Radiofrequency Denervation of the Lumbar Zygapophysial Joints Targeting the Best Practice

Transcription

1 Blackwell Publishing IncMalden, USAPMEPain Medicine American Academy of Pain Medicine? RESEARCH ARTICLESTechnique of Lumbar Zygapophysial Joint Radiofrequency DenervationGofeld and Faclier PAIN MEDICINE Volume 9 Number Radiofrequency Denervation of the Lumbar Zygapophysial Joints Targeting the Best Practice Michael Gofeld, MD, FIPP, and Gil Faclier, MD, FRCPC Pain Management Program, Department of Anesthesia, Sunnybrook Health Sciences Center, Toronto, Ontario, Canada ABSTRACT ABSTRACT Objective. Radiofrequency denervation of the zygapophysial (facet) joint is a frequently performed procedure for chronic low back pain. Several studies have shown considerable efficacy of the procedure, but none of the randomized controlled trials performed to date has used anatomically correct placement of radiofrequency cannula parallel to the target nerve. Three main techniques have been utilized for many years: North American, European, and Australian. Each has conceptual and technical limitations. This review analyzes these techniques and proposes a standardized technique of radiofrequency denervation of lumbar zygapophysial joints. Design. Current techniques of radiofrequency neurotomy were analyzed with respect to anatomic and technical accuracy. Step by step guidelines were developed using a combination of previously described techniques along with newly elaborated technical hints and details. Conclusion. We believe that the technique using tunnel vision with anatomically appropriate cannula placement and use of a large-bore, curved needle with a 10-mm active tip may improve the results of radiofrequency denervation of lumbar zygapophysial joints. Providing a detailed description of an anatomically accurate technique should be of value to those seeking to perform this procedure in a valid manner. Key Words. Low Back Pain; Zygapophysial Joint; Radiofrequency; Denervation; Neurotomy Introduction R adiofrequency denervation of the zygapophysial or facet joint is a frequently performed procedure for chronic low back pain. Numerous small studies of the procedure have been reported, with widely varying but mostly positive results. These observational studies prompted randomized controlled trials to assess the efficacy of the procedure, four of which have been published to date [1 4]. Unfortunately, they all contained methodological and technical flaws and therefore failed to clarify the efficacy of radiofrequency neurotomy for the treatment of zygapophysial pain. None of these studies used comparative or placebo-controlled blocks, nor did they use an Reprint requests to: Michael Gofeld, FIPP, 2075 Bayview Avenue, Toronto, Ontario, Canada M4N 3M5. Tel: ; Fax: ; michael.gofeld@ sunnybrook.ca. anatomically validated technique (i.e., placement of the cannula parallel to the target nerve [5,6]). A recent review addressed these issues but focussed mainly on the selection of patients for treatment [7]. An accompanying editorial elaborated the issue of diagnosis [8], but neither the review nor the editorial fully explored the problem of correct technique for lumbar radiofrequency neurotomy. In only one study [9] of those included in the review was the correct anatomic technique applied for radiofrequency denervation of zygapophysial joints. However, this was a nonrandomized cohort study with a small number of patients available for follow-up. The nerve supply of the lumbar facet joints has been described in detail elsewhere [5,10]. This anatomy predicates the accuracy of radiofrequency neurotomy. Techniques can be considered anatomically correct and methodologically acceptable only if radiofrequency cannulae are American Academy of Pain Medicine /08/$15.00/ doi: /j x

2 Technique of Lumbar Zygapophysial Joint Radiofrequency Denervation 205 Table 1 Major differences in the three accepted techniques of radiofrequency denervation of lumbar zygapophysial joints Cannula Fluoroscopic view Position of the tip to nerve Skin-bone distance Learning curve European 21,22-gauge, 5 mm active tip Oblique Perpendicular or at an angle Short Steep North American 10-mm curved tip; size varies Posterolateral Parallel Intermediate Long Australian 16-gauge Ray Decline Parallel Long Intermediate placed adjacent and parallel to the target nerve. In this regard, Bogduk and colleagues [5,10] described the anatomy of only the lumbar medial branches and the fifth lumbar dorsal ramus. Others have proposed that ascending branches of the first sacral (S1) dorsal ramus also contribute to the innervation of the L5 S1 zygapophysial joint [11]. The present article provides an analysis of the respective merits and disadvantages of the several ways in which lumbar radiofrequency neurotomy is currently practiced. It draws on the available anatomic and technical data with the view to defining a standardized technique for this procedure. The North American technique uses tangential insertion of a curve-tipped cannula in parallel with the nerves [13] (Figure 2). It allows longitudinal contact between the cannula and the nerve, which results in a larger lesion. Rotation of the cannula results in a further increase in lesion size. Notwithstanding its purported success, this approach has one major drawback. Because the cannula is placed under uniplanar AP fluoroscopic view, successful placement of the cannula depends on the operator s feeling for three-dimensional depth and direction. There is a long learning curve to excel in this technique, and results vary between operators. Methods Analysis of Radiofrequency Techniques Three radiofrequency techniques have been in use over the past three decades. We refer to them by name of continents where each was originally described: European, North American, and Australian. The major differences between them are highlighted in Table 1. The European technique relies on the so-called Scotty dog radiological appearance of the lumbar vertebrae and use of the eye of the Scotty dog as a target point. A detailed manual for this technique was recently published [12]. Although popular and easily learned, this technique suffers from two technical flaws. The first is inadequate exposure of the cannula tip to the target nerve. The straight cannula tip is placed perpendicular or at an angle to the target point, at the base of the transverse process or at the groove on the sacral ala, using tunnel vision, in an oblique view of the target point, some 10 to 20 from the true radiological anteroposterior (AP) projection. Consequently, only the cannula tip makes contact with the target nerve, if any contact is made at all (Figure 1). The second flaw is that the recommended 22-gauge cannula with a 5-mm active tip generates a lesion that may be too small to ablate the nerve sufficiently even if placed parallel to the nerve. A B Figure 1 European technique. (A) Anteroposterior view shows cannulae positioned for radiofrequency denervation of the third and fourth lumbar medial branches and the fifth lumbar dorsal ramus. (B) Lateral view shows cannulae making perpendicular contact with the target nerves (shown in black).

3 206 Gofeld and Faclier Some authors contend that the S1 dorsal ramus contributes to the innervation of the L5 S1 zygapophysial joint [11]. A technique for coagulating this nerve has been described [15], but no study has been published that included this step in the treatment of zygapophysial joint pain. Figure 2 North American technique. Fluoroscopic posteroanterior view showing position of the cannulae. The Australian technique uses an anatomically based approach to the nerve supply of the zygapophysial joints. This method uses the concept of a declined view, in which a steep caudocephalad axial tilt of the fluoroscopy beam, along with a 20 lateral tilt, allows the cannula to hug the anterolateral aspect of the base of the superior articular process (SAP). This technique has been described in detail recently [14]. Although the cannula can be positioned precisely with this method, two potential disadvantages exist. First, because of the rather unusual appearance of vertebral structures with this view, a spinal needle must be inserted first, as for medial branch block. This needle is needed to guide the placement of the cannula. Second, the steep caudocephalad declination, coupled with lateral rotation, may result in a long distance to the target site that cannot be reached by commonly used 10-cm cannulae. This technique may require the use of a 14.5-cm cannula. Nevertheless, Dreyfuss et al. [9] used an adaptation of this technique and reported substantial and lasting effect. The cannula used (16 gauge, 5-mm active tip) was significantly larger than conventional cannulae, and a second lesion was executed after the cannula was withdrawn by 4 5 mm. The combination of strict selection criteria, positioning of the cannula tip parallel to the target nerve, and large lesion size ultimately resulted in impressive and lasting pain relief. Lau et al. [10] published additional data regarding the surgical anatomy and radiology of the Australian technique, but while they focused on radiological landmarks, the authors did not describe the technique of cannula placement in satisfactory details to be useful to inexperienced interventionalists. Analysis of Equipment Just as there is controversy about which of the different techniques is most appropriate, no consensus has been reached about the size of cannula and length of the active tip that should be used for lumbar facet neurotomy. A large-bore cannula with a longer active tip is considered the best choice to produce a large lesion, which should result in greater success of denervation. However, no further increase in lesion size was seen with an increase in diameter from 18-gauge to 15-gauge [16]. A curved cannula may be more beneficial for two reasons: the tip of the cannula can be positioned closer to the target nerve, hugging the base of the SAP; and rotating the tip of a curved cannula will further increase the size of lesion [17]. Despite these findings, a straight 22-gauge cannula with a 5-mm active tip remains the tool most commonly used for radiofrequency denervation of the facet nerves. Over a period of more than two decades, a vast amount of data about techniques and equipment and the results of many clinical trials have accumulated in the medical literature. Although several of these reports were descriptive (rather than experimental) and although the technique was performed incorrectly in some of the studies, the number of publications is impressive, and the method became widely acceptable in the clinical practice. In his review, Slipman [18] called for a prospective, randomized controlled trial that would use a standardized treatment, uniform outcome measures, and have an adequate follow-up duration of at least 12 months. The standard procedure must implement large-bore cannula placed correctly parallel to the target nerve [8]. Critical analyses of the techniques and knowledge of surgical anatomy and physics of radiofrequency have made it possible to develop the technical guidelines for successful radiofrequency denervation of the lumbar zygapophysial joints. We believe that a modified Australian technique, using the sublaminar oblique tunnel view and allowing anatomically correct placement (i.e., parallel to the path of the target nerve) of a largebore, curved cannula with 10-mm active tip, may improve the results of radiofrequency denervation

4 Technique of Lumbar Zygapophysial Joint Radiofrequency Denervation 207 Technique The neuroanatomy of the zygapophysial joints and the final position of cannulae for the lower lumbar medial branches and the L5 and S1 dorsal rami are shown in Figure 3. Short-acting intravenous agents (fentanyl, midazolam) are usually administered, supplementing local anesthesia of the skin and underlying tissues. A local anesthetic is injected before the lesioning. A local anesthetic is also useful to decrease electrical tissue impedance and thus homogeneous oval-shaped lesion is created. We recommend an 18-gauge cannula with a 10- mm curved active tip (Baylis Medical Company Inc, Montréal, Quebec, Canada). Two 75-second cycles at 80 C are applied. Rotating the tip after completion of the first cycle increases lesion size substantially (Figure 4). The duration of 75 seconds was derived from observations made during generation of radiofrequency lesions using the Pain Management Generator (PMG, Baylis Medical Company). The cannula temperature reaches the appropriate level after 15 seconds, and an additional 60 seconds is required to generate the lesion. Therefore, the total time at 80 C is 120 seconds, the period previously recommended to obtain sufficient lesion size [14]. Similarly, although lesion size depends on temperature, raising the temperature to 90 C did not result in larger lesions [14]; hence, a temperature of 80 C seems sufficient. Technique for Neurotomy of S1 Dorsal Ramus For denervation of the S1 dorsal ramus, a true AP fluoroscopic image of the lower lumbar spine must be obtained. The image intensifier is rotated cephalo-caudad until no double contour of the L5 S1 endplates can be seen (Figure 5). Figure 3 The third and fourth medial branches and the fifth lumbar and the first sacral vertebra (S1) dorsal rami with cannulae placed parallel to the nerves (S1 inset). of zygapophysial joints and serve as a methodologically correct technique for further clinical trials. Figure 4 Lesion area is increased by rotating the tip of the cannula (18-gauge, 100-mm cannula with 10-mm curved active tip) 180, as shown on a chicken breast model. Lesion size is enlarged by 2 mm when the cannula size is increased from 20 gauge (A) to 18 gauge (B). Figure 5 Diagrammatic representation of fluoroscopy angles and target site of the cannula (inset) for denervation of the first sacral dorsal ramus.

5 208 Gofeld and Faclier Figure 6 Anteroposterior view of the lumbosacral segment. White dotted line = upper border of anterior foramen; white curved line = lower border of posterior foramen; arrows = upper border of posterior foramen (rarely seen). The posterior S1 sacral foramen will then be visible. Some degree of medial or lateral oblique rotation is usually required to obtain a clear view. Two semilunar lines, similar to the shape of an eye, can be usually seen: the upper convex line is the upper border of the anterior foramen, and the smaller, lower concave line is the lower border of the posterior foramen (Figure 6). The upper border of the posterior foramen is rarely seen. It can be visually reconstructed by drawing on the skin using a metal pointer and permanent marker. The skin of the low back and the sacral area is prepared and draped using usual sterile technique. A skin weal is created and soft-tissue anesthesia is applied through a 25-gauge spinal or hypodermic needle. Special care should be taken not to instill local anesthetic into the foramen. The 18-gauge cannula with 10-mm curved activetip is inserted with the tip facing cephalad, toward the edge of the posterior S1 foramen, at the 12 o clock position, until contact with bone is made; this procedure is performed with the aid of tunnel view fluoroscopy. The cannula is then rotated caudad and advanced into the outer portion of the foramen. Finally, the tip is rotated cephalad, with retention of contact with bone (Figure 7). A lateral view will show the cannula tip positioned at the roof of the sacral canal but not entering the canal (Figure 8). Before commencing radiofrequency generation of the lesion, the tip is rotated laterally 45 from midline. Two cycles are applied at 80 C for 75 seconds, with 90 medial rotation of the tip between cycles, to increase the size of the lesion. Figure 7 The cannula is inserted into the left first sacral vertebra foramen with the tip facing cephalad; this procedure is performed with tunnel vision fluoroscopy. Technique for Neurotomy of L5 Dorsal Ramus and First to Fourth Lumbar Medial Branches First, a lateral image of the lumbar spine is obtained. The scout image of the target segment is printed, and the angles between the upper border of the vertebra and the base of the SAP are measured (Figures 9, 10). This measurement will give the exact angle of caudocephalad axial rotation needed for correct positioning of the cannula in the sagittal plane. This angle is typically 20 to 40, depending on individual lordotic curvature. Next, a true AP view of the corresponding area should be found. The image intensifier is then rotated in the caudocephalad direction to the previously measured angle to achieve a sublaminar view. Insertion of the cannula parallel to the base of the SAP in the sagittal plane ensures placement along the course of the target nerve. The last Figure 8 The cannula is positioned in the outer portion of the first sacral vertebra foramen with the tip rotated cephalad. Arrowheads = cannula tip within sacral foramen.

6 Technique of Lumbar Zygapophysial Joint Radiofrequency Denervation 209 Figure 11 Sublaminar and slightly oblique view clearly shows the junction of the superior articular processes with the corresponding transverse processes of the fourth and fifth lumbar vertebrae and the sacral ala (white dots). Figure 9 Line drawing of angles between superior endplates of the fifth lumbar (L5) and first sacral (S1) vertebrae and corresponding base of superior articular processes (SAP) of L5 and S1. The angle denominates the degree of caudocephalad tilt. 1 = L5 SAP; 2 = S1 SAP; solid lines drawn parallel to vertebral endplate; dashed lines drawn parallel to base of the SAPs; arrowed curve = angle between 1 and 2. alignment is a lateral rotation. Dynamic fluoroscopy is used to rotate the image ipsilaterally by 10 to 20 until the junction of the superior proximal edge of the transverse process and the SAP is visualized (Figure 11). This fluoroscopy view can be termed the sublaminar oblique view. The iliac crest may interfere with the visualization of L5 level if excessive lateral tilt is adopted. Therefore, the image intensifier is rotated just until the lateral edge of S1 SAP is clearly visible. The degree of obliquity depends on the extent to which the iliac crest obscures the target point. This angle will vary between 5 and 10. The schematic presentation is shown in Figure 12. Figure 10 Radiographic lateral view of lumbar spine. Prone positioning with a pillow usually partially alleviates lumbar lordotic curvature. Black lines represent the transverse planes of the superior endplates of the vertebrae. White lines are plotted through the bases of the superior articular processes (SAP) fourth lumbar and first sacral vertebrae. Radiofrequency cannula is placed according to the previously measured angle parallel to the base of the fifth lumbar vertebra SAP. Figure 12 Diagram showing patient s position and cannulae placement for radiofrequency neurotomy of the third (L3) and fourth medial branches and the fifth dorsal ramus.

7 210 Gofeld and Faclier Figure 13 Lateral view of two cannulae positioned at the base of the fifth lumbar (L5) and first sacral vertebrae superior articular processes. White lines represent the fourth lumbar medial branch and the dorsal ramus of L5. When the target point is identified, skin marks are made. The skin is prepared and draped using sterile technique. Local anesthetic is administered through a 25-gauge spinal needle to diminish pain from insertion of the large-bore cannula. Cannulae are inserted using the tunnel view technique (i.e., the target point, the tip, and the hub of the cannula should appear as a dot ). The cannula, with the tip facing caudad, is inserted until contact is made with bone. The tip of the cannula is rotated cephalad and advanced a further 3 5 mm until contact with bone is lost. The cannula is then rotated caudad and mesiad to expose the full length of the cannula tip to the target site. To confirm proper placement, lateral fluoroscopy is performed. A true lateral view is obtained when the bony landmarks are superimposed, eliminating the double contour appearance of vertebral bodies. In that view the cannula tip should reach the anterior two-thirds of the base of the SAP. This position ensures maximal contact between the active tip of the cannula and the target nerve. Deeper positioning may cause inadvertent contact with an exiting dorsal root ganglion or ventral ramus. A more posterior position means that the active tip is lying on the mammillo-accessory ligament and cannot produce a substantial lesion of the nerve (Figure 13). If the procedure is to be successful, the lesion should be created between the posterior border of the intervertebral foramen and the mammilloaccessory ligament, where the target nerves hug the base of the SAP. The cannula position should be confirmed in the AP view. The active tip should make contact with the bone to ensure maximal exposure to the target nerve. The cannula must be positioned snugly against the base of the SAPs (Figure 14). Subsequent steps in the procedure are similar to those for denervation of the S1 dorsal ramus, described above. The first lesion is created with the cannula tip rotated caudally, following the curvature of the bone. A second lesion is generated after dorsal rotation of 90, moving up the SAP, to increase the size of the overall lesion. This rotation is warranted because variations in the path of the nerve occur in the caudocephalad direction on the surface of the SAP. Conclusions We use well-known anatomic principles in radiofrequency denervation of the lumbar zygapophysial joint. With modification of anatomically correct and logical Australian technique and use of the suitable cannula, this technique can be recommended for use in a busy clinical practice. The sublaminar oblique view is easily accomplished and clearly recognizable even by an inexperienced operator. Clinical audit of this technique in our department has shown similar outcomes relative to the ones that used to be standard North American approach. In addition, trainees and visiting physicians have found it easy to master the technique. Figure 14 Anteroposterior view of two cannulae positioned at the base of the fifth lumbar and first sacral vertebrae superior articular processes.

8 Technique of Lumbar Zygapophysial Joint Radiofrequency Denervation 211 References 1 van Kleef M, Barendse GA, Kessels A, et al. Randomized trial of radiofrequency lumbar facet denervation for chronic low back pain. Spine 1999;24: Gallagher J, Petriccione di Vadi PL, Wedley JR, et al. Radiofrequency facet joint denervation in the treatment of low back pain: A prospective controlled double-blind study to assess its efficacy. Pain Clin 1994;7: Leclaire R, Fortin L, Lambert R, Bergeron YM, Rossignol M. Radiofrequency facet joint denervation in the treatment of low back pain: A placebocontrolled clinical trial to assess efficacy. Spine 2001;26: van Wijk RM, Geurts JW, Wynne HJ, et al. Radiofrequency denervation of lumbar facet joints in the treatment of chronic low back pain: A randomized, double-blind, sham lesion-controlled trial. Clin J Pain 2005;21: Bogduk N, Long DM. The anatomy of the so-called articular nerves and their relationship to facet denervation in the treatment of low-back pain. J Neurosurg 1979;51: Bogduk N, Long D. Percutaneous lumbar medial branch neurotomy: A modification of facet denervation. Spine 1980;5: Hooten WM, Martin DP, Huntoon MA. Radiofrequency neurotomy for low back pain: Evidencebased procedural guidelines. Pain Med 2005;6: Bogduk N. Diagnosing lumbar joint pain. Pain Med 2005;6: Dreyfuss P, Halbrook B, Pauza K, Joshi A, McLarty J, Bogduk N. Efficacy and validity of radiofrequency neurotomy for chronic lumbar joint pain. Spine 2000;25: Lau P, Mercer S, Govind J, Bogduk N. The surgical anatomy of lumbar medial branch neurotomy (facet denervation). Pain Med 2004;5: Paris S. Anatomy as related to function and pain. Orthop Clin North Am 1983;14: Sluijter ME. Radiofrequency. Part 1: The Lumbosacral Region, 1st edn. Meggen, Switzerland: FlivoPress SA; Kline MT, Yin W. Radiofrequency techniques in clinical practice. In: Waldman SD, ed. Interventional Pain Management. Philadelphia, PA: W.B. Saunders Company; 2001: Bogduk N, ed. ISIS Guidelines. Practice Guidelines for Spinal Diagnostic and Treatment Procedures. San Francisco, CA. International Spine Intervention Society; Ray CD. Percutaneous radio-frequency facet nerve blocks: Treatment of mechanical low back pain. In: Radionics Procedure Technique Series. Burlington, MA: Radionics; Goldberg SN, Gazelle GS, Dawson SL, et al. Tissue ablation with radiofrequency: Effect of probe size, gauge, duration, and temperature on lesion volume. Acad Radiol 1995;2: Racz G, Stanton-Hicks M. Lumbar and thoracic sympathetic radiofrequency lesioning in complex regional pain syndrome. Pain Pract 2002;2: Slipman CW, Bhat AL, Gilchrist RV, et al. A critical review of the evidence for the use of zygapophysial injections and radiofrequency denervation in the treatment of low back pain. Spine J 2003;3:310 6.