The lateral transpsoas approach to the lumbar spine has become an increasingly popular method to achieve fusion. Although this approach requires less tissue dissection, a smaller incision, decreased operative time, reduced blood loss and postoperative pain, and shorter hospital stay, it carries the potential for serious neurologic and visceral complications. This article reviews these complications in detail and proposes mechanisms for their avoidance.
Key points
- •
The lateral transpsoas approach to the lumbar spine employs a true lateral position to laterally approach the midposition of the treatment disc through the psoas major muscle using fluoroscopy and tube dilators.
- •
The advantages to this approach include smaller incisions, less tissue dissection and blood loss, shorter operative time and hospital stay, and reduced postoperative pain.
- •
The main disadvantage is the fact that common fusion levels, particularly L4-5 and L5-S1, are often inaccessible.
- •
This approach carries a unique set of complications. The most significant of these can be divided into approach-related (eg, lumbar plexus injury, genitofemoral nerve trauma, psoas weakness, retroperitoneal hematoma, posterior abdominal wall hernias) and instrumentation-related (eg, graft subsidence, vertebral body fracture, pseudoarthrosis).
Introduction: nature of the problem
The minimally invasive lateral transpsoas approach to the lumbar spine, also known as extreme lateral interbody fusion (XLIF) or direct lateral interbody fusion (DLIF), has become an increasingly popular approach for achieving interbody fusion over the past decade. This approach differs from other interbody fusion techniques in many ways. Instead of the prone or supine position, the lateral transpsoas technique employs a lateral decubitus position. The approach then utilizes a retroperitoneal dissection followed by splitting of the psoas muscle to gain access to the lateral aspect of the spine.
Reported advantages of this technique include a smaller incision and less blood loss compared with open procedures, leading to decreased operative times and shorter hospital stays as well as less postoperative pain. Because the procedure utilizes a lateral (rather than anterior) retroperitoneal corridor, it also offers less risk of injury to peritoneal contents and the hypogastric sympathetic plexus when compared with more anterior minimally invasive approaches. Furthermore, the XLIF/DLIF technique has been shown to significantly improve regional, segmental, and global coronal balance in patients with degenerative lumbar disease and has been proven to be a feasible technique for achieving fusion in adult degenerative scoliosis. In an in vitro setting, the direct lateral approach has been proven to be biomechanically equivalent to the anterior approach.
Despite these advantages, the DLIF/XLIF technique carries a unique set of complications with the potential for significant neurologic morbidity. Because the technique differs from the others, mainly in its lateral transpsoas approach to the spine, the most significant complications of the technique are approach-related. Hardware- and instrumentation-related complications are also possible (as with any interbody fusion technique), and these will also be discussed in this article.
Introduction: nature of the problem
The minimally invasive lateral transpsoas approach to the lumbar spine, also known as extreme lateral interbody fusion (XLIF) or direct lateral interbody fusion (DLIF), has become an increasingly popular approach for achieving interbody fusion over the past decade. This approach differs from other interbody fusion techniques in many ways. Instead of the prone or supine position, the lateral transpsoas technique employs a lateral decubitus position. The approach then utilizes a retroperitoneal dissection followed by splitting of the psoas muscle to gain access to the lateral aspect of the spine.
Reported advantages of this technique include a smaller incision and less blood loss compared with open procedures, leading to decreased operative times and shorter hospital stays as well as less postoperative pain. Because the procedure utilizes a lateral (rather than anterior) retroperitoneal corridor, it also offers less risk of injury to peritoneal contents and the hypogastric sympathetic plexus when compared with more anterior minimally invasive approaches. Furthermore, the XLIF/DLIF technique has been shown to significantly improve regional, segmental, and global coronal balance in patients with degenerative lumbar disease and has been proven to be a feasible technique for achieving fusion in adult degenerative scoliosis. In an in vitro setting, the direct lateral approach has been proven to be biomechanically equivalent to the anterior approach.
Despite these advantages, the DLIF/XLIF technique carries a unique set of complications with the potential for significant neurologic morbidity. Because the technique differs from the others, mainly in its lateral transpsoas approach to the spine, the most significant complications of the technique are approach-related. Hardware- and instrumentation-related complications are also possible (as with any interbody fusion technique), and these will also be discussed in this article.
Therapeutic options and surgical technique
The lateral transpsoas procedure differs from anterior lumbar interbody fusion (ALIF), posterior lumbar interbody fusion (PLIF), and transforaminal lumbar interbody fusion (TLIF) in several important aspects ( Table 1 ).
| ALIF | PLIF | TLIF | DLIF/XLIF | |
|---|---|---|---|---|
| Access | Open or laparoscopic | Open or minimally invasive | Open or minimally invasive | Minimally invasive |
| Approach | Anterior abdominal (retroperitoneal or transperitoneal) | Midline posterior incision with laminectomy/laminotomy and nerve root retraction | Offset posterior incision with access through intervertebral foramen | Lateral retroperitoneal approach to anterior spine with specialized retractors |
| Advantages |
|
|
|
|
| Drawbacks |
|
|
|
|
While ALIF utilizes supine positioning with an anterior abdominal approach and PLIF, and TLIF uses prone positioning with posterior approaches, DLIF/XLIF is unique in that lateral decubitus positioning is used for a true lateral retroperitoneal approach to the spine. Neurologic monitoring, especially electromyography (EMG), is then employed via placement of electrodes that correspond to the L2-L5 myotomes with stimulation to confirm adequate twitch strength. This allows for accurate reproducible EMG recordings, which are mandatory throughout the DLIF/XLIF procedure, because the psoas muscle-splitting approach exposes the lumbar plexus to potential injury.
After positioning and initiation of neurologic monitoring, a lateral radiograph is obtained to confirm a truly lateral position and to center the planned incision over the treatment level. An incision is made on the lateral aspect of the abdomen directly over the spine, and blunt dissection is used to identify a retroperitoneal corridor to the psoas muscle. A series of tubes and dilators are then used to direct the surgeon to the midposition of the treatment disc, taking care not to enter the peritoneal space, and the initial dilator is inserted through the psoas muscle. At this point, the fibers of the psoas muscle are separated using a system of sequential dilators, and the neural monitoring system is used to evaluate proximity to the lumbar plexus via a stimulator electrode. Stimulation typically localizes the lumbar plexus to the posterior and inferior quadrants of the dilator tubes, so the safest docking site after continued dilation to enter the disc space is a position slightly anterior to the midpoint of the disc. After the final dilator is placed, a retractor is introduced and fixed to the operating room table and then opened to reveal the disc space ( Fig. 1 ). Neural monitoring should then be checked again to ensure that the lumbar plexus is not being stretched across the operative field.
After the index disc has been accessed, it can be incised and removed utilizing fluoroscopy to ascertain the appropriate resection depth. After disk resection and end plate preparation, an appropriately sized interbody graft is placed. Due to the lateral approach and access, which obviates dural and nerve root retraction, DLIF/XLIF typically allows for a larger interbody implant than either PLIF or TLIF. If needed, segmental instrumentation can then be placed percutaneously from a posterior approach, which allows for an overall minimally invasive operation ( Fig. 2 ).
The DLIF/XLIF approach is most commonly used for the treatment of a single disc level, although multilevel procedures are certainly feasible. The main limitation to this approach is inaccessibility of the most caudal segments. The presence of the sacrum and pelvis renders the L5-S1 disc space inaccessible from a lateral corridor, and the L4-5 interspace can be similarly obscured nearly half the time. Additionally, approaching a lumbarized sacrum has been described as a relative contraindication to this approach.
Clinical outcomes
Most DLIF/XLIF procedures are performed for degenerative conditions, including spondylolisthesis, herniated disc, degenerative disc disease, postlaminectomy kyphosis, adjacent segment disease, and degenerative scoliosis. There are rare case reports of the procedure being used for treatment of spinal osteomyelitis or metastasis.
Several large series detailing the outcomes and complications of this approach have been published recently, the majority of these being retrospective reviews involving procedures performed at 1 to 2 levels with supplemental posterior element fixation. Rodgers and colleagues assessed both patient outcomes and fusion rates in 66 patients after surgery. In 1-year follow-up, 96.6% of levels were judged as fused on computed tomography (CT) scan, and nearly 90% of patients responded as “satisfied or very satisfied” on a written survey. Ozgur and colleagues reported a 91% fusion rate and a favorable change in Oswestry Disability Index (ODI) in 75% of patients at 2 years after undergoing XLIF.
There have also been multiple retrospective analyses of postoperative radiographic outcomes, including both local segmental parameters and global sagittal and coronal balance. Through restoration of disc height via placement of a large interbody graft, it has been proposed that DLIF/XLIF can expand the intervertebral foramen and thus indirectly decompress the nerve roots. Olveira and colleagues analyzed such indirect decompression in 15 patients undergoing XLIF for degenerative disc disease and stenosis. Using plain radiographs and magnetic resonance imaging (MRI), the authors demonstrated substantial dimensional improvement in disc height, foraminal height, and foraminal area, thus implying adequate neural decompression.
In recent years, the indications for XLIF/DLIF have expanded to include treatment of degenerative spinal deformity, which tends to involve several levels of correction and fixation. In 2008, Anand and colleagues demonstrated the feasibility of XLIF for this indication in 12 patients with a mean of 3.64 corrected segments achieving an average of 13° of correction per patient. Their 2-year outcomes demonstrated a 100% fusion rate with maintenance of immediate postoperative correction.
Karikari and colleagues analyzed both radiographic and clinical outcomes in 22 patients treated with lateral interbody fusion for a variety of conditions (including degenerative scoliosis, tumors, thoracic disc herniations, and vertebral osteomyelitis). In those treated for degenerative scoliosis, the mean coronal Cobb angle went from 22° preoperatively to 14° postoperatively, and a substantial clinical benefit was observed in 95.5% of patients.
In a similar fashion, Dakwar and colleagues reported 25 patients who underwent XLIF for thoracolumbar degenerative deformity. Although they found no significant correction in sagittal balance at 11-month follow-up, clinical outcomes were acceptable, and there were minimal long-term complications. Wang and colleagues published a similar series in which a mean sagittal correction of 20° was obtained, and a fusion rate of over 97% was achieved at 2-year follow-up.
Acosta and colleagues performed a thorough radiographic analysis of the changes in coronal and sagittal plane alignment following XLIF for degenerative scoliosis and noted favorable results for correction of both parameters. It was concluded that at 2-year follow-up, the direct lateral transpsoas approach (when combined with posterior segmental fixation) resulted in significant improvement in segmental, regional, and global coronal plane alignment. Although there were no statistically significant changes in global sagittal alignment or lumbar lordosis, there was a significant improvement in ODI at 2-year follow-up.
Complications and concerns
The DLIF/XLIF technique carries a unique set of complications due to the lateral transpsoas approach utilized to access the spine. These approach-related complications include neurologic morbidity from damage to the nearby lumbar plexus, psoas weakness, and retroperitoneal complications such as hematomas and surgical hernias ( Table 2 ).
Approach-Related Complications
The incidence of approach-related complications reported in the literature seems to vary significantly. The largest series of XLIF procedures was reported by Rodgers and colleagues, with 600 patients and an exceedingly low (1%) approach-related complication rate and a nearly equally low (6.2%) overall complication rate. They additionally found that there was no significant difference in complication rate with obese patients. Such low incidence of approach-related morbidity in these large series seems to differ widely from that reported in other smaller series published around the same time (see Table 2 ). This is likely because of differences among authors with regards to methods of reporting and tracking these complications. By far the most common reported complications to the XLIF/DLIF technique are ipsilateral anterior thigh pain and psoas major muscle weakness, both of which are likely related to splitting the psoas muscle fibers upon approaching the spine and/or minor trauma to the lumbar plexus. Anand and colleagues reported these symptoms in nearly 75% of patients, while Cummock and colleagues (when specifically analyzing postoperative thigh symptoms) found them in nearly two-thirds of patients. Other authors have broken down postoperative thigh/psoas symptoms into specific components (pain vs weakness vs numbness) and reported them as separate incidences. Moller and colleagues reported a 36% rate of immediate postoperative psoas weakness, with 25% and 23% incidences of anterior thigh numbness and pain, respectively.
Of note, most patients who suffer from psoas and thigh-related approach complications seem to have resolution of these symptoms in long-term follow-up. Moller and colleagues noted 84%, 69%, and 75% complete resolution of psoas weakness, anterior thigh numbness, and pain, respectively, at 6-month follow-up. Pumberger and colleagues tracked the course of these complications at 6-week, 12-week, 6-month, and 12-month intervals and found that the incidence of anterior thigh pain, anterior thigh numbness, and psoas flexion weakness tended to diminish in a nearly linear fashion over time (see Table 2 ).
Despite reporting such a low incidence of approach-related complications in their 2 large series, Rodgers and colleagues noted that, in fact, thigh pain and psoas weakness were both “nearly universal” and “always transient” in their patient population and were therefore not directly reported as approach-related complications. This likely explains the significant discrepancy of reported approach-related morbidity between their 2 large series and those of other authors.
Although less common, DLIF/XLIF also carries the risk of more permanent neurologic sequelae from its unique lateral approach (see Table 2 ). These complications are more likely caused by more severe or direct injury to the lumbar plexus, which lies underneath the psoas muscle and is at risk for traumatic injury from the surgical approach.
In a series of 58 patients treated with XLIF for degenerative conditions, Knight and colleagues reported 2 patients who suffered from ipsilateral L4 distribution motor deficits that persisted 1 year postoperatively. Other studies analyzing DLIF/XLIF for degenerative conditions reported neurologic approach-related sequelae that were transient.
When utilized for degenerative scoliosis, however, the incidence of longer-lasting postoperative lumbar radiculopathy or plexopathy seems to be higher. Out of a series of 23 patients treated for adult spinal deformities, Wang and Mummaneni reported a single patient who required a long-term assistive ambulatory device due to postoperative numbness, pain, and dysesthesias ipsilateral to the approach side. Tormenti and colleagues reported 5 out of a series of 8 patients treated using DLIF/XLIF for adult degenerative scoliosis who had persistent sensory radiculopathy and a single patient with persistent motor radiculopathy.
Isaacs and colleagues studied this particular patient population further with a prospective nonrandomized analysis of 107 patients undergoing DLIF/XLIF for symptomatic adult thoracolumbar scoliosis. Similar to other studies, they found that isolated proximal hip weakness was the most common complication and was transient 86% of the time. Seven out of the 107 patients, however, suffered from severe and protracted hip weakness that persisted 6 months postoperatively.
| Study | n | DLIF/XLIF Approach Complications | Rate |
|---|---|---|---|
| Anand et al, 2008 | 12 | 3 thigh dysesthesias (resolved in 6 wk) 1 transient quadriceps weakness (resolved in 6 wk) | 4/12 (25%) |
| Knight et al, 2009 | 58 | 2 ipsilateral L4 nerve root injuries 6 irritation of femoral cutaneous nerve resulting in meralgia paresthetica 1 significant psoas muscle spasm extending hospital stay | 9/58 (15.5%) |
| Anand et al, 2010 | 28 | 17 thigh dysesthesias (resolved in 6 wk) Several patients with transient hip flexor weakness and pain 2 quadriceps palsies with weakness of vastus medialis (complete recovery) 1 intraoperative retrocapsular renal hematomas | 20/28 (71%) |
| Tormenti et al, 2010 | 8 | 6 sensory radiculopathies 2 motor radiculopathies 2 pleural effusions necessitating chest tube placement 1 bowel perforation | 6/8 (75%) |
| Dakwar et al, 2010 | 25 | 3 transient ipsilateral anterior thigh numbness | 3/25 (12%) |
| Wang et al, 2010 | 23 | 7 ipsilateral, pain, weakness, and dysesthesias 1 pneumothorax (requiring chest tube) | 8/23 (35%) |
| Oliveira et al, 2010 | 21 | 3 psoas weaknesses 1 psoas hematoma | 4/21 (19%) |
| Rodgers et al, 2011 | 600 | 1 incisional hernia 1 subcutaneous hematoma 3 quadriceps weaknesses 1 anterior tibialis weakness | 6/600 (1%) |
| Rodgers et al, 2010 | 432 | 4 nerve injury 1 incisional hernia | 5/432 (1%) |
| Youssef et al, 2010 | 84 | 1 ipsilateral psoas weakness and numbness | 1/84 (1%) |
| Isaacs et al, 2010 | 107 | 1 pleural effusion 1 kidney laceration 7 motor deficits 2 pleural cavity violations requiring chest tube 1 sensory deficit | 12/107 (11%) |
| Dakwar et al, 2011 | 568 | 10 abdominal paresis | 10/568 (1.8%) |
| Moller et al, 2011 | 53 | 19 subjective hip flexor weaknesses | 25% |
| 13 new thigh/groin numbness ipsilateral to approach | 23% | ||
| 12 new thigh/groin pain ipsilateral to approach | 23% | ||
| Cummock et al, 2011 | 59 | 37 thigh numbness 14 hip flexor weaknesses 4 knee extension weaknesses | 37/59 (63%) |
| Tohmeh et al, 2011 | 102 | 28 hip flexor weaknesses 18 upper medial thigh numbness 1 quadriceps weakness 1 dorsiflexion weakness | 27.5% 17.6% |
| Pimenta et al, 2011 | 36 | 5 psoas weaknesses | 13.8% |
| 3 anterior thigh numbness | 8.3% | ||
| 1 weakness of ipsilateral leg | 2.8% | ||
| Sharma et al, 2011 | 43 | 15 anterior thigh pains | 34.8% |
| 11 hip flexor weaknesses | 25.6% | ||
| 4 quadriceps weaknesses | 9.3% | ||
| 1 retroperitoneal hemorrhage | 2.3% | ||
| Kepler et al, 2011 | 13 | 3 transient hip flexion weaknesses 1 hypoesthesia in anterior thigh | 4/13 (31%) |
| Papanastassiou et al, 2011 | 14 | 2 injuries to contralateral psoas and neural elements | 2/14 (14%) |
| Berjano et al, 2012 | 97 | 4 transient L4 distribution weaknesses 3 transient L4 distribution numbness 9 transient thigh symptoms 1 psoas hematoma | 17/97 (18%) |
| Pumberger et al, 2012 | 235 | Sensory deficits anterior groin/thigh | |
| 6 wk: 70 | 28.7% | ||
| 12 wk: 32 | 13.1% | ||
| 6 mo: 14 | 5.7% | ||
| 12 mo: 4 | 1.6% | ||
| Pain anterior groin/thigh | |||
| 6 wk: 101 | 41% | ||
| 12 wk: 39 | 16% | ||
| 6 mo: 9 | 3.7% | ||
| 12 mo: 2 | 0.8% | ||
| Psoas mechanical flexion deficits | |||
| 6 wk: 32 | 13.1% | ||
| 12 wk: 9 | 3.7% | ||
| 6 mo: 7 | 2.9% | ||
| 12 mo: 4 | 1.6% | ||
| Lumbar plexus-related motor deficits | |||
| 6 wk: 12 | 4.9% | ||
| 12 wk: 12 | 4.9% | ||
| 6 mo: 7 | 2.9% | ||
| 12 mo: 7 | 2.9% | ||
| Sofianos et al, 2012 | 45 | 10 iliopsoas weaknesses 8 anterior thigh numbness 3 quadriceps weaknesses 3 radiculopathies 1 foot drop | 18/45 (40%) |
| Le et al, 2013 | 71 | 14 transient ipsilateral thigh numbness 2 ipsilateral iliopsoas weaknesses 2 paresthesias/radiculopathy | 14/71 (19.7%) |
| Cahill et al, 2012 | 118 | 2 femoral nerve injury | 1.7% |
| 5 abdominal flank bulges (injury to abdominal wall motor innervations) | 4.2% | ||
| Malham et al, 2012 | 30 | 5 ipsilateral leg dysesthesias 1 ipsilateral leg motor deficit 1 bowel injury | 7/30 (23%) |
Related posts:
Current Techniques in the Management of Cervical Myelopathy and Radiculopathy
Minimally Invasive Treatment of Thoracic Disc Herniations
Minimally Invasive Transforaminal Lumbar Interbody Fusion (MI-TLIF)
Percutaneous Pedicle Screw Fixation for Thoracolumbar Fractures
Miniopen Pedicle Subtraction Osteotomy
Lateral Transpsoas Lumbar Interbody Fusion
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
