Since its initial description in 2006,¹ the lateral transpsoas approach to the lumbar spine has evolved into a workhorse approach for lumbar arthrodesis for various pathological conditions, with an acceptable safety profile.^2–4 This approach provides indirect, minimally invasive decompression of the neural elements.⁵ Its advantages over more traditional open posterior techniques include less disruption of the posterior elements and musculature, less blood loss, and less pain.⁶ Moreover, larger-footprint implants help to restore desired spinal alignment while minimizing the risk of subsidence.^4,7

Lateral lumbar interbody fusion (LLIF) is often supplemented by posterior pedicle screw fixation. Traditionally, LLIF is performed with the patient in the lateral decubitus position. The patient is then repositioned prone to achieve posterior fixation. There is increasing interest in single-position surgery with the patient in the lateral position to access both the anterior and posterior elements of the spine.^8–10 Single-position surgery is appealing because it avoids the need for staged surgery and reduces overall procedure time.^11,12 However, posterior access to the spine with the patient in the lateral decubitus position can be challenging and time-consuming because it is less ergonomic, which makes performing a decompression difficult, and it has been associated with inaccurate pedicle screw placement.¹³

More recently, authors have reported the feasibility of a single-position lateral transpsoas approach with the patient in the prone position.^14–16 This procedure has numerous advantages: 1) the patient can be placed on the Jackson table with the abdomen hanging freely, which facilitates natural lordosis; 2) the prone position is customary and familiar; 3) the surgeon has access to the anterior and posterior column in the same setting; 4) the procedure can easily be converted to a posterior fusion if the anatomy is not favorable; 5) repositioning is not necessary; 6) the lordotic position increases the distance between the ribs and the iliac crest; and 7) the iliopsoas and lumbar plexus may be more posterior in lordosis (as shown in preliminary experience). To date, reports on the prone transpsoas approach have been cadaver studies or small clinical case series.^14–16 The senior author of the current study (R.W.P.) has performed this procedure for more than 1 year. Herein, we describe the operative technique and present a consecutive case series of patients who have undergone single-position prone transpsoas LLIF. Surgical, clinical, and radiographic outcomes are reported.

Methods

Operative Technique

After induction of general anesthesia, somatosensory evoked potentials are initiated, including monitoring the saphenous nerve ipsilateral to the retroperitoneal approach. Next, electromyography electrodes are applied to the patient in all relevant dermatomes that emanate from the lumbosacral plexus. The patient is then placed in the prone position on a Jackson table. The ipsilateral hip pad is positioned just distal to the posterior superior iliac spine to create working distance between the 12th rib and the iliac crest. A bolster may be placed on the contralateral side of the table for use as a backstop during graft placement and to allow some degree of coronal bend (Fig. 1A). Tape is placed across the thorax and pelvis to secure the patient to the table, allowing for some table rotation. Tape is also placed on the lateral thorax and pelvis to provide both rostral and caudal tension to open the space between the bottom of the ribs and the top of the iliac crest (Fig. 1B). The side of the approach is determined by various factors, including surgeon preference.

Key operative steps of the single-position prone lateral transpsoas approach. A: A contralateral bolster is placed to serve as a backstop during graft placement. B: Tape is placed on the patient’s lateral thorax and pelvis to provide both rostral and caudal tension to open the space between the bottom of the ribs and the top of the iliac crest. C: The operative level is marked using standard fluoroscopy. D: The retractor arm is placed on the ipsilateral side of the bed. Draping includes the ipsilateral operative corridor as well as the posterior spine. Figure is available in color online only.

The operative level is marked using either standard fluoroscopy or intraoperative CT-based navigation (Fig. 1C).¹⁷ True anteroposterior (AP) and lateral images are obtained before skin incision, with the endplates and pedicles parallel on the lateral view and with the spinous process bisecting the pedicles on the AP view. The prone position also allows for posterior midline marking on the lumbar spine to identify the appropriate level of surgery and to angle down to the disc space, pedicles, and spinal canal. After the skin incision has been marked, the patient is prepared, and the operative field is draped. Draping includes the ipsilateral operative corridor as well as the posterior exposure. Meticulous stapling of the drapes around the lateral exposure is essential to prevent the hip pad and tape from creeping into the sterile field. During surgery the surgeon can stand or sit on a stool next to the patient, can easily move the fluoroscope C-arm from the AP to the lateral position, and can obtain retroperitoneal access using standard techniques. Caveats with the prone position are that the incision is made more posterior than in the lateral decubitus position and that the degree of bending in the coronal plane is less.

After the psoas is identified, the initial dilator can be guided into the retroperitoneum along the index finger until it is pinned on the lateral spine to avoid trapping retroperitoneal fat or bowel against the lateral spine. Triggered electromyography stimulation can then be initiated in a 360° fashion to ensure that the lumbosacral plexus is posterior to the dilators and retractor. In the senior author’s experience, the lumbosacral plexus is typically more posterior than with the standard lateral approach because of the additional lordosis achieved when the patient is in the prone position. Subsequently, a K-wire should be placed in the posterior one-third of the disc space, with AP and lateral fluoroscopy used to confirm placement. With the patient in the prone position, gravity tends to move both the dilators and the exposure toward the anterior portion of the disc space and toward the floor throughout the procedure. The surgeon must remain continually aware of this tendency and should use fluoroscopy to monitor for anterior slippage.

After the dilators are in position, the retractor is placed over them with the K-wire in the disc space, and the angle and rotation of the retractor are confirmed on fluoroscopy. The ideal placement puts the retractor perpendicular to the disc space and the handle rotation parallel with it. After the retractor location is confirmed, the retractor arm is attached to the ipsilateral side of the bed. We have found the midthigh level to be the ideal location for angling the retractor arm away from the patient to create a trapezoidal configuration (Fig. 1D). This configuration allows for the greatest degree of freedom to manipulate the fluoroscope while minimizing the tendency to push the retractor toward the floor. After the retractor is secured, additional AP and lateral radiographs should be obtained to confirm positioning. Next, we remove the dilators, position the lights while leaving the K-wire in place, and inspect inside the retractor after slightly opening the craniocaudal and AP blades. This maneuver allows visual inspection of the lateral retroperitoneal space to ensure that the bowel and peritoneum are not trapped against the disc space or pinned by the K-wire. Furthermore, one can then stimulate with the ball-tipped probe to ensure that there are no motor nerves in the operative field. Finally, the lateral disc space should be inspected to ensure that the genital femoral nerve is not in the operative field before fully dissecting the psoas muscle to expose the lateral annulus. The lateral disc space then comes into view and is ready for placement of the intradiscal shim, which can be placed in the posterior third of the disc space after confirming that the retractor is angled parallel with the disc space “down the barrel” (Fig. 2). Standard annulotomy, discectomy, endplate preparation, trialing, and graft placement are all performed with the surgeon seated. Operating while seated allows for a more ergonomic and less physically demanding operative experience.

Artist’s illustration demonstrating lateral retractor placement at L4–5 disc for the single-position prone lateral transpsoas approach. m. = muscle. Used with permission from Barrow Neurological Institute, Phoenix, Arizona.

With the prone lateral approach, posterior decompression and pedicle screw fixation can be performed without repositioning the patient. Pedicle screws are typically placed via a percutaneous approach using fluoroscopy, CT navigation, or robotic assistance. Direct decompression or osteotomy performed from the posterior approach can also be achieved as indicated. Our typical operative workflow, all conducted without repositioning the patient, is to first place pedicle screws with robotic assistance, followed by direct posterior decompression as needed, then to place lateral cages, and finally to reduce and lock down the rods.

Results

Illustrative Case

A 75-year-old woman presented with low-back pain and left lower-extremity radiculopathy. The patient had previously undergone an L4–S1 laminectomy and an L4–5 microdiscectomy 5 years earlier. She had subsequently been managed with physical therapy, medication, and multiple epidural corticosteroid injections, which failed to alleviate her intractable symptoms. Preoperative MRI (Fig. 3A) and a standing lateral radiograph (Fig. 3B) demonstrated degenerative disc disease and spondylolisthesis at L4–S1. The patient underwent a prone left lateral transpsoas interbody fusion at L4–5 and a minimally invasive transforaminal lumbar interbody fusion (TLIF) at L5–S1, with percutaneous pedicle screw fixation in a single position. Postoperatively, the patient’s VAS–back pain scores improved from 7 to 4, and her VAS–leg pain score improved from 10 to 3. A postoperative radiograph demonstrated a reduction in the spondylolisthesis (Fig. 3C).

Imaging obtained in a woman who presented with persistent radiculopathy that failed to improve despite prior microdiscectomy and conservative treatment. Preoperative sagittal T2-weighted MR image (A) and standing lateral radiograph (B) demonstrating spondylosis and spondylolisthesis at L4–S1. The patient subsequently underwent a prone LLIF at L4–5 via a transpsoas approach and a minimally invasive TLIF at L5–S1 with percutaneous pedicle screw fixation, all in a single prone position. Postoperative standing lateral radiograph (C) showing good reduction of the spondylolisthesis.

Discussion

LLIF via a transpsoas approach is increasingly considered a vital part of the modern spine surgeon’s armamentarium. This minimally invasive approach has been shown to safely and successfully produce arthrodesis for degenerative and deformity-related spine diseases.^2,3 Traditionally, this procedure is performed with the patient in the lateral decubitus position, and the patient is then repositioned for posterior surgery in a second stage. However, this change in positioning increases the patient’s time under anesthesia and decreases surgical efficiency. Interest is therefore growing in investigating the viability and nuances of performing single-position surgery. Its purported benefit is that both lateral and posterior approaches can be performed without repositioning the patient. Several reports of single-position surgery in the lateral position have documented the feasibility of this approach and the resultant increase in operative workflow and efficiency.^10–12 One study showed that patients treated in a single lateral position had decreased blood loss, complications, and length of stay compared to patients treated with traditional 2-stage (i.e., 2-position) surgery.⁸

However, single-position surgery with the patient in the lateral decubitus position does have drawbacks. Approaching the posterior aspect of the spine in this position is less familiar to the spine surgeon. Pedicle screw placement in this position also requires changes to traditional techniques, resulting in a high breach rate.¹³ Additionally, posterior decompression is difficult to achieve in the lateral position. In 2020, Pimenta et al.¹⁴ reported a novel single-stage approach that involved performing both the lateral and posterior approaches with the patient in the prone position. To demonstrate the technical feasibility of this patient positioning, Pimenta et al. described a case in which this approach was used. Lamartina and Berjano¹⁶ further validated this single-stage approach in a prospective study of 7 patients treated with single-level lateral fusion in the prone position. No major complications were observed, and patients had postoperative improvements in ODI scores and VAS–back and VAS–leg pain scores. When Lamartina and Berjano¹⁶ compared these 7 patients to 10 patients treated with the traditional 2-stage lateral approach followed by the posterior approach, they found a decrease in overall surgical time (133 vs 182 minutes, not significant) with the single-position approach. Last, Godzik et al.¹⁵ reported a cadaveric feasibility study of this approach and a retrospective series of 12 patients. Eleven patients were treated successfully at 14 levels; for 1 patient, the single-position approach was converted to standard lateral positioning. The authors concluded that, in their early experience, this technique appeared to be safe and feasible.

Our consecutive case series of 28 patients treated by 1 surgeon who recently adopted this innovative single-position prone lateral approach is the largest reported series to date demonstrating the successful use of this technique. Thirty-nine levels were treated in the 28 patients, who presented with a variety of common clinical indications. Overall, these patients experienced significant improvement in their self-reported clinical outcome scores, including VAS–back pain, VAS–leg pain, ODI, and SF-36 scores. At the 12-week follow-up, these patients also demonstrated acceptable radiographic outcomes.

We have noted several benefits to this single-stage approach over the past year. The first benefit is that the entire procedure can be performed without changing the patient’s position. Maintaining the patient in one position improves operative efficiency, as reported previously by other authors. Placing the patient in the prone position on the Jackson table allows the spine to fall into lordosis before the procedure is started. Additionally, the prone position is the most familiar to the spine surgeon, and it facilitates easier placement of posterior pedicle screw fixation. Moreover, the patient is well positioned for a posterior decompression if a concern exists or arises about relying on indirect decompression alone. Last, attempted lateral transpsoas access may not be achievable with the patient in the prone position. In these cases, the surgeon may choose to reposition the patient in the lateral decubitus position to perform the lateral approach. However, with the patient initially in the prone position, the surgeon has the option to convert to a TLIF without having to reposition or re-drape the patient. In only 1 of 29 cases in our consecutive series was lateral access aborted (because of a high iliac crest). In this case, the surgical plan was quickly and efficiently changed to a TLIF, which was performed successfully. The patient had consented preoperatively to this possibility. Thus, in cases in which the surgeon believes there is clinical equipoise between a lateral fusion and a posterior fusion, the TLIF is readily available as a bailout strategy if lateral access fails. A TLIF can also be performed at the L5–S1 segment from the posterior approach and combined with lateral access at the rostral segments, as described in our case example.

Intraoperative complications with this single-stage approach included cage repositioning in 1 patient (4%) and inadvertent ALL rupture in 2 patients (7%). As previously reported, gravity pushes the retractor toward the floor, therefore ventrally in the patient. This anterior shift of the retractor puts the patient at risk for ALL rupture if the surgeon is not constantly aware of its position within the disc space. A slightly more posterior incision may help mitigate this complication.¹⁴ When ALL rupture does occur, lateral plating should be performed. Intraoperative subsidence, which was noted in one of our patients, and other complications may be minimized by continued modification of this technique. No neurovascular injuries occurred in this preliminary surgical experience; in fact, the senior author noted that the lumbosacral plexus appears to be more posteriorly displaced in this position because he has not experienced any “anteriorly located” nerves at the L4–5 level like those observed in the lateral position.

This study is not without limitations. First, its design is that of a retrospective observational series. Therefore, there is no control group for comparisons, such as for patients who underwent standard 2-stage lateral prone procedures or single-position lateral procedures. The 28 patients also have limited follow-up because of the relatively new nature of this technique. However, this consecutive case series is currently the largest to date in the neurosurgery literature. The patients’ outcomes demonstrate the feasibility of this approach for a spectrum of common clinical indications. As this technique is increasingly adopted, larger prospective series with longer-term follow-up will be an important addition to the literature. Furthermore, the experiences of surgeons beyond early adopters will facilitate improvements in operative efficiencies.

Conclusions

Recent studies have demonstrated the feasibility of the prone transpsoas approach for LLIF. This single-surgeon consecutive case series demonstrates the good clinical and radiographic outcomes achievable with this novel technique. Our results indicate that this procedure is well tolerated and has an acceptable rate of complications. Larger patient series with longer follow-up are needed to further elucidate the safety profile and long-term outcomes of this approach.

Acknowledgments

We thank the staff of Neuroscience Publications at Barrow Neurological Institute for assistance with manuscript preparation.

Disclosures

Dr. Porter is the founder and owner of The Medical Memory, Inc., and receives consulting fees from Globus Medical, Inc.

Author Contributions

Conception and design: all authors. Acquisition of data: Naeem, Bhargava. Analysis and interpretation of data: all authors. Drafting the article: Farber. Critically revising the article: all authors. Reviewed submitted version of manuscript: all authors. Statistical analysis: Farber. Administrative/technical/material support: Porter. Study supervision: Porter.