INTRODUCTION

Automated percutaneous lumbar discectomy (APLD) was introduced in 1985 by Gary Onik et al.^1–3 Since the 1960s, many different techniques for percutaneous removal of the nucleus pulposus or its protruding components have been proposed; they may achieve the goal in different ways, with different types of instruments, with or without fiberoscopic vision, or different types of energy (radiofrequency [RF], laser, coblation, etc.).^4–12 The basic principle, shared by most percutaneous intradiscal decompressive procedures, including APLD, is that in an enclosed space a reduction in volume, even partial, confers a much greater reduction in pressure; this leads to decreased pressure upon the nerve root, and relief of sciatica, even without a radiographically evident reduction in total disc volume.¹³ After weeks or months, the partial vacuum causes the protruded portion of nucleus pulposus (or other disc material) to move away from the nerve root back towards the center of the disc, pushed by partially intact fibers and ligaments of the outer anulus; this process, along with regeneration of a more fibrous nucleus pulposus, favors restoration of the inner fibers of the anulus and decreases the tendency to further protrusion towards the spinal canal. The success of the procedure depends to a great extent on selecting lesions to treat: the protruding nucleus pulposus must be at least partially contained by the external fibers of the disc, without a large extrusion and migrated or sequestrated fragments.^14–16

For decades, minimally invasive treatments for disc protrusions have been opposed by the surgical community, despite the high preference of patients to undergo a less intrusive intervention. APLD seems to have suffered the drawback of having been the first nonchemical, nonmanual procedure to be used worldwide, as the technique was met with fierce opposition. It would be fair to state that 20 years ago, surgeons were not ready to embrace percutaneous procedures. Now that the neurosurgical and orthopedic communities have accepted the concepts of intradiscal decompression and minimally invasive procedures,¹⁷ other techniques less effective than APLD dominate the field. In all likelihood, this relationship stems from the fact that APLD is still burdened with old, biased, and superficial judgments that are in part substantiated by poorly conducted studies.^18,19 In most published series good results range from 60% to 85%,^20–26 depending on patient selection criteria, while poor results are reported in the only two randomized and controlled studies.^18,19 In 1993, Revel et al. reported a 37% success rate at 1 year in a study comparing APLD and chemonucleolysis.¹⁸ Chatterjee et al., 2 years later, found a 29% success rate with APLD when compared to open surgery.¹⁹ However these studies, like others reporting low percentages of good outcomes,^27–29 have limitations and features that make the patient populations and technical conditions not really suitable for a comparison, and their results unreliable. First, the numbers of patients are low: 32 treated by one operator¹⁹ and 69 treated by many operators in a multicenter study.¹⁸ The authors do not state how experienced the operators were, i.e. how many APLD procedures each had already performed. The technical learning curve for APLD is longer than one might expect. It is only after many procedures are performed that the surgeon can obtain sufficient quantities of nucleus pulposus, and from the correct location of the disc. For example, the L5–S1 level is approached safely and reliably only by operators having performed a minimum of 40–50 procedures at higher levels. It is highly likely that the operators in the two studies mentioned above were much more experienced in open surgery or chemonucleolysis. As well, these investigators did not have access to some of the technique modifications that are described later in this chapter.

Another critical issue relates to the population of patients that are being compared. It is undeniable that adhering to specific inclusion and exclusion criteria is crucial to obtain good results with APLD, chemonucleolysis, and open surgery. However, there is little overlap between the indications for open surgery or chemonucleolysis and those for APLD. Consequently, there is a huge inherent limitation in randomized trials attempting to compare APLD to either chemonucleolysis or open surgery. The patients recruited in the two trials mentioned above are likely to have extruded, noncontained protrusions, which are not good indications to perform APLD.

In almost 20 years, a minimum of 170 000, and very likely many more, procedures have been performed. It is reasonable to state that APLD ‘opened the way’ to the concept of minimally invasive spine surgery, and that the concept and the technique itself have stood the test of time. APLD is an effective and safe method to obtain an intradiscal decompression, for relief of discogenic radicular or low back pain. It remains the percutaneous procedure that removes the largest amounts of nuclear material from within the intervertebral disc. Another great advantage, when comparing APLD with physical techniques that blindly destroy the disc (such as laser, RF or Coblation®), is that the surgeon can verify directly and visually the quantity of disc material removed, and its ‘quality’ as well. The extracted nucleus pulposus can be observed as it passes through the transparent tubing that connects to a filter. How much nucleus is taken out and how degenerated it is are important procedural and prognostic pieces of information. For example, viewing the quantity of removed nuclear tissue and comparing that to the amount that was anticipated to be extracted from interpreting the preoperative imaging provides critical information to determine whether the probe worked in the correct intranuclear location. Observing blood coming from the disc could suggest the presence of unexpected degeneration, or of painful granulation tissue inside the disc, or prompt arrest of the procedure so as not to damage the endplate cartilage. Aspirated disc material can also be sent for histology or microbiology in selected cases.

APLD achieves a very good compromise between low invasiveness and the need to obtain discal decompression. Its clinical results remain among the most satisfying when dealing with minimally invasive percutaneous treatments.

SAFETY

One of the appeals of percutaneous procedures, other than obtaining a high proportion of good results in properly selected cases, is the limited associated tissue destruction. While open surgery is effective, it has well-known disadvantages, including epidural scarring, damage to bone, denervation of paraspinal muscles with consequent segmental instability, long postoperative inactivity, and the feared ‘failed back-surgery syndrome.’ Patients who experience this latter phenomenon are often untreatable and can be severely disabled. Indeed, they represent the best advertisement for the benefits of minimally invasive procedures, particularly given the high tendency of disc protrusions to self-heal.

When considering the issue of side effects and complications, it appears that APLD is an extremely safe technique. Of course, that assumes that the surgeon performing the technique has the experience and ability to convert it into a safe procedure. If the nucleotome is improperly placed, it can easily cut dura, nerve roots, vessels, and other soft tissues. However, once the nucleotome is safely within the disc, i.e. isolated from surrounding neural and vascular structures, it is unable, unlike other devices, to cut its way out of the disc space to cause injury to those structures. The operator must at every moment be absolutely sure of the anatomical position of the operating instruments. Obviously, a key prerequisite is the ability to view two-dimensional fluoroscopic images and reconstruct a three-dimensional anatomic picture in the mind’s eye of the operator. To achieve this objective the operator must have a perfect knowledge of radiological projections and a large amount of experience.

The mortality rate of the procedure is zero. Lesions of nerve roots, vessels, or the ureter are possible;^30–32 however, as previously emphasized, with thorough knowledge of and attention to radiographic landmarks for proper probe positioning, vascular, neural, or dural injuries are very unlikely. The only major reported injury following APLD occurred in Mexico and resulted in a cauda equina direct lesion. It is quite likely that there was little if any attention directed to the radiographic landmarks that allow the surgeon to stay out of a potentially harmful pathway;³³ moreover, the procedure was performed under general anesthesia, definitely contraindicated, for the reasons explained later.

A posteriorly placed colon can insinuate behind the psoas muscle.^34,35 For this reason the preoperative imaging studies, both computed tomography (CT) or magnetic resonance imaging (MRI), must be carefully examined to exclude the presence of such an anatomical condition, since bowel in the path of the instruments could be perforated, with the risk of peritoneal or disc infection or local abscess formation. If not available, a planning CT scan of the whole abdomen through the disc space of interest with large field of view (FOV) must be obtained. When the L5–S1 disc is being removed, two scan slices (at the L4–5 and L5–S1 levels) should be obtained because the entry point for the L5–S1 placement is at the L4–5 level to avoid the iliac crest. In addition, this preoperative, planning CT can provide other valuable information. At the L5–S1 level, special attention should be paid to the bifurcation of the iliac vessels; at upper levels, the scan ensures that the lower pole of the kidney or the sulcus of the pleural space will not be traversed.

Beginning in June 1987, in our institution more than 1250 patients (accounting for more than 1450 discs) were treated. We observed and reported an overall complication rate of less than 0.9%.³⁶ There were no injuries to nerve roots, dura mater, ureters, major vessels, or bowel. We suspect this extremely low complication rate stemmed from our singular use of only local anesthesia, with or without light sedation, and the avoidance of general anesthesia. There was one acute hematoma in the iliopsoas that occurred following injury to a small artery and which resolved without sequelae in approximately 1 month. Among the side effects observed were two cases of discitis, resulting in a rate of 0.16 %, similar to the rate published in large series of discography.^37–39 Since discitis is a major complication, special care must be taken during skin prep and draping. We also use prophylactic antibiotics and typically give 2 g of intravenous cephalosporin to cover Streptococcus epidermidis.

It should be emphasized that some of the potential complications and side effects one could observe following open surgery have never been reported following APLD. Advantages of APLD when compared to other percutaneous technologies are the internal cutting action of the device and its blunt external portion that obviates damage to structures other than the nucleus pulposus. Of course, these factors dictate that the inherent morbidity of APLD is lower than that of other percutaneous disc removal methods.

INSTRUMENTATION

Automated percutaneous lumbar discectomy utilizes a probe called Nucleotome® (Fig. 28.1), manufactured by Clarus Medical, LLC, for removal of the nucleus pulposus. The probe tip, excluding the handle, is 20.2 cm long and has an outer diameter of 2.2 mm. The blunt tip is an extremely important safety feature. Once the probe is inserted, the lack of a sharp end prevents it from piercing through the outer limits of the disc, even with an inadvertent hard push. This feature is unique and not a component of other instruments such as laser, RF probes, or manual biopsies, thus essentially removing the risk of lesion of vessels or other abdominal structures. The negative pressure for aspiration is generated by a console. A vacuum is created that draws nuclear material into the side port, which is located a few millimeters proximal to the distal tip of the probe. The cutting blade for fragmentation of nucleus pulposus aspirated through the port, works with a reciprocal, not rotatory motion. This type of movement is a safety feature because the ‘guillotine’ blade is contained within the probe. Consequently, only the nuclear material that is drawn into the port can be cut. The blade is pneumatically driven by a pressure pulse, generated by the same console that creates the vacuum that draws nuclear material into the side port. The console also controls the cut rate and the flow of irrigation fluid to the probe. Internal irrigation with sterile saline is a vehicle for easy aspiration. The reciprocal movement of the internal cutting blade also sequences the introduction of liquid inside the disc, to prevent accumulation of nuclear material and consequent clogging inside the probe, or an excess infusion of fluid within the disc. The cutting rate knob on the console allows for adjustment of between 60 and 180 cuts per minute. At the beginning of the procedure, the maximum cutting rate should be used to cut smaller pieces of disc and prevent the instrument from clogging. As the decompression proceeds, the amount of disc material aspirated diminishes, allowing the surgeon to ratchet down the cutting rate. This will allow more time for the negative pressure to draw disc material through the port before it is resected.

Fig. 28.1 The Nucleotome® probe has a blunt, rounded tip. Internal irrigation and cutting functions are incorporated. The aspirated nucleus pulposus enters the side port, and is resected by a pneumatically driven ‘guillotine blade’, which has a reciprocal, not rotary movement.

The fluids and solid material aspirated from the inner disc and exiting through the metallic probe are ultimately deposited into a filter in a disposable collection bottle. To reach that location the extracted nuclear tissue traverses a transparent plastic tube. Throughout this journey the nuclear material is clearly visible. As alluded to previously, there are tangible benefits that can be realized by real-time monitoring of the disc material as it flows through the transparent aspiration line, mainly the possibility to verify in which positions of the probe inside the disc the largest quantities of nuclear material are extracted. Moreover, the nucleus pulposus collected in the filter is available for quantitative and macroscopic qualitative evaluation, or even for histology examination. The operator must also use the transparent line to check for air: an excessive amount mixed with little or no nuclear material indicates a leak somewhere in the system, which decreases the effectiveness of the suction. The most common cause is an inadequate seal on the cannula, which allows air to be sucked into the disc space. This usually does not happen because anulus and extradiscal soft tissues make a seal around the probe and prevent air from being sucked into the disc. However, if anulus and tissues have less resilience because the patient is older, air will be sucked through the cannula preventing aspiration of the disc and thus rendering the procedure less effective (or longer); in these cases the straight cannula can be replaced with the larger, curved one.

If the cannula is not the cause, the air leak may be emanating from the fluid delivery line or even the probe. If a check of the fluid delivery line does not reveal the problem and all maneuvers fail to stop the leak, place the trocar back through the cannula and change both the cannula and the probe.

A sequence of devices is used for introduction of the probe inside the disc. After local anesthesia, the first device is positioned in the center of the disc. It is a flexible, 18-gauge stainless steel guide wire with a trocar point. Unlike bevel-pointed needles, a trocar point does not have a sharp cutting effect, thus limiting the risk of vessel or nerve injuries; these structures are more likely deviated by the trocar point rather than resected. In addition, unlike beveled needle tips, the symmetric trocar point follows a straight trajectory, without deflections that would render a precise positioning more difficult. This is particularly applicable when driving the instrument in relatively resistant tissues such as the anulus fibrosus. Moreover, the blunt trocar device is more difficult to push through soft tissues, which gives the surgeon a much better feeling of the tissue actually met and traversed, and allows for easy recognition of the muscular fasciae, anulus fibrosus, and nucleus pulposus.

Once the guide wire is positioned correctly, a cannula with a dilator inside is passed over the guide wire. The dilator is designed to protect soft tissues from surgical trauma; this prevents both bleeding and postoperative muscle spasm. Each single kit contains both a straight cannula, with an outer diameter of 2.8 mm, and a curved cannula, with an outer diameter of 3.8 mm. The reason for a larger diameter in the latter is that it is internally coated by a Teflon layer, which reduces friction and favors sliding of the flexible but straight probe. Once the cannula is positioned against the outer fibers of the anulus, the dilator is removed from the cannula and replaced by a trephine. An incision in the anulus fibrosus is made by means of the trephine, which is a few millimeters longer than the cannula; the same flexible trephine is designed to function with both the straight and curved cannula.

PATIENT POSITIONING AND SELECTION OF ENTRY ROUTE

The procedure can be performed with the patient in either the prone or lateral decubitus position. When the prone position is used, bolsters are placed underneath the patient’s abdomen to open the disc spaces posteriorly. This will provide for easy access and for better transmission of the decreased pressure in the center of the disc, consequent to the aspirating action of the probe, to the herniating nucleus pulposus. For the same reason, the patient is flexed when in the lateral decubitus position, which is the approach of preference for addressing pathology of the L5–S1 disc.

The entry route, as usual with every lumbar percutaneous approach to the disc space, is posterolateral (Fig. 28.2). Correctly positioning the guiding trocar is crucial to the result. The trocar must be placed with its tip in the midline in frontal view, at the junction of the middle and posterior thirds of the disc in lateral view, where the normal nucleus lies (Fig. 28.3). This location corresponds to the initial working position of the probe. In cases of large, posterior protrusions invading the spinal canal, it is preferable to aim for a more posterior position of the probe.

Fig. 28.2 The correct path of the devices from the skin entry point to the center of the disc is anteriorly tangent to the superior articular facet (A, axial view; B, oblique view; C, lateral view).

Fig. 28.3 The tip of the trocar must be placed at the junction between middle and posterior third of the disc in lateral view. The anteroposterior view must confirm the correct position in the midline. In addition, the trocar must be ideally parallel to the endplates in both views.

The best and safest way to accurately place the trocar is to choose a path that passes anteriorly and laterally tangent to the posterior zygapophyseal complex, i.e. with the trocar touching the anterolateral surface of the superior articular process (see Fig. 28.2). The usual skin entry point for such a route is 10 cm from the spinous processes. The operator can use the preoperative CT scan to accurately and easily geometrically derive the proper skin entry site. Once the skin target has been determined, a skin wheal is raised. Immediately thereafter, a 22–25-gauge spinal needle is inserted more posteriorly than the anticipated route to the disc, with the aim to touch the posterior facets. Local anesthetic is injected at this location and the needle is gradually withdrawn toward the skin, allowing for anesthetization of the underlying spinal musculature. It is critical that local anesthetic is not deposited anterior to the articular processes, which can anesthetize the exiting root. General anesthesia is contraindicated for this procedure since the chance of nerve root injury is much greater under those circumstances.

The 22-gauge, 15 cm long needle can also be used to test if the trajectory from the skin to the anulus is accurate.

Which side to begin the procedure depends upon the side of the patient’s symptoms. We usually select the side of the patient’s symptoms. There are two compelling reasons to do so: first, to place the nucleotome probe as close to the herniation as possible, particularly when the herniation is lateralized; second, to avoid the possibility of bilateral symptoms in the event of complications. In rare circumstances a contralateral approach is preferred. First, if a correct trocar positioning inside the disc becomes impossible. That situation can arise when the trocar repeatedly abuts the exiting nerve root, precluding safe advancement into the anulus. Anatomic circumstances that can lead to this scenario are a conjoined nerve root, flattened root compressed by the herniation, and in particular at L5–S1, the normal position of the exiting L5 root. When a conjoined nerve root is present, both roots can exit from a single foramen. The superior root takes a more caudal and inferior course, which places it in the line of a correctly placed trocar. In addition, in case of a far lateral herniation, it is possible that the nerve root is in an abnormal position and pushed posteriorly to close the space between the nerve and the anterior surface of the facet. The second reason to use the contralateral side is when the patient does not tolerate lying with the ipsilateral side superior. A third factor is when an anatomical variant precludes entry. For instance, an L5–S1 approach is not possible from the side of a monolateral transitional vertebra, with sacralization of the transverse process, or if there is a risk of posteriorly placed colon, behind the psoas muscle, as shown by the preoperative CT scan with large field of view (FOV). In the latter case, considering that the left-sided sigmoid colon is more mobile, the surgeon can let gravity assist, positioning the patient on his/her right side; under this condition the left colon falls away from the entry route of the instruments. It must be emphasized that the prone position increases the risk of piercing posteriorly placed bowel.

The procedure is monitored fluoroscopically, with use of a C-arm. A comprehensive knowledge and extremely precise use of the radiological projections and landmarks is a prerequisite to performing this procedure. Such knowledge will dramatically impact the outcome and the probability of realizing a side effect or complication. An incorrect projection means that the probe is actually working away from the place where it is supposed to, not effectively aspirating nucleus or, worse, damaging vital or functionally important structures.

The crucial radiographic rules are as follows:

Fig. 28.4 Effect of the parallax created by the diverging X-rays on the position of objects seen under fluoroscopy. Two radiopaque objects are seen. The one in the central beam is projected where it actually is. The off-center one is projected lateral to its actual position (indicated by the dotted line and arrow).

NUCLEOTOME PLACEMENT AND ASPIRATION OF THE NUCLEUS PULPOSUS

As with the needle for local anesthesia, the operator should start, in lateral view, with a posterior trajectory until the trocar hits the bone of the articular facets; then, the trocar is worked anteriorly with tiny movements, until it slides along the anterior surface of the articular complex (actually, over the anterolateral surface of the superior articular facet) (see Fig. 28.2). The closer to this superior articular process the instrumentation is, the less likely the operator is to touch the nerve root. Then, the trocar is pushed toward the posterior profile of the disc, corresponding to the posterior somatic walls, always parallel to the vertebral endplates. When the tip of the trocar reaches the posterior profile of the disc, a gritty, hard but elastic (definitely different from bone and soft tissues) tactile sensation is felt, corresponding to the anulus fibrosus (Fig. 28.5). If the operator does not feel the anulus, the trocar must be withdrawn and replaced more posteriorly; under no circumstances should the operator push the trocar anterior to the posterior profile of the disc. In this situation, the path of the trocar would be too anterior, rendering it impossible to achieve correct trocar placement. Moreover, the risk of injuring nerve root, bowel, inferior vena cava, aorta, or iliac arteries would be escalated. If one attempts to position the trocar more posteriorly, but is precluded because of the articular complex, then a more lateral skin entry point must be used.

Fig. 28.5 The nerve root exits underneath the pedicle at the top of the foramen and courses anteriorly and inferiorly. Under fluoroscopy in lateral view, the trocar’s tip must be at the posterior vertebral body line (PVBL) when the anulus is felt.

The described trajectory brings the needle and eventually the nucleotome dorsal to the nerve, which is coursing from the upper portion of the foramen anteriorly and inferiorly (see Fig. 28.5). If the patient is going to experience radicular pain from the needle placement, it will usually occur when the needle is placed too high in the foramen or anterior to the posterior vertebral body’s line. If the nerve root is touched, the patient experiences radicular symptoms, usually a sensation described as a sudden ‘electrical shock’ which may be experienced as distal as the foot, depending on the root that has been abutted. In contrast, the pain originating directly from the nociceptive fibers of the external anulus is less intense and never goes down distal to the knee. If the patient describes a radicular sensation, the trocar must be withdrawn and replaced, even if that requires the skin entry point be modified. In general, moving the entry point 1–1.5 cm in either direction is enough to change the trajectory and obtain a painless trocar placement. All major redirections of the trocar require that it be withdrawn into the subcutaneous space before readvancement, because the fascial planes create a point of fixation that does not allow for major path corrections. Consequently, if the trocar is not withdrawn a sufficient distance, further attempts of trocar placement will only result in trocar bending.

When the anulus is reached, as determined by its tactile quality, and the 18-gauge trocar touches the posterior vertebral body’s line and is midway between and parallel to the vertebral body endplates, the anteroposterior view should be checked (Fig. 28.6) and this represents a major safety feature. It confirms that the 18-gauge trocar is outside the spinal canal and consequently will not traverse the thecal sac as it is advanced into the disc space. In this anteroposterior view, the tip of the trocar should be lateral to a vertical line connecting the medial borders of the pedicles, thereby confirming its correct position outside the spinal canal. Once confirmed, the trocar can be safely advanced into the center of the disc using this anteroposterior view. The fluoroscope is then repositioned to obtain a lateral view, allowing the surgeon to confirm that the trocar’s tip is correctly placed on both anteroposterior and lateral views (see Fig. 28.3). If the trajectory is too anterior, the trocar tip is visible in the center of the disc on the anteroposterior view, but extends ventral to the center of the disc on the lateral view. Since we want to be as close as possible to the disc herniation, no placement that has an anterior trajectory is acceptable.

Fig. 28.6 In the anteroposterior view, after the trocar has touched the anulus at the posterior vertebral body line, as shown in Figure 28.5, the trocar’s tip must absolutely lie lateral to the vertical line joining the medial borders of the pedicles; this confirms that the trocar is outside the spinal canal and it will not traverse the thecal sac when pushed to advance into the disc space.

When the trajectory is posterior, the trocar tip will appear to be in the center of the disc on the anteroposterior view, but posterior to the center of the disc on the lateral view. Since the nucleus pulposus is situated slightly posterior to the center of the disc and we want to be as close to the herniation as possible, a posterior trajectory placement is not only acceptable, but preferred (as previously stated, ideally at the junction between middle and posterior third of the disc).

Once the trocar is in the correct position, the dilator and cannula are placed over the trocar and advanced until the anulus is reached. The dilator is then removed, and the cannula is pushed the extra few millimeters to rest on the anulus. At this point, the biopsy stop is brought down to the skin. Later in the procedure, this marker will indicate whether the cannula has been withdrawn or pushed into the anulus. The fluoroscopic beam is now reoriented perpendicular to the cannula, in an oblique view, to confirm that the cannula has reached the anulus (Fig. 28.7). Due to the oval shape of the disc, the tip of the cannula could overlap the margins of the disc on the anteroposterior and lateral views and still not be against the anulus. In this circumstance, if the trephine is used and the cannula is not absolutely against the anulus, the adjacent nerve root could be inadvertently injured. When the oblique view confirms that the cannula is resting on the anulus, the dilator is exchanged with the trephine while keeping cannula and trocar in place. During this maneuver the cannula must be pushed firmly against the anulus to prevent the root from insinuating itself between the cannula and the anulus. The trephine is placed against the anulus, which is then incised. Just prior to using the trephine, lightly tapping the anulus with the trephine ensures that a portion of nerve root has not been trapped by the cannula. Patients with a chronic disc disease may experience intense, nonradicular pain when the cannula and the trephine are pushed against the annulus, probably because of sprouting of nociceptive fibers that can accompany degeneration of a disc. If the patient cannot tolerate such pain, before incision with the trephine, a long, 22-gauge needle is inserted in the space between cannula and trocar, and 0.5 ml of 1% plain lidocaine are injected at the site of incision of the anulus.

Fig. 28.7 An oblique X-ray view, perpendicular to the cannula, confirms that the cannula itself is abutting the anulus fibrosus, without any gap in which the nerve root could slide and be cut by the trephine.

Now that the incision is made, the trocar and trephine can be removed. While accomplishing this, the cannula must be held firmly so as not to lose the newly created hole. The nucleotome is then placed into the disc.

Once the aspiration probe is seen to be in the correct position in anteroposterior and lateral views, the port of the instrument is rotated toward the area of the herniation, the nucleotome console is turned on, the footswitch pressed, and the disc aspiration is initiated. As much disc material as possible is aspirated, while gradually moving the probe back and forth. Once no more additional disc material can be retrieved with the port in the direction of the herniation, turn the port to a new area and repeat the back-and-forth motions along the entire length of the nucleus. The cannula can be angled to obtain material from other areas of the disc. During the initial 5–7 minutes, use the straight cannula to aspirate centrally and anteriorly. A curved cannula should be utilized for the rest of the aspiration phase of the procedure, which usually does not take more than 12–15 minutes. The curved cannula that is provided with the surgical kit is particularly helpful in reaching the L5–S1 disc, especially when the approach is partially covered by the iliac crest. It also enhances the range of movement of the probe inside a disc, resulting in the aspiration of greater amounts of nucleus pulposus. This advantage is extremely important when in the more posterior or posterolateral positions, where the nerve root is compressed. For this reason the curved cannula is used at every discal level. During aspiration, the probe is constantly moved in a double rotation-type maneuver, therefore creating successively larger and larger arcs of motion. During this action, the port is maintained in a position that is external to these virtual circles. Ultimately, at least two large tunnels are excavated in the disc.

Since the disc is avascular, the fluid flowing from the aspiration line should not be bloody. If the fluid has a red tinge, the most likely explanation is the presence of a gap between the cannula and the anulus that allows blood to be sucked into the disc from the surrounding soft tissues. Be aware that you will aspirate blood whenever you retrieve the probe into the cannula or during each exchange of the probe or cannula (for instance, after withdrawing the probe to purge clogged extracted tissue or exchanging the curved for the straight cannula). One final cause of a haemorragic return is enlargement of the hole in the anulus that may occur near the end of the procedure; angling of the cannula to reach various portions of the disc can widen the initial annular opening. It is through this increased aperture that blood is aspirated inside the disc.

When treating discogenic back pain, aspiration is never stopped before blood starts coming from the disc (from granulation tissue or endplate cartilage). Observing blood in the tube suggests the possibility of subsequent scarring. It is always possible to achieve this result when using the ‘double rotation’ technique. It is likely the good results of APLD for discogenic pain are achieved by lowering the pressure within the disc and by stimulating scar formation. The latter ultimately reinforces the anulus with more fibrous tissue and diminishes the motion in cases of symptomatic microinstability. It is reasonable to assume that discogenic pain can be improved or cured thanks to removal of nodules of innervated, painful ‘granulation’ tissue. These areas may be evident on MRI as zones of high signal, usually in the posterior annulus;⁴⁰ since granulation tissue is vascularized, early bleeding occurs during aspiration. Bleeding is not, however, necessary when treating patients for relief of root compression; aspiration is stopped when the amount of disc material coming from the probe greatly decreases. When treating radicular pain, at the end of the procedure, the side port of the probe is positioned close to the protruding parts of the disc in the neural foramen in an attempt to aspirate the protruding material itself. This ‘topographical’ criterion must guide the aspiration. It is important to keep in mind that the probe is not positioned to work only in the center of the disc, but should be directed to look for the main bulk of the nucleus pulposus and its protruding components. As long as the preprocedural MRI or multiplanar reformatted CT scan indicate that the height of the disc is preserved, at least a moderate amount of nuclear material must be removed. In general, the less degenerated, i.e. yellowish, the nucleus material appears, the more quantity must be aspirated. If, after an apparently correct first positioning of the probe inside the disc, nuclear material does not come freely, the probe must be moved to locate it. Treatment proceeds until nucleus is found and removed.

If the instrument becomes obstructed, the probe must be removed from the cannula and flushed with normal saline to remove the clogging material.

The probe is withdrawn into the cannula once no more material can be aspirated. Both the cannula and probe are removed, as a single unit. This maneuver prevents the port of the probe from inadvertently injuring any structures during removal. In addition, by retracting the probe into the cannula first, the dissipation of any negative pressure that was created in the disc is prevented.

After withdrawal, 80 mg methylprednisolone acetate (Depo-Medrol®) and 1 ml bupivacaine 0.5% are injected in the disc. This is easily accomplished by passing an 18-gauge needle over an 0.8 mm Kirschner wire that had been left in place as a guide. The author’s experience is that adding an intradiscal glucocorticoid infusion results in a more rapid convalescence, and postoperative pain relief will be realized much earlier.

Since introduction of the probe requires a miniscule skin incision, a suture is not necessary to end the procedure.

PERCUTANEOUS DISCECTOMY AT THE L5–S1 LEVEL

The height of the iliac crests is the crucial factor determining the difficulty of a percutaneous approach at the L5–S1 level. High iliac crests cover the disc space, and consequently the entry route must come from a more cephalad starting point. There are instances in which straight instrumentation will not enter the disc correctly, i.e. parallel to the vertebral bodies’ endplates. For this reason the procedure is usually easier in women because of their wider and gently sloping iliac crests. Three other situations in which the procedure is more difficult should be anticipated. First, the narrower the disc space, the more difficult the procedure becomes. Second is the presence of a spondylolisthesis. Lastly, the smaller the space between the transverse process of L5 and the sacrum, the more difficult the procedure. Prior to beginning a decompression at the lumbosacral junction, the anteroposterior radiography should be reviewed to ascertain whether there is a transitional vertebra, an enlarged transverse process, the degree of disc height decrement, and the relationship of the iliac crest to the annular entry site.

Performing the procedure with the patient lying in the lateral decubitus position increases the probability of correctly entering the L5–S1 disc. A soft silicon gel cushion or other similar prop wedged just superior to the iliac crest will laterally flex and lower the iliac crest on the entry side, thus opening an access trajectory to the L5–S1 disc (Fig. 28.8).

Fig. 28.8 The L5–S1 disc space is more easily approached in the lateral decubitus position. A soft silicon gel cushion is positioned under the patient’s hip and used to tilt the pelvis and to bend the lumbar spine in order to lower the iliac crest on the entry side, thus uncovering the disc from the iliac crest.

Owing to the upward convex curve of the iliac crests, the more lateral the skin entry point is selected, the steeper from above (i.e. the less parallel to the endplates) the trajectory to the disc space will be. However, if too medial a starting point is chosen, the nucleotome will terminate in a too ventral (anterior) location within the disc. If the correct intradiscal position cannot be achieved with a straight cannula, the curved one should be substituted for it.

Although placement of the curved cannula is technically demanding, it offers a palpable benefit. When oriented correctly, the curve of the cannula allows the trocar and the probe to enter the plane of the disc (i.e. parallel to the endplates), despite an angled approach from above of the devices (Fig. 28.9).

Fig. 28.9 The angle to the L5–S1 disc space is too great, because of the slope of the iliac crest: a straight cannula would lead the trocar to a course not parallel to the endplates, entering high and going down to touch the superior endplate of the sacrum (left). Properly maneuvered, as described in the text, the curved cannula allows a more favorable positioning of the devices in the disc space, i.e. acceptably parallel to the endplates (right).

When it is anticipated that the trocar will enter the disc following a suboptimal path, either because the trajectory is too anterior or not parallel to the endplates, the curved cannula can be brought down over the trocar to the anulus. Then, the trocar can be withdrawn and the cannula can be slid along the anulus or angled to the anulus at the appropriate degree to achieve the correct trajectory. The trocar can then be reinserted into the disc. If the angle of the trocar in relation to the disc space is too steep, it is not possible to place the trocar within the disc. In this case, the trocar can be firmly anchored against the anulus or endplate and the curved cannula advanced to the disc. Again, if it is kept against the disc or endplate, the cannula can be appropriately moved or angled to allow the trocar to enter the disc.

Once the trocar is correctly placed, the dilator is removed and the trephine is placed over the trocar and through the cannula. The trephine and trocar are then both removed and the probe is placed through the cannula into the disc. If the curved cannula was required to deposit the probe parallel with the endplates, it can be subsequently used to reach posterior aspects of the disc (Fig. 28.10). Indeed, the reach it achieves is superior to that offered by a straight cannula. It is sometimes difficult to pass the curved cannula through the lumbar fascia, because of the larger diameter and bigger step down from the edge of the cannula to the dilator. If so, it may be necessary to dilate the tract first with the straight cannula and dilator.

Fig. 28.10 After positioning of the probe in the L5–S1 disc space as described in the text and in Figure 28.9, the curved cannula can be turned with an anterior convexity, so that the probe will be directed to work and aspirate in a position more posterior than it would be allowed by a straight cannula.

At L5-S1, the disc is aspirated using the same concepts and methods that apply to more cephalad intervertebral discs.

In favorable anatomy, an alternative method for placement of the trocar in the L5–S1 can be used. The path to the disc is found centering the X-ray beam over the L5–S1 disc space in the anteroposterior view. The fluoroscopic unit is then angled in the cephalocaudal direction until the endplates of the disc are visible as single superior and inferior entities, indicating that the entry angle is parallel to the endplates. The X-ray beam is then angled toward the lateral view, and as it is moved in an oblique orientation, the L5–S1 facet joint moves across the disc space and the iliac crest starts to overlap the anterior portion of the disc. When the beam is at approximately a 45° angle, a triangular window at the center of the disc space is seen (Fig. 28.11). This triangle is bounded laterally by the iliac crest, medially by the anterior surface of the superior articular process of S1, and superiorly by the inferior endplate of the L5 vertebra. Once this view is obtained and centered on the screen, the trocar, with use of a holder, is inserted in bull’s-eye fashion through the middle of the triangle. When done properly the trocar will appear as a single dot. The trocar should then be advanced using real-time imaging until it touches the outer annular fibers.

Fig. 28.11 In an oblique view, a triangle free from bony structures indicates the path to the L5–S1 disc; the triangle is bounded by the iliac crest, the anterolateral surface of the superior articular facet of S1, and the inferior endplate of L5. The needle, appearing in the central beam as a single dot, will be pushed easily in the center of the disc space.

POSTOPERATIVE CARE

We routinely perform APLD on an outpatient basis. At the conclusion of the procedure, observation is needed for about 2 hours before discharging the patient home. Prescriptions are provided for a 2-week supply of a nonsteroidal antiinflammatory agent, and for diazepam at bedtime for 2–3 days. Patients are encouraged to move, stand, and walk on day three. Driving or prolonged sitting is proscribed for a 2–3-week interval.

After APLD, early activity is not only possible but also useful. It is imperative to avoid muscle atrophy and general deconditioning. Repetitive forward flexion, prolonged car driving, prolonged sitting, and lifting heavy weights is prohibited for 3–4 weeks. Limb pain resolution may take weeks, owing to ‘remodeling’ of the disc and regression of inflammation at the surgical site. Progressive return to heavy activities or sports is usually possible at 4–6 weeks.

A procedure that does not result in substantial relief of pain should not be considered a failure until at least 6 weeks have passed. During the convalescence phase, rehabilitation measures applied by experienced physical therapists may be useful, not before 3 weeks after the intervention. Among the concepts that need to be learned are maintaining a positive attitude, recognizing the difference between symptoms of a residual herniation and those of a healing process, and proper biomechanics.

The only noteworthy side effect is the possibility of increased back pain. Most patients with a surgical wound have pain, and that applies to percutaneous discectomy. Injury to skin, muscle, fascia, and anulus will occur; a correct operative technique usually avoids injury to the endplates, but this may occur when treating patients for back pain. Patients are warned that they may experience new back pain for up to 3–4 weeks. Patients should be encouraged to maintain as much mobility as possible despite the presence of back pain.

SUMMARY

Automated percutaneous lumbar discectomy is a minimally invasive intradiscal decompressive technique proposed in 1985 by Gary Onik, referred to as ‘automated’ since it is characterized by the use of a mechanical probe, working by means of a ‘suction and cutting’ action for removal of the nucleus pulposus.

The method achieves a high degree of clinical efficacy for relief of not only radicular but also of low back pain. Results are particularly favorable in some subgroups of patients, such as elderly people, patients previously operated on, and patients suffering from ‘discogenic’ low back pain.

Complication rates are extremely low (less than 1%) and usually mild from the clinical point of view, followed by complete recovery; a correct operative technique rules out the risk of any major, particularly neurological, complication.

A thorough technique is also mandatory to obtain a good intradiscal and root decompression, and the learning curve is longer than usually believed. A ‘topographic’ criterion must guide the operator, who must maneuver the aspiration probe inside the disc looking for the nuclear material where it is and not ending the procedure until a significant amount of it is extracted. In this respect a great advantage of Onik’s method is the possibility to visually and directly verify quantity and quality of extracted nuclear material.

Among minimally invasive percutaneous treatments, APLD probably achieves the best compromise between low invasiveness and entity (quantitative and topographic) of discal decompression.