Stand-alone prone lateral interbody fusion: a retrospective case series

Abstract

Background

Prone lateral spine surgery has gained traction as a single-position approach due to advantages including reduced operative time, and improved lordosis restoration compared with traditional decubitus lateral approaches. Stand-alone anterior lumbar interbody fusion (ALIF) and direct lateral fusion (DLF) techniques have demonstrated reduced operative time, blood loss, length of stay, and cost compared with posteriorly instrumented constructs, though with higher reported rates of subsidence. While stand-alone instrumentation has been described for ALIF and DLF, its use in the prone lateral transpsoas approach has not previously been reported. This study presents a retrospective case series evaluating the feasibility, perioperative outcomes, and radiographic characteristics of stand-alone prone lateral lumbar interbody fusion.

Case Description

Eight adult patients underwent elective, single-level stand-alone prone lateral lumbar interbody fusion performed by a single surgeon at a large academic institution between May 2020 and September 2025. All patients had prior multilevel lumbar fusion and American Society of Anesthesiologists (ASA) classifications of 2 to 3. Static and expandable cages with integrated lateral plating were utilized. Operative time ranged from 80 to 194 minutes (median 112 minutes), estimated blood loss ranged from 5 to 50 mL (median 50 mL), and length of hospital stay ranged from 1 to 3 days (median 2 days). Three of eight patients required revision surgery (37.5%); one experienced postoperative numbness due to inadequate decompression and two required revision due to subsidence. Radiographic analysis demonstrated median increases in anterior disc height (6.54 mm), posterior disc height (2.75 mm), and foraminal height (2.0 mm). Median change in lumbar lordosis on both immediate postoperative and long-term imaging was ten degrees and seven degrees, respectively. Median change in segmental lordosis on immediate and long-term postoperative imaging was ten degrees and five degrees respectively.

Conclusions

This case series of stand-alone prone lateral interbody fusion demonstrates the technical feasibility in select patients. While outcomes pertaining to disc height and lordotic restoration were favorable, surgeons considering this technique should be aware of the complications and revisions observed in this series. Further studies with larger cohorts and longer follow-up are needed to better define the influence of bone quality, indications, radiographic durability, and long-term outcomes of this approach.

Highlight box.

Key findings

• This is the first reported case series describing stand-alone prone lateral lumbar interbody fusion, expanding the application of the prone transpsoas approach. Stand-alone prone lateral surgery demonstrated favorable perioperative efficiency, including low blood loss, short operative times, brief hospital stays, but a moderate postoperative complication rate in a medically complex population.

What is known and what is new?

• Stand-alone anterior instrumentation and indirect decompression are viable techniques in select patients in anterior lumbar and conventional lateral lumbar surgeries.

• Stand-alone constructs may also be a practical approach in suitable patients undergoing prone lateral lumbar surgery.

What is the implication, and what should change now?

• These early findings support stand-alone prone lateral fusion as a viable option in carefully selected patients, warranting further study to define optimal indications and long-term outcomes.

Introduction

After its initial cadaveric description earlier this decade, prone lateral spine surgery continues to steadily gain in popularity in the United States and worldwide, with research growth trajectories surpassing those of traditional lateral fusion during comparable developmental stages (1-3). When compared to traditional decubitus direct lateral fusion (DLF), prone lateral surgery offers potential advantages of easier single-position access, shorter operative times, and improved lordosis restoration (4,5). Complication profiles are similar between the procedures, demonstrating no differences in overall complications, nerve palsy, vascular injury, or instrumentation failure (5).

There has been growing interest in anterior lumbar interbody fusion (ALIF) and DLF techniques that rely on stand-alone anterior fixation, without supplemental posterior pedicle screw and rod instrumentation to reduce operative time, blood loss, and postoperative complications, especially in highly comorbid patients (6,7). Comparative analyses of these approaches have demonstrated similar overall outcomes. When relying on anterior interbody support and fixation alone-overall costs, operative times and blood losses were consistently lower (8-10). Clinical outcomes scores remained similar between groups with the exception of a small increase in subsidence rates and non-unions rates in the stand-alone groups in comparison to the percutaneously instrumented groups (8-10).

In theory, the minimally invasive nature of stand-alone constructs should synergize well with the positional advantages of a prone transpsoas approach, such as allowing for gravity facilitated lordosis and disc space opening. Further, the stand-alone constructs allow small incisions, just wide enough for the interbody cage, and reduce the overall cost of the procedure. Yet, there exists no literature on the viability of stand-alone cages in a prone transpsoas approach (10). Therefore, we present eight early cases of stand-alone prone lateral procedures, providing descriptive data and demonstrating feasibility. We present this article in accordance with the PROCESS reporting checklist (11) (available at https://jss.amegroups.com/article/view/10.21037/jss-2026-0101/rc).

Case presentation

Methods

Elective, prone lateral cases from a single surgeon at Duke University were retrospectively collected between May 2020 and September 2025. Inclusion criteria were adult patients undergoing surgery involving a single-level lateral lumbar interbody fusion using a prone transpsoas approach with placement of a stand-alone cage. Exclusion criteria encompassed any patients who underwent multi-level fusions or multiple types of fusion (i.e., prone lateral adjacent to ALIF), did not have stand-alone cages (i.e., received unilateral or bilateral rod placement), and those with infective, traumatic, or oncologic indications. 8 patients met inclusion criteria.

Surgical approach was offered upon the basis of patient anatomy, previous procedures, and joint decision making with the patients on a case-by-case basis. All patients were induced under general anesthesia. Neuromonitoring devices were connected and transcranial electrical motor evoked potentials (TCMEP) and somatosensory evoked potential (SSEP) were monitored throughout the procedure. Operative levels were identified using instrument tracking (Track X, Hillsborough, NC, USA) technology. All approaches were conducted from the left side in a standard prone-lateral fashion. Annulotomy, discectomy, and cage insertion were conducted using a combination of a disc knife, straight and angled cobbs, cage trials, pituitary rongeurs, and straight and angled curettes. ELSA model expandable cages were anchored in position using an integrated lateral interbody plate and screws-securing the cage between the adjacent vertebral bodies.

Radiographic measurements were conducted on preoperative, earliest postoperative, and most recent postoperative standing radiographs. If patients had subsequent spine surgery, the most recent radiograph was limited to the time between the surgery of interest and the additional surgery (Tables 1,2).

Table 1. Patient demographic data and baseline clinical characteristics.

Variable	Value
Previous fusion	8 (100.0)
History of osteoporosis/osteopenia	2 (25.0)
Gender
Male	5 (62.5)
Female	3 (37.5)
ASA status 2	2 (25.0)
ASA status 3	6 (75.0)
Age, years	74 (70.75, 78.75)
BMI, kg/m2	27.83 (24.27, 31.90)
Hounsfield, HU	128 (114.50, 133.84)

Data are presented as median (Q1, Q3) or n (%). ASA, American Society of Anesthesiologists; BMI, body mass index; HU, Hounsfield units.

Table 2. Cage implant characteristics and dimensions used in this cohort.

Variable	Value
Implant characteristics
ELSA 8–17 mm, 20×60 mm, 5–20°	5 (62.5)
ELSA 8–17 mm, 20×55 mm, 5–20°	2 (25.0)
ELSA 8–17 mm, 20×50 mm, 5–20°	1 (12.5)
Cage dimensions
Cage height (mm)	14.15 (11.65, 15.8)
Cage angle (degrees)	8 (4, 11)

Data are presented as median (Q1, Q3) or n (%). ELSA, expandable lateral lumbar interbody spacer.

Imaging data was collected from preoperative full length spine radiographs, at the date closest to the operation. All patients had pre-operative imaging collected within one year of surgery. Immediate post-operative imaging was collected from full length spine X-rays less than one month from the procedure date. Long-term imaging was collected from any X-ray of the entire spine that was more than two months after the procedure date, and prior to any other spine surgeries. Long term imaging was available for four patients with a median of 154 days from the procedure. Seven patients had a preoperative computed tomography (CT) where Hounsfield units were calculated and averaged from the levels involved in the fusion in accordance with previous literature (12).

The data was analyzed utilizing descriptive statistics. As the study did not include any comparative data, the study focused on quantifying outcomes within this patient population. Continuous variables were summarized by mean and standard deviation and categorical variables reported using counts and percentages (Tables 3-5).

Table 3. Perioperative metrics including complications.

Metric	Value
Length of surgery (minutes)	112 (96.25, 124.25)
Estimated blood loss (mL)	50 (23.75, 50)
Length of stay (days)	2 (1.75, 2.25)
Complications
Postoperative numbness	1 (12.5)
Revision	2 (25.0)
Total	3 (37.5)

Data are presented as median (Q1, Q3) or n (%).

Table 4. Duration of time elapsed in days of imaging both preoperatively and postoperatively.

Radiographic data	Time from procedure in days
Preoperative imaging	124.5 (76.5, 189.25)
Earliest postoperative imaging	2 (1, 8.5)
Most recent imaging (n=5)	154 (130.75, 302)

Data are presented as median (Q1, Q3).

Table 5. Postoperative changes in radiographic vertebral, spinopelvic, and disc-related parameters, reported as median values with interquartile ranges.

Metrics	Short term	Long term
Anterior disc height change (mm)	6.54 (2.66, 8.18)	3.02 (1.18, 5.69)
Posterior disc height change (mm)	2.75 (1.93, 3.48)	1.31 (0.84, 1.97)
Foraminal height change (mm)	2 (1.6, 5.0)	3 (1.2, 4.13)
Lumbar lordosis change (degrees)	10 (8, 13)	7 (3, 10)
Segmental lordosis change (degrees)	10 (7.5, 11.5)	5 (4.25, 6.5)
Disc angle change (degrees)	8 (7, 8)	8.5 (6, 10.25)
Pelvic incidence change	0 (−1.5, 2.5)	1.5 (0, 3.5)
Pelvic tilt change	−2 (−2.5, 0.5)	−1 (−2, 0.25)
Sacral slope change	4 (−1.5, 6.5)	−1 (−2.5, 1.75)
Pelvic incidence-lumbar lordosis mismatch change	−8 (−11.5, −7)	−3 (−6.25, −0.75)

Data are presented as median (Q1, Q3).

Results

Patient demographics

Eight patients were identified as meeting inclusion criteria, having undergone single-level fusion with a standalone interbody cage using a prone transpsoas approach. Median age at time of surgery was 74 (70.75, 78.75) years and median body mass index (BMI) was 27.83 (24.27, 31.90) kg/m². Five patients were male (62.5%). All patients had previous multilevel spinal fusions. One patient previously had a L3 to L5 fusion, one patient with a previous L3 to S1 fusion, and six patients with a previous L4 to S1 fusion. Two patients (25%) had a history of osteopenia or osteoporosis. All patients had an American Society of Anesthesiologists (ASA) status of between 2 and 3 (Table 1).

Cage utilization

All patients received expandable ELSA cages (Globus Medical, Audubon, PA, USA), allowing for adjustment across both height and lordotic angle following placement. Five patients received ELSA 8–17 mm 20×60 mm 5–20D model cage, two patients received the ELSA 8–17 mm 2×55 mm 5–20D model cage, and one patient received the ELSA 8–17 mm 20×50 mm 5–20D model cage. For these cages, the median expanded height and angle of the cage was set to 14.15 (11.65, 15.8) mm and 8 (4, 11) degrees respectively.

Perioperative metrics

Operative times ranged from 80 to 194 minutes and the median length of surgery of 112 minutes. Operative blood loss ranged from 5 to 50 mL with a median of 50 mL. Patient length of hospital stay ranged from 1 to 3 days with a median of 2 days.

Imaging metrics

Preoperative imaging ranged from 50 to 343 days before the procedure with a median of 124.5 days. Postoperative imaging was captured between 1 and 22 days with a median of 2 days after the procedure. Long-term imaging was captured a median of 154 days after the procedure (Table 4).

Two patients were subsequently diagnosed with osteoporosis following the procedure with average Hounsfield units of adjacent vertebrae measuring 110 and 98. Overall, Hounsfield units had a median of 128 with a minimum of 98 and a maximum of 156.

Immediate change in anterior disc height ranged from 0.2 to 11.7 mm with median change of 5.7 mm (Figure 1; Table 5). Immediate change in posterior height ranged from 1.0 to 5.2 mm with a median of 2.7 mm. Immediate change in foraminal height ranged from 0.2 to 11.5 mm with a median change of 2 mm. Long-term follow-up measures for anterior disc height ranged from 0.8 to 8.5 mm with a median of 3.0 mm. Long-term measures for posterior disc height ranged from −0.2 to 3.7 mm with a median of 1.3 mm. Long-term measures for foraminal height ranged from −1.2 to 4.5 mm with a median of 3.0 mm.

Figure 1.

Spinopelvic parameters averages are plotted with error bars representing the standard deviation.

Immediate change in lumbar lordosis both ranged from 2 to 14 degrees with a median of 10 degrees, with long term change ranging from 0 to 10 degrees with a median of 7. For segmental lordosis, immediate differences ranged from 5 to 14 degrees with a median of 10 degrees; long-term measures ranged from 2 to 11 degrees with a median of 5 degrees. Change from baseline disc angle in the immediate post-operative X-ray ranged from 5 to 14 degrees with a median of 8 degrees; for long-term radiographs, the range was 3 to 11 degrees with a median of 8.5 degrees.

Pelvic incidence-lumbar lordosis (PI-LL) mismatch ranged from −17 to 16 degrees in immediate post-operative radiographs with a median of −6.5 degrees; for long-term follow-up, PI-LL mismatch ranged from −8 to 22 degrees with a median of 4 degrees.

Complications

Three patients experienced complications following the procedure leading to repeat operation. One patient experienced new-onset postoperative right anterior thigh numbness and right quadriceps weakness, requiring a minimally invasive laminectomy one month postoperatively. One had improvement following the procedure with the reduction in neuropathic pain, but with continued pain with standing for more than five minutes with magnetic resonance imaging (MRI) indicating underestimated severity and insufficient lift from the stand-alone cage placement that required laminectomy and classic posterior fusion. Finally, one patient experienced transient improvement in the symptoms with worsening pain developing five days postoperatively, with imaging indicating cage subsidence, that required laminectomy and classic posterior fusion.

Ethical consideration

All procedures performed in this study were in accordance with the Declaration of Helsinki and its subsequent amendments. This study was approved by the Institutional Review Board of the Duke University (IRB # Pro00071025). Written informed consent was obtained from the patients or their legal guardians for publication of this case series and accompanying images. A copy of the written consent is available for review by the editorial office of this journal. All patient data were treated as confidential according to Health Insurance Portability and Accountability Act standards. Any protocol amendments were submitted to the IRB for review and approval.

Discussion

To our knowledge, this case series represents the first instance in literature of stand-alone prone lateral constructs laying the foundation for the potential synergy between prone lateral approaches and minimal footprint of stand-alone expandable cages. While our early observations demonstrate the feasibility and reproducibility of this operation and suggest it may be of benefit in specific patient indications. Long term data presented warrants cautious interpretation as three of the patients within this cohort required reoperation after their stand-alone prone lateral surgery.

All patients in the study had prior multi-level fusion of at least two levels, requiring reoperation due to development of adjacent segment disease (ASD). Previous research has identified stand-alone constructs in lateral approaches as providing strongly positive clinical and radiographic outcomes, with significantly improved visual analog scale-back scores, segmental lordosis, disc height, and PI-LL mismatch (13). A minimally invasive approach may expand surgical eligibility to patients who might otherwise be considered suboptimal candidates for more extensive procedures. In this series, patients demonstrated preoperative risks profiles such ASA class 2 to 3. As such, this technique may be more appropriate for specific patients with comorbidities, in which, minimizing operative duration and physiologic stress is a priority.

Conventional stand-alone procedures yield numerous operative and immediate postoperative benefits, such including decreased blood loss, surgical times, lengths of stay, and correspondingly decreased costs. When compared to traditionally instrumented prone transpsoas approaches, our data suggests similar trends. One study notes mean operative times of 226.2±147.9 minutes, mean estimated blood loss of 138.4±215.6 mL, and mean hospital stays of 2.7±2.2 days (14). Another study identifies 204.7±83.3 minutes, 187.9±211 mL, and 3.0 days as means for the same respective metrics (15). In comparison, our findings of 112 minutes in median operative time, 50 mL in median blood loss, and 2 days in median length of stay are all notably lower; this seems to further highlight the minimally invasive advantage of maintaining smaller incisions in inserting stand-alone cages. While our sample size is relatively small, our data reflects a proof of concept for this surgical technique may provide greater operative efficiency and reduced surgical footprint.

Regarding radiographic measurements, particularly the amount of lordotic and segmental restoration, stand-alone cages showed promising results in our case series. Anterior disc heights, posterior disc heights, disc angles, lumbar lordosis, segmental lordosis, and PI-LL mismatch all seemed to offer broad improvements across both immediate post-op and long-term follow-up imaging (Figure 2). Numerous studies have found that the prone transpsoas approach lends itself to significantly greater postoperative lumbar lordosis and decreased PI-LL mismatch, though the postoperative timing of radiographs is not well defined (16-18). While limited by power, demographic results in our study are consistent with prior data on posteriorly instrumented prone transpsoas surgeries, yielding promising radiographic outcomes (5,8).

Figure 2.

Standing pre-operative (A), standing post-operative (B), and 6-month standing-post-operative (C) films of stand-alone prone lateral construct.

With any stand-alone anterior spinal instrumentation, increased interbody load is of consideration (19-22). Stand-alone cages yield significantly higher rates of reoperation and cage movement, and subsidence (23). In regards to radiographic parameters, conventional stand-alone constructs may show improved early radiographic results with loss of lordosis and disc height occurring at longer-term follow-up with cage settling (8,19). Descriptively, some of the same trends were observed in our patients (Figure 1). In this cohort, one patient required reoperation for subsidence following the procedure (Figure 3). This patient demonstrated immediate post-operative pain relief, but five days after the procedure had significantly worsening pain. On postoperative CT after the onset of pain, there minimal improvement in foraminal height, with a further reduction in foraminal height on long-term imaging, which is associated with increased risk of early intervention (24). Two additional patients required reoperation with one patient experiencing insufficient lift following the procedure, and another for worsened neurologic symptoms immediately postoperatively.

Figure 3.

X-rays of the patient with subsidence following the procedure are shown, with preoperative X-ray (A), imaging two weeks after the procedure (B), and six months after the procedure (C).

Patients in this study did not undergo a preoperative dual-energy X-ray absorptiometry (DEXA) scan, however all but one patients had a preoperative CT scan (25). Of the two patients that had postoperative subsidence, one patient was diagnosed with osteoporosis postoperatively with an average Hounsfield of 99 and the other had an average of 142. Further, it should be noted that all but one patient were within previously proposed osteoporosis interval values of 90.9 to 138.7 (12). In these patients, standalone instrumentation may provide insufficient support to prevent subsidence.

This case series has multiple limitations limiting generalizability. While serving as a proof-of-concept, comparative efficacy to instrumented cases cannot be concluded. Additionally, due to the limitations of a retrospective review, variability in imaging follow-up intervals introduces heterogeneity into radiographic outcome monitoring. The cohort is highly selected, as all patients had undergone prior multilevel fusion and presented with ASD; this introduces bias and limits applicability to primary surgical candidates without pre-existing instrumentation. The small sample size is a fundamental constraint of this study, significantly limiting analysis to descriptive statistics limits and without direct comparison limits causal analysis. Due to the nature of the study and the revision rate, long term follow up is limited with only a subset of patients having extended radiographic assessment, which weakens analysis regarding the durability of radiographic improvements. Fusion status and segmental stability at the treated level were not formally evaluated in this series, which is a notable gap given that the absence of supplemental posterior instrumentation in stand-alone constructs may increase the risk of pseudarthrosis or construct instability; CT-based fusion assessment or dynamic flexion-extension radiographs would strengthen future iterations of this work. Bone quality assessment was limited by the absence of routine preoperative DEXA scanning, and all but one patient in this study were within the previously defined interval for diagnosis of osteoporosis based on HU limiting the direct comparison based on bone density. Future studies should consider defining evidence-based HU thresholds or formal DEXA criteria as potential contraindications for stand-alone prone lateral constructs to better guide patient selection.

Conclusions

Stand-alone prone lateral lumbar interbody fusion is surgical option that can be considered in patients that minimizing operative duration is a priority, as the perioperative efficiency metrics are encouraging. However, it is important to note the 37.5% revision rate from our cohort when determining the most appropriate approach during preoperative planning. While this case series is small, it provides a preliminary description of stand-alone prone lateral interbody fusion. Time for the operation and hospital length of stay were shorter than that of the traditional prone transpsoas approach, with similar radiographic outcomes. However, this should be weighed against the high complication and revision rate in this series. Future studies are required to determine the utilization of the approach and to identify which characteristics to may improve patient selection. Investigators may consider including examination of Hounsfield cut offs, larger cohorts, and longer follow-up as part of their metrics.

Supplementary

The article’s supplementary files as

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. All procedures performed in this study were in accordance with the Declaration of Helsinki and its subsequent amendments. This study was approved by the Institutional Review Board of the Duke University (IRB # Pro00071025). Written informed consent was obtained from the patients or their legal guardians for publication of this case series and accompanying images. A copy of the written consent is available for review by the editorial office of this journal.