Minimally Invasive Posterior Approaches to the Spine

Published on 17/03/2015 by admin

Last modified 22/04/2025

Print this page

This article have been viewed 2575 times

CHAPTER 56 Minimally Invasive Posterior Approaches to the Spine

Choll W. Kim, MD, PhD, James D. Schwender, MD, Kevin Foley, MD

The term minimally invasive spine (MIS) surgery describes a variety of surgical techniques that employ key concepts in spinal surgery. The guiding principles of MIS are to (1) decrease muscle crush injuries during retraction, (2) avoid injury of osseo-tendinous attachments important for spinal stability, (3) maintain the integrity of the dorsolumbar fascia, (4) limit bony resection, (5) avoid injury to neurovascular supply of muscle compartments by using known anatomic planes, and (6) decrease the size of the surgical corridor to coincide with the area of the surgical target site. These principles can be used in a variety of settings. In the lumbar spine, the most significant application of these principles has been for posterior approaches to the lumbar spine.

Surgical Anatomy of the Posterior Paraspinal Muscles

The posterior lumbar paraspinal muscles are responsible for maintaining spinal posture in its neutral position. Furthermore, the paraspinal muscles guard the spine from excessive bending that would otherwise endanger the integrity of the intervertebral discs, facet joints, and ligaments.¹ It is the body’s dynamic stabilizing system that prevents pain and injury to spinal column due to the repetitive loads during the course of daily activities. The posterior paraspinal muscles are composed of several muscle groups that run along the thoracolumbar spine and attach caudally to the sacrum, sacroiliac joint, and iliac wing.

Posterior spine surgery using midline approaches inherently causes damage to surrounding muscles.^2,³ Muscle injury leads to long-term muscle atrophy, which in turn leads to decreased force production capacity of the muscle.^2,⁴^–¹² The multifidus muscle is most severely injured when using this approach for several reasons. First, its medial location inherently requires it be displaced most during retraction. This predisposes the muscle to greater retraction pressures and makes it more vulnerable to disruption of its neurovascular supply.^2,⁷ Of equal importance, the midline posterior approach inevitably leads to disruption of the multifidus tendon attachment to the spinous process, as well as the integrity of the dorsolumbar fascia.

Key Concepts for Minimally Invasive Spine Retraction Systems

A key advancement in minimally invasive surgery came from Foley and colleagues ¹³ with the development of the tubular retractor. A cylindrical retractor allows the surgical corridor to be opened via serial dilation using sequentially larger concentric tubes. This decreases the need for muscle stripping during the exposure. Furthermore, a tubular retractor maximizes the surface contact area, which in turn minimizes the pressure per unit area. Another key concept in MIS is use of a retractor holder mounted to the table instead of using a “self-retaining” mechanism. In a self-retaining retractor system, constant pressure on the tissues must be exerted to hold the retractor in place. Studies show that the maximum intermuscular pressure around a tubular retractor decreases by 50% within 3 seconds.¹⁴ Thereafter, the pressure is undetectable. With self-retaining retractors, the pressure remains unchanged.

Minimally Invasive Spine Surgical Corridor

The guiding principle of MIS posterior lumbar surgery is to avoid injury to the multifidus tendon attachment to the spinous process and maintenance of the dorsolumbar fascia integrity. This is accomplished by using paramedian approaches rather than midline approaches. Decompression, microdiscectomy, interbody fusion, posterolateral fusion, and pedicle screw instrumentation can be accomplished through this surgical corridor. Emerging techniques for advanced reconstruction including posterior corpectomies and strut fusion for burst fractures, tumors, and infections are also possible.

Posterior Lumbar Approaches

Tubular Microdiscectomy

The treatment of herniated discs via MIS tubular microdiscectomy is the most common technique currently used in the United States. This system, developed by Foley and Smith, consists of a series of concentric dilators and thin-walled tubular retractors of variable length. The use of the tubular retractor, rather than blades, allows the retractor itself to be thin walled (0.9 mm). The tube circumferentially defines a surgical corridor through the erector spinae muscles. The appropriate depth of retractor prevents the muscle from intruding into the field of view. The retractor allows for the appropriately sized working channel to permit spinal decompression. The typical retractor size is 14 to 18 mm for microdiscectomy (Fig. 56–1). Surgery is typically performed using an operating microscope. Several randomized controlled trials have been performed to compare traditional open microdiscectomy with minimally invasive tubular microdiscectomy.¹⁵^–¹⁷ These studies all show that tubular microdiscectomy is safe and efficacious compared with well-established traditional techniques. Clinically significant superiority was not shown, likely reflecting the difficulty is demonstrating differences between the two already successful procedures.

FIGURE 56–1 Intraoperative photographs and fluoroscopic images of the tubular microdiscectomy technique. A, Intraoperative photograph and B, corresponding to the lateral fluoroscopic image of the surgical target site showing the lateral pars and infero-medial transverse process after removal with a fine Kerrison rongeur. This maneuver allows for palpation of the pedicle and assists identification of the exiting nerve root, adjacent to the ball-tip probe.

Lumbar Decompression

An important goal of minimally invasive posterior surgery is maintaining the tendinous attachment of the multifidus to the spinous process. During a traditional laminectomy, the spinous process is removed and the multifidus muscle is retracted laterally. Upon wound closure, the multifidus origin can no longer be repaired to the spinous process. The midline approach affords a symmetric view of the posterior elements, which allows for safe resection of the lamina, ligamentum flavum, and medial facets. The symmetric view allows the surgeon to readily identify and orient the surgical corridor. However, a thorough decompression can be achieved without need for removal of the spinous process. In a technique originally described by McCulloch and colleagues,¹⁸ the spinal canal can be approached through a unilateral portal via a hemilaminectomy technique. Decompression of the central canal and contralateral recess can be achieved by angling the tubular retractor dorsally to view the undersurface of the spinous process and contralateral lamina (Fig. 56–2). The dural tube can be gently pushed down, and the ligamentum flavum and contralateral superior articular process resected to achieve a bilateral decompression.

FIGURE 56–2 Magnetic resonance axial (A) and sagittal (B) image in a patient with spinal stenosis. Intraoperative fluoroscopic navigation AP (C) and lateral (D) images showing insertion of the initial dilator. The surgical corridor is overlaid with the tubular retractor in solid white and surgical target sites in dashed lines. Intraoperative photomicrograph of the surgical target site. E, Initial bony exposure of the base of the spinous process and facet joint line. Top is medial. F, Visualization of the dural tube to the contralateral side using a unilateral approach.

The efficacy and safety of minimally invasive posterior lumbar decompression have been assessed in multiple studies.¹⁸^–²⁴ In a review by Asgarzadie and Khoo,²⁵ this technique provides long-term symptomatic relief equivalent to traditional open surgery but with significant reductions in operative blood loss, postoperative pain, hospital stay, and narcotic usage. The effect of the MIS learning curve remains a significant concern as increased complication rates are seen during the initial series of patients.²⁶ Despite the learning curve, the overall complication rates remain low, even in patients who are elderly or medically frail.²⁷^–²⁹

It is important to consider the anatomic variation of the lower lumbar spine with the upper lumbar spine with this particular technique. At L3 and above, the lamina between the spinous process and facet joint can be narrow (Fig. 56–3). With a unilateral approach, it may be difficult to reach the ipsilateral recess without excessively removing the ipsilateral inferior articular process. An option is to use a bilateral cross-over technique to reach the right lateral recess from a left-sided hemilaminectomy and vice versa. Anatomically, the lateral recess is more accessible through a contralateral approach when using the unilateral approach. In a preliminary study of four patients and seven levels of decompression, the total operating time was 32 minutes per level and the estimated blood loss was 75 mL. The average postoperative stay was 1.2 days. All patients had resolution of neurogenic claudication and there were no complications.³⁰

FIGURE 56–3 Magnetic resonance cross-sectional image through disc space of A, L2-3, B, L3-4, C, L4-5, and D, L5-S1. The outline of the tubular retractor is overlaid on each image. Note the proximity of the ipsilateral facet joint at the higher lumbar levels. At L3-4 and above, care must be exercised to avoid inadvertent injury to the ipsilateral facet joint. A bilateral approach can be used (E) to decompress the lateral recess from the contralateral sides (F).

Posterior Lumbar Interbody Fusion

An extension of the minimally invasive hemilaminectomy technique is transforaminal lumbar interbody fusion (MIS TLIF). The unilateral approach is used to perform the analogous decompression and is combined with a complete facetectomy. The surgical corridor is in the neurovascular plane between the multifidus and longissimus muscles (Fig. 56–4). A complete facetectomy allows for decompression of the spinal canal from the ipsilateral to the contralateral side (Fig. 56–5). Access to the disc space is through a window bordered medially by the dural tube, proximally by the exiting nerve root, and distally by the pedicle and superior endplate of the caudad vertebra, thus forming within the Kambin triangle (Fig. 56–6). Angled curettes are used to perform a subtotal discectomy from a unilateral approach. If necessary, an osteotome is used to remove the overhanging rim of the posterior vertebral endplate during discectomy. Fusion is performed using interbody spacers that can be placed anteriorly for maximum lordosis correction. A second cage may be inserted by using the smooth trials to push the first cages to the far side of the disc space. Dual cage constructs may be desirable when there is significant osteoporosis or at L5-S1 in a multilevel fusion.

FIGURE 56–4 The minimally invasive spine transforaminal lumbar interbody fusion surgical corridor is described. A, Magnetic resonance cross-sectional image through L4-5 disc space showing the neurovascular plane between multifidus and longissimus muscles (arrow). B, Intraoperative photomicrograph showing the lateral aspect of the multifidus muscle and the neurovascular bundle (arrow). C, Intraoperative photograph and D, corresponding illustration showing the technique of manual blunt dissection to the surgical target site. E, Photographs of spine model with the surgical target site in red and black hash lines. F, The view of the surgical target site through a minimally invasive retractor.

FIGURE 56–5 The minimally invasive spine transforaminal lumbar interbody fusion decompression is shown. A, Photographs of a spine model with removal of the inferior articular process of the cephalad level and overhang of the superior articular process of the caudad level is shown. B, Corresponding view through a minimally invasive retractor. C, Intraoperative photomicrograph of the dural tube with a Penfield probe reaching to the contralateral side. D, Postoperative CT scan with overlay of the surgical corridor.

FIGURE 56–6 The minimally invasive spine transforaminal lumbar interbody fusion discectomy is shown. A, Illustration and B, corresponding photomicrograph of the surgical target site after facetectomy to expose Kambin triangle, which is bounded medially by the dural tube, proximally by the exiting nerve root and distally by the pedicle and superior endplate of the caudad vertebra. C, Photographs of a spine model after facetectomy. Kambin triangle is colored in red. D, Corresponding view through a minimally invasive retractor. E, Intraoperative image during discectomy showing a paddle distractor and sizer and F, during cage insertion.

The clinical safety and efficacy of this technique has been well established. Schwender and colleagues ³¹ reported on 49 patients who underwent MIS TLIF through a paramedian, muscle-sparing approach using an expandable tubular retractor system. Of these patients, 26 patients had degenerative disc disease (DDD) with herniated nucleus pulposus (HNP), 22 had spondylolisthesis, and 1 had a Chance-type fracture as their primary diagnosis. The minimum follow-up was 18 months with a mean follow-up of 22.6 months. Operative time averaged 240 minutes (110 to 310 minutes), and average estimated blood loss (EBL) was 140 mL (50 to 450 mL). No patients required a blood transfusion, and there were no intraoperative complications. Length of hospital stay was 1.9 days on average (1 to 4 days). All 45 patients who had preoperative radicular symptoms had resolution of their symptoms. All patients with mechanical low back pain (LBP) had postoperative improvement of their pain. Four complications were noted postoperatively (two from malpositioned screws, one from graft dislodgement causing new radiculopathy, and the last from radiculopathy caused by contralateral neuroforaminal stenosis). Visual Analog Scale pain scores improved from 7.2 to 2.1, and Oswestry Disability Index scores improved from 46% to 14% at last follow-up.

Numerous studies have since confirmed the safety and efficacy of this technique.³²^–⁴⁰ These studies show that MIS TLIF can achieve results comparable with traditional open techniques but with less postoperative pain, decreased blood loss, and shorter hospital stays, particularly when compared with anterior-posterior circumferential fusion.⁴¹

Percutaneous Pedicle Screw Instrumentation

Insertion of pedicle screws through a midline approach requires massive retraction of the multifidus muscle, subjecting the muscle to high retraction pressures and disruption of its osseo-tendinous attachments and neurovascular supply. The rationale for MIS pedicle screw insertion lies in the preservation of multifidus muscle function. Pedicle screw insertion can be performed percutaneously or via a paramedian mini-open technique. With the percutaneous technique, the pedicle is entered using a Jamshidi-type trocar needle under fluoroscopic control (Fig. 56–7). Once the needles are within the pedicle, the stylets are removed and guidewires inserted. The guidewire is then used to direct cannulated taps and screws into the pedicle (Fig. 56–8). Sequential soft-tissue dilators are used to create a path for the tap and screw. The outermost dilator can be used as a protective sleeve during pedicle tapping. A cannulated pedicle screw is then placed over the guidewire. Rods are inserted percutaneously to minimize soft tissue trauma.

FIGURE 56–7 Percutaneous pedicle screw insertion requires scrupulous intraoperative imaging. A, E, I, The trocar needle is inserted using a perfect en face anteroposterior image of the pedicle. The endplates should be flat, and the top of the pedicle is in line with the superior endplate. The needle is inserted from a lateral to medial direction until the tip reaches the medial border of the pedicle (I and J). The C-arm is then brought under the table for a lateral view (B, F, J). If the needle is correctly inserted into the pedicle, the tip should be past the posterior vertebral body line (K and L). A-D, Shows the initial position of the needle in the anteroposterior, lateral, and axial planes, respectively. E-H, shows the needle halfway across the pedicle. I-L, Shows the needle at the medial border of the pedicle. Once past the posterior vertebral body line, the needle can be inserted another 5 mm, in preparation for insertion of the guidewire.

FIGURE 56–8 Percutaneous pedicle screw insertion uses guidewires to guide the taps and screws through the pedicle and into the vertebral body in a reliable manner. A, Once the Jamshidi needle is safely passed through the pedicle and into the vertebral body, the inner stylet is removed and the guidewire inserted. The cannula is then removed and soft tissues opened with serial dilators (B). The tap is then passed over the guidewire and into the dilator sleeves, which is monitored on lateral fluoroscopic image (C). It is critical to maintain control of the guidewire to avoid inadvertent advancement into the abdominal cavity. After tapping, cannulated pedicle screws with extension sleeves are inserted over the guidewire (D). Again, lateral image is used to monitor the guidewire and to insert the tulips to a consistent depth. Once all the screws are in place, a contoured rod is inserted under the dorsolumbar fascia and through the rod sleeves (E). The rod sleeves are then used to reduce the rod to the tulips and insert the locking caps (F).

In the mini-open technique, a longitudinal, paramedian incision is placed slightly lateral to the lateral edge of the pedicles. Dissection is performed through the intermuscular plane between the multifidus and longissimus muscles. A tubular retractor system is subsequently deployed after tissue dilation is performed. The pars interarticularis and the mammillary processes of the cephalad and caudal levels are exposed with gentle electrocautery. A high-speed bur to create a starting point and pedicle probes are used to enter the pedicle. Cannulated or noncannulated pedicle screws can be used with this technique. The exposure allows for decortication of the pars, facet joint, and transverse processes for bone grafting and fusion. A standard rod and end caps can then be placed.

The mini-open technique offers several advantages over the percutaneous method. It allows for direct visualization of the anatomy and the choice of using either cannulated or noncannulated pedicle screw systems. The mini-open technique also allows for greater access for bone grafting posteriorly. On the other hand, the mini-open technique threatens the medial branch of the dorsal rami, which extends downward to the transverse process of the caudal level, where it passes between the mammillary and accessory processes. It then curves posteriorly, where it branches to supply the multifidus muscle, the intertransverse muscles and ligaments, and the facet joint of the cephalad level. As a result, insertion of a pedicle screw through the mammillary process at one level can cause injury to the medial branch nerve of the dorsal rami (MBN) that supplies the adjacent cephalad level. In a cadaveric study comparing these MIS techniques, Regev and colleagues found that the mini-open technique causes injury to the medial branch of the dorsal rami more frequently than the percutaneous technique.⁴² They recommend that pedicle screw insertion at the cephalad level be performed percutaneously if one desires to minimize denervation of the multifidus complex at the cephalad adjacent level.

The overall safety and accuracy of minimally invasive pedicle screw insertion has been shown in several studies. Ringel and colleagues ⁴³ assessed a total of 488 pedicle screws implanted in 103 patients via a percutaneous technique. They found that only 3% of screws were rated as unacceptable, leading to nine screw revision surgeries. These results mirror a growing body of evidence that reflects the safety and efficacy of minimally invasive posterior spinal instrumentation.^31,^44,⁴⁵ These results are comparable with pedicle screws inserted via a traditional open approach. In a meta-analysis of 130 studies and 37,337 pedicle screws placed, the overall screw accuracy was 91.3%.⁴⁶

Limitations and Drawbacks

Radiation Exposure

Several techniques for minimally invasive posterior screw insertion exist, but the percutaneous pedicle screw technique is the least tissue disruptive and is currently adapted for single or multilevel fusions. Its use, however, depends on intraoperative fluoroscopy. In the past, fluoroscopy was mainly used for lateral fluoroscopic pedicle screw guidance in open surgery. However, multiplanar fluoroscopic techniques are necessary for minimally invasive spinal instrumentation. Obtaining multiple views in several planes increases accuracy but increases operating times. The operative time for two screws on the same vertebra level reaches 10 minutes or longer using advanced fluoroscopic techniques, whereas lateral-only fluoroscopic methods require less than 5 minutes per level.⁴⁷^–⁴⁹ With increased insertion times associated with advanced fluoroscopic guidance, the cumulative exposure to radiation increased concomitantly.

Studies have shown that fluoroscopically guided pedicle screw placement exposes surgeons to 10 to 12 times the dose of radiation required when compared with nonspinal musculoskeletal procedures.⁵⁰ Despite these concerns, the convenience of the C-arm, combined with a high degree of accuracy, has made intraoperative fluoroscopy an increasingly necessary part of advanced spinal surgery. The addition of navigation technology is a promising means of decreasing radiation exposure to the surgical team. Kim and colleagues⁵¹ showed that the use of navigation-assisted fluoroscopy for MIS TLIF markedly decreases direct exposure to radiation. In addition to reducing radiation exposure, navigation eliminates the need for cumbersome protective lead gear and clears the surgical field by removing the C-arm during surgery.

Learning Curve for Minimally Invasive Spinal Surgery

The barriers to widespread adoption of minimally invasive techniques appear to be related to technical difficulties of the procedures and a lack of adequate training opportunities. In a survey of spinal surgeons, Webb and colleagues ⁵² showed that most surgeons perceive MIS to be efficacious and want to perform more MIS procedures. However, most have not pursued MIS surgery because of concerns with technical difficulties of the procedure and a lack of adequate training opportunities. The technical difficulty of the procedure, combined with inadequate training, is evident in initial studies of MIS surgery. Nowitzke evaluated the learning curve for tubular decompression and noted that 3 of the first 7 but none of the subsequent 28 cases required conversion to open.⁵³ Villavicencio and colleagues⁴¹ noted a higher rate of overall perioperative complications, Dhall and colleagues⁵⁴ found a higher rate of instrumentation complications, and Peng and colleagues⁵⁵ noted longer operative times when comparing MIS TLIF versus open TLIF.