High neurological complication rates for extreme lateral lumbar interbody fusion and related techniques: A review of safety concerns.

BACKGROUND: There are frequent reports of lumbosacral plexus and other neurological injuries occurring with extreme lateral interbody fusions (XLIF) and other related lateral lumbar techniques.
METHODS: This review focuses on the new neurological deficits (e.g. lumbosacral plexus, root injuries) that occur following minimally invasive surgery (MIS) XLIF and other related lateral lumbar techniques.
RESULTS: A review of multiple articles revealed the following ranges of new postoperative neurological complications for XLIF procedures: plexus injuries 13.28%; sensory deficits 0-75% (permanent in 62.5%); motor deficits 0.7-33.6%; anterior thigh pain 12.5-25%. Of interest, in a study by Lykissas et al., the frequency of long-term neural injury following lateral lumber interbody fusion (LLIF) with BMP-2 (72 patients) was much higher than for LLIF performed with autograft/allograft (72 patients). The addition of bone morphogenetic protein led to persistent sensory deficits in 29 vs. 20 without BMP; persistent motor deficits in 35 with vs. 17 without BMP; and persistent anterior thigh/groin pain in 8 with vs. 0 without BMP. They should also have noted the unacceptably high incidence of neural injury occurring with LLIF alone without BMP.
CONCLUSION: This review highlights the high risk of neural injury (up to 75% for sensory, 33.6% for motor, and an overall plexus injury rate of 13.28%) utilizing the XLIF and other similar lateral lumbar approaches. With such extensive neurological injuries, is the XLIF really safe, and should it still be performed?

INTRODUCTION

This review focuses on the frequency of neural injuries resulting from extreme lumbar interbody fusion (XLIF) or similar lateral procedures (e.g., direct lumbar interbody fusion (DLIF) and lateral lumbar interbody fusion (LLIF)). The multiple studies analyzed cited the following ranges of new postoperative neurological complications attributed to XLIF: new sensory deficits 0–75% (permanent 62.5%), new motor deficits 0.7–33.6%, new anterior thigh pain 12.5–25%, and a plexus injury rate of 13.28%. Because it is well-known that the transpsoas approach places the lumbosacral plexus at risk, some authors even concluded these frequent deficits should not be considered “complications,” as they were simply “anticipated” consequences of the procedure. Here, we emphasize the unacceptably high rate of neurological injuries attributed to XLIF and related lateral lumbar procedures, and question whether, with this safety record, it should still be performed?

HIGH FREQUENCY OF NEW NEUROLOGICAL INJURIES WITH XLIF AND RELATED LATERAL LUMBAR PROCEDURES

Multiple authors have reported a high frequency of neurological injuries, including anterior thigh pain, contralateral femoral nerve palsies, new sensory/motor root deficits, and lumbar plexopathies attributed to XLIF and other lateral lumbar procedures. In 2011, Sharma et al. evaluated the 1-year outcomes of 43 LLIF addressing degenerative disc disease (DDD), degenerative spondylolisthesis (DS), or scoliosis; 25% of the patients experienced new postoperative anterior thigh pain [Table 1].[11] Further, in 2011, Papanastassiou et al. observed 2 (6.25%) new contralateral femoral nerve root injuries occurring out of a series of 32 XLIF procedures [Table 1].[10] One of the latter injuries was attributed to the contralateral displacement of an endplate fragment due to “overzealous endplate removal and breaking of the osteophytes in the opposite corner of the intervertebral disc”; the second was due to a failure to pay close attention to cage placement, which extended into the opposite foramen.[10] In 2013, Ahmadian et al. reviewed 18 studies involving 2310 patients undergoing XLIF; 304 (13.2%) patients exhibited plexus injuries, while root injuries led to motor (0.7–33.6%) and sensory (0–75%) deficits.[1] Of interest, here the authors concluded neural complications for XLIF procedures were being underreported. In 2014, Hrabalek et al. analyzed the complication rates for minimally invasive (MIS) anterior lumbar interbody fusions (ALIF) (20 patients) vs. MIS XLIF (88 patients) addressing T12–L5 level pathology [Table 1].[7] MIS ALIF complications included post-sympathectomy syndrome (19 patients; 15.8%) and 35 minor intra/postoperative complications (32 patients; 26.6%). For XLIFs, 26 complications occurred in 22 patients (25%); one transient L5 root injury (1.1%) and 20 of 25 root/plexus injuries involving transient left groin/anterior thigh pain (11 patients; 12.5%) or numbness (9 patient; 10.2%). Formica et al., in 2014, evaluated the complications of 39 consecutive XLIF performed for degenerative or post-traumatic lumbar pathology over an average postoperative interval of 16 months; 10 (26%) patients exhibited “mild, transient” motor quadriceps deficits that completely resolved [Table 1].[5] Khajavi et al. in 2015 further assessed the outcomes for 160 patients (197 levels) undergoing MIS XLIF for degenerative lumbar disease [Table 1].[8] Patients experienced anterolateral thigh/groin sensory deficits (14%) and iliopsoas weakness (9%); notably, they considered these “minor complications.” In a retrospective cohort analysis of the neurological sequelae of 120 XLIF and DLIF, Cheng et al., in 2015, found that 31 (25.8%) patients experienced one or more adverse event (AEs).[2] Twenty-two patients (18.3%) experienced 24 neurological AEs; 15 (12.5%) complained of anterior/lateral thigh dysesthesias, and there were 6 (5.0%) instances of radiculopathy and 3 (2.5%) cases of postoperative weakness. Interestingly, AEs occurred more frequently following DLIF (28%) vs. XLIF (14.2%). In 2016, Epstein commented on the much higher frequency of root/plexus injuries occurring with MIS XLIF (23.8%) vs. ALIF (15.8%) reported in the Spine Patient Outcomes Research Trial by Desai et al.[34] Interestingly, these numbers far exceeded those quoted for open diskectomy (0.13% to 0.25%), open laminectomy/stenosis with/without fusion (0%), and open laminectomy/stenosis/degenerative spondylolisthesis with/without fusion (2%). Performing a retrospective review of outcomes 1 year following 108 XLIF, Grimm et al. (2016) found a 23% complication rate (25 complications); 21 (19.4%) involved thigh pain paresthesias, whereas the remaining 4 (3.7%) addressed a vertebral body fracture, a contralateral root injury, quadriceps weakness, and residual stenosis; notably 3 patients required additional surgery.[6] Interestingly, the authors also commented that some patients experienced severe femoral nerve palsies despite the use of intraoperative neurophysiological monitoring (IONM). Certainly, the overall high frequencies of neural/plexus injuries occurring with XLIF and related lateral lumber procedures should lead to questions regarding the safety of these procedures and whether they should be continued.

Table 1

Neurological (e.g. Root and Plexus/Other Injuries) Attributed to XLIF

EXTREME LATERAL INTERBODY FUSION COMPLICATIONS INCREASED WITH THE ADDITION OF BONE MORPHOGENETIC PROTEIN-2

The study by Lykissas et al. in 2014 critically pointed out the potential direct damage of rhMP-2 to the lumbosacral plexus when utilized for LLIF [Table 1].[9] The authors recognized that bone morphogenetic proteins-2 (BMP-2) contributed to neurological deficits when used for ALIF or transforaminal lumbar interobody fusion (TLIF), however, questioned in their 6-year retrospective series, whether it would have similar adverse effects (e.g., lumbosacral plexus injuries and/or pain) following LLIF. The study included two clinically comparable controlled cohort LLIF populations; 72 LLIF utilized BMP-2 vs. 72 LLIF performed with autograft/allograft. They documented that rhBMP-2 combined with LLIF directly damaged the lumbosacral plexus uniformly in the long-term; sensory deficits occurred in 29 fused with rhBMP-2 vs. 20 without; motor deficits in 35 fused with rhBMP-2 vs. 17 without; and anterior thigh/groin pain in 8 fused with rhBMP-2 vs. 0 without. In addition, careful reassessment of the extremely high rate of neurological injury occurring secondary to the LLIF procedures alone, even with intraoperative neural monitoring, should lead to reconsideration of whether these procedures should still be offered to patients [Table 1].

COMBINING EXTREME LATERAL INTERBODY FUSION WITH OTHER PROCEDURES FOR ADULT SCOLIOSIS

XLIF procedures combined with additional procedures to address adult scoliosis increased the complication rates [Table 1].[12] To address adult degenerative thoracolumbar scoliosis, Tormenti et al. in 2010, compared the efficacy of performing 8 XLIF, TLIF, and pedicle screw fixation procedures versus 4 TLIF/pedicle screw procedures alone [Table 1].[12] Of the 8 patients undergoing XLIF/TLIF/pedicle screws, 1 sustained an intraoperative bowel injury requiring a colon resection, 2 (25%) developed new motor radiculopathies (1 resolved; 1 continued at 3 months), whereas 6 (75%) exhibited new thigh paresthesias (note 5 of 6 persisted at 10.5 postoperative months). Of the 4 undergoing TLIF/pedicle screw fixation, 1 required revision of the instrumentation. The authors rightfully concluded the combined procedures carried “significant risks that require further evaluation and proper informed consent.” However, how many surgeons inform their patients prior to XLIF with/without other procedures that they have such a substantial risk of sustaining new postoperative neurological deficits? Let us conclude it is very unlikely because it would certainly curtail the number of XLIF being performed.

CONCLUSION

This review focused on the high frequency of neural injuries resulting from XLIF and related lateral lumber interbody procedures. New neurological injuries included sensory deficits in 0–75% of cases (permanent 62.5%), motor deficits in 0.7–33.6% of patients, anterior thigh pain in 12.5–25%, and a 13.28% plexus injury rate. When is the neurological injury rate for XLIF and its related procedures considered unacceptable? And when do we question why, with this safety record, it should still be performed?

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.