints could support the diagnosis of stenosis and facet arthropathy, respectively. A positive discogram without evidence of annular disruption in the presence of a prominent Schmorl’s node may point to the node as the pain generator rather than the anulus. Short of a radical placement of the IDET catheter tip in the node itself, conventional IDET is not indicated. Performing the high heat protocol with a standard catheter placement within the anulus has been shown in a histologic study not to affect the adjacent endplate.¹ Fractures of the pars, significant central canal stenosis, large disc protrusions (>5 mm extrusion beyond the posterior border of the vertebral body), extrusions, and sequestered fragments are all relative contraindications for the procedure.

Patients with prior history of fusion who return with adjacent symptomatic discs or those with multiple symptomatic levels may be suitable candidates (Table 26.2). In experience of the authors, patients tolerate single-visit, three-level IDETs without prolongation of the recovery period. Patients who have had prior partial discectomy and fusion may also be candidates though the fusion mass or instrumentation may prevent an acceptable approach to the disc. Similarly, prior chemonucleolysis, laser decompression, or nucleoplasty at the symptomatic level does not preclude IDET as long as other inclusion criteria, especially adequate preservation of disc height, are present. These intradiscal procedures may increase annular fibrosus with consequent difficulty of catheter placement.

Table 26.2 IDET Potential Indications

A repeat IDET procedure on the same disc may be considered in certain cases. Patients who initially respond with greater than 50% relief, but then have return of similar axial low back pain or those whose lack of response is felt to be due to less than ideal catheter placement may benefit from a second IDET at the same level. With increased experience in performing the procedure, the surgeon finds fewer patients filling the latter category. Further studies are needed to judge the efficacy of repeat procedures, but in the authors’ limited experience, selected patients often experience clinically significant decreases in symptoms. Six months postprocedure is the minimum period required to judge clinical outcome and is necessary prior to consideration of a repeat IDET.

The symptomatic disc should have preserved disc height >50% with reproduction of concordant pain at low pressure and volume on provocative discography. Postdiscography CT reveals an abnormal discogram with disc disruption associated with a radial tear extending to the outer anulus (see Fig. 26.13A below). Extravasation of contrast dye into the extra-annular space indicating full-thickness annular disruption is not uncommon and is not a contraindication for IDET. Patient selection should not rely solely on the results of the discogram. Discography is just one step in the evaluative process and invariably follows a thorough history and physical exam after failure to improve with rehabilitation measures.

Fig. 26.13 Case history 1. (A) Postdiscogram CT scan of L5–S1 disc showing midline posterior annular tear. (B) Lateral and (C) CT views of intradiscal catheter placement.

In general, IDET should be considered in patients in whom fusion is to be avoided. The procedure fills a notable gap in the treatment of discogenic low back pain between nonoperative care outlined above and spinal fusion. Further clinical studies are needed to better define the exact patient subsets that are responsive to IDET.

PREOPERATIVE PREPARATION

The patient should be instructed to stop all nonsteroidal antiinflammatory drugs (NSAIDs) and aspirin-containing compounds 1 week prior to the procedure. The use of anticoagulant medications must be stopped prior to the performance of IDET. An exception is the prophylactic use of 81 mg of aspirin daily in high-risk cardiac patients where the risk of a cardiac event is considered to be greater than the risk of bleeding. Whether the risk of discontinuation of these agents is worth the potential benefit of the procedure is a decision that requires input from the physician managing the relevant medical problem, the surgeon planning to perform IDET, and the patient. In general, when a decision has been made to proceed, the patient is instructed to stop coumadin, antiplatelet drugs such as clopidogrel, and aspirin 5 days prior and nonselective NSAIDs including COX-2 selective drugs 48 hours prior to the procedure. Low molecular weight heparins such as enoxaparin need to be discontinued 24 hours prior. An INR of 1.1 or less and a platelet count of greater than 50 000 should be documented in the chart on the day of the procedure. Bleeding times for chronic antiplatelet medication use are advisable. Medications that treat hypertension, diabetes, and cardiac disease may be taken on the day of the procedure.

Comorbid medical conditions must be under acceptable control. Acute infections of any type are a contraindication to the procedure. Allergies to iodine or latex should be clearly marked on the chart so that gadolinium¹⁰ and latex free gloves and syringes can be used, respectively.

The patient is allowed liquids but no solid food on the day of the procedure. The use of sedative medications precludes driving until the day following the procedure. Arrangements should be made for a relative or friend to accompany the patient to and from the hospital.

PATIENT POSITION AND MARKING THE SKIN

Prone positioning is preferred to allow access to both sides of the spine if so needed. A pillow is placed under the belly of the patient to reduce lumbar lordosis and facilitate a clear approach to lower lumbar segments.

The patient is placed on the table so that the chosen side of approach is away from the physician. This is done so that the camera is not interposed between the physician and the operative field. A change of approach mid-procedure will result in this awkward positioning which is dealt with by moving the camera in and out to adjust the needles.

After the patient has been placed prone on the table, anteroposterior (AP) scout fluoroscopic views are taken to confirm the target level. Wide views to include T12 may be needed to count down to the correct interspace. Cross-table lateral views can also be helpful, especially if there is a pronounced lordosis. Once the target level is in view, it is necessary to ‘square up’ the angle of view with the interspace. The inferior endplate at the target level should be seen on end so that it appears as a thin horizontal line. The camera is tilted toward the head (‘cephalic tilt’) to align the proximal and distal edges of the endplate to obtain this view. The amount of tilt will be greater with lower levels and increasing lordotic curve.

A radiopaque straight edge is then laid on top of the inferior endplate and a line is drawn across both sides of the back (Fig. 26.3). This will facilitate contralateral introducer placement and cut down on fluoroscopic exposure should a bilateral approach become necessary. Next, the camera is tilted towards the side of approach to obtain oblique views of the spine showing the classic ‘Scotty Dog’ appearance of the facet joints, pedicle, and transverse process. The camera is rotated until the superior articular process (SAP), the ‘ear’ of the dog, is approximately in the middle of the disc space (Fig. 26.4). At the L5–S1 level, the iliac crest will be obstructing the approach to the disc in this view. To open up a window, the camera needs to be rotated back towards the AP view and/or tilted in a cephalic direction until an upside-down radiolucent triangle appears composed of the edges of the superior endplate, the SAP, and the iliac crest (Fig. 26.5).

Fig. 26.3 AP fluoroscopic view of straight edge across inferior endplate of L5 vertebral body.

Fig. 26.4 Oblique fluoroscopic view showing ‘Scotty Dog’ appearance of lumbar posterior elements with radiopaque marker ventral to L5 SAP.

Fig. 26.5 Oblique fluoroscopic view of introducer ventral to L5 SAP. Note radiolucent triangle formed by superior articular process of S1, the inferior endplate of L5, and the iliac crest.

The skin is marked slightly lateral to the SAP in the oblique view. Marking immediately lateral to it may prevent adequate medial placement of the introducer within the disc. This difficulty can be reduced by aligning the SAP within the middle of the disc on the oblique view.

It will be noted that the skin markings for L4–5 and L5–S1, when approaching from the same side, will often be within a few centimeters or less of each other. Indeed, it is sometimes possible to use a single puncture to approach both levels. The economy of this approach, however, must not be chosen if it will compromise optimal placement of the introducer as this is the key to a successful approach to the difficult L5–S1 disc.

PATIENT PREP, SEDATION, AND ANESTHESIA

The patient should be prepped and draped in sterile fashion. Both sides of the patient’s back around the line marked through the inferior endplate should be cleaned with three passes of a betadine sponge. Small under-drapes are placed to form a rectangular operative field so that there is an adequate window of access to both sides of the back. The larger over-drape contains a smaller, precut rectangle which should be positioned so that the marked approach is centered in the surgical field. Should a change in the side of approach become necessary, the wide placement of the under-drapes allows easy exposure of the opposite side by simply moving the over-drape. Prior to covering the patient with the over-drape, a grounding pad is placed on the thigh. The chosen site should be cleaned with alcohol to ensure good contact between skin and pad. Patients with abundant body hair may need to have the grounding site shaved. The head of the C-arm and the side of the table are also draped to prevent contamination of the surgical field with cross-table lateral positioning of the camera.

The patient is now given mild i.v. sedation. The choice of medications should be those the anesthesiologist is comfortable using and invariably include midazolam and, often, propofol. General anesthesia is contraindicated and not given under any circumstances. It is an essential safeguard of the procedure that the patient be able to self-monitor for and report signs of nerve root irritation. During the course of the procedure, i.v. fentanyl may be used for analgesia as needed.

The skin is anesthetized using a short-acting amide local anesthetic such as lidocaine. The bevel of a 27-gauge needle is inserted in the epidermis and approximately 0.5 cc of 1% lidocaine is injected to raise a small weal. Using the needle in a ‘down-the-tube’ approach following the angle of the camera to the target, several milliliters are used to anesthetize the dermis. A longer, 3.5″, 25-gauge needle can then be used to anesthetize the deeper fascia and muscle tissue down to the level of the superior articular process (SAP). Care must be taken in especially thin patients in whom the neuroforamen may be reached by a 2.5″ needle not to anesthetize the exiting nerve root or spinal nerve. This is avoided by touching down on the SAP as opposed to placing the needle tip lateral to it. Several milliliters of local anesthesia are injected with the needle at its full depth and several more may be used as it is slowly withdrawn. Clinical experience has shown that the application of generous amounts of local anesthetic (4–6 mL of 1% lidocaine) to the paraspinal tissues makes passage of the larger introducer needles much more comfortable, thereby reducing the need for excessive intravenous sedation.

PLACING THE INTRODUCER

The introducer comes in either a beveled or diamond-shaped tip. An arrow on the hub indicates the face of the bevel. The diamond-shaped tip (Gen II – Smith & Nephew/Oratec) was developed out of concern that the bevel tip (Gen I) increased the risk of spinal nerve damage (Fig. 26.6). Both are 17-gauge with a screw-in stylet and come in 6″ and 9″ sizes. The larger size is designed for use with an extra-long catheter.

Fig. 26.6 Photo of SpineCATH, top, with Gen II 17-gauge introducer needle and screw-in stylet, bottom.

Making a slight bend in the needle increases the ability to direct or ‘drive’ it into the optimum location in the disc. This is done by holding the introducer with a gauze 4 × 4 with the forefinger behind the tip and the thumb below it on the shaft. Pressing gently but firmly with the thumb will bend the tip slightly upwards. The direction of bend should be toward the arrow on the introducer hub. With the bend, the stylet should still be able to be removed without difficulty. A number 11 scalpel is used with a short stab over the skin mark to facilitate passing the introducer through the tough epidermis without undue force.

The introducer is also inserted along the angle of the camera with the bend facing the patient’s midline. The larger gauge of the introducer allows a greater feel of the tissues it passes through. The operator will be able to clearly identify dermis, subcutaneous fat, fascia, and muscle. Using the oblique fluoroscopic view as a guide, the introducer is aimed at the SAP. To avoid the exiting spinal nerve which passes laterally in oblique fashion across the superior endplate and disc, the introducer should hug the SAP’s ventral edge in order to stay medial to the nerve. To accomplish this, the needle is advanced until it gently touches down on bone. By withdrawing slightly, rotating 180° in a caudal direction away from the exiting nerve root so that the bend faces away from the spine, and readvancing, the tip can be made to slide off the ventral edge. Once it has cleared the SAP, the introducer is rotated back, again in a caudal direction, completing a turn around the facet joint (see Fig. 26.5). It is important to always rotate the needle in a caudal direction to minimize the chance of contact with the exiting nerve root.

Before advancing the introducer, the placement must be checked on an AP view. The tip should be seen lateral to the medial border of the inferior pedicle. Otherwise, there is danger of puncturing the thecal sac. If it does so, withdrawal and lateral redirection are necessary. A cross-table, lateral, view is checked next for cephalocaudad orientation. The alignment of the introducer should be parallel to the disc space. Any deviation from the midpoint between the two endplates is noted and corrected as the introducer is advanced using the lateral view. At this point in the procedure, the patient should be awake and alert for any symptoms of nerve root irritation. A 17-gauge needle can, obviously, be injurious to the anulus and exiting nerve root. Proper alignment of the introducer prior to entering the disc will minimize the chance of nerve injury as well as the need for multiple passes through the anulus.

As the tip contacts the posterior anulus, resistance to advancement will increase slightly, together with a ‘gummy’ feel indicating penetration into annular fibers. The patient may experience discomfort or pain in the form of deep localized axial back pain as the anulus is breached and this reaction can be taken as confirmatory. Reports of sharp leg pain below the knee indicates close proximity to the exiting spinal nerve. This is more likely to occur with superolateral positioning of the tip on its approach. The introducer should be withdrawn and redirected in a medial and/or inferior manner before further advancement. From this point, AP and lateral fluoroscopic views are checked frequently while advancing into the nucleus of the disc. Unlike standard discographic technique, the tip of the introducer should be placed slightly more anterior and off center so that it lies within the ipsilateral anterior quadrant of the disc. Optimal positioning will be seen as tip placement just lateral to the midline of the disc on AP and approximately three-fourths of the way across the diameter of the disc on lateral views (Fig. 26.7). This will allow for the passage of the catheter into the mid- to outer annular rings.

Fig. 26.7 AP and lateral fluoroscopic views of correct needle position within disc (A,B). Illustration showing correct positioning of the tip in the ipsilateral anterior quadrant of the disc (C).

Correct introducer placement is one of the keys to the IDET procedure. All subsequent steps in the procedure will be affected by it. Suboptimal positioning of the introducer leads to suboptimal catheter positioning. Taking the time to place the introducer correctly will be rewarded by increasing the chance of a successful outcome.

Placement at the L5–S1 level is complicated by the intervening iliac crest. Advancing the introducer between the crest, SAP, and inferior endplate requires a steeper cephalocaudal angle with lesser angulation away from the sagittal midline than at more superior levels. The net result is increased difficulty securing adequate medial positioning within the disc. Increasing the bend placed in the needle will help. Once the tip is past the SAP, the needle is rotated so that the bend curves medially. Marking the skin as lateral from the SAP as the confines of the radiolucent triangle will also facilitate greater medial placement. In addition, the steepness of the approach limits the window for passage into the outer anulus between the endplates. Osteophytic spurs of the endplates or hypertrophy of the ring apophysis have greater potential to obstruct advancement. Again, the bend in the needle is used to navigate around them. Often, rotation to take advantage of the bend along with leverage applied to the heavy-gauge needle will enable the tip to glance off a bony prominence. Make sure the bevel opening of the needle is facing posteriorly within the plane of the disc by turning the arrow on the needle hub. Once optimal positioning is achieved, unscrew and remove the stylet in preparation for placing the catheter.

THREADING THE CATHETER

The SpineCath (Smith & Nephew/Oratec) comes in two sizes (SpineCATH, SpineCATH XL). Both have working distal lengths of 5″. The proximal portion is either 6.8″ or 9.8″ for overall lengths of 11.8″ (30 cm) for the SpineCATH and 14.8″ chest (37.5 cm) for the SpineCATH XL. The SpineCATH XL used with the corresponding Gen II XL introducer can be very helpful in larger patients. The longer catheter and introducer are distinguished by their black hubs.

Both versions have a plastic-coated wire core which serves as the heating element and which is surrounded by insulation proximal to the distal heating coil. The width of the working portion is 0.038″ (approximately 1 mm). The distal heating coil is marked by radiopaque marks 2.2″ (5.5 cm) apart. Prior to performing the heating protocol, both marks must be clearly seen on the fluoroscope distal to the end of the introducer to avoid heating the introducer and damaging the tissue through which it passes. There are hash marks on the proximal portion of the catheter, approximately a centimeter apart, two of which are thicker than the others and mark the points at which the radiopaque marks emerge from the tip of the introducer. These hash marks should not be used, however, in lieu of fluoroscopic visualization to determine when the radiopaque marks of the heating coil are safely out of the introducer.

Before using the SpineCATH, it should be tested to ensure that it is functional. On the sterile field, connect the catheter to its cable so that the white stripe on the hub aligns with the white dot on the cable plug. There will be an audible click when the cable plug is correctly inserted into the catheter hub. The other end of the cable is handed off the sterile field and plugged into the generator. Make sure the generator is plugged in and the power is turned on using the On/Off switch on the rear panel. The thumbwheel on the rear panel needs to be set on 2 for SpineCATH procedures. (The 20S generator designed to replace the 50S has probe recognition and setting the thumbwheel is not needed [Fig. 26.8]). Check that the Actual Temperature is recording the temperature of room air (18–26°). The impedance should be between 85 and 230 ohms. Disconnect the catheter from its cable before placing it in the introducer.

Fig. 26.8 The 20S Electrothermal Generator with catheter attached.

Both versions of the SpineCATH have an angled tip to facilitate driving within the disc. The direction of the angle corresponds to a white stripe on the long axis of the hub. The hub may be used to facilitate rotation of the catheter tip. It should not be used to advance the catheter. Holding the catheter by the proximal shaft when advancing limits the amount of traction that may be developed and decreases the incidence of kinking.

The catheter is inserted into the introducer with the white stripe facing the patient’s spine. Initial AP views are best to show the catheter advancing towards the contralateral anterior anulus before turning posteriorly with the contour of the disc. It is important to make sure that the tip of the catheter exits the introducer in a manner that is parallel to the disc space (Fig. 26.9). Once tracks are made in the anulus, the catheter will follow the path of least resistance so that redirecting the catheter that is heading towards the endplate will be difficult. Any deviation of the catheter in a cephalic or caudal direction may be adjusted with rotation of the white stripe on the hub in the appropriate direction.

Fig. 26.9 AP fluoroscopic view of catheter advancement. Proximal radiopaque mark is still within the introducer (A) and exited (B).

Advancing the catheter is a slow, gentle process accomplished in 1 cm increments with repeated fluoroscopic images to check the progress. As the catheter passes the anterolateral corner, the direction of advance will be more or less ‘head-on’ into the camera on an AP view. Switching to a cross-table lateral view will be more helpful in showing the progress of the catheter through the disc. Similarly, returning to an AP view as the catheter passes the posterolateral corner will provide a better view of catheter advancement. With greater experience, the catheter may be safely advanced mostly by feel, reducing the number of camera angles needed. Otherwise, it is important to maintain a clear view of the catheter tip as it moves through the disc.

Again, holding the catheter by the proximal shaft as it is advanced will increase sensitivity to changes in resistance to catheter passage and reduce the incidence of kinking when obstacles are encountered. A kink is a permanent bend in the shaft of the catheter (see Fig. 26.15A below). Kinks result from continued force against resistance preventing catheter advancement. Any change in resistance should prompt a fluoroscopic view in several planes to adequately assess the advancement and position of the tip. Bowing of the catheter may occur parallel to the needle and is poorly visualized on a single plane. Typical locations of increased resistance within the disc include the posterolateral corner, the posterior midline, and sites of annular tears. Loss of resistance usually indicates escape of the catheter tip into the extra-annular or epidural space (Fig. 26.10). This occurs in excessively degenerated discs where the general integrity of the anulus is compromised. Focal annular tears can serve as paths of least resistance and may redirect the catheter along the lesion into inappropriate areas. Obstacles include the superior and inferior endplates, the anterior and posterior longitudinal ligaments, and scar tissue within the disc.

Fig. 26.15 Case history 2. Left side approach. AP fluoroscopic (A) view showing catheter tip at posterior midline with a kink in the catheter. Lateral fluoroscopic view (B) showing catheter tip extending into extra-annular space. Note the absence of clearly visible kinking on the lateral view.

Fig. 26.10 Lateral fluoroscopic view of catheter tip beyond posterior border of epidural space.

When meeting increased resistance, the initial strategy to avoid kinking the catheter will be a gentle rotation of the catheter back and forth a few degrees while advancing a few millimeters. This maneuver should be followed with a fluoroscopic view to check for tip advancement even if there is little resistance. Should there be no advancement or if continued resistance is encountered, needle repositioning will be required.

Repositioning the needle involves withdrawing the catheter, rotating the tip, and readvancing. Unfortunately, the ability to drive the catheter within the disc is limited due to the confined space inside the disc, disc anatomy, and equipment design. The blunt catheter tip has limited ability to puncture the anulus and most often follows the nuclear/annular interface. The tougher anulus deflects and serves to guide the catheter along this interface. Where the anulus is discontinuous due to defects such as radial tears, it loses this guide function, allowing the catheter, as mentioned, to track along paths of lesser resistance. It is quite common, for example, for a radial tear to catch the catheter and direct it external to the anulus or into the PLL, especially when the approaching tip forms an obtuse angle with the tear.

The nature of the disc space limits the ability to drive around these points in cephalocaudal directions as it is most limited in this plane. Changing the catheter’s position in the AP plane is, generally, more productive. To accomplish this usually involves a considerable withdrawal of several centimeters or complete retraction of the tip back within the introducer followed by rotation and readvancement. Failing this, it may be necessary to change the point of attack of the tip on the annular/nuclear interface by repositioning the introducer within the disc. In general terms, retracting the introducer towards the center of the disc will cause the catheter to travel a smaller diameter whereas advancing it away from the disc center will create a path of larger diameter. Despite these maneuvers, it may be impossible to correct the catheter position or bypass an obstacle.

Prior to abandoning an approach, if the catheter is judged to be safely positioned within a portion of the posterior anulus, a heating protocol may be performed. (Care must be taken to make sure that both radiopaque marks have exited the introducer.) The benefit of heating before trying a contralateral approach is twofold. First, this portion of the anulus may not be accessible from the contralateral side and, second, following heating, the surgeon may find that the catheter can be successfully advanced, thus obviating the need for the contralateral approach.

Several unsuccessful attempts to traverse such an obstacle should prompt consideration of a 180° change in the direction of the catheter’s advancement. Most commonly this is done by changing the side of approach to the disc. Having prepped the contralateral side prior to beginning the procedure will facilitate this change by simply moving the sterile drapes to frame the new operative field. Locate the new entry site by measuring a similar distance from the midline as that of the initial approach along the horizontal line through the inferior endplate drawn at the beginning of the procedure. The introducer is guided through the anulus on the opposite side and into the nucleus. As mentioned, the camera will come between the physician and the operative field when obliqued to view the extrapedicular approach. Moving the camera in and out to allow needle adjustments may be necessary. The catheter is then inserted and advanced in the opposite direction along the posterior anulus.

Although changing the side of approach is the most common method, there are case reports of the so-called ‘pig-tail’ technique in which a 180° redirection of the catheter is done without moving the introducer.¹¹ The catheter is made to turn in a sharply acute angle when exiting the introducer so that it doubles back and advances in the opposite direction. The sharp turn out of the introducer gives a curled appearance reminiscent of a pig’s tail, prompting the name. This technique has the advantage of preserving a unilateral approach and avoiding the necessity of a double puncture of the posterior anulus. The acute angle between the catheter and introducer may, however, create a problem with catheter removal with increased risk for catheter shearing. Concern over these potential complications limit the use of the technique, especially in inexperienced hands.

Optimal placement of the catheter is obtained when its position is deemed adequate on two planes of view, the AP and lateral. A specialized view that somewhat replicates the axial view of the disc seen on a standard CT of the spine is obtained with extreme caudal tilt of the fluoroscopy camera. The truest axial views are seen at the lowest lumbar segments with this technique. Known as the ‘CT view,’ it is helpful in defining the extent of circumferential coverage of the anulus. The heating coil, bounded by the distal and proximal radiopaque marks, should be seen to traverse the posterior anulus and its two posterolateral corners (Fig. 26.11).

Fig. 26.11 AP (A) and CT scan (B) fluoroscopic views of intradiscal catheter placement at L4–5 showing optimal coverage of the posterior anulus on CT view (B).

Postdiscography CTs of the target disc are helpful when assessing catheter position and should be available for review during the procedure. Lateral fluoroscopic views may show the catheter to be dorsal to the posterior edge of the vertebral bodies, raising concern for epidural placement. Lateral CTs, by showing the extent of disc bulging, can confirm safe placement within the disc. X-ray exposures of the cross-table fluoroscopic view can help to assess the location of a poorly visualized catheter and rule out placement in the epidural space (Fig. 26.12). Axial CTs may be consulted when considering placement across radial tears, especially in the hard to reach far lateral corners of the posterior anulus. If the heating coil of the catheter cannot be placed to cover the entire posterior anulus, yet covers a significant portion, consideration should be given to performing the heating protocol with the catheter in place prior to withdrawing and using a bilateral approach.

Fig. 26.12 Lateral X-ray exposure view showing catheter tip beyond posterior border of vertebral body.

With the catheter in place, preparation is made for the delivery of thermal energy. The generator (model Ora-50™ S) is turned on. The AutoTemp mode is used if the generator is so equipped (AutoTemp label will be seen in the center of the generator faceplate). If not, the Temperature Control mode is used. The mode button is pressed once for AutoTemp mode and twice for Temperature Control mode. In the AutoTemp mode, the Set Power button will display P90, indicating a peak temperature of 90°C. Peak temperature can be adjusted to 80, 85, or 90 degrees (P80, P85, or P90) using the Set Power button. All the heating protocols start at 65°C and increase 1 degree every 30 seconds until the maximum temperature is reached. The maximum temperature is sustained for 4–6 minutes. Lower heat protocols reach the maximum temperature sooner and are shorter in duration. The P90 protocol lasts 17 minutes.

The cable is reattached to the hub of the catheter. It is important to check that both radiopaque marks are clearly outside the introducer. Check that impedance is in the normal range (85–230 ohms) indicating that the power circuit is complete. Impedance should be checked periodically during the procedure. Pressing the RF button once or, if using a foot pedal, stepping on the pedal once will initiate the delivery of energy and commence the heating protocol. The generator will emit an audible tone and a blue light on the panel will light when thermal energy is delivered. It is good practice to note that Actual Temperature readings coincide with Set Temperatures. Temperature may be adjusted using the Set Temperature buttons. Note that the temperature cannot exceed peak temperature.

During the procedure, the patient should be monitored for pain. It is extremely important that the patient be alert and able to self-monitor for signs of nerve root irritation during the procedure. General anesthesia is contraindicated. It is expected that the procedure will reproduce familiar axial back pain. Severe back pain or pain radiating into the lower extremity may indicate dangerously high temperatures in the outer wall of the anulus adjacent to the thecal sac and/or nerve root. Generally speaking, tolerable back pain with radiation into the buttock and thigh but not below the knee is of little concern. Pain radiating below the knee is a sign of nerve root or cauda equina injury and will require reducing the set temperature or repositioning the catheter. The patient should be instructed, when questioned, on the importance of making this distinction between above and below the knee prior to the procedure. Do not attempt to move the catheter in any direction while RF energy is being delivered. In the event of severe pain or signs of nerve root irritation, the heating protocol should be stopped immediately. Pressing the RF button or stepping on the pedal once will pause the delivery of energy. The catheter should be repositioned before resuming the procedure. If symptoms of nerve root irritation persist, repositioning by changing the side of approach can be trie d if not already attempted. Otherwise, the procedure should be abandoned.

At the conclusion of the heating protocol, the generator will cease emitting sound and the blue light will go out, indicating cessation of energy delivery. The cable is disconnected from the catheter hub. First the catheter is removed, followed by the introducer. The technique for removal is a back and forth rotation of the catheter as it is retracted through the introducer (‘a lot of wiggle with a little pull’). Resistance to retraction of the catheter may indicate the needle biting into the catheter. To avoid shearing the catheter, it should be advanced a centimeter, rotated, and then gently withdrawn. Continuing resistance will require removing the catheter and introducer as a single unit.

The special case of a catheter kink causes concern due to weakening of the catheter at the kinked section. Further advancement or manipulation of the kinked catheter is ill-advised. Careful withdrawal of the catheter from the disc is indicated. Prior to withdrawal, a heating protocol may be performed without further damage to the integrity of the catheter if it is deemed to be in an acceptable treatment position. When treating with a kinked catheter, it is necessary to verify catheter function by confirming normal ranges for Impedence and Actual Temperature readings.

To withdraw a kinked catheter, the introducer tip should be carefully retracted a few millimeters while ensuring that the tip remains in the disc. This is followed by cautious withdrawal of the catheter. Stop immediately if any resistance is felt. If there is resistance on withdrawal, the introducer and catheter should be withdrawn as a unit. Using excessive force on withdrawal or continued manipulation may sever the catheter within the disc.

At the conclusion of the procedure, sterile compression dressings should be placed over the entry site. The patient is discharged home accompanied by a friend or family member following a period of postprocedure observation.

POSTPROCEDURE

Many patients will experience an increase in their typical back pain following the procedure. NSAIDs and mild opiates (hydrocodone) are used for symptomatic relief. This procedure-related discomfort subsides over the course of several days to a week with most patients returning to their typical level of symptoms by the end of the first or second week. Generally, symptom improvement is gradual and occurs over the course of 3–5 months. Preparing the patient for an extended recovery period prior to the procedure will help the patient to remain calm and optimistic.

A brief period (1–3 days) of bed rest is followed by relative rest in which vertical sitting or walking is limited to 10–20 minutes. Driving is not allowed in the first 48 hours. Resumption of driving begins with short rides of 20–30 minutes for the first 6 weeks. Return to work in 1 week with these restrictions is permitted for sedentary work.

Recovery typically takes 6–12 weeks during which time the activity of the patient must be restricted to maximize healing in the treated disc. A corset-type lumbosacral orthosis is worn during the day for approximately the first 6 weeks to remind the patient of proper body mechanics and to limit activities and range of movement (ROM). The patient should be instructed to avoid back flexion, extension, and rotation. Lifting is limited to 10 pounds. No sports activities including running, biking, golfing, tennis, skiing, etc. are allowed in the first month.

Physical therapy is not usually started before 6–8 weeks. Daily exercises initially include hamstring and piriformis stretching. Pulling the knees to the chest is done with each knee in turn as well as both together. If the patient tolerates extension and prone lying, butt/glut squeezes are performed prone over a pillow. Progression to stabilization exercises includes proper pelvic bracing while performing ‘dying bug’ and ‘superman’ routines, abdominal curls, and wall slides. Patients are encouraged to begin a graduated walking program. Backstroke-only swimming and standard stationary bike riding are also permitted. The patient is graduated to advanced stabilization exercises for the back and progressive resistance exercises for the limbs 3–4 months postoperatively. Resumption of athletic pursuits begins 4–6 months postoperatively and is graduated on an individual basis.

Case history 2

A 47-year-old man presented with a 7-year history of persistent left-sided axial back pain without radiation to the legs aggravated by lifting, prolonged sitting, and standing. His symptoms were unresponsive to therapeutic exercise or injections. He received only minimal relief from a combination of NSAIDs and narcotics. An MRI showed Grade I L5–S1 spondylolisthesis with high-intensity zones at the L3–4 and L4–5 discs. Reproduction of concordant pain at low pressure and volume was found at L3–4. Postdiscography CT scan revealed a left posterolateral tear extending to the outer anulus.

Proper introducer needle placement was obtained using a standard extrapedicular approach from the right side. The catheter was passed circumferentially into the disc with a sudden loss of resistance after passing the left posterolateral corner. The tip of the catheter was clearly within the epidural space on lateral views, confirming the impression of catheter escape through a radial tear. Retraction and redirection of the catheter was unsuccessful in traversing the tear. The catheter was then reinserted after retraction of the introducer in an effort to bypass the tear by finding a more interior line of passage within the disc. This maneuver produced the same result.

Finally, the catheter was retracted so that the tip was just contained within the posterolateral corner of the disc. The proximal radiopaque mark was concealed within the introducer in this position. The introducer was then withdrawn sufficiently to clearly reveal both marks within the disc (Fig. 26.14). A medium heating protocol was then performed beginning at 65°C and progressing to a maximum of 85°C.

Fig. 26.14 Case history 2. Right side approach. AP and lateral (A,B, respectively) fluoroscopic views showing retraction of the introducer needle and both radiopaque marks of the catheter within the disc.

It was determined that a less than optimal coverage of the annular tear had been obtained from the right side and an approach from the left side was attempted. The catheter was threaded as far as the posterior midline before further advancement was met with insurmountable resistance. Repeated attempts to advance past the midline resulted in kinking of the catheter (Fig. 26.15A). Lateral fluoroscopic views appeared to show the catheter positioned just beyond the posterior border of the vertebral body (Fig. 26.15B). A heating protocol was begun cautiously. Within the first minute, the patient experienced sharp back and foot pain. The protocol was stopped and the procedure was discontinued. The catheter was removed via the introducer without incident.

Following the procedure the patient reported no symptoms in the lower extremity and was discharged home after a stable neurologic exam. One month postprocedure, the patient reported a decrease in visual numeric back pain score from 8 to 2, a decrease in Roland Morris Disability score from 18 to 3, with increases in sitting, standing, and walking tolerances. He no longer required medications for pain symptoms.