Reverse Lumbar Pedicle Screw in Oblique Lateral Interbody Fusion: A Novel Concept to Restrict Cage Subsidence

Abstract

Objective

Cage subsidence is a common morbidity after oblique lumbar interbody fusion (OLIF), with risk of compromising clinical and radiographic outcomes. The study aims to describe an innovative reverse lumbar pedicle screw (RLPS) technique in OLIF and compare its effect on restricting cage subsidence with classical lateral fixation (LF) in radiological and clinical evaluation.

Method

Consecutive patients having undergone single‐level OLIF‐LF/RLPS from 2018 to 2020 were retrospectively reviewed. In OLIF‐RLPS, the upper entry point was determined at the intersection between one horizontal line (1 cm above inferior endplate) and one vertical line (dissecting anterior and middle thirds of the vertebra) while the inferior entry point between one horizontal line (5 mm below superior endplate) and the same vertical line. Trajectories were from vertebrae reverse into contralateral pedicle. Radiological evaluation included disc height (DH) and segmental lordosis (SL); cage subsidence was evaluated by DH loss. Clinical assessment included visual analogue scale (VAS) and Oswestry disability index (ODI). Student t or Mann–Whitney U test was used for continuous variation according to normality analysis while Chi‐square test for category variation.

Results

A total of 29 patients had been enrolled in the study including 14 cases in the RLPS group and 15 cases in the LF group. The DH in the OLIF‐RLPS group had increased from the preoperative 9.07 ± 1.73 mm to 13.73 ± 1.83 mm postoperatively, without significant difference compared with the OLIF‐LF group during the perioperative, but decreased to 12.53 ± 1.74 mm in 3 months and maintained at 12.00 ± 1.45 mm in 12 months, significantly higher than the OLIF‐LF group (p < 0.05). At the last follow‐up, 7.1% (1/14) cases in the OLIF‐RLPS group had shown subsidence of grade I, significantly less than 46.7% (7/15) cases in the OLIF‐LF group. Pain and disability had improved similarly in two groups, without significant difference detected between two groups at the last follow‐up.

Conclusion

RLPS technique with modified entry points and prolonged trajectory could effectively restrict cage subsidence in OLIF postoperatively compared with traditional lateral fixation.

Depicting reverse lumbar pedicle screw technique. (A) Red spot represented the upper entry point at the intersection between one horizontal line about 1 cm above the inferior endplate and one vertical line dissecting the anterior‐middle thirds of the vertebrae. Blue spot represented the inferior between one horizontal line about 5 mm below the superior endplate and the same vertical line above; (B) The red arrow represented trajectory pointed to the contralateral pedicle with an angle of 30°.

Introduction

The oblique lumbar interbody fusion (OLIF) technique has been increasingly accepted as a standard minimally invasive protocol in treating degenerative diseases for advantages of restoring disc height (DH) and segmental lordosis (SL), enlarging foraminal area and spinal canal, correcting sagittal alignment, without disturbance to posterior structure. ¹ , ² Nevertheless, manipulation in the lateral position had made intraoperative reposition inevitable to achieve rigid posterior pedicle fixation which led to prolonged operative time and position‐related risk, until lateral‐only fixation on vertebrae was increasingly developed and accepted as common practice. However, due to the absence of direct neurological decompression, OLIF essentially achieved indirect neurological decompression through distracting DH to restore foraminal volume which made the maintenance of initial decompression extremely vulnerable to cage subsidence, particularly in cases with lateral‐only instrumentation. Subsidence was the radiological description of a cage, secondary to axial mechanical compression on endplate‐cage interface, breaking adjacent endplates and submerging into vertebrae resulting in shrinkage of foraminal height and canal volume. ³ , ⁴ Though controversaries remained as to whether mild subsidence should be regarded as normality of endplate remodeling secondary to mechanical overloading or abnormality of cage invading into adjacent vertebrae, as mild subsidence had been confirmed a common presence after OLIF surgery without significant clinical worsening reflected in visual analogue scale (VAS) scores in long‐term follow‐up, ⁵ severe subsidence was recognized as a main contributor to initial decompression failure via triggering re‐narrowing of the foramen and spinal canal, re‐loosening of the facet joint capsule, which eventually led to radiculopathy recurrence, ⁶ , ⁷ as severe subsidence (>4 mm) had been reported tightly associated with worse clinical outcomes in VAS and Oswestry disability index (ODI) evaluation along with lower fusion rate (64.5% vs. 83.3%) compared with the mild (2–4 mm), Therefore, great efforts should be made to decrease the incidence of severe cage subsidence in performing OLIF in treating degenerative spondylitis.

Various precautions had been taken to prevent subsidence including meticulous endplate preparation, avoiding over‐distraction, backward cage position and bionic cage application, ⁸ , ⁹ , ¹⁰ without qualified evidence validating their efficacy. Therefore, this study aimed to: (i) illustrate an innovative fixation technique with modified entry points and prolonged trajectories applied in OLIF; and (ii) to evaluate the validity of the present technique in restricting cage subsidence and assess the clinical and radiographic outcomes compared with classical lateral fixation.

Materials and Methods

Patients

This retrospective study was approved by the ethics committee of our institution ((B)KY2022161). Patients who had undergone single‐level oblique lumbar interbody fusion with lateral fixation (OLIF‐LF) or oblique lumbar interbody fusion with reverse lumbar pedicle screw (OLIF‐ RLPS) during a 3‐year period from January 2018 to December 2020 were enrolled in the study based the following inclusion criteria: (i) discogenic low back pain with mild–moderate radiculopathy or dynamic back pain with segmental instability; (ii) lumbar spondylolisthesis (Meyerding grade < II); and (iii) lumbar spinal stenosis (Schizas grade C) with intermittent claudication. Cases meeting the following criteria were excluded: (i) severe radiculopathy; (ii) severe spinal stenosis (Schizas grade D) or spondylolisthesis (Meyerding grade > III); and (iii) severe osteoporosis requiring posterior fixation. Consequently, a total of 29 (8 males/21 females) consecutive patients were included in this study. The aetiologic diagnosis included chronic discogenic pain (15 cases), lumbar spondylolisthesis (12 cases), lumbar spinal stenosis rated grade C (two cases). The distribution of level was from L2‐L5, the age ranged from 45 to 70 years, all operations had been performed by a single senior surgeon.

Surgery Procedure

Decompression and reconstruction

Standard OLIF procedure prior to instrumentation was performed in both the OLIF‐LF and OLIF‐RLPS groups. Patients were posed in the right lateral decubitus position without breaking of the operating table required. An oblique 4–5 cm incision was made 3 cm ventral to the anterior border of targeted level. The exterior/internal abdominal oblique and transverse abdominal muscles were bluntly dissected with fingers to expose the underlying extraperitoneal fat. With meticulousness, the peritoneum was prudently separated ventrally to allow palpation along the quadratus muscle, transverse process, psoas muscle until the opening between the aorta and psoas was reached. Peritoneal context was pushed further to leave enough space for establishing a working corridor under direct visualization, after which radiography was used for level confirmation with a guide pin inserted into the targeted disc, followed by dilators and retractors placed sequentially. Various curettes and rongeurs were used alternatively to perform discectomy with care to avoid damaging the endplate prior to the insertion of a peek cage (50 mm width, 22 mm length, 13 mm height, 0 angled; Johnson & Johnson, New Brunswick, NJ, USA) filled with allograft bone infiltrated with autologous bone marrow into interbody space by an orthogonal maneuver.

Instrumentation

Classical lateral‐only instrumentation in the OLIF‐LF group was performed according to studies, ¹¹ with entry points located at the center of the vertebrae and trajectories vertical to the sagittal plane (Figure 1A,C). In contrast, the instrumentation in the OLIF‐RLPS group was performed with modified entry points (Figure 1B) and adjusted trajectories (Figure 1D). In detail, the upper entry point was determined at the intersection between one horizontal line about 1 cm above the inferior endplate and one vertical line dissecting the anterior‐middle thirds of the vertebrae. For the inferior entry point, it was at the intersection between one horizontal line about 5 mm below the superior endplate and the same vertical line above (Figure 2A). With entry points confirmed, a pilot hole was drifted with an awl, followed by a gearshift treading at a medial angle of 30° towards the contralateral pedicle (Figure 2B). During the procedure, the C‐arm was used to confirm screws entered the contralateral pedicle without disturbing the spinal canal. The rationale bore a resemblance to percutaneous vertebroplasty technique as following: when the posterior rim of vertebra was reached in lateral view (Figure 3A), the tip of the gearshift should have crossed the medial edge of the pedicle without exceeding the lateral edge in anterior‐posterior view (Figure 3B). Finally, a screw was inserted and progressed along the trajectory until the cortical margin was reached (Figure 3C,D).

Fig. 1.

3‐Dimension sketch map comparing the OLIF‐LF and OLIF‐RLPS techniques. (A) Entry points in the OLIF‐LF technique; (B) Entry points in the OLIF‐RLPS technique; (C) Trajectory in the OLIF‐LF technique; (D) trajectory in the OLIF‐RLPS technique. Abbreviations: OLIF‐LF: oblique lateral interbody fusion with lateral fixation. OLIF‐RLPS: oblique lateral Interbody fusion with reverse lumbar pedicle screw.

Fig. 2.

Sketch map depicting RLPS technique. (A) The red spot represents the upper entry point at the intersection between one horizontal line about 1 cm above the inferior endplate and one vertical line dissecting the anterior‐middle thirds of the vertebrae. The blue spot represents the inferior between one horizontal line about 5 mm below the superior endplate and the same vertical line above; (B) The red arrow represented trajectory pointed to the contralateral pedicle with an angle of 30°.

Fig. 3.

The rationale of fluoroscopy‐based RLPS technique: when the posterior rim of vertebra (green line) was reached in lateral view, the tip of gearshift (black line) should have crossed the medial edge of pedicle (yellow line) without exceeding the lateral edge in anterior‐posterior view (red line). Then a screw was placed along the trajectory until the cortical margin was reached. (A) Sagittal view; (B) the coronal view; (C, D) the transverse view.

Radiographic Evaluation

All patients had received radiological assessment including CT scan and X‐ray. The CT scan was taken in the preoperative planning and postoperative evaluation whereas X‐ray was used for radiographic assessment of DH and SL during the perioperative and follow‐up (3/6/12 months). DH was determined as the average of the anterior and posterior height of interbody space (Figure 4A). SL was determined as the angle between the superior and inferior endplates (Figure 4B). Radiological measurement of DH and SL was analyzed with Software Infinitt PACS (Infinitt Healthcare, Seoul, South Korea) independently by two spinal surgeons blinded to the research. Cage subsidence was rated by the classification presented by Marchi et al. ⁵ based on DH loss at the final follow‐up compared with the postoperative (0–24% was defined as grade 0; 25%–49% as grade 1; 50%–74% as grade 2; 75%–100% as grade 3). Radiological assessment was manually made by two surgeons blinded to the study and the average value was used. A third senior reviewer was available if any dispute occurred.

Fig. 4.

Illustration demonstrating the measurement of DH and SL. (A) DH was determined as the average of a and p; (B) SL was determined as the angle between the superior and inferior endplates. a: anterior disc height; p: posterior disc height. Abbreviation: DH: disc height. SL: segmental lordosis.

Clinical Evaluation

The VAS and ODI score systems were used for clinical assessment at the pre‐operative and last follow‐up.

Statistical Analysis

Data was presented as mean ± standard deviation for various variation, number and percentage for category variation. Statistics evaluation was made with SPSS (Version 18.0; SPSS Inc., Chicago, IL, USA). Student t‐test and Mann–Whitney U test were used for continuous variation according to normality analysis and chi‐square test for category variation. p < 0.05 was regarded significant difference.

Results

Demographic evaluation

As shown in Table 1, a total of 29 cases (eight males and 21 females) were enrolled in this study with an average age of 56.5 ± 7.1 years old including 15 cases in the OLIF‐LF group and 14 cases in the OLIF‐RLPS group. No statistical difference was detected in demographic distribution including sex, aging, follow‐up duration or etiology and level distribution (p > 0.05).

TABLE 1.

Demographical data.

	OLIF‐LF	OLIF‐RLPS	p value
Male/Female	5/10	3/11	0.682
Age (years)	57.5 ± 6.7	55.5 ± 7.5	0.465
Etiology diagnosis			1.000
Lumbar spinal stenosis	1	1
Lumbar spondylolisthesis	6	6
Discogenic lumbar pain	8	7
Fusion level			0.349
L2/3	1	0
L3/4	2	0
L4/5	12	14
Follow‐up (months)	13.3 ± 1.6	13.2 ± 1.9	0.638

Surgical safety evaluation

As shown in the Table 2, the mean operative time and estimated blood loss in the OLIF‐RLPS group were 157.2 ± 29.1 mins and 196.4 ± 84.3 ml, without significant difference compared with 143.7 ± 23.5 mins and 170.0 ± 116.4 ml OLIF‐LF group (p > 0.05). The OLIF‐LF group had reported 26.6% (4/15) cases with complications after surgery including transient psaos/quadriceps weakness (one case), sympathetic chain irritation (one case) and pain/numbness in front of thighs (two cases). Similarly, 28.6% (4/14) cases in the OLIF‐LF group had shown complications consisting of psaos/quadriceps weakness (two cases) and pain/numbness in front of thighs (two cases). All postoperative complications above were significantly alleviated after 2‐week symptomatic treatment. No cases had reported symptom recurrence requiring revision.

TABLE 2.

Surgical safety evaluation.

	OLIF‐LF	OLIF‐RLPS	p value
Operation duration	143.7 ± 23.5	157.2 ± 29.1	0.178
Estimated blood loss	170.0 ± 116.4	196.4 ± 84.3	0.493
Complication profile
Transient psaos/quadriceps weakness	1	2
Sympathetic chain irritation	1	0
Pain and numbness in front of thigh	2	2
Vascular injury	0	0
Visceral injury	0	0

Radiological evaluation

As shown in Table 3, the DH in the OLIF‐RLPS group had increased from 9.07 ± 1.73 mm to 13.73 ± 1.83 mm postoperatively, then decreased to 12.53 ± 1.74 mm in 3 months, 12.09 ± 1.48 mm in 6 months and maintained at 12.00 ± 1.45 mm at the last follow‐up. In contrast, the DH in the OLIF‐LF group was increased from 8.18 ± 1.45 mm to 12.44 ± 1.68 mm after surgery and decreased to 9.43 ± 1.51 mm at the last follow‐up. Though no significant difference was detected during the perioperative between the two groups (p > 0.05), the DH in the OLIF‐RLPS group was significantly higher than the OLIF‐LF group after 3 months (p < 0.05), and the distinction had been maintained until the last follow‐up of 12 months (12.00 ± 1.45 mm vs. 9.43 ± 1.51 mm) (p < 0.05). Based on subsidence classification, only 7.1% (1/14) of cases in the OLIF‐RLPS group had shown more than a 25% loss of DH and were rated as grade 1, significantly less than the OLIF‐LF group with 46.7% (7/15) cases. No cases in both groups had shown subsidence rated grade 2 or more.

TABLE 3.

Radiographic and clinical evaluation.

	OLIF‐LF	OLIF‐RLPS	p value
Radiographic assessment
Disc height (mm)
Pre‐operation	8.18 ± 1.45	9.07 ± 1.73	0.143
Post‐operation	12.44 ± 1.68	13.73 ± 1.83	0.057
3 months	9.90 ± 1.45	12.53 ± 1.74	0.000
6 months	9.59 ± 1.46	12.09 ± 1.48	0.000
12 months	9.43 ± 1.51	12.00 ± 1.45	0.000
Segmental lordsis (°)
Pre‐operation	6.83 ± 3.94	8.35 ± 4.28	0.330
Post‐operation	10.25 ± 3.63	9.90 ± 4.64	0.820
3 months	8.94 ± 2.50	9.21 ± 4.29	0.835
6 months	8.47 ± 2.64	9.07 ± 4.21	0.650
12 months	8.27 ± 2.65	8.88 ± 3.88	0.727
Cage subsidence
Grade 0	8	13
Grade 1	7	1
Grade 2	0	0
Grade 3	0	0
Clinical evaluation
VAS‐leg
Pre‐operation	2.13 ± 0.74	2.29 ± 0.47	0.518
Last follow‐up	0.87 ± 0.52	0.64 ± 0.50	0.245
VAS‐back
Pre‐operation	5.00 ± 0.65	4.79 ± 0.70	0.402
Last follow‐up	1.53 ± 0.64	1.21 ± 0.58	0.172
ODI
Pre‐operation	49.20 ± 8.15	45.57 ± 4.88	0.161
Last follow‐up	11.80 ± 3.90	12.29 ± 3.75	0.735

Notably, the SL in the OLIF‐RLPS group was increased from 8.35° ± 4.28° to 9.90° ± 4.64° postoperatively and reduced to 8.88° ± 3.88° at the last follow‐up. Comparably, the SL in the OLIF‐LF group increased from 6.83° ± 3.94° to 10.25° ± 3.63° after surgery and decreased to 8.27° ± 2.65° at last follow‐up. No significant difference had shown in SL during the perioperative and through the whole follow‐up. A representative case was illustrated in Figure 5.

Fig. 5.

A representative case applying RLPS technique. (A, B) Preoperative X‐ray film in coronal and sagittal view; (C, D) Preoperative sagittal and transverse view of MRI; (E, F) Postoperative X‐ray in coronal and sagittal view; (G, H) Coronal and sagittal view of X‐ray at last follow‐up; (I, J) Postoperative CT scan in sagittal view and 3‐dimonsion reconstruction; (K, L) Transverse view of the fixed levels by CT scan.

Clinical evaluation

The mean VAS scale for back and leg in the OLIF‐RLPS group were 4.79 ± 0.70 and 2.29 ± 0.47 respectively and reduced to 1.21 ± 0.58 and 0.64 ± 0.50 at last follow‐up. Comparably, the mean VAS scale for back and leg in the OLIF‐LF group were 5.00 ± 0.65 and 2.13 ± 0.74, decreased to 1.53 ± 0.64 and 0.87 ± 0.52 at the last follow‐up. No significant difference had been shown in VAS‐back or leg throughout the whole follow‐up. Meanwhile, the mean ODI in the OLIF‐RLPS group decreased from 45.57 ± 4.88 to 12.29 ± 3.75, without marked difference compared with the OLIF‐LF group from 49.20 ± 8.15 to 11.80 ± 3.90 (Figure 6).

Fig. 6.

Radiological and clinical assessment between OLIF‐LF and OLIF‐RLPS. (A) DH changes between the two groups at the preoperative, postoperative, 3 months, 6 months, and 12 months; (B) SL changes between the two groups at the preoperative, postoperative, 3 months, 6 months, and 12 months; (C) VAS score between the two groups at the preoperative and last follow‐up; (D) ODI score between the two groups at the preoperative and last follow‐up. *: p < 0.05.

Discussion

The study had illustrated an innovative fixation technique with screw orientation reverse into the contralateral pedicle, charactered with prolonged trajectory and more compact structure. Consequently, the DH loss in the RLPS group was significantly less than the LF group after 3 months postoperatively, resulting in less cage subsidence at the last follow‐up. Despite no significant difference had been detected in patient‐reported indicators between two groups, the present technique should be regarded as an effective alternative to traditional lateral fixation in OLIF surgery with enhanced mechanical rigidity against postoperative cage subsidence.

Cage Subsidence: A Threat to Traditional OLIF Surgery

OLIF technique has been increasingly performed with great popularity in various lumbar pathologies in recent years. Retroperitoneal access to targeted segments had greatly minimized approach‐oriented injuries to posterior musculoligamentous complex. Moreover, neurological decompression was achieved indirectly by accommodating a large cage without irritation to the spinal canal, significantly decreasing the risk of neurological injury. ¹² , ¹³ Meanwhile, utilization of natural opening between the psoas muscle and the artery obviated the risk of lumbar plexus injury. ¹⁴ Nevertheless, the classical OLIF technique generally required intraoperative position flipping to prone position for posterior pedicle fixation, inevitably resulting in increased operative time due to re‐antisepsis and re‐draping. As an alternative, a single‐position OLIF‐LF technique applying lateral fixation on vertebrae free of intraoperative reposition has been increasingly developed with satisfactory clinical results reported. However, as neurological decompression was achieved by adequately distracting interbody space with an oversized cage with excessive footprint, leading to enlarged foraminal area and increased disc height, the OLIF‐LF technique was incapable of direct circumferential neurological decompression as in the posterior approach, resulting in extreme reliance on the maintenance of distracted DH while decreased DH was reported to directly compromise initial decompression postoperatively and associated with radiculopathy recurrence. ¹⁵ Therefore, efforts had been taken to investigate the etiology underlying postoperative subsidence in which multiple patient‐oriented and procedure‐related factors had been reported, including reduced bone mineral density, improper cage size and position, endplate violation, and so on. ¹⁶ , ¹⁷

Traditional LF Was Vulnerable to Postoperative Cage Subsidence

It had been reported recently that the instruments played a vital role in the development of cage subsidence. Han et al. ¹⁸ compared the incidence of cage subsidence in anterior cervical discectomy fusion (ACDF) with and without instrumentation and concluded the absence of instrumentation to be responsible for higher risk of cage subsidence. Similarly, Pinder et al. ¹⁹ also demonstrated increased subsidence in ACDF without reliable instrumentation. Meanwhile, Jin et al. ²⁰ further investigated the effect of zero‐profile cages and conventional cage‐plate combination on postoperative subsidence and showed increased subsidence by the former, demonstrating the incidence of subsidence to be associated with instrument's rigidity. Therefore, it seemed plausible to conclude that both instrument's presence and strength played an important role in the initiation of subsidence. As for the lumbar pathologies, despite the paucity of direct research on instrument‐related subsidence, biomechanical comparison of different instrumentation in OLIF had been well conducted, in which lateral instrument was shown to provide stronger rigidity than standalone technique, but weaker than bilateral pedicle screw‐rod system. ²¹ Furthermore, Zindrick et al. had described a “windshield wiper” effect due to the rotation center shifting to the distal screw tip secondary to asymmetrical purchase between the cortical end and cancellous end, which would expose the distal end of the screw to excessive mechanical loading, leading to risk of cage subsidence, ²² indicating screws' structural susceptibility to cage subsidence. Therefore, we thought the inferior rigidity of the vertebral screw due to shorter trajectory combined with uneven purchase between two ends had been an important factor associated with cage subsidence in lateral fixation and it was reasonable to modify the current lateral fixation technique for stronger rigidity, to effectively decease the incidence of postoperative cage subsidence. Nevertheless, to our knowledge, no research had been reported to have achieved significant subsidence restriction via altering screw strategies with modified entry points and prolonged trajectory.

RLPS: An Innovative Fixation Technique with Modified Entry Points and Prolonged Trajectory

As a resolution, RLPS technique aimed to achieve profound improvement through the following aspects: (i) longer trajectory into the pedicle: longer trajectory allowed longer screw accommodation. As screws in 45–50 mm length was the common option in classical lateral fixation, RLPS could accommodate screws up to 55 mm with increased purchase against pullout and decreased pressure allocation along the whole trajectory; (ii) balanced mechanics—different from the “cortical‐cancellous” structure in traditional vertebrae fixation, traversing from the cortical vertebral surface into cancellous vertebral body and into the cortical pedicle lateral wall would constituted a “cortical‐cancellous‐cortical” structure producing a more balanced resistance to mechanical pressure; and (iii) more direct support effect—it had been well illustrated that cage subsidence was more likely to occur on the superior endplate than the inferior. ⁹ Therefore, closer cage‐screw intervals in the RLPS technique would facilitate more direct support to cages which was particular advantageous in cases with confirmed endplate violation, as closer proximity between the endplate and screws would let the embedded cage instantly load on the inferior screw and be directly supported by the more balanced mechanical structure with strengthened purchase. Based on the advantages above, the RLPS technique was supposed to provide stronger mechanical rigidity than lateral fixation in restricting postoperative cage subsidence.

Advantages of RLPS in Restricting Postoperative Cage Subsidence Compared with LF

Consequently, though significant improvement had been shown in both groups after surgery compared with the baseline without obvious statistical difference between two groups, DH loss in the RLPS group was significantly less than the LF group after 3 months postoperatively, as the DH had decreased from 13.73 ± 1.83 to 12.53 ± 1.74 mm in the RLPS group and from 12.44 ± 1.68 to 9.90 ± 1.45 mm in the LF group (p < 0.05). Nevertheless, no significant progress had been shown in further follow‐up in both groups as the DH had been maintained at 12.00 ± 1.45 mm in the RLPS group and 9.43 ± 1.51 in the LF group after 12 months, indicating the postoperative subsidence was not a continuous phenomenon aggravating with time. In another indicator based on the subsidence extent, compared with nearly half cases rated as grade 1 in the LF group, only 7.1% (1/14) cases in the RLPS group had shown similar subsidence extent, indicating significant restriction on postoperative subsidence extent. Interestingly, except for cases rated grade 1, mild subsidence (grade 0) was found a common presence in both groups regardless of precautions applied, without endplate violation confirmed intraoperatively. Therefore, we inclined to regard mild subsidence as a prevailing natural phenomenon after cage insertion due to mechanical remodeling at the cage‐endplate interface. Notably, we had not detected any subsidence rated more than grade 2 or cases presented with radiculopathy recurrence, which had been reported in previous literature applying the OLIF technique. One reason was that we had excluded patients with severe osteoporosis or endplate breakage confirmed preoperatively from the indications for the RLPS and LF techniques for fear of instrumentation failure, which were more suitable for posterior pedicle fixation; secondly, as over‐distraction was increasingly recognized as an important risk factor responsible for postoperative cage subsidence, we commonly performed interbody distraction with cages no more than 13 mm to avoid over‐distraction. Lastly, according to stiffness distribution in endplates, ²³ we are accustomed to placing cages more backwards to obtain stiffer support. Those precautions above may account for the absence of more severe subsidence in our study. Notably, with respect to perioperative SL, our results did not show significant difference between the two groups through the whole follow‐up. This result may be ascribed to the application of a 0° cage in both groups rather than 8°, which was better at facilitating SL restoration; in addition, placement more backward had furtherly limited the improvement of SL by interbody distraction.

Controversies remained whether subsidence would directly contribute to radiculopathy recurrence and lead to enhanced risk of revision surgery. As illustrated previously, ⁶ severe cage subsidence (>4 mm) was tightly associated with worse patient‐reported indicators including VAS and ODI, leading to lower fusion rate in long‐term follow‐up. Even the moderate subsidence (2–4 mm) would result in transient clinical worsening compared with the mild (<2 mm). In contrast, our results did not show significant difference in terms of VAS and ODI between the two groups at the last follow‐up, though nearly half cases in the LF group had shown DH loss over 24%, with just 7.1% in the RLPS group. Given the fact that subsidence in both groups were all in mild–moderate extent, with adequate foraminal height and canal volume counteracting the negative effect by DH loss, it was reasonable to suggest mild–moderate subsidence was insufficient to produce significant negative impact on clinical prognosis. Further studies are warranted to investigate whether severe subsidence (grade 2–3) would directly trigger radiculopathy. Nevertheless, given the risk of neurological worsening associated with severe cage subsidence reported before, ⁶ , ⁷ the RLPS technique, effectively restricting postoperative cage subsidence compared with traditional lateral fixation technique, had provided an effective alternative to traditional lateral fixation.

Strengths and Limitations

Though the study has provided preliminary evidence of the RLPS technique on restricting subsidence extent in OLIF surgery, we acknowledge that the retrospective design is a limitation and it is worth noting that the RLPS technique had only been applied in one institution and practiced in a small sample while more studies are warranted in multiple centers with a larger sample to further assess its feasibility and validity. In addition, though the present technique had significantly restricted subsidence extent in OLIF with lateral fixation, its rigidity was still regarded inferior to the posterior dual screw‐rod system. In cases with severe osteoporosis, lateral instrumentation was not an optimal option while posterior fixation with dual‐screw instrument would be a better alternative.

Prospects of Clinical Applications

The technique could be widely applied in degenerative spondylitis including lumbar disc herniation and spondylolisthesis as an effective alternative to traditional lateral fixation, especially in cases with significant risk of postoperative cage subsidence.

Conclusion

The RLPS technique could effectively decrease the extent of cage subsidence postoperatively, due to the prolonged trajectory and more compact structure, compared with classical lateral fixation. Despite no significant difference being shown in patient‐reported indicators, the technique might be regarded as an efficient alternative with enhanced mechanical rigidity in cases with risk of cage subsidence.

Author Contributions

Jinyue He: Data curation; Writing‐Original draft preparation; Writing‐Review and Editing; Software.

Fei Luo: Review; Methodology. Qing Fang: Investigation; Resources. Jianzhong Xu: Review; Editing.

Zehua Zhang: Conceptualization; Supervision; Funding acquisition; Writing‐Review and Editing.

Conflict of Interest Statement

On behalf of all authors, the corresponding author states that there is no conflict of interest.

Ethical Approval

The study was conducted in accordance with the principles of the Declaration of Helsinki and approved by the ethics committee of Southwest Hospital, the Army Medical University.

Consent to participate and publish. Informed consent was obtained from all individual participants included in the study.

Data Availability Statement

All the data and materials have been available and have not been under consideration prior to the submission.