Abstract
Background Lateral lumbar interbody fusion is a widely used technique to address degenerative lumbar conditions but can be associated with injury to the psoas, lumbar plexus, and abdominal wall owing to retractor usage. We describe a minimally invasive endoscopic lateral lumbar interbody fusion (ELLIF) procedure that aims to reduce these complications by avoiding prolonged muscle retraction, preparing the disc space under direct endoscopic vision, and shortening the surgical time.
Methods Between 2019 and 2024, 35 patients underwent ELLIF at a single center. Discectomy, endplate preparation, and iliac crest harvest were done via a working-channel endoscope without expandable retractors. Neurophysiological monitoring was used to minimize nerve injury. Outcomes included complications, visual analog scale scores for pain, and Oswestry Disability Index (ODI).
Results Of the 35 patients (mean age 60 years), 26 had preoperative radicular pain and 9 had neurological deficits. Six minor complications occurred in 4 patients (11.4%), all managed conservatively without permanent deficits. No patients developed new radiculopathy or paresis, and there were no infections or reoperations. ODI improved by 57% at 1 month and by 88% at 1 year (both P < 0.001). By the 3-year follow-up in 9 patients, ODI scores remained near normal, and visual analog scale was reduced by 93% from baseline.
Clinical Relevance We present a minimally invasive, ELLIF, and decompression technique that provides patients with minimal complications and excellent functional recovery.
Conclusion ELLIF offers a safe, minimally invasive alternative for patients with lumbar degenerative disease. This technique minimizes direct retraction on the psoas and lumbar plexus, resulting in a low complication rate and substantial functional recovery at short- and medium-term follow-up.
Level of Evidence 4.
Introduction
Lateral lumbar interbody fusion (LLIF) has become an increasingly popular minimally invasive surgical option for lumbar disc disease, primarily because it reduces posterior soft tissue disruption compared with traditional approaches.1 Despite these benefits, complications such as psoas muscle injury and lumbosacral plexus dysfunction can arise when using a lateral retractor system.2 These complications include numbness, weakness, and, in rare cases, more severe injuries such as vascular perforation.3,4
Endoscopic LLIF (ELLIF) has emerged as a potential means of further reducing these complications by minimizing or eliminating the need for sustained psoas retraction.4–6 Motor and sensory deficits in the distribution of the lumbosacral plexus have been documented after ELLIF; however, they are often transient postoperatively.7 We previously described a modification in which prone positioning is used, facilitating easier instrumentation and further decreasing the risk of compression neuropathies associated with lateral decubitus positioning.8 This approach also allows harvesting of an autologous iliac crest graft through a small working portal while maintaining endoscopic visualization. Interlaminar and transforaminal decompression are also possible through this approach, as well as working on the anterior ligament to improve the lordotic angle and disc space preparation under direct endoscopic vision.
In the present study, we expand our initial experience to 35 patients who underwent this ELLIF procedure at a single center. We analyze perioperative data, complication rates, and functional outcomes over follow-up intervals up to 3 years.
Methods
This retrospective study included 35 patients who underwent ELLIF at a single institution between 2019 and 2024. All patients selected for surgery had a clinical diagnosis of lumbar instability, as evidenced by radiographic findings of degenerative disc disease and instability on flexion-extension views, as well as intractable mechanical low back pain unresponsive to conservative management. Among these patients, 26 presented with radicular pain and 9 had additional pre-existing neurological deficits. Severe osteoporosis (T score ≤ –3.5), active infection, and previous lumbar fusion at the same levels were considered exclusion criteria. Institutional Review Board approval was obtained for this study (HM Hospitales IRB Codigo CElm: 25.02.2483-GHM). Informed consent was obtained from each of the 35 patients included in the study, with their information deidentified.
Surgeries were performed either in the prone position (31 patients) or by using a transiliac approach (4 patients) under general anesthesia with continuous neurophysiological monitoring. For those placed prone, the patient was positioned on a radiolucent table, ensuring that all bony landmarks of the lumbar region and pelvis were clearly visible under fluoroscopy (Figure 1). A 2- to 3-cm incision was made lateral to the disc space, avoiding the neurovascular structures and root emergence zone. The incision point coincided exactly with the point in the lateral view of the disc, concretely in the union of the anterior third and the middle third of it (Figure 2A). Blunt finger dissection was carried out to enter the retroperitoneal space, palpating the disc and working corridor. If a free trajectory was not clearly palpable through the incision, a second incision was made inferiorly in order to palpate the trajectory (Figure 2B).
(A) Lateral approach through a single incision with the patient in the prone position. (B and C) Iliac crest tunneling and graft harvesting. (D and E) Disc space approach in anteroposterior and lateral views. (F) Discectomy. (G, H, and I) Cage preparation and implantation through the original incision over a blunt-tip guidewire.
(A) The point on the lateral x-ray image of the disc used to mark the initial skin incision. (B) Second, an inferior incision made to assist in palpating the trajectory for the cannula and endoscope.
Successive dilators were then placed, culminating in a 7-mm beveled tubular retractor, through which a water-based working-channel endoscope was introduced. Under endoscopic visualization, the surgeon performed discectomy and endplate preparation; in cases where the anterior longitudinal ligament required release, this was similarly achieved under endoscopic guidance (Figure 3). Additional contralateral foraminal decompression was performed when clinically indicated.
Representative intraoperative endoscopic views. (A) Disc preparation with (B) corresponding fluoroscopy image. (C and D) Two views of endplate work, in preparation of cage placement. (E) Anterior longitudinal ligament release with (F) corresponding fluoroscopy image. (G) Iliac crest cancellous bone harvesting with (H) corresponding fluoroscopy image.
In most patients, iliac crest bone was harvested through the same skin incision. A Jamshidi needle was inserted into the iliac crest, and sequential dilators up to 16 mm were introduced to create an access channel for the working endoscope. Under direct endoscopic view, cancellous bone was obtained using endoscopic graspers and then packed into a titanium cage (Joimax EndoLIF, 35 × 14 mm). The size of the cage is guided by the successive dilators, which prepare the disc for extraction and subsequent cage placement. The point between the middle third and anterior third of the endplate is sought for optimal cage location as it is safest owing to the distance from the emerging nerve root and vascular structures anterior to the spine. This graft-filled cage was delivered through the psoas under fluoroscopic guidance by advancing it over a superelastic nitinol blunt-tip guidewire. No expandable-blade retractor systems were used during the procedure, thereby minimizing sustained retraction on the psoas and lumbar plexus. Finally, percutaneous pedicle screws were inserted using standard anatomical landmarks and fluoroscopic verification (Figure 4). An overview of the surgical procedure is depicted schematically in Figure 5.
(A) Minimally invasive, posterior pedicle screw approach over k-wires. (B and C) L5 pedicle screw placement. (D) L4 pedicle screw placement. (E and F) Final construct in lateral and anteroposterior views via intraoperative fluoroscopy. (G and H) Final construct in lateral and AP views via postoperative x-ray images. (I) Incision healing progress shown on postoperative day 20.
Schematic representation of major surgical steps in operative order.
Outcome measures included the Oswestry Disability Index (ODI) and the visual analog scale (VAS) for pain, which were administered preoperatively, immediately postoperatively, at 1 month, 3 months, 1 year, and, where possible, at 3 years. Patients did not undergo follow-up computed tomography imaging if they were clinically stable to limit cost and unnecessary radiation exposure. Statistical significance of pre- to postoperative changes was assessed using paired t tests, with significance set at P < 0.05. Postoperative complications were documented, with particular attention to distinguishing pre-existing radiculopathy or paresis from new or worsened deficits.
Results
A total of 35 patients were studied, of whom 18 were men, and the mean patient age was 60 years (range 39–85 years). All 35 patients had mechanical lumbar pain and radiographic instability; 26 patients additionally presented with radicular pain, and 9 patients had notable neurological deficits at the time of surgery (Table 1). Of the 35 patients, 31 underwent the prone approach, while 4 required a transiliac trajectory, typically due to anatomic considerations or difficulties in achieving an optimal lateral path. Tables 2 and 3 contain the full breakdown of treatment characteristics.
Characteristics of patients who underwent ELLIF (N = 35).
Specific treatment modalities utilized per patient’s individual needs within ELLIF procedures (N = 35).
On average, the time to prepare and implant the first cage was approximately 40 minutes; for the 9 cases involving a second cage, a further 25 minutes were required. Thirty-one patients had standard titanium cages placed, and 4 required expandable cages. The series encompassed both single- and 2-level fusions, determined by the extent of degenerative disease and instability, with a total of 44 cages implanted.
Over a mean follow-up interval of 16.8 months, 6 minor complications occurred in 4 patients (11.4%). These events included sacroiliitis with adjacent disc injury in 1 patient, isolated adjacent disc injury in another, cage subsidence in 2 patients, and delayed fusion in 1 patient. All of these complications were managed conservatively, and no patient required revision surgery. All 4 patients recovered without permanent symptoms. Follow-up computed tomography scans were obtained after 12 to 15 months in 3 patients to confirm fusion. This, however, was not ordered routinely for all patients as clinical evolution was satisfactory, and there were no signs of pseudarthrosis. Notably, 6 of 26 patients who had radicular pain preoperatively continued to report radicular complaints postoperatively, and there were no cases of new radiculopathy. Similarly, there were no cases of new paresis or worsened neurological deficits. There was a 76.9% reduction in patients experiencing radicular pain and a 50% reduction in paresis postoperatively. No infections, new neurological deficits, cage migration, or hardware failures occurred in this cohort during the study period (Table 3).
Postoperative complications in patients who underwent ELLIF (N = 35).
Functional outcome measures indicated substantial improvement at an average follow-up length of 16.8 months (Table 4). The mean ODI score declined from 41.6 ± 3.8 (severe disability) preoperatively to 17.9 ± 8.7 (minimal disability) at 1 month in all 35 patients, corresponding to a 57% improvement (P < 0.001). By 1 year, the mean ODI was 5 ± 7.6, representing an 88% reduction relative to preoperative levels (P < 0.001). In the 9 patients who reached 3-year follow-up, the average ODI improved further to 2.3 ± 3.7, which is classified as minimal or negligible disability (Figure 6). Most patients had their procedure within 3 years of data collection, and as such, had not yet been measured at this timepoint. The VAS for back pain demonstrated a similar trend: patients reported an average 78% reduction in pain at 1 year and a 93% reduction at 3 years (P < 0.001).
Functional recovery monitored by Oswestry Disability Index and visual analog pain scale from postoperative to 3-year follow-up.
Graphical representation of the mean Oswestry Disability Index (ODI) at each successive follow-up timepoint with ODI interpretation overlaid, depicting the rapid improvement from severe, to moderate, to minimal disability. Error bars display the SD at each timepoint.
Discussion
This study of 35 consecutive patients undergoing prone ELLIF demonstrates that endoscopic approaches, coupled with careful patient positioning, can successfully address both axial back pain and radicular symptoms while minimizing complications traditionally associated with lateral interbody fusion. One of the critical advantages of ELLIF lies in the reduced or eliminated need for sustained retraction on the psoas and lumbar plexus, which often contributes to transient thigh numbness, muscle weakness, or plexus injuries in standard LLIF.9,10 The absence of new neurological deficits and persistent complications in this cohort suggests that limiting direct mechanical traction may indeed mitigate such risks. By contrast, the established rates of transient and persistent neurological complications in standard LLIF vary widely in the literature, often falling between 3% and 5% for permanent deficits and above 20% for transient deficits.10–14 Our observed cases of persistent radiculopathy were all self-limiting, suggesting that they were due to irritation, either from prior compression, manipulation, or distraction. The height gain in the levels loads tension on the roots, and this can temporarily irritate them. Regarding subsidence, the cages used in this initial series are similar to standard transforaminal lumbar interbody fusion and posterior lumbar interbody fusion cages. One of the strengths of this technique is the ability to introduce larger cages as they become available. At the current center, these larger cages are being introduced as the next evolution to further prevent subsidence.
Another notable benefit is the ability to harvest autologous iliac crest bone via the same incision used for the endoscopic approach, obviating the need for a separate harvest site. The iliac crest bone graft can then be directly placed within the interbody cage under endoscopic visualization. In addition to simplifying workflow, these factors may translate into reduced operative times, as reflected in the mean of 40 minutes to cage implantation. While the reported timing is quite efficient, it is likely attributable to a combination of surgeon expertise, optimized patient selection, and the integrated procedural workflow with a single-position approach.
Our functional outcome data, particularly the impressive improvements in ODI and VAS at 1 year and beyond, reinforce the viability of this approach. In addition to the significant improvement in radicular pain and paresis, patients experienced immediate improvement from severe to moderate disability with continued recovery at each subsequent follow-up (Figure 6). While both ODI and VAS are subjective scales, they are long validated and are of particular utility when assessing patients with similar conditions and tracked over time.15,16 A further benefit of the modified ELLIF is the ability to operate with the patient in the prone position, which reduces the risk of compression neuropathies seen in the lateral decubitus position. Recent studies published by Pimenta et al have also demonstrated the benefits of single-position prone LLIF, with favorable complication profiles and significant segmental lordosis gain compared with traditional LLIF and lateral-then-prone LLIF.17,18 This technique also provides the opportunity for a transiliac approach, which was performed in 4 patients in our series.
A series of 70 patients who underwent endoscopic-assisted LLIF, with the endoscope used in phases of the surgery, showed excellent outcomes with a transient complication rate of 12.9%. These included transient psoas muscle weakness, transient thigh sensory disturbance, retroperitoneal injury, and 4 cases of cage migration requiring cage replacement. There were no permanent deficits.4 Our center utilized a similar assisted approach in the early phase of developing the technique, before transitioning to a fully percutaneous and endoscopic procedure for the 35 patients reported herein.
An earlier series of 60 patients with endoscopic transforaminal decompression and fusion in 2012 reported improved motor function in all patients, with minor residual deficits and complications seen in 12 patients (20%).19 These included numbness, lumbar extension discomfort, S1 nerve root irritation, and pedicle screw loosening. Mean operating time was 2 hours, 54 minutes.19 A study describing 14 patients who underwent endoscopic-assisted oblique lumbar interbody fusion also found no persistent complications, with only 2 patients experiencing transient leg numbness and 2 with cage subsidence (28.5%).20 Patients in this series had impressive functional improvement, with a mean ODI improvement from 42.5 preoperatively to 13.4 postoperatively.20 Similarly, we observed 2 cases of subsidence; however, there were no cases in our series of psoas weakness or sensory disturbances in the lower extremities. We did report an overall lower transient complication rate, seen in only 4 patients (11.4%), with a permanent complication rate of 0%. We have also shown that this technique is safe and effective in a wide age range, with an average age in our cohort of 60 years and a maximum of 85 years. It should also be noted that 9 patients underwent 2-level fusion, with 2 cages implanted (Table 5). This brings the total number of cages to 44. Classically, complications and risks are associated with each cage, which further highlights the safety and low complication rate in our series.
Presentation and treatment by vertebral level.
Although direct comparisons to other techniques are hampered by differences in patient populations and study designs, the relatively low complication rate seen here is encouraging. Nevertheless, a 35-patient series is still a modest sample size; longer-term follow-up and larger comparative cohorts would be instrumental in definitively establishing the safety and efficacy profile of ELLIF. It should be noted that the learning curve for performing a truly endoscopic lumbar fusion in the prone position can be steep. The surgical team’s experience in endoscopic spine techniques, efficient setup with neuromonitoring, and diligent preoperative planning all likely play integral roles in achieving favorable outcomes. Future investigations may explore how these variables, along with patient-related factors such as bone quality, obesity, and underlying comorbidities, influence complication rates and functional recovery.
Conclusion
Prone ELLIF provides an alternative to conventional LLIF by circumventing prolonged retraction of the psoas and lumbar plexus while allowing for direct endoscopic visualization of the disc space. In this series of 35 patients and 44 cages, the approach was associated with minimal complications, no new neurological deficits, and significant functional improvement sustained through 1 year of follow-up and beyond. Although these findings are promising, further research involving larger patient cohorts and controlled study designs is warranted to validate the broader applicability of ELLIF and confirm its advantages over established fusion techniques.
Footnotes
Funding The authors received no financial support for the research, authorship, and/or publication of this article.
Declaration of Conflicting Interests The authors report no conflicts of interest in this work.
Disclosures The authors have no disclosures to report.
Ethics Approval Institutional Review Board and Research Ethics Committee approval were obtained for this work (HM Hospitales IRB Codigo CElm: 25.02.2483-GHM). Informed consent was obtained from each of the 35 patients included in the study, with their information deidentified.
Disclosures The authors have no disclosures to report.
- This manuscript is generously published free of charge by ISASS, the International Society for the Advancement of Spine Surgery. Copyright © 2025 ISASS. To see more or order reprints or permissions, see http://ijssurgery.com.
References
- 1.↵Resnick DK, Choudhri TF, Dailey AT, et al. Guidelines for the performance of fusion procedures for degenerative disease of the lumbar spine. Part 7: intractable low-back pain without stenosis or spondylolisthesis. J Neurosurg Spine. 2005;2(6):670–672. 10.3171/spi.2005.2.6.0670
- 2.↵Le TV, Burkett CJ, Deukmedjian AR, Uribe JS. Postoperative lumbar plexus injury after lumbar retroperitoneal transpsoas minimally invasive lateral interbody fusion. Spine (Phila Pa 1976). 2013;38(1):E13–20. 10.1097/BRS.0b013e318278417c
- 3.↵Assina R, Majmundar NJ, Herschman Y, Heary RF. First report of major vascular injury due to lateral transpsoas approach leading to fatality. J Neurosurg Spine. 2014;21(5):794–798. 10.3171/2014.7.SPINE131146
- 4.↵Kyoh Y. Minimally invasive endoscopic-assisted lateral lumbar interbody fusion: technical report and preliminary results. Neurospine. 2019;16(1):72–81. 10.14245/ns.1938024.012
- 5.↵Relvas-Silva M, Pinto BS, Sousa A, Loureiro M, Pinho AR, Pereira P. Is endoscopic technique an effective and safe alternative for lumbar interbody fusion? A systematic review and meta-analysis. EFORT Open Rev. 2024;9(6):536–555. 10.1530/EOR-23-0167
- 6.↵Casal Grau R, Ortega Carnero Á, González Murillo M, et al. Evolución de la endoscopia de columna. ¿de dónde venimos y a dónde vamos. Rev Esp Artrosc Cir Articul. 2023;30(3):162–169. 10.24129/j.reaca.30380.fs2209015
- 7.↵Lykissas MG, Aichmair A, Hughes AP, et al. Nerve injury after lateral lumbar interbody fusion: a review of 919 treated levels with identification of risk factors. Spine J. 2014;14(5):749–758. 10.1016/j.spinee.2013.06.066
- 8.↵Casal Grau R, Sánchez Benitez de Soto FJ, Barhouse P, et al. Endoscopic lateral lumbar interbody fusion: technical note and case series. Int J Spine Surg. 2024;18(1):101–109. 10.14444/8572
- 9.↵Grunert P, Drazin D, Iwanaga J, et al. Injury to the lumbar plexus and its branches after lateral fusion procedures: a cadaver study. World Neurosurg. 2017;105:519–525. 10.1016/j.wneu.2017.06.027
- 10.↵Pumberger M, Hughes AP, Huang RR, Sama AA, Cammisa FP, Girardi FP. Neurologic deficit following lateral lumbar interbody fusion. Eur Spine J. 2012;21(6):1192–1199. 10.1007/s00586-011-2087-9
- 11.↵Salzmann SN, Shue J, Hughes AP. Lateral lumbar interbody fusion—outcomes and complications. Curr Rev Musculoskelet Med. 2017;10(4):539–546. 10.1007/s12178-017-9444-1
- 12.↵Cahill KS, Martinez JL, Wang MY, Vanni S, Levi AD. Motor nerve injuries following the minimally invasive lateral transpsoas approach. SPI. 2012;17(3):227–231. 10.3171/2012.5.SPINE1288
- 13.↵Hijji FY, Narain AS, Bohl DD, et al. Lateral lumbar interbody fusion: a systematic review of complication rates. Spine J. 2017;17(10):1412–1419. 10.1016/j.spinee.2017.04.022
- 14.↵Ghobrial GM, Williams KA, Arnold P, Fehlings M, Harrop JS. Iatrogenic neurologic deficit after lumbar spine surgery: a review. Clin Neurol Neurosurg. 2015;139:76–80. 10.1016/j.clineuro.2015.08.022
- 15.↵Fairbank JCT, Pynsent PB. The oswestry disability index. Spine (Phila Pa 1986). 2000;25(22).
- 16.↵Delgado DA, Lambert BS, Boutris N, et al. Validation of digital visual analog scale pain scoring with a traditional paper-based visual analog scale in adults. JAAOS Glob Res Rev. 2018;2(3):e088. 10.5435/JAAOSGlobal-D-17-00088
- 17.↵Pimenta L, Pokorny G, Amaral R, et al. Single-position prone transpsoas lateral interbody fusion including L4-L5: early postoperative outcomes. World Neurosurg. 2021;149:e664–e668. 10.1016/j.wneu.2021.01.118
- 18.↵Pimenta L, Amaral R, Taylor W, et al. The prone transpsoas technique: preliminary radiographic results of a multicenter experience. Eur Spine J. 2021;30(1):108–113. 10.1007/s00586-020-06471-y
- 19.↵Osman SG. Endoscopic transforaminal decompression, interbody fusion, and percutaneous pedicle screw implantation of the lumbar spine: a case series report. Int J Spine Surg. 2012;6(1):157–166. 10.1016/j.ijsp.2012.04.001
- 20.↵Heo DH, Kim J-S. Clinical and radiological outcomes of spinal endoscopic discectomy–assisted oblique lumbar interbody fusion: preliminary results. J Neurosurg. 2017;43(2):E13. 10.3171/2017.5.FOCUS17196
