INTRODUCTION

Endoscopic disc surgery is evolving rapidly because of improvements in surgical technique, endoscope design, adjunctive surgical tools, and instrumentation. New endoscopes and complementary surgical devices enhance the endoscopic spine surgeon’s ability also to probe spinal anatomy in a conscious patient. The surgeon can then evolve his or her diagnostic and surgical skills with this newly found ability to evaluate pathologic, anatomic, and physiologic processes causing the patient’s pain. Now that diagnostic spinal endoscopy can be performed, conditions previously not even considered for surgery may be probed, evaluated, and surgically treated, and perhaps with greater accuracy. We believe that our understanding of discogenic back pain is enhanced by this ability to endoscopically visualize lesions not previously seen intradiscally and in the ‘hidden’ extra-foraminal zone.¹ Spinal endoscopy is poised to parallel the development and evolution of knee, shoulder, and ankle arthroscopy.² Without endoscopy, spine surgeons must depend heavily on imaging systems that, while extremely sensitive in identifying pathologic conditions, do not always correlate that condition with the patient’s pain.

Interventional pain management complementing endoscopic surgery

Foraminal epidurography and foraminal therapeutic injections can be performed using a needle technique that mimics the surgical approach to the foramen and spinal canal.³ The approach differs from traditional interventional pain management ‘down the tunnel’ approaches because it is the same far lateral approach used by endoscopic spinal surgeons to access the foramen for endoscopic surgery. The tip of the needle is directed into Kambin’s triangle between the exiting and traversing nerve root, rather than at the shoulder of the exiting nerve root. If non-ionic contrast agent (i.e. Isovue 300) is injected to outline the foramen, it should be able to identify the location and position of the traversing and exiting nerves that cross each spinal level (Fig. 133.1). Anomalous configurations such as conjoined nerves or furcal nerves provide additional information regarding the anatomy of the lumbar spine. In fact, these anomalies are rarely observed with magnetic resonance imaging (MRI) but are routinely visualized with spinal endoscopy in the ‘hidden zone’ or far lateral zone of the foramen.^1,18,19 Diagnostic and therapeutic information gleaned from these modified injection procedures can be used by the surgeon to better select patients for surgical intervention. Patients with symptomatic disc protrusions, annular tears, and foraminal stenosis may first be diagnosed by discography followed by foramino-epidurography, then get relief with the therapeutic foraminal injection. If the response is favorable but short lived, the diagnostic and therapeutic process will help the surgeon with patient selection. As a bonus, the epidurogram will help determine the ease and feasibility of endoscopic surgery because it allows for a trial needle placement into the foramen before surgical intervention.

Fig. 133.1 This foraminal epidurogram outlines the traversing nerve and exiting nerve. Needle placement is at the axilla of the exiting nerve root, between the exiting and traversing nerve root. The exiting nerve is partially obstructed by a foraminal osteophyte causing lateral recess stenosis.

INDICATIONS

In general, the indications for diagnostic and/or therapeutic endoscopy require a pathologic lesion that is accessible, visible, treatable, or requires endoscopic confirmation through the foramen.⁵ Limitations of the surgeon’s skill and/or experience with the endoscopy or difficult spinal anatomic are the primary contraindications, exclusive of comorbidities such as infection or uncontrolled coagulopathy.⁶ For herniations from T10 to L4, the foraminal approach provides excellent access to the disc and epidural space (Fig. 133.2). At L5–S1, anatomic restrictions may cause the surgeon to opt for the posterior transcanal approach. As surgeon experience increases, anatomy that previously precluded spinal endoscopy is no longer a barrier.⁷ Anatomic structures within reach of the spine endoscope transforaminally are illustrated in Figure 133.3.

Fig. 133.2 Anatomy of the posterolateral foraminal portal from L2 to S1. Only in the L5–S1disc space is access to the spinal canal restricted due to the pelvis and the relatively wide facet (gray hubbed needle in the L5–S1 disc). High lumbar disc herniations from L1 to L3 are easier to reach endoscopically through the posterolateral foraminal portal. L4–5 provides ample room for either approach. Note the furcal nerve branches entering the psoas muscle.

Fig. 133.3 The dome. Spinal structures in the foramen accessible to visualization and surgical intervention and probing via the posterolateral approach.

Disc herniation

Symptomatic disc herniation is the most common indication, with surgical decompression limited only by the accessibility of endoscopic instruments to the herniated fragment.^8,9 The posterolateral foraminal approach is ideal for a far lateral, extraforaminal disc herniation. Traditional surgical approaches to far lateral, extraforaminal disc herniations are more difficult, requiring a paramedian incision through the vascular intertransverse ligament. This surgical area is often termed the ‘hidden zone.’¹ Although the lateral zone of the disc can be accessed with a paramedian incision, in our view, it is easier to access the extraforaminal zone endoscopically via the posterolateral portal. A typical foraminal view of nucleus pulposus extruded past the posterior annulus is shown (Fig. 133.4). In our experience, this is the preferred approach for disc herniations in the upper lumbar and lower thoracic spine since the transcanal approach will require more extensive laminectomy that may destabilize the spinal segment if the herniation is above L3–4.

Fig. 133.4 Foraminal view of the decompressed traversing nerve after extraction of a large extruded subligamentous disc herniation through the posterior annulus at L5–S1. The indigo carmine dye stained the nucleus pulposus and disc fragment for easier identification and extraction. The fragment clearly extruded through the posterior annulus and was extracted from the ventral portion of the traversing nerve.

Discitis

Endoscopic excisional biopsy and disc space debridement is ideal for surgically debriding infectious discitis (Fig. 133.5).¹⁰ With this technique the surgeon will not have to be overly concerned about creating dead space for the inflamed or infected disc material to spread into the dead space created by a posterior approach. The clinical results are dramatic and, we believe, tissue biopsy is more accurate than needle aspiration in identifying the cause of discitis. In our experience, even sterile discitis will benefit from intradiscal debridement and irrigation.

Fig. 133.5 Intradiscal view of discitis after partial debridement. Inflammatory tissue and loose endplate cartilage are usually visualized and readily removed from the disc space with pituitary rongeurs and automated shavers. Pain relief is immediate, and abundant tissue is available for laboratory analysis. Discitis is often sterile, or nonsuppurative, even when sufficient inflammatory tissue is removed for culture. The incidence of discitis following endoscopic disc surgery is about 0.3%.

Lateral recess and central stenosis

Endoscopic foraminoplasty by endoscopic techniques can be accomplished by experienced endoscopic surgeons.^6,11,12 Although trephines, rasps, and burrs can be utilized, the Ho:YAG laser has enhanced the procedure technically, as laser is a very precise cutting tool for visually controlled soft tissue and bone ablation. Endoscopic laser foraminoplasty (ELF) has an inherent advantage over classic surgery as it will not produce further instability (Fig. 133.6).¹³ With endoscopy, the foramen can be enlarged up to 45.5% versus the 34.2% attainable with the standard posterior technique of removing only the medial one-third of the facet. As well, posterior decompression of the lamina with removal of the medial one-third of the facet will produce increased extension and axial rotation postoperatively.¹³ In comparison, endoscopic foraminoplasty has not been shown to cause increased instability, even in spondylolisthesis.¹² The technique is most useful for lateral recess stenosis. In central spinal stenosis, when there is concomitant posterior disc protrusion, decompression of the spinal canal can be effectively accomplished by resecting the bulging annulus in a collapsed disc, thus lowering the floor of the foramen without destabilizing the spinal segment. In isthmic spondylolisthesis, when there is more leg pain than back pain, this is usually due to impingement on the exiting nerve by the pars pseudoarthrosis defect. The goal in this instance is to decompress the compromised exiting nerve by elevating the dome formed by the undersurface of the superior articular facet and lamina without further destabilizing the spinal segment (Fig. 133.7).

Fig. 133.6 The technique of endoscopic foraminal decompression. The superior articular process can be decompressed by rotating the open face of the cannula against the base of the facet while the ventral half of the cannula protects the exiting nerve. The decompression continues cephalad toward the tip of the facet and the exiting nerve, resecting the ligamentum flavum attachment that can impinge on the axilla, causing symptomatic lateral recess stenosis. Just resecting the capsule and releasing the ligamentum flavum may be enough to enlarge the foramen and free the traversing as well as the exiting nerve. This foraminal endoscopic view at L5–S1 demonstrates how a diamond burr is used to decompress the ventral surface of the superior facet. The exiting nerve is at 9 o’clock.

Fig. 133.7 Illustration of side firing laser ablating bone under the superior articular process.

CURRENT IMAGING METHODS

In the authors’ experience, our imaging studies are only about 70% accurate and specific for predicting pain.^5,12,14–17 Conditions such as lateral annular tears, rim tears, endplate separation, small subligamentous disc herniations, anomalous nerves in the foramen, and miscellaneous discogenic conditions are cumulatively missed about 30% of the time. These conditions are diagnosable and often treatable with spinal endoscopy. Tears that are in the lateral and ventral aspect of the disc are sometimes missed by MRI studies. Very small disc herniations that protrude past the outer fibers of the annulus are also missed, since the fragment may be flattened against the posterior longitudinal ligament or nerve, appearing on the MRI as a thickened or bulged annulus, but really containing a subligamentous herniation. In rare circumstances, when the nerve root appears ‘swollen’ or enlarged, the MRI may not be capable of distinguishing a swollen nerve from a conjoined nerve or a nerve with an adherent fragment of disc. Endoscopy allows for direct visualization of such a ‘swollen root,’ thereby enhancing our ability to diagnose and subsequently treat this entity. We have observed that when the disc tissue is in direct contact with a nerve, the nerve can be irritated and a painful inflammatory membrane forms. When an inflammatory membrane is present, the pain pattern can be confusing. Diagnostic spinal endoscopy has confirmed ‘nondermatomal’ pain in scores of patients with proximal thigh, buttock, and groin pain at levels distal to the root origin of the anatomic area. Removal of the source of irritation can diminish or abate painful symptoms.

Future considerations

It is conceivable that the spine scope will eventually be used for all conditions where endoscopic visual inspection is desired. The authors have utilized spinal endoscopy to inspect a spinal nerve suspected to be irritated by orthopedic hardware adjacent to the pedicle, to remove suspected recurrent or residual disc herniations that do not show up on imaging studies, to decompress the lateral recess by endoscopic laser foraminoplasty, to remove osteophytes and facet cysts that cause unrelenting sciatica, and to locate painful lateral annular tears or small disc herniations not evident on physical examination or on MRI. Some of these correctable lesions are responsible for failed back surgery syndrome (FBSS), especially with recurrent disc herniations and with lateral recess stenosis. The lateral ‘hidden zone’ is rarely visualized by surgeons. It has been demonstrated that, with endoscopy, it is possible to do isolated disc and annulus surgery using a visualized thermal modulation procedure (Fig. 133.8), challenging the old concept that disc surgery is merely nerve decompressive surgery. For example, discogenic pain from annular tears is currently being evaluated and correlated with the pathoanatomic conditions visualized.^18,22–32

Fig. 133.8 (A) Annular tears. Grade V annular tears open into the epidural space, extraforaminal zone, or psoas muscle, sensitizing the nerve and forming an inflammatory membrane that allow the ingrowth of nerves and capillaries. An inflammatory membrane next to the dorsal root ganglion of the exiting nerve, psoas muscle, or traversing nerve can cause nondermatomal pain out of proportion to the MRI study. If patients with annular tears get relief from foraminal epidural blocks after a painful positive discogram, more lasting relief is probable with selective endoscopic discectomy and thermal annuloplasty. The procedure removes degenerative nuclear material keeping the tear open, thermally closes the tear, and allows for ablation of the inflammatory tissue. (B) Grade IV annular tear. Degenerative interpositional disc tissue is embedded in the annulus. The annulus is contracted by a bipolar radiofrequency flexible probe or laser, removing the interpositional disc tissue that is preventing the annular tear from healing. Intraoperative chromo-discography will stain the degenerative disc and annulus.

(C) This is intradiscal view of a large grade V annular tear. Inflammatory and granulation tissue surrounds the tear extending into the epidural space. The presence of inflammatory and granulation tissue is what makes annular tears painful. Ablation of the inflammatory membrane with radiofrequency may provide relief of varying periods, and may depend on the size and location of the tear. (D) Ellman bipolar radiofrequency trigger-flex probe thermal modulating a grade IV annular tear that has stained the outer fibers of the annulus. The more annular layers remain, the better the prognosis. If a radial tear or annular defect is large, thermal contraction sometimes causes the tear to enlarge, especially if there is an associated disc herniation. These large tears associated with large disc herniations do not create the chronic lumbar discogenic pain syndrome, which is still not completely understood.

TECHNIQUE

PROBLEMS AND COMPLICATIONS

As with arthroscopic knee surgery, the risks of serious complications or nerve injury are low, about 1–3% in the authors’ experience.³³ The usual risks of infection, nerve injury, dural tears, bleeding, and scar tissue formation are always present, as with any spinal surgery. Fenestration past the anterior annulus is a potential hazard, creating a bowel or vascular injury. Although this is a rare complication because the thickness of the anterior annulus will usually prevent fenestration, it must be recognized as a potential risk if the annulus is weakened or fenestrated by an anterior disc herniation. This risk is also present with the posterior approach. One limitation of the endoscopic technique is the need to use some instruments in a ‘blind’ fashion. This occurs because the size of shavers, pituitary rongeurs, and basket forceps are too large to fit into the working channel of the endoscope. Of course, their placement and use must be monitored with fluoroscopy. The cannulae are designed to protect vital structures by utilizing windows as surgical portals. Spinal nerves may be adherent to the disc and annulus, and can be extracted along with the disc or annulus by shavers or cutting instruments. In addition, the authors have identified anomalous autonomic and peripheral nerves in the foramen (furcal nerves), buried in the annular fat, that connect with the sacral plexus or the traversing nerve. These nerves are described in the medical literature and can be symptomatic but are typically small and not appreciated during open surgery. These furcal nerves are readily visualized endoscopically and probing them can produce pain in the awake patient. (Fig. 133.9). The inflammatory membrane may contain tiny nerves and blood vessels that contribute to severe discogenic pain (Fig. 133.10).

Fig. 133.9 Anomalous nerves (A) Furcal nerve. (B) Autonomic nerve. (C) Biopsy of autonomic nerve shown in (B). Anomalous nerves are seen in the annular fat in the ‘hidden’ foraminal zone. When found in the foramen, it is considered an anomalous branch, but furcal nerves are common branches from the exiting nerve entering the psoas muscle. These communicating branches are described as furcal nerves in the anatomy and literature. Small sympathetic nerves are occasionally seen. Ablation or resection of these nerves may be associated with dysesthesia, but removal may also decrease chronic discogenic lumbar pain.

Fig. 133.10 Neoangiogenesis and neoneurogenesis is commonly present in the inflammatory membrane adjacent to annular tears in patients who have severe discogenic pain and sciatica, but a rather benign MRI.

Dysesthesia, the most common postoperative complaint, occurs about 5–15% of the time, but is almost always transient. Its cause is still incompletely understood and may be related to manipulation of furcal nerves or the anulus. Another possibility is simply delayed nerve recovery as the symptoms typically occur days or weeks after surgery. Finally, this side effect may be due to irritation of the dorsal root ganglion from manipulation or postoperative bleeding. This condition cannot be completely avoided, as neuromonitoring with dermatomal SEP and continuous EMG, the most sensitive means of monitoring, has not identified intraoperative irritation as the major cause of dysesthesia.³⁴ In some cases the symptoms can be severe and similar to complex regional pain syndrome (CRPS), but usually without the skin changes that accompany CRPS. Stimulation of the dorsal root ganglion of the exiting spinal nerve can also result in dysesthesia when foraminoplasty is performed, even when the exiting nerve was clearly identified and protected during endoscopy.

The endoscopic technique, because of its approach, may be accompanied by risk for iatrogenic injury. We suspect spinal endoscopy is safer than traditional surgery, as the patient is awake and able to provide immediate input to the surgeon when pain is generated. Since the authors prefer patients to be reasonably alert, we never use propofol for ‘sedation.’ Neuromonitoring techniques can warn the surgeon of nerve irritation. About 66% of the time, there is EMG activity recorded that warns the surgeon that there is nerve irritation.³⁴ Neuromonitoring may give the surgeon more intraoperative feedback, but in a subsequent study comparing surgical cases without neuromonitoring, it has been demonstrated to be just as safe using of dilute local anesthetic and conscious sedation. As previously stated, the patient’s ability to feel pain provides the surgeon with a safety net. Consequently, the authors only use a dilute solution of local anesthetic (0.5% Xylocaine or its equivalent) to minimize the risk of deeply anesthetizing the spinal nerve. As well, the ability of the patient to report pain during the probing component of the procedure will enable the recognition of pain generators in the spine. When there is documented postoperative improvement in nerve conduction velocities and improvement of abnormal preoperative EMGs immediately postoperatively, this is clear, objective evidence of clinical improvement.³⁴ If a patient is not improved following surgery when electrodiagnostic studies show an improvement, it is likely due to the progression of the disease process itself.

NEW HORIZONS

There will soon be an explosion of new imaging systems, endoscopes, and endoscopic instruments. These new devices will enhance the precision of the endoscopic surgeon. As well, with more physicians performing these techniques there will necessarily be a refinement in the techniques used. This will culminate in improved outcomes and ultimately result in the universal acceptance of this method of treatment.

KEY POINTS