Percutaneous and Minimally Invasive Approaches to Decompression and Arthrodesis of the Thoracolumbar Spine




Summary of Key Points





  • Minimally invasive surgery (MIS) has become increasingly popular among both spine surgeons and patients.



  • Since the early 2000s, MIS technology (i.e., retractors, instrumentation, interbody cages, pedicle and facet screws) has advanced at a rate that has exceeded the literature on the topic.



  • The fundamental premise of MIS surgery is that it is better for the patient because it reduces the amount of tissue trauma associated with open procedures.



  • Short-term results indicate a benefit for patients following decompression and fusion surgery in regard to narcotic use and hospital stays.



  • There is a paucity of articles that define long-term outcomes.



  • Studies have demonstrated that open midline spine approaches are associated with paraspinal muscle damage, and proponents of MIS surgery use this as a springboard to promote MIS techniques.



Although still considered a burgeoning field, minimally invasive surgery (MIS) of the spine boasts a storied history spanning several decades. Since its inception, many MIS techniques have been developed to treat a wide variety of spinal pathologic conditions. The history of this field, the philosophy and rationale for its use, and an overview of clinical outcomes achieved are thoroughly covered elsewhere in this book. Additionally, Chapter 70 in this book is devoted to MIS of the cervical spine. In this chapter we examine surgical approaches limited to the lumbar spine and the thoracic spine, and we analyze their clinical outcomes.




Lumbar Spine


Dorsal Approaches to Arthrodesis


Minimally Invasive Posterior Lumbar Interbody Fusion


The traditional approach for posterior lumbar interbody fusion (PLIF) requires a large dorsal midline exposure with substantial manipulation of overlying ligamentous and muscular processes. When compared to other approaches to the lumbar spine, PLIF results in the highest complication rate, possibly owing to the broad exposure that is traditionally used. This rate makes the possibility of successful minimally invasive approaches particularly attractive. An increasing number of studies have examined MIS PLIF, which has been shown to decrease blood loss, hospitalization time, and pain when compared to open PLIF.


One of the first and most illustrative reports of the use of MIS to limit collateral soft tissue damage during PLIF was presented in 2002 by Khoo and colleagues. Briefly, small incisions were made between 2 and 4 cm from the midline bilaterally at the level of interest ( Fig. 90-1 ). Under fluoroscopic guidance, sequentially larger dilators were then placed over a K-wire to ultimately place a tubular retractor that provided a working corridor 15 to 20 mm in diameter. Decompression, hemilaminotomy, and discectomy were then performed endoscopically. During interbody graft placement, the authors used endoscopic visualization to ensure neural element retraction while working within the tubular system.




Figure 90-1


Posterior lumbar interbody fusion (PLIF) dilation/tubular retraction. METRx Tubular Retraction System for PLIF.

(Images provided by Medtronic, Inc., Minneapolis, MN)


Percutaneous pedicle screw-rod systems were used in concert to complete the instrumented arthrodesis. Generally (and as reported in all patients in the cited study), cannulation of the pedicles can be performed by using the same incisions that are made for the purpose of decompression after removal of the tubular retractor and endoscope. Fluoroscopy was utilized while a Jamshidi needle was advanced to the pedicle and subsequently swapped for a K-wire using the Seldinger technique. In lieu of traditional fluoroscopy, a fluoroscopic navigation system was used to track the needle in the sagittal, coronal, and axial planes. A tap was passed over the K-wire, and the multiaxial pedicle screws (Sextant, Medtronic Sofamor Danek, Memphis, TN) were attached to screw extender sleeves and passed over the K-wire into the screw pathway. The remainder of the screw-rod assembly involved mating of the ipsilateral pedicle pairs using Sextant precontoured rods.


Clinical data from this series of three patients undergoing L4-5 MIS PLIF for treatment of bilateral L5 radiculopathies clearly demonstrate the steep learning curve necessary for the MIS procedure. The first operative case took 6 hours and 15 minutes to complete, but time spent in the final case had been “whittled down” to 4 hours and 30 minutes. Furthermore, blood loss decreased from 208 to 110 mL, as did the length of hospital stay for each patient (from 5.4 to 3 days). The authors cited 100% fusion at the 8-month follow-up and an observational decrease in the amount of morphine equivalence units with the MIS approach when compared to previous unpublished open PLIF data.


Another clinical series in 2002 reported positive outcomes in 10 patients who underwent MIS PLIF with pedicle screw fixation. Again, operative time was found to be greatest during the familiarization period and declined with time. Half of the patients were discharged on postoperative day 1 or 2, the remainder being discharged on the third postoperative day. Solid fusions were documented in all patients at 13.8 months of mean follow-up time. In 2007, Park and Ha compared 32 patients who had been treated with traditional one-level PLIF with 29 who had undergone MIS PLIF. Consistent with the aforementioned comparison, there was no statistical difference in fusion rates at the 1-year minimum follow-up. The MIS group did show statistically significant differences in postoperative and intraoperative metrics. These included decreased blood loss, postoperative pain, recovery time, and hospital stay.


Interestingly, another study in 2009 examined the MIS placement of percutaneous pedicle screws using a miniature robotic system. In this cohort, 31 patients received a PLIF with decompression, discectomy, and polyetheretherketone (PEEK) cage implantation along with percutaneous pedicle screw fixation. The SpineAssist (MAZOR Surgical Technologies Ltd., Israel) system was utilized, in which an image-correlated framing system provided an edifice for spinous process clamping and screw guidance. The authors used a modified frame to support pedicle screw fixation and computer-based trajectories. In 29 of 31 cases, seamless percutaneous pedicle screw placement was achieved, with deviations from surgical trajectories of less than 2 mm in the majority of cases. Although this study was an interesting proof of concept and demonstrated precision in percutaneous screw placement, it utilized a modified MIS approach and did not provide outcomes for the patient population. Further work with this or similar guided systems may be of interest in the future to expand MIS PLIF approaches.


In 2011, Kim and associates reported the clinical effectiveness on MIS PLIF for multilevel fusions. They described 42 patients who underwent MIS PLIF using the percutaneous screw fixation system while assessing the outcomes of the Low Back Outcome Score (LBOS) and visual analog score (VAS) and other secondary outcomes such as radiologic fusion and postoperative complications.


In this study, the authors used expandable tubular retractors at each level to perform a central decompression and foraminal decompression on the contralateral side. The authors reported no failed fusion or neurologic deficits during their follow-up period. Of the 42 patients in this study, 32 underwent fusion of two levels and 10 had a three-level fusion. These patients had a mean follow-up of 25.3 months and a mean increases of LBOS and VAS. The authors consequently concluded that the vertical axis and detachable screw extender for multilevel PLIF effectively and safely improves patient outcomes with multilevel disease.


Additionally, in 2012, Cheung and coworkers reported the long-term outcomes of patients who were treated with open PLIF versus those who underwent MIS PLIF. Of these 52 patients, 43 were followed for a minimum of 3 years with 22 undergoing PLIF and 21 undergoing MIS PLIF. Each patient completed a visual analog scale and the Short Form (36) Health Survey (SF-36) quality of life measurement and had repeat radiographic imaging. At 12 months follow-up, the MIS PLIF group had less listhesis than the PLIF group, but this difference was no longer seen at long-term follow-up. In both groups, 100% fusion rates were reported, whereas both groups also reported improved SF-36 quality of life scores at 12 months and at long-term follow-up. Thus, the authors concluded that both PLIF and MIS PLIF are viable options for treating spondylolisthesis and can sustainably improve physical function and decrease pain over a long follow-up period.


Transforaminal Lumbar Interbody Fusion


As a variation of dorsal lumbar fusion, transforaminal lumbar interbody fusion (TLIF), utilizes a far lateral, or transforaminal, approach that can be applied throughout the lumbar spine. Additionally, complications related to neural element retraction are minimized in comparison to PLIF as TLIF does not necessitate retraction of the traversing nerve root during the discectomy and cage/graft placement. Furthermore, TLIF is not limited to levels L3-4 and below, as is the case with PLIF. When compared to MIS PLIF, MIS TLIF uses a similar but more lateral approach, giving way to an eventual facetectomy, along with complete resection of the inferior and superior articulating facets of the inferior and superior vertebrae, respectively. Discectomy, preparation of end plates, pedicle/facet screw placement, and interbody graft placement are then performed in the usual manner.


Schwender and colleagues first reported the results of 49 patients who had undergone TLIF with percutaneous pedicle screw placement. Fusion was reported in all patients at a minimum follow-up of 18 months, and narcotic use was discontinued on average between 2 and 4 weeks postoperatively. Other clinical indicators were similarly favorable, with an average blood loss of 140 mL and an average Oswestry Disability Index (ODI) score that dropped from 46 preoperatively to 14 at last follow-up. Jang and Lee also presented a cohort of 23 patients undergoing MIS TLIF with both ipsilateral pedicle screw and contralateral facet screw fixation. In this series, 21 of 24 patients achieved fusion at follow-up (mean of 19 months), with an average blood loss of 310 mL and ODI scores dropping from 33.1 to 7.6 after surgery.


In 2009, Peng and colleagues presented one of the most comprehensive comparisons of MIS versus open spine surgeries, comparing 29 MIS TLIF procedures to 29 open TLIF procedures. The benefits of MIS TLIF were most pronounced in the immediate postoperative period and included less blood loss, a shorter hospitalization period, diminished postoperative pain, and subsequent lower analgesic use ( P < .05 for all parameters). However, the MIS procedure also required longer operative and fluoroscopic exposure time. Importantly, follow-up at 6 months and follow-up at 2 years were statistically equivalent when quality-of-life metrics and fusion rates were compared. The authors concluded that MIS TLIF had “retained” positive long-term outcomes associated with TLIF but with fewer immediate postoperative complications (13.8% versus 6.9%, open versus MIS). It should be noted that three quarters of complications in the open group likely arose from immobility (two urinary tract infections and one case of atelectasis) and were not directly attributable to the surgical procedure.


In 2014, Tumialán and coworkers investigated the utility of applying an expendable provisional to MIS TLIF with a primary outcome of its ability to revive disc height in 30 patients. The authors analyzed radiographic data in preoperative, intraoperative, and postoperative patients undergoing one- to two-level MIS TLIF while using an ipsilateral expandable provisional. The device was tested in the first three cases using an open midline incision, and the following 27 cases were performed using minimally invasive expandable retractors.


The authors used four low profile pedicle screws in both open and MIS cases before decompression was preformed. All of the 30 patients in this series revived a minimum follow-up of 6 months with a mean follow-up of 9.4 months with no evidence of interbody migration in any case. Additionally, 27 of 30 patients showed radiographic fusion within the interbody device without instability with flexion or extension. Thus, the authors concluded that a provisional ipsilateral rod is a viable option to achieve and maintain increased disc height correction through an MIS TLIF technique.


Ventral and Lateral Approaches to Arthrodesis


Lateral Interbody Fusion


In 2004, Bergey and colleagues described an endoscopic lateral transpsoas approach to the lumbar spine in 21 patients undergoing discectomy and fusion. A modification of McAfee’s endoscopic approach, the true lateral approach provides a retroperitoneal approach to the lumbar spine without trespassing into the peritoneum or dissecting the great vessels. The exposure typically requires the patient to be in the right lateral decubitus position ( Fig. 90-2 ). An incision is made along the lateral border of the lumbar paraspinal muscle, and a finger dissection is performed through the retroperitoneal space to the psoas. Fluoroscopy is used to determine the proper starting point along the flank, and a counterincision is made directly lateral to the disc space(s) of interest. The dilators are guided to the psoas using the initial incision, and a corridor is developed through the psoas to the disc space of interest. Continuous electromyogram monitoring is utilized to avoid injuring the exiting lumbar nerve roots, which also pass through the psoas muscle. A final expandable retractor is then placed over the final dilator, providing adequate access to the disc space. The discectomy and placement of the interbody graft are then followed by placement of a side plate or dorsal fixation of some kind (i.e., pedicle screws, translaminar or facet screws).




Figure 90-2


Extreme lateral interbody fusion procedure.

A, Incision over the psoas. B, Sequential dilation through the retroperitoneal space. C, Insertion of the implant.


Although a lateral approach decreases the likelihood of complications arising from anterior lumbar interbody fusion, it presents a unique set of challenges. First, it is essential to note the location of the lumbosacral plexus, as it migrates ventrally from L1 to L5 and is at significant risk when approaching the L4-5 interspace. Injury to the genitofemoral nerve is still possible and was noted in 30% of patients in Bergey and associates’ 2004 study. This complication was not found in a 2006 study, but staying within the ventral third of the psoas and visualizing the genitofemoral nerve has been recommended. The far lateral approach provides excellent access to the L1-4 levels and ability to restore disc height, but further dissection is necessary to expose any caudal disc spaces and may even require removal of a portion of the iliac crest. Clinical outcomes of extreme lateral interbody fusion are also notable; one study showed MIS extreme lateral interbody fusion, unlike traditional open lumbar procedures, to be no more likely to result in complications in the obese patient.


In 2013, Watkins and colleagues compared anterior, lateral, and transforaminal approaches and their ability to correct sagittal deformities. In this study, 309 operative levels on 220 patients were analyzed after MIS anterior lumbar interbody fusion (ALIF) (184 levels), MIS lateral lumbar Interbody fusion (LLIF) (84 levels), and MIS TLIF (39 levels). Patients were followed for an average of 585 days postoperatively with sequential radiographs measuring segmental lordosis, disc height, and positive spondylolisthesis. Comparison of preoperative and postoperative radiographs of all three surgical approaches revealed they all significantly reduced spondylolisthesis, whereas ALIF had improved rates of lordosis reduction when compared to LLIF and TLIF. Additionally, they found there was no statistically significant difference in lordosis restoration between LLIF (2.2 degrees) versus TLIF (0.8 degrees). The authors postulated that this finding is a result of their inability to completely remove the anterior longitudinal ligament during LLIF and TLIF. They also demonstrated a statistically significant difference in improved disc height with LLIF (2.2 mm) and ALIF (2 mm) compared to TLIF (0.5 mm). Thus, the authors concluded that although all three techniques (ALIF, LLIF, and TLIF) improve spondylolisthesis, considerations for surgical approach can have important implications for disc height restoration and lordotic reduction.


In 2013, Ahmadian and colleagues investigated the utility of MIS LIF for patients with L4-5 spondylolisthesis. In this series, 31 patients with grade I (26) and grade II (5) listhesis underwent extreme lateral interbody fusion (XLIF). The patients in this series had a mean blood loss of 94 mL, an average hospitalization time of 3.5 days, and a mean follow-up of 18.2 months. After 6 months, all patients demonstrated radiographic and clinical evidence of fusion with plain radiographs or computed tomography (CT) scan. Additionally, every patient experienced improvement of anterolisthesis, whereas 27 of 31 patients had complete resolution of deformity. However, 4 patients had postoperative regression of listhesis. Improvement of clinical outcomes of related disability (ODI), pain (VAS), and health-related quality of life (SF-36) indices at 6 months suggested an overall improvement of postoperative symptoms and increase in patient satisfaction. Thus, the authors concluded that MIS LIF is a safe and effective way to correct anterolisthesis, despite the necessity to navigate the lumbar plexus.


Although minimally invasive approaches have garnered praise for improved quality-of-life metrics, they are not without their fair share of complications. To investigate the rates and types of complications associated with lateral approaches, Meredith and coworkers analyzed complications in patients undergoing XLIF with particular attention to early pulmonary complication and nonpulmonary complications in addition to rates of arthrodesis. In their study, 18 patients undergoing modified XLIF had a mean estimated blood loss of 577 mL with 7 patients experiencing delayed extubation postoperatively. The most common complication encountered in these patients was pleural effusion, which usually occurred after postoperative day 2. This occurred in 8 patients requiring chest tube placement, which had a mean time to removal of 2 days and average stay of 3 days in a monitored care setting. Nonpulmonary complications included new cardiac arrhythmias in two patients and two patients with incidental durotomies, with one requiring revision. Additionally, one patient developed a fracture with pullout of instrumentation requiring revision and an extension of the fusion. Radiographic fusion was obtained in 17 patients. This study highlighted that although arthrodesis was obtained in the majority of cases, the risk of postoperative complications is substantial enough that the surgical team should be vigilant in the perioperative period.


Ventral Limited, Laparoscopic Approaches


Ventral laparoscopic approaches to the lumbar spine have fallen out of favor with the spine surgery community. As late as the mid-1990s, prospective studies demonstrated equivalent rates of complications between laparoscopic and open approaches, encouraging further study. However, in the early 2000s it was revealed that complication rates for video-assisted procedures were higher than those in open surgeries. In one series, the laparoscopic approach, utilizing techniques similar to those of abdominal laparoscopy, resulted in higher rates of retrograde ejaculation and new-onset radicular pain. In 11% of cases, the laparoscopic approach required conversion to a more open approach for reasons that included major vessel lacerations and peritoneal tears. The laparoscopic techniques also favored the L4 vertebra and below, as higher approaches were deemed even more technically demanding. Therefore, the authors of that study advocated abandonment of the laparoscopic technique for ventral lumbar fusion.


Nonlaparoscopic, MIS approaches to the lumbar spine have faired far better. In 1997, Mayer introduced a muscle-splitting approach that allowed easy access to the entire lumbar spine. After exposure of the target level, disc removal and graft placement can proceed. In his study, Mayer reported uniform fusion at follow-up, low intraoperative blood loss and postoperative morbidity, and a short recovery period. A 2004 study retrospectively compared 33 patients who had undergone the traditional extraperitoneal approach to 23 who had undergone Mayer’s minimally invasive approach. First, unlike the case with most MIS procedures, the authors reported a significantly shorter operative time when compared to the open procedure. Blood loss was significantly less in the MIS group, and complication and fusion rates were similar. In light of these findings and a similar study showing higher complication rates in the open group, it is reasonable to conclude that the MIS approach may be preferred to the traditional extraperitoneal approach. A variety of grafting options can be used for an MIS ALIF approach. The Lumbar Tapered (LT)-Cage (Medtronic Sofamor Danek, Memphis, TN) is a stand-alone tapered titanium cage that has been approved by the Food and Drug Administration (FDA) for single-level ALIF with the concomitant use of bone morphogenetic protein-2 (BMP-2) (INFUSE Bone Graft, Medtronic Sofamor Danek, Memphis, TN), with fusion rates reported at 94.5% at 12 months. Other graft options for MIS (analogous to those in open procedures) include the autogenous iliac crest corticocancellous graft, allograft (e.g., femoral ring), and PEEK, as well as carbon fiber, bioabsorbable, and other metallic interbody cages. These materials, when used as stand-alone devices (i.e., no dorsal fixation), result in suboptimal fusion rates and clinical outcomes. Therefore, it is recommended that they be used in conjunction with dorsal fixation.


To address the question of ideal grafting substrates, a 2014 study by Flouzant-Lachaniette and colleagues investigated the bone union rates in patients undergoing MIS ALIF with recombinant human bone morphogenic protein-2 (rhBMP-2) versus autologous iliac bone graft (ICBG) in PEEK cages. In this series, 51 patients underwent one (40 cases) or two (11 cases) level video-assisted MIS ALIF. In each case, radiolucent cages were inserted with two chambers. In each cage, one chamber was filled with autologous ICBG and the other with 6 mg of rhBMP-2. A CT scan was preformed in the early postoperative period and at the 1-year follow-up to assess bone union rates using thin-slice multiplanar reconstruction (MPR). At the 1-year follow-up, the rate of acquired fusion for rhBMP-2 was 71% compared to 88% in the ICBG group, whereas 4.8% experienced pseudarthrosis. Although the authors could not prove there was no interaction between the two grafts only separated by a central beam, they concluded that ICBG was superior to rhBMP-2 for inducing quality bone union in MIS ALIF with PEEK cages and that 6 mg was an appropriate dose to achieve adequate fusion.


Minimally Invasive Decompression


A thorough treatise on MIS lumbar decompression could fill several volumes. The use of laser dehydration, chemical application, percutaneous extraction, and MIS implants have been evaluated repeatedly and found to be inferior to direct decompression. Although implants may be an attractive alternative for select patients, it is still not entirely known how patient selection affects outcome, and several studies have shown high failure rates or inconsistent radiographic and clinical factors predicting clinical outcomes. Therefore, this chapter focuses on MIS surgical intervention for decompression of the thoracolumbar spine.


The approach to MIS decompression (i.e., laminotomy, laminectomy, foraminotomy) is similar to that of MIS PLIF in that muscle-splitting techniques are used with tubular retractor systems to provide targeted access to the affected vertebra. The placement of the tubular or expandable retractor relative to midline is predicated on the type of decompression being performed (e.g., ipsilateral foraminotomy versus bilateral decompression via a unilateral approach; Fig. 90-3 ). A microscope or endoscope can be used for visualization; however, the latter has ergonomic restraints, particularly in working through a narrow retractor. In 2008, Rahman and associates retrospectively compared 38 patients who had undergone MIS decompression with 126 patients who had been treated with open laminectomy. MIS procedures were associated with shorter operative times, less blood loss, shorter length of hospital stay, and fewer complications. Several other studies have also reported excellent clinical results in single-level and multilevel MIS laminotomy or midline decompression. A biomechanical study examined lumbar segments with regard to axial compression, flexion, extension, and lateral bending after intervention of isolated fenestration, bilateral decompression via unilateral approach, and medial or total facetectomy.


Feb 12, 2019 | Posted by in NEUROSURGERY | Comments Off on Percutaneous and Minimally Invasive Approaches to Decompression and Arthrodesis of the Thoracolumbar Spine

Full access? Get Clinical Tree

Get Clinical Tree app for offline access