INTRODUCTION
Lumbar foraminal stenosis (LFS) is a common degenerative condition among older adults, primarily resulting from complex spondylotic changes. Its pathophysiology involves multiple factors, including herniated discs, reduced disc height, hypertrophy of the facet joint or ligamentum flavum (LF), syndesmophyte formation, and pedicle impingement [1,2]. These concurrent changes can compress the exiting nerve root (ENR) within the foraminal zone and, in some cases, extend into the extraforaminal or MacNab hidden zones, leading to radiating pain in the lower extremities [3].
Traditionally, fusion surgery has been used to decompress the ENR. However, this approach has notable drawbacks, including the loss of segmental motion and the potential need for extended fusion due to adjacent segment degeneration. Consequently, alternative non-fusion techniques, such as microscopic foraminotomy using the Wiltse approach, were developed to preserve motion at the affected segment [4]. More recently, full endoscopic foraminotomy has been introduced as an alternative method and has been investigated in various approaches by multiple researchers [5-7].
Transforaminal endoscopic lumbar foraminotomy (TELF), which utilizes a standard transforaminal approach through Kambin triangle, has demonstrated favorable clinical outcomes compared with microsurgery, even in patients with LFS accompanied by spondylolisthesis [8-11]. In the transforaminal approach, selecting an appropriate entry angle is critical to achieving adequate foraminal decompression while minimizing the risk of ENR injury. To prevent postoperative segmental instability and reduce facet joint injury, TELF typically involves partial resection of the superior articular process (SAP).
In contrast, we developed a surgical method that employs a more medial skin incision than TELF, creating a steeper approach angle for lumbar foraminal decompression through total resection of the SAP. This procedure, termed selective superior articular process resection (SSAPR), permits the use of a larger endoscope than TELF, which operates within the confined space of the lumbar foramen. Consequently, SSAPR provides greater surgical versatility and facilitates the use of diverse instruments for direct visualization of the ENR.
In this study, we aim to describe in detail the surgical differences between traditional TELF and this novel SSAPR technique. We also compare the clinical outcomes and safety of total SAP resection with those of partial SAP resection, which is conventionally performed to prevent segmental instability during lumbar foraminal decompression. To the best of our knowledge, this is the first comparative study evaluating total versus partial SAP resection in the treatment of LFS.
MATERIALS AND METHODS
1. Study Design and Subject
This study had a retrospective cohort design. Institutional Review Board (IRB) approval was obtained before the study initiation (CHOSUN 2024-09-017). Patient data were collected from a single center from March 2018 to September 2022. Data were gathered retrospectively from the medical records and imaging studies of all patients.
1) Inclusion criteria
(1) Adults ≥18 years, diagnosed with symptomatic LFS.
(2) Demonstrable nerve root compression owing to LFS on lumbar magnetic resonance imaging (MRI) corresponds to radicular pain in the lower extremities.
(3) Lower extremity radicular pain that persisted despite at least 3 months of conservative treatment.
(4) Patients who had undergone full-endoscopic lumbar foraminotomy surgery for the LFS.
2) Exclusion criteria
(1) MRI findings of nerve root compression due to LFS without corresponding lower extremity radicular pain.
(2) Patients with predominant low back pain (LBP) rather than lower extremity radicular pain.
(3) Grade 2 or higher spondylolisthesis or segmental instability.
(4) Grade 3 or higher central canal stenosis [12].
(5) Presence of other pathological conditions such as infection, trauma, tumor, or congenital LFS.
(6) Cases involving extrusion or sequestration of a lumbar disc.
(7) Diseases causing lower extremity pain unrelated to the lumbar spine.
(8) Insufficient medical records and imaging data for analysis.
2. Surgical Procedure
The patients were placed in the prone position with a flexed lumbar posture that reduced lordosis by bending the spine, hips, and knees to widen the lumbar foraminal space. All procedures were performed under epidural anesthesia with minor sedation while monitoring the vital signs.
1) Transforaminal endoscopic lumbar foraminotomy
Under C-arm fluoroscopy guidance, the skin entry point is usually located 10–12 cm lateral to the midline, depending on the size of the patient’s waist. An endoscopic system with an outer diameter of 6.3 mm and a working channel diameter of 3.7 mm and optical angle of 30° was used. A bevel-ended working sheath with an endoscope was docked onto the lateral side of the SAP over the Kambin triangle as the target landing point. Using endoscopic burrs, the SAP and isthmus were minimally partially resected from caudal to cranial and from lateral to medial, following the anticipated course of the ENR. Subsequently, the soft tissue and foraminal ligament compressing the ENR were partially removed. After gently retracting the ENR with the working sheath, the protruding disc fragments were extracted using a pituitary forceps or cutting pituitary forcep.
2) Selective superior articular process resection
The skin entry point was positioned approximately 2 cm lateral to the lateral border of the pedicle. Surgery was performed using an endoscopic system with an outer diameter of 10 mm, working channel diameter of 6 mm, and an optical angle of 15°. A bevel-ended working sheath with an endoscope was docked at the junction of the transverse process (TP), the isthmus, and the SAP. The surgical approaches for TELF and SSAPR are shown in Figure 1 as a schematic figure. Bone work was initiated at the TP junction of the upper vertebra using an endoscopic drill. By drilling, the inferior part of the TP junction, lateral part of the isthmus, and tip of the SAP of the lower vertebra were subsequently removed. The hypertrophied SAP was resected from the tip to the base using an endoscopic drill or osteotome. The area of bone resection through the bone work of the 2 surgical techniques is illustrated in Figure 2. After total resection of the SAP, the LF between the upper and lower pedicles within the foramen area appeared in the top-down view angle. Additionally, the endoscopic device can be easily advanced into the space where the SAP is placed. The exposed LF was removed almost completely using a curette and Kerrison punch. After completing full posterior decompression of the ENR, the protruding disc was treated with annuloplasty using a radiofrequency coagulator, but the disc itself was not removed. Pre- and postoperative computed tomography (CT) images of total SAP resection are shown in Figure 3. Supplementary Video Clip 1 shows the overall surgical procedure for SSAPR.
3. Outcomes Measurement and Evaluation
We collected baseline patient information such as age, sex, type of surgery, operative time, spinal level addressed, number of segments operated on, duration of follow-up, and the presence of accompanying spondylolisthesis. The visual analogue scale (VAS) was used to assess changes in the severity of LBP and radiating leg pain, and the Oswestry Disability Index (ODI) was used to evaluate the impact on activities of daily living. These were measured at specific intervals before surgery, 1 week after, 3 months after surgery, and at the final follow-up. Patient satisfaction with treatment was measured using the MacNab criteria. The outcomes were grouped into 2 categories: favorable, comprising 'excellent' and 'good' ratings, and unfavorable, which consisted of 'fair' and 'poor' ratings. The frequency and proportion of each category were analyzed to provide an overall view of treatment satisfaction.
We also performed a detailed review to identify and quantify the types and occurrence rates of postoperative complications during the follow-up period, including incidental durotomy, motor weakness, infection, and recurrence of foraminal stenosis. In particular, we assessed whether postoperative instability occurred by measuring vertebral slippage at 3 different time points: before surgery, immediately after surgery, and 6 months postoperatively. Vertebral slippage was assessed using plain lateral radiographs in the flexed posture. Additionally, we investigated the incidence and severity of postoperative dysesthesia (POD) one month after surgery and at the final follow-up. Severity was categorized based on the grading system introduced in a previous study [13]. Finally, we integrated these data to comprehensively analyze whether significant differences existed in the clinical outcomes between TELF and SSAPR.
4. Grading System of POD
• Grade 1: Dysesthesia due to compression before surgery. Minimal radiating pain was sustained, similar to preoperative pain in a compressed root-innervated region. The symptoms were not concordant but were similar to the preoperative pain. The symptoms were limited to the follow-up duration.
• Grade 2: Dysesthesia due to dorsal root ganglion (DRG) retraction. Moderate to severe dysesthetic pain or burning dysesthesia in a properly manipulated DRG-innervated region without definite motor deficits. The symptoms were not concordant with preoperative pain, and other characteristics of dysesthesia were present. Symptoms were usually limited to the follow-up duration.
• Grade 3: Dysesthesia caused by DRG injury. Dysesthetic pain is accompanied by motor deficits or atrophic changes in properly manipulated DRG-innervated regions. The patient’s symptoms persisted. This differs from reflex sympathetic dystrophy. Patients with persistent preoperative motor deficits were not classified as POD grade 3.
5. Statistical Analysis
Data were analyzed using R ver. 4.2.2 (R Foundation for Statistical Computing, Austria). For continuous variables, paired t-tests were used based on data normality. Categorical variables were examined using Pearson's chi-square test after verifying the expected frequencies. Statistical significance was set at p<0.05.
RESULTS
1. Baseline Characteristics
A total of 79 patients with 86 surgical levels were included in this study. The mean age of all participants was 61.9±14.2 years. The cohort comprised 39 men (49.4%) and 40 women (50.6%). There were no significant differences in age (p=0.06) or sex (p=0.552) between the groups. In the TELF group, 9 of 43 segments (17.3%) had accompanying spondylolisthesis, whereas in the SSAPR group, 8 of 34 segments (23.5%) presented with the condition. The difference in the prevalence of accompanying spondylolisthesis between the 2 groups was not statistically significant (p=0.666). Regarding surgical levels, the TELF group had a higher proportion of procedures performed at L3–4 and above compared with the SSAPR group. Specifically, in the TELF group, there were 2 cases (3.8%) at L1–2, 7 cases (13.5%) at L2–3, and 16 cases (30.8%) at L3–4, whereas in the SSAPR group, there were no cases at L1–2, 4 cases (11.8%) at L2–3, and 4 cases (11.8%) at L3–4. The proportion of surgeries at L4–5 was similar between the 2 groups, with 22 cases (42.3%) in the TELF group and 14 cases (41.2%) in the SSAPR group. However, the SSAPR group had a significantly higher proportion of surgeries at L5–S1 compared with the TELF group—12 cases (35.3%) versus 5 cases (9.6%), respectively. The average follow-up period was comparable between groups, at 14.2±2.6 months for TELF and 14.9±2.1 months for SSAPR, with no significant difference (p=0.448). The demographic details and baseline characteristics of the 2 groups are summarized in Table 1.
2. Clinical Outcomes
Analysis of the VAS revealed no significant difference in preoperative scores between the TELF and SSAPR groups (p=0.659). However, at the 1-week follow-up, the TELF group demonstrated significantly lower VAS scores compared with the SSAPR group (2.94±0.84 vs. 3.50±0.72, p=0.002). By the 3-month and final follow-up visits, the VAS scores had converged, showing no significant differences between the groups (p=0.917 and p=0.603, respectively).
The ODI also showed a significant preoperative difference, with the TELF group exhibiting higher disability scores than the SSAPR group (76.0±8.4 vs. 69.8±10.6, p=0.008). This trend persisted at the 1-week follow-up, where the TELF group showed lower ODI scores compared with the SSAPR group (30.1±6.2 vs. 33.1±6.5, p=0.046). At the 3-month and final follow-up evaluations, however, no significant differences were observed between the groups (p=0.880 and p=0.776, respectively).
According to the MacNab criteria, a high proportion of patients in both groups achieved favorable outcomes—85.1% in the TELF group and 96.9% in the SSAPR group—with no statistically significant difference between them (p=0.133). The detailed clinical outcomes are presented in Table 2.
3. Complications
Table 3 summarizes the complications associated with each surgical method. In the TELF and SSAPR groups, infections and incidental durotomies were not observed in any patient. Motor weakness was rare, occurring in only one case (2.1%) in the TELF group. Four patients (7.7%) experienced recurrence in the TELF group, whereas none in the SSAPR group.
The grade of dysesthesia was also evaluated. At the 1-month follow-up, the number of patients with a dysesthesia grade ≥2 was similar between the groups: TELF (7 patients, 14.9%) and SSAPR (4 patients, 12.5%) with p=1.000. At the final follow-up, the TELF group had 6 patients (12.8%) with a dysesthesia grade ≥2, whereas the SSAPR group had no patients (0%) with a dysesthesia grade ≥2 (p=0.07). Although this difference did not reach statistical significance, it is noteworthy that the SSAPR group had a relatively lower severity of dysesthesia, with no patients having a dysesthesia grade ≥2 at the final follow-up. This finding is illustrated in Figure 4.
Changes in the length of vertebral slippage were evaluated in the patients with preoperative spondylolisthesis. In the TELF group, the preoperative slippage was 5.49±3.64 (mm), which increased to 7.48±4.78 (mm) postoperatively and further to 8.75±6.78 (mm) at the 6-month follow-up. The net difference in slippage from preoperative to the 6-month follow-up was significant, with an increase of 3.25±3.35 (mm) (p=0.019). In the SSAPR group, the preoperative slippage was 3.67±3.57 (mm), which changed to 4.34±3.25 (mm) postoperatively and 3.86±3.17 (mm) at the 6-month follow-up. The net difference in slippage over the follow-up period was minimal, with a change of 0.2±2.96 (mm) (p=0.858) and was not statistically significant. These changes are shown in Figure 5.
DISCUSSION
1. Traditional Concept: Partial SAP Resection
Prior to the development of endoscopic spine surgery, total resection of the facet joint was required to relieve compression of the ENR caused by LFS. However, this approach frequently resulted in segmental instability, necessitating additional instrumentation and fusion surgery. Consequently, both patients and healthcare providers faced several short-term disadvantages, including higher complication rates, prolonged recovery periods, and increased healthcare costs [14-16]. Long-term drawbacks also emerged, such as the permanent loss of segmental mobility and an elevated risk of adjacent segment degeneration, which often led to extended fusion procedures [17,18]. The advent of endoscopic spinal surgery for LFS was intended to overcome these limitations by decompressing the ENR within the foraminal area while preserving the facet joint.
In this approach, the facet joint is bypassed to access the lumbar foraminal area, resulting in a skin incision positioned more laterally than that used in traditional fusion surgery or the Wiltse approach [19]. The surgical trajectory is deliberately designed at an oblique angle. The endoscope’s diameter is limited by the narrow confines of the lumbar foramen to ensure safe manipulation and minimize the risk of ENR injury. In addition, the endoscopic instruments must be sufficiently long to reach the lumbar foramen from the lateral skin incision, even in patients with moderate obesity. This precise design led to the development of a slender and elongated endoscope capable of accommodating fine, long surgical tools, forming the basis of the modern TELF technique.
However, the concept of partial SAP resection—introduced to preserve the facet joint—places considerable limitations on surgical efficiency and instrument maneuverability. Surgeons must operate through a narrow working channel dictated by the endoscope’s small diameter, which restricts visualization and tool handling. Moreover, the risk of ENR injury remains a significant concern [20-23]. These challenges increase procedural complexity and contribute to the steep learning curve associated with full endoscopic spine surgery [24].
2. New Concept: Total SAP resection
The SSAPR technique introduced in this study utilizes skin incisions and an approach method similar to those described in the Wiltse approach and the full endoscopic paramedian approach by Nam et al. [19,25]. However, it departs from the established principle of minimizing facet joint resection—a concept consistently emphasized in the development of surgical techniques for lumbar foraminal pathologies—by nearly completely resecting the SAP. Because the SAP is not preserved, the surgical trajectory no longer requires the shallow approach angle used in TELF, allowing the skin incision to be positioned more medially. This medial repositioning of the incision, combined with the removal of the SAP obstruction, provides greater operative freedom and yields benefits that exceeded initial expectations. The technical differences between TELF and SSAPR are summarized in Table 4.
3. Main Advantages of Total SAP Resection
First, SSAPR allows the incision to be placed closer to the lumbar foramen. With the removal of the SAP, a larger working space is created, eliminating the need to operate within the confined foraminal area typical of TELF. Consequently, endoscopes with larger diameters and shorter lengths can be employed. The shorter endoscope facilitates easier instrument manipulation and exchange during surgery, thereby reducing surgeon fatigue. Additionally, the larger working channel accommodates more versatile surgical instruments, including larger drills, Kerrison punches, and bipolar coagulators, which enable efficient resection of pathological tissue and effective bleeding control. The benefits of surgeon-friendly devices of this type have been well documented in the literature, as reported by Lim et al. [26]. Collectively, these advantages contributed to the shorter operative time observed for SSAPR compared with TELF in this study. Moreover, in L5–S1 cases, the iliac crest—often a significant anatomical obstacle when planning a TELF approach—poses minimal interference during SSAPR.
Compared with TELF, complete resection of the SAP provides the surgeon with an almost perpendicular and wider field of view of the entire foraminal region. Specifically, SSAPR secures a direct line of sight and an operative range extending from pedicle to pedicle and from the dural sac to the extraforaminal area (Figure 6). This enhanced visualization facilitates the identification of the ENR, dural sac, and disc anatomy, allowing easier removal of the LF from above. Furthermore, by exposing the full height of the intervertebral disc—from the superior to the inferior endplate—SSAPR enables safer and more complete disc procedures for conditions such as bulging, protrusion, or extrusion.
Another major advantage of this expanded surgical field is increased safety with respect to ENR injury. Because SSAPR eliminates the need for endoscopic manipulation within the restricted foraminal space, the surgeon can visualize the ENR and disc from a broader perspective within the cavity created by SAP resection. This significantly reduces the risk of ENR irritation, as supported by this study’s finding that SSAPR resulted in a lower rate of POD compared with TELF. Although the approach described by Nam et al. [25] also provides a vertical surgical view, it involves only partial SAP resection, suggesting that SSAPR offers superior advantages in terms of ease of disc manipulation and reduced risk of ENR injury.
4. Regarding Risk of Postoperative Instability
Despite these advantages, SSAPR in this study did not fall behind TELF in terms of clinical outcomes such as pain reduction. The primary consideration was whether the total resection of the SAP led to segmental instability. Based on the findings of this study, total resection of the SAP did not appear to cause segmental instability or postoperative LBP. Unexpectedly, this study demonstrated that TELF rather than SSAPR was associated with a greater increase in vertebral slippage. Upon investigating the underlying causes of this paradoxical result, the authors concluded that deeper discectomy performed in the TELF group likely altered the disc structure, contributing to the observed outcome. Postdiscectomy instability is often attributed to damage to the annulus fibrosus during surgery, which can compromise the structural integrity of the intervertebral disc [27,28]. It is presumed that the remaining portion of the facet joint, the joint surface formed by the inferior articular process of the upper vertebra, and the pars of the lower vertebra may be sufficient to maintain segmental stability.
Additionally, in TELF, the working sheath is typically docked at the caudal and lateral base of the SAP, and bone work is performed from caudal to cranial and from lateral to medial. Although only partial SAP resection was intended, a greater-than-expected amount of bone may have been removed from the ventral and caudal portion of the SAP during this process. In contrast, during SSAPR, the working sheath is docked more cranially and medially than in TELF. After trimming the lower portion of the isthmus and TP until the facet joint becomes visible, only the upper tip of the SAP is completely removed. While total resection of the SAP tip was intended, the bone forming the articular surface was likely preserved to a greater extent. Given these technical differences, SSAPR may not be inferior to TELF in terms of facet preservation. A future three-dimensional CT-based volumetric comparison would help clarify the actual extent of bone resection between the 2 techniques.
To summarize the points discussed above, spinal surgeons may need to reconsider the longstanding notion of minimizing SAP resection to address foraminal pathologies in spinal surgery. We believe that except in specific cases where SAP preservation is warranted, SSAPR offers ease of surgery and safety.
5. Limitations
This was a retrospective analysis based on medical records, which presents certain limitations in terms of scientific evidence. However, large-scale studies are needed to provide further evidence. Additionally, this study included only cases of unilateral total SAP resection specifically restricted to foraminal pathology. Further research is needed to evaluate the stability of bilateral total resection or determine its broader applicability, extending to paracentral and central pathologies. Furthermore, additional in-depth studies are needed to further validate the impact of discectomy performed during foraminotomy on vertebral stability.
CONCLUSION
SSAPR offers an effective alternative to TELF by enabling comprehensive decompression with improved visualization and instrument versatility without increasing segmental instability. This approach, which eliminates the need to preserve the SAP, results in shorter operative times and lower ENR irritation, highlighting its potential as a rational choice for LFS.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation 
Print 
