Comparison of short-term clinical outcomes and muscle injury in patients with lumbar spinal stenosis undergoing arthroscopic-assisted uni-portal spinal surgery, unilateral biportal endoscopic surgery, and percutaneous interlaminar lumbar discectomy: a six-month follow-up

Abstract

Objective

This study aims to assess and compare the six-month postoperative clinical outcomes of Arthroscopic-assisted Uni-portal Spinal Surgery (AUSS), unilateral biportal endoscopy (UBE), and percutaneous interlaminar endoscopic discectomy (PEID) for lumbar spinal stenosis (LSS). Additionally, muscle injury associated with these procedures is evaluated by analyzing changes in creatine kinase (CK) and C-reactive protein (CRP) levels.

Methods

A total of 288 patients diagnosed with single-segment unilateral LSS and treated between January 2021 and June 2024 were included in this study. Patients were assigned to the AUSS group (n = 129), UBE group (n = 86), or PEID group (n = 73). Surgical parameters, including operative time, incision length, intraoperative blood loss, and postoperative facet joint preservation rate, were recorded. Clinical outcomes were assessed preoperatively and at 3 days, 3 months, and 6 months postoperatively using the Visual Analog Scale (VAS) for back and leg pain, Oswestry Disability Index (ODI), EQ-5D-5 L quality of life index, and the modified Macnab criteria. The extent of muscle injury was quantified through serum CK and CRP levels measured preoperatively and on postoperative days 1, 3, 5, and 7. Descriptive statistics and multiple comparison analyses were used to assess clinical parameters among the three groups. Longitudinal data were analyzed using a generalized mixed linear model.

Results

The AUSS group demonstrated significantly shorter operative times and smaller incision lengths compared to the UBE and PEID groups (P < 0.001). Postoperative VAS scores decreased significantly in all three groups, with the greatest improvement observed in the AUSS group (P < 0.05). ODI and EQ-5D-5 L scores indicated superior postoperative quality of life in the AUSS group compared to the UBE and PEID groups (P < 0.05). However, no statistically significant differences were observed in the modified Macnab criteria outcomes or postoperative complication rates among the three groups (P > 0.05). The PEID group exhibited the lowest postoperative CK and CRP levels as well as the least intraoperative blood loss (P < 0.05), suggesting minimal muscle injury. The AUSS group showed lower muscle injury levels than the UBE group (P < 0.05).

Conclusion

Arthroscopic-assisted Uni-portal Spinal Surgery (AUSS), unilateral biportal endoscopic (UBE) technique, and percutaneous endoscopic interlaminar discectomy (PEID) are all effective minimally invasive approaches for lumbar spinal stenosis (LSS). AUSS offers shorter operative time and better preservation of anatomical structures. PEID minimizes intraoperative tissue damage and reduces inflammatory response, while UBE balances visualization with operational flexibility. All three techniques demonstrate good short-term clinical outcomes. The surgical approach should be tailored to each patient’s symptoms and individualized needs. Although AUSS demonstrated certain intraoperative advantages, this study did not show that AUSS is superior to UBE or PEID in clinical efficacy or complication control. As a novel technique, AUSS may improve postoperative pain and quality of life, providing a valuable addition to the minimally invasive treatment options for lumbar spinal stenosis.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13018-025-06088-1.

Introduction

The incidence of lumbar degenerative diseases has been steadily increasing in China, driven by an aging population and changes in lifestyle. Among these conditions, lumbar spinal stenosis (LSS) has shown a particularly notable rise in prevalence among the elderly [1]. This disorder primarily results from degenerative changes in bone and soft tissue, leading to ligamentum flavum hypertrophy, facet joint proliferation, and intervertebral disc herniation. These pathological changes contribute to neural compression within the spinal canal, intervertebral foramen, or lateral recess, resulting in lower back pain, neurogenic claudication, and gait dysfunction, ultimately reducing patients’ quality of life [2–5]. Surgical intervention remains the primary treatment for alleviating radicular pain and intermittent claudication in LSS patients [6]. Traditional open surgery often requires extensive soft tissue dissection and significant removal of the lamina and facet joints. These procedures can compromise anatomical support structures, including muscles and ligaments, and may lead to surgical trauma to surrounding tissues, increasing the risk of postoperative back pain, muscle atrophy, and lumbar instability [7, 8]. With the advancement of minimally invasive spine surgery, endoscopic spinal techniques have gained popularity due to their advantages, such as reduced intraoperative blood loss, shorter hospital stays, and minimal tissue trauma. However, percutaneous endoscopic interlaminar discectomy (PEID) has limitations, including a restricted surgical field, constrained instrument maneuverability, and the potential for incomplete decompression [9, 10]. Similarly, unilateral biportal endoscopy (UBE) presents challenges such as a steep learning curve, difficulty in stabilizing surgical instruments, and high visualization demand [11, 12]. To address these limitations, a modified technique—Arthroscopic-assisted Uni-portal Spinal Surgery(AUSS)—was developed based on UBE. This approach enhances surgical field visualization and procedural efficiency. However, its effectiveness in LSS treatment remains insufficiently studied [13]. Previous retrospective studies have shown that UBE and PEID yield comparable clinical outcomes in lumbar discectomy [14], but no comparative studies have evaluated AUSS to date.This study employs a case-control design to systematically assess the short-term (six-month) clinical outcomes of AUSS, UBE, and PEID in LSS treatment. Additionally, preoperative and postoperative levels of creatine kinase (CK) and C-reactive protein (CRP) are analyzed to evaluate muscle injury associated with these three surgical techniques.

Study design and inclusion/exclusion criteria

Patients diagnosed with lumbar spinal stenosis (LSS) and treated between January 2021 and June 2024 were included in this study. A total of 288 patients met the inclusion criteria, among whom 129 underwent the AUSS technique (AUSS group), 86 received the UBE technique (UBE group), and 73 were treated using the PEID technique (PEID group). All procedures were performed as single-stage surgeries, and electronic informed consent was obtained from all participants before surgery. Demographic and surgical data, including age, sex, weight, height, body mass index (BMI), surgical duration, disease stage, intraoperative incision length, and estimated blood loss, were collected. Preoperative and postoperative (day 3) computed tomography (CT) imaging was analyzed to assess facet joint loss. Clinical outcomes of the three surgical techniques were evaluated through preoperative and postoperative assessments at multiple time points, using visual analog scale (VAS) scores for back and leg pain, Oswestry Disability Index (ODI) scores, EQ-5D-5 L index scores, and the modified Macnab criteria. Additionally, serum creatine kinase (CK) and C-reactive protein (CRP) levels were measured preoperatively and on postoperative days 1, 3, 5, and 7 to assess muscle injury. All surgeries were performed by a single experienced spine surgeon with extensive expertise in minimally invasive lumbar decompression techniques. The study received ethical approval from the Institutional Review Board of Qinghai Red Cross Hospital and adhered to the principles of the Declaration of Helsinki.The inclusion criteria were as follows: [1] patients aged 40 to 80 years; [2] those with single-segment, unilateral foraminal stenosis and narrowing of the lumbar spinal canal confirmed by standard imaging; [3] patients presenting typical intermittent claudication along a single-side nerve root without significant improvement following three months of conservative treatment.Exclusion criteria included: [1] patients with a history of revision surgery at the affected site; [2] those diagnosed with other spinal diseases, such as ankylosing spondylitis, spinal tumors, fractures, or spinal tuberculosis; [3] patients with multi-segment central canal lumbar stenosis as confirmed by imaging; [4] those with evidence of any mental or neurological disorder; and [5] patients with other conditions deemed unsuitable for the study by the surgeon [6]. Loss to follow-up.

Surgical techniques

AUSS The AUSS technology (Arthroscopic-assisted Uni-portal Spinal Surgery (AUSS), also known as Uni-portal Non-coaxial Spinal Endoscopic Surgery (UNSES), is a minimally invasive spinal surgical system proposed by Professor Song En in 2021.)involves a percutaneous two-handed operation through a single incision. One hand holds the endoscope, and the other hand uses surgical instruments, making the working channel and the observation channel within the same incision. In the surgical operation, the AUSS technology combines the triangular operation of the arthroscope with the coaxial translation and swinging techniques of the coaxial spinal endoscope. At the same time, it has developed a unique non-coaxial rotation technique specific to the AUSS technology. Based on this, according to the intraoperative situation, the arthroscope and operation tools can be interchanged between the two hands, further expanding the operating space and making it more flexible. This approach offers several advantages, including an open field of vision, increased freedom in the surgical space, and compatibility with various spinal surgical techniques and instruments. AUSS is an advancement and upgrade of foraminal endoscopy and an improvement over the unilateral dual-channel technique.The procedure consists of the following steps: After confirming the patient’s identity, they are placed in a prone position under general anesthesia, with the surgical area leveled, the abdomen suspended, and the lumbar region slightly flexed. Using C-arm fluoroscopy, the target intervertebral space is identified, and the incision point is marked at the intersection of the interlaminar and pedicular lines. The surgical field is then disinfected and draped. A 2 cm incision is made at the marked location, through which a dilator and dissector are introduced to establish an operative channel. The arthroscope and plasma knife are positioned, and equipment functionality is verified.With the endoscope, the soft tissue overlying the intervertebral space is cleared using a 90° plasma knife. A drill is then employed to remove portions of the lower and upper edges of the lamina and partial facet joint structures, exposing the ligamentum flavum insertion. After achieving hemostasis with radiofrequency, the posterior ligamentum flavum is excised, and bone is removed from the contralateral facet joint, revealing the dura mater and bilateral nerve roots. Compressive intervertebral discs and hypertrophic tissue are removed to achieve decompression. The annular defect is evaluated and repaired via electrocautery-induced thermal contraction. Once decompression is confirmed, with visible pulsation of the dura and relaxed nerve roots, gelatin sponges are applied to the dura mater. Finally, instruments and sponges are counted, the incision is closed in layers, and the wound is dressed with sterile adhesive. (see Fig. 1).

Fig. 1.

A:The inferior edge of the upper lamina: B:▲The upper edge of the lamina removed by grinding;C:★The superior edge of the lower lamina;D:#The ipsilateral inferior articular process; E:*The attachment point of the ligamentum flavum;F:☆The ipsilateral nerve root;G:☆The dural sac

UBE

After general anesthesia was induced, the patient was positioned prone, with the surgical area leveled, the abdomen suspended, and a slight flexion applied at the waist bridge. Using C-arm fluoroscopic guidance, the target intervertebral space was located. The target spinous process was aligned with the upper vertebral lamina, and the entry point for the endoscopic channel was marked 1.5 cm proximal to the intersection of a line extending from the pedicle’s inner edge. Routine disinfection and sterile draping were then completed, and the endoscope, radiofrequency electrode knife, burr, and irrigation system were set up.A skin incision was made, extending through the deep fascia to allow for the insertion of a progressively expanding cannula, separating soft tissue within the multifidus muscle triangle. Additional separation was achieved using a UBE dissector, after which the endoscopic channel was inserted and the UBE endoscope positioned. Under endoscopic visualization, instruments were introduced through the operative channel, allowing for flexible instrument exchange as needed. A 405 knife was used to clear soft tissue from the multifidus triangle, exposing the lower edge of the upper vertebral body, the inferior articular process, the upper edge of the lower vertebral body, and the laminar space.A 4.0 cm diamond burr was employed to thin the lower edge of the upper vertebral lamina, and Kerrison rongeurs were used to resect the ligamentum flavum. Further bone removal within the foramen was performed with the burr and rongeurs, eliminating any remaining ligamentum flavum to reveal the shoulder of the traversing nerve root. A nerve dissector was applied to carefully separate and release the traversing nerve root, which was then gently retracted medially using a UBE nerve retractor, exposing the herniated disc.The 301 knife was utilized to remove any superficial blood vessels, followed by an incision of the annulus fibrosus to extract the herniated nucleus pulposus. An irrigator was inserted into the intervertebral space to flush out any residual disc fragments. Finally, the 405 knife was applied to contract the annular tear. The nerve root decompression was confirmed, and the procedure concluded with wound closure and the application of a sterile dressing over the surgical site.

PEID

Following induction of general anesthesia, the patient was positioned prone, with the surgical area leveled, the abdomen suspended, and slight flexion at the lumbar bridge. Using fluoroscopic guidance with a C-arm X-ray, the target intervertebral space was identified, and an appropriate puncture point was marked at the lateral upper edge of the laminar space. Skin markings were applied accordingly, followed by standard preoperative disinfection and the application of a sterile film and drape. The endoscope, radiofrequency electrode knife, burr, and irrigation system were then connected.With anteroposterior and lateral fluoroscopic views, a puncture was made into the lateral upper edge of the laminar space on the right side of the target intervertebral space. A guidewire was inserted, and the skin was incised at the puncture site. A guide rod and sequential dilator cannulae were then placed over the guidewire, with the final insertion of the working cannula. The correct placement of the working cannula was confirmed using the C-arm, after which the guidewire and rod were removed.The intervertebral foramen endoscope was introduced into the working cannula, and optimal fluid flow was adjusted. Soft tissues such as muscle and fat within the visual field were resected, allowing visualization of the superior and inferior laminae and the facet joint. The ligamentum flavum and the medial edge of the dorsal superior articular process within the lateral recess were excised under direct vision, decompressing the spinal canal and lateral recess to expose the dural sac and nerve root. Hemostasis was achieved at bleeding points using radiofrequency.Upon inspection, a herniated nucleus pulposus was observed protruding posteriorly from the target intervertebral space. The herniated nucleus was gradually extracted using grasping forceps, and the intervertebral space was thoroughly cleared. A final examination confirmed a relaxed nerve root, adequate expansion of the dural sac, and no remaining nucleus pulposus within the spinal canal. The annulus fibrosus was sutured, the working cannula was withdrawn, and the wound was closed and covered with a sterile dressing.

Postoperative pain management

Postoperative pain management was consistently administered across all three patient groups. Following surgery, diluted lidocaine was injected at the wound site, with additional oral analgesics provided only if patients reported inadequate pain control. This protocol was designed to prevent both peripheral and central sensitization, aiming to reduce postoperative stress and inflammatory responses.To verify the surgical level and decompression extent, follow-up lumbar CT and MRI scans were conducted, enabling periodic postoperative comparisons. Patients were encouraged to begin mobilizing with a lumbar brace 3–4 h after surgery and were typically discharged within one week. Strengthening exercises targeting the lumbar and back muscles were recommended to commence 4–6 weeks postoperatively, while high-intensity labor was restricted for one year.Patients were scheduled for follow-up visits at 3, 6, and 12 months.

Evaluation method

Clinical indicators were collected at admission and at 3 days, 3 months, and 6 months postoperatively to evaluate primary and secondary outcomes. The primary outcome focused on changes in Visual Analog Scale (VAS) scores, while secondary outcomes assessed clinical efficacy through the Oswestry Disability Index (ODI), EQ-5D-5 L index scores, modified Macnab criteria, and joint preservation rates across the three surgical techniques.The Visual Analog Scale (VAS) was employed to measure postoperative pain levels, targeting lumbar and leg pain at preoperative baseline and at 3 days, 3 months, and 6 months post-surgery. The VAS is scored from 0 (no pain) to 10 (worst pain), with higher scores indicating more severe pain.The Oswestry Disability Index (ODI) is a specialized questionnaire assessing disability and quality of life (QOL) related to lower back function. It includes 10 items, each offering 6 response options scored from 0 to 5. The total ODI score, ranging from 0 to 100, indicates the extent of disability, with higher scores reflecting worse outcomes and a lower quality of life.The EQ-5D-5 L, a general tool for QOL assessment, consists of 5 questions, each with 5 response options. Scores are converted into a single EQ-5D-5 L index value ranging from 0 to 1, where higher values signify better quality of life.The modified Macnab criteria provided a further evaluation of therapeutic efficacy, categorizing outcomes as Excellent, Good, Fair, or Poor. “Excellent” indicates complete symptom relief and a return to normal work and life; “Good” reflects mild residual symptoms with minimal activity limitations and no impact on daily life; “Fair” indicates symptom reduction with activity limitations affecting work and life; and “Poor” denotes no significant improvement or a worsening of symptoms following treatment.

Statistical analysis

Descriptive statistics were applied to assess the general characteristics of the patients. The normality of continuous variables was evaluated using the Shapiro-Wilk test, while homogeneity of variance was assessed with Levene’s test. For normally distributed data, analysis of variance (ANOVA) was performed, whereas the Kruskal–Wallis test was used for non-normally distributed data. Categorical variables were expressed as frequencies and percentages and compared between groups using Pearson’s chi-square test or Fisher’s exact test.Multiple comparisons were conducted for the Visual Analog Scale (VAS), Oswestry Disability Index (ODI), EQ-5D-5 L index scores, and bilateral joint retention rates. Treatment efficacy was determined based on the modified Macnab criteria and postoperative complication rates.Longitudinal data analysis was performed using a mixed-effects model with repeated measures to estimate overall between-group differences. Follow-up time points were treated as categorical variables, with time-treatment interactions and patient ID incorporated as random effects. Given that VAS scores represented repeated measures within the same individuals and were not independent, a generalized linear mixed model (GLMM) was applied for multivariate analysis to examine the effects of the three surgical approaches. The model included VAS as the dependent variable and accounted for fixed effects, including sex, age, body mass index (BMI), disease stage, surgical duration, and bilateral facet joint loss rate.Post hoc exploratory analyses were performed to assess the interactions among potential confounders and their influence on primary outcomes. The primary outcomes were reported with 95% confidence intervals (CIs). Continuous data were expressed as mean ± standard deviation (SD). A p-value of < 0.05 was considered statistically significant. All statistical analyses were conducted using SPSS version 27.0.

Results

Comparison of patient basic data and clinical surgical indicators

A total of 288 consecutive patients were retrospectively analyzed in this single-center study. The patients were divided into three groups: the AUSS group (129 patients), the UBE group (86 patients), and the PEID group (73 patients). Table 1 summarizes the basic data of these groups, including age, sex, height, weight, BMI, and disease stage. No significant differences were observed among the groups (P > 0.05).The average surgical times for the AUSS, UBE, and PEID groups were 90.09 ± 8.73 min, 121.02 ± 22.02 min, and 102.22 ± 8.17 min, respectively. The AUSS group had a significantly shorter surgical time than both the FEID and UBE groups (P < 0.001).A comparison of the postoperative facet joint destruction rates revealed significant differences between the groups, both on the surgical side (23.26 ± 4.82 vs. 25.70 ± 2.74 vs. 22.47 ± 2.81) and the contralateral side (17.14 ± 5.50 vs. 19.92 ± 4.15 vs. 16.88 ± 2.71) (all P < 0.001). The PEID group exhibited a significantly lower joint destruction rate on the surgical side compared to the AUSS and UBE groups (P < 0.001), with this advantage also evident on the contralateral side (P < 0.001).Intraoperative blood loss was significantly lower in the PEID group compared to the other two groups (P < 0.001). Additionally, the AUSS group had a significantly smaller surgical incision compared to the other groups (P < 0.001).

Table 1.

Comparison of basic demographic information and clinical surgical indicators among AUSS, UBE, and PEID patients

	AUSS(129)	UBE(86)	PEID(73)	P-value
Age	51.07 ± 18.50	49.66 ± 18.06	51.81 ± 18.13	0.748
BMI	23.85 ± 2.76	23.87 ± 3.13	24.19 ± 3.42	0.693
Weight	67.37 ± 10.21	67.09 ± 10.60	67.74 ± 11.29	0.929
Height	168.03 ± 8.96	167.38 ± 9.23	169.04 ± 7.89	0.493
Gender(n%)				0.946
Male	63(49)	42(49)	34(47)
Female	66(51)	44(51)	39(53)
Operation time (min)	90.09 ± 8.73	121.02 ± 22.02	102.22 ± 8.17	<0.001
Surgical segments(n%)				0.235
L3-4	11(9)	7(8)	8(11)
L4-5	76(59)	52(60)	52(71)
L5-S1	42(32)	27(32)	13(18)
Intraoperative blood loss (mL)	59.73 ± 5.21	64.77 ± 10.77	36.30 ± 3.88	<0.001
Incision length (cm)	1.51 ± 0.17	3.15 ± 0.18	2.29 ± 0.53	<0.001
Facet loss rate(%)
Advancing side	23.26 ± 4.82	25.70 ± 2.74	22.47 ± 2.81	<0.001
Opposite side	17.14 ± 5.50	19.92 ± 4.15	16.88 ± 2.71	<0.001

Primary outcomes

There were no significant differences in preoperative VAS scores for back and leg pain across the groups (P > 0.05). Postoperatively, all three groups showed a significant reduction in pain, with VAS scores decreasing markedly compared to preoperative levels (see Figs. 2 and 3). However, at 3 days postoperatively, the VAS scores in the AUSS group were significantly lower than those in the UBE and PEID groups (P < 0.05) (see Table 2).

Fig. 2.

The changing trends in back pain scores at different time points before and after the three surgical procedures

Fig. 3.

The changing trends in leg pain scores at different time points before and after the three surgical procedures

Table 2.

Comparison of clinical outcomes between AUSS, UBE, and PEID

	AUSS(129)	UBE(86)	PEID(73)	P-value
Lumbar VAS score
pre-operation	6.66 ± 0.80	6.65 ± 0.79	6.74 ± 0.85	0.744
3 days after operation	3.88 ± 0.79	3.91 ± 0.84	4.45 ± 0.50	<0.001
3 months after operation	2.63 ± 0.77	2.70 ± 0.81	2.81 ± 0.79	0.298
6 months after operation	2.41 ± 0.63	2.48 ± 0.66	2.55 ± 0.63	0.338
Leg VAS score
pre-operation	7.37 ± 0.75	7.37 ± 0.75	7.29 ± 0.79	0.716
3 days after operation	3.36 ± 0.57	3.45 ± 0.61	3.67 ± 0.77	0.005
3 months after operation	2.57 ± 0.66	2.62 ± 0.65	2.99 ± 0.74	<0.001
6 months after operation	2.00 ± 0.79	2.07 ± 0.78	2.23 ± 0.77	0.128
ODI (%)
pre-operation	65.94 ± 6.53	65.93 ± 6.72	65.40 ± 5.47	0.822
3 days after operation	27.58 ± 4.22	28.29 ± 4.44	30.92 ± 5.39	<0.001
3 months after operation	21.60 ± 4.96	22.12 ± 4.92	24.11 ± 5.36	0.003
6 months after operation	16.33 ± 2.97	16.63 ± 2.92	17.27 ± 2.78	0.085
EQ-5D-5 L index score
pre-operation	0.412 ± 0.082	0.406 ± 0.084	0.393 ± 0.086	0.283
3 days after operation	0.631 ± 0.071	0.616 ± 0.077	0.593 ± 0.075	0.002
3 months after operation	0.702 ± 0.036	0.699 ± 0.038	0.686 ± 0.041	0.017
6 months after operation	0.750 ± 0.037	0.742 ± 0.037	0.735 ± 0.040	0.026
Facet preservation rate(%)
Advancing side	76.74 ± 4.82	74.30 ± 2.74	77.53 ± 2.81	<0.001
Opposite side	82.86 ± 5.50	80.08 ± 4.15	83.12 ± 2.71	<0.001

Secondary outcomes

Excluding the preoperative assessment, the ODI scores in the AUSS group were significantly lower than those in the control groups within 3 months postoperatively (P < 0.05) (see Table 2). Furthermore, ODI scores significantly decreased in all three groups postoperatively, with values notably lower than preoperative levels (see Fig. 4). Six months postoperatively, no significant differences in ODI scores were observed between the groups.There were no significant differences in preoperative EQ-5D-5 L index scores among the groups (P > 0.05). However, the EQ-5D-5 L index scores in the AUSS group were significantly higher than in the control groups at all postoperative time points (P < 0.05) (see Fig. 5).Postoperative bilateral facet joint retention rates were significantly higher in the PEID group compared to the other two groups, with a statistically significant difference (P < 0.001) (see Fig. 6).Six months postoperatively, clinical efficacy was assessed using the modified Macnab criteria. No significant differences were found in treatment outcomes for spinal stenosis between the AUSS, UBE, and PEID groups (P > 0.05). Additionally, the incidence of postoperative complications was similar across the three groups (P > 0.05) (see Table 3).Table 4 shows the changes in CK and CRB values at 1, 3, 5, and 7 days postoperatively for each surgical method. The PEID group demonstrated significantly lower values at all stages compared to the other two groups (P < 0.05).

Fig. 4.

The changing trends in ODI scores at different time points before and after the three surgical procedures

Fig. 5.

The changing trends in EQ-5D-5 L scores at different time points before and after the three surgical procedures

Fig. 6.

A: The length of bilateral facet joints before surgery.B: The length of bilateral facet joints after surgery.Advancing side: The length of the facet joint on the surgical side.Opposite side: The length of the facet joint on the contralateral side.Facet joint preservation rate (%): The length of the facet joint after surgery / The length of the facet joint before surgery × 100%

Table 3.

Treatment effect evaluation and complications in the AUSS, UBE, and PEID

	AUSS(129)	UBE(86)	PEID(73)	P-value
Therapeutic effect				0.685
Excellent	49	29	24
good	65	41	34
Fair	13	14	12
Poor	2	2	3
Complications				0.885
Transient nerve injury	1	1	3
dural laceration	2	2	3

Table 4.

Comparison of Pre- and Post-Operative CK and CRP levels across three surgical techniques

	AUSS(129)	UBE(86)	PEID(73)	P-value
CK
pre-operation	103.51 ± 8.76	103.81 ± 8.68	103.49 ± 8.61	0.864
1 days after operation	164.26 ± 23.57	180.13 ± 7.19	131.29 ± 6.60	<0.001
3 days after operation	113.73 ± 10.51	120.26 ± 4.90	100.91 ± 5.08	<0.001
5 days after operation	106.90 ± 13.31	115.54 ± 5.55	89.73 ± 4.09	<0.001
7 days after operation	90.17 ± 13.46	99.40 ± 3.37	71.79 ± 2.63	<0.001
CRB
pre-operation	2.65 ± 1.39	2.66 ± 1.30	2.66 ± 1.43	0.999
1 days after operation	20.96 ± 5.13	22.68 ± 4.30	18.86 ± 5.45	<0.001
3 days after operation	75.47 ± 13.42	85.56 ± 7.93	61.74 ± 3.23	<0.001
5 days after operation	51.77 ± 8.94	57.93 ± 6.09	43.73 ± 5.06	<0.001
7 days after operation	30.47 ± 5.93	34.05 ± 4.20	25.29 ± 3.33	<0.001

Post-Hoc exploratory analysis

The post-hoc exploratory analysis revealed no significant correlations between sex, age, BMI, surgical time, disease stage, or the loss rate of bilateral facet joints and the VAS scores for legs and back during the follow-up period (P > 0.05) (see Supplementary Materials: Tables 5 and 6).

Discussion

Research indicates that the posterior column structure plays a critical role in maintaining spinal stability, and its functional integrity directly influences postoperative outcomes [15]. The posterior column includes the interspinous ligaments, facet joints, and joint capsules. However, traditional open surgery often causes extensive damage to surrounding skin and muscle tissues, leading to instability of the posterior column. This, in turn, can result in further intervertebral disc degeneration, contralateral neuropathy, perioperative nerve injury, stress shift, and bone destruction [16]. Consequently, minimizing damage to the paraspinal muscles and posterior column structure is essential for long-term outcomes [17].In recent years, minimally invasive surgery has gained popularity among spine surgeons due to its benefits, such as preserving normal vertebral structure, minimizing surgical trauma, reducing operative time, and facilitating faster postoperative recovery [18]. Previous studies have shown that the UBE approach, which offers a large surgical angle, provides a decompression effect similar to that of open surgery. It causes minimal damage to the paraspinal muscles, significantly enhancing spinal stability [19, 20]. Its efficacy in treating lumbar spinal stenosis has been widely researched and confirmed. Furthermore, UBE provides a broader field of view compared to PEID, with separate observation and working channels, allowing for greater freedom in instrument handling. When using a 30° endoscope, the contralateral lamina gap is easily visible, making the technique particularly convenient, as spinal instruments are easier to operate than with tubular endoscopes [10]. However, the steeper learning curve extends the training period for novice surgeons [21, 22], and the high water pressure during surgery may result in postoperative headaches.In contrast, PEID has a shorter learning curve compared to UBE, as noted in previous studies [23, 24]. However, it has a smaller operative field, lower instrument efficiency, and a more limited decompression range. This is especially problematic for central lumbar disc herniation (LDH), where removing a laterally displaced disc often requires rotating the cannula. This may compress the contralateral nerve and dural sac, leading to postoperative perineal numbness or dorsiflexion weakness [25]. Additionally, when the nerve root is retracted and protected by the working channel, patients undergoing PEID may experience more severe, even intolerable, intraoperative pain, which increases the risk of iatrogenic nerve root injury [26, 27]. Our study also found that the incidence of transient nerve injury following PEID was significantly higher than in the other two groups.AUSS is an improvement on the UBE technique and provides a new approach to treating lumbar degenerative diseases. The anatomical trajectories of AUSS, UBE, and PEID are similar to those of traditional posterior laminectomy. All three techniques involve gradually stripping muscle layers and removing a section of the lamina to establish a pathway between the intervertebral spaces. However, AUSS can be combined with traditional posterior decompression instruments used in spinal surgery, achieving decompression results comparable to open posterior surgery while ensuring surgical efficiency [28].In our study, PEID demonstrated the least facet joint loss, likely because it uses a single working channel that primarily targets the intervertebral space and disc area. The interference with the facet joints and lamina is minimal. The procedure is relatively focused and specific, primarily addressing disc herniation or intraspinal lesions, with minimal involvement of the facet joint area. In contrast, UBE, when dealing with conditions such as facet joint hypertrophy or lateral recess stenosis, often requires more dissection and exploration around the lamina and facet joints to fully expose the surgical field and facilitate instrument manipulation. This sometimes involves additional grinding or resection of the facet joints to achieve sufficient decompression and expand the nerve root canal, increasing the risk of facet joint damage.AUSS, however, may reduce facet joint loss compared to UBE. This is due to the use of the annular bone resection technique in AUSS, which precisely defines the boundaries for bone resection along the superior and inferior edges of the lamina and the facet joint root. This technique expands the spinal canal volume while preserving the bony connection structure of the facet joint, thus maintaining the biomechanical stability of the posterior column. Additionally, dynamic visual control and a step-by-step approach are key safeguards. The panoramic visualization provided by the single-port endoscope system enables the surgeon to monitor the depth and range of bone resection in real-time. This is combined with a standardized procedure of “bone decompression first, followed by ligament treatment,” which involves laminectomy, lateral recess decompression, and the resection of hypertrophic ligamentum flavum, followed by full exposure of the nerve root and dural sac. Radiofrequency knives are used for precise ligament-bone interface separation, effectively avoiding blind bone grasping due to limited visualization and reducing unnecessary facet joint resection.In summary, while there are subtle differences in facet joint resection among the three techniques, all three preserve the integrity of the facet joints, spinous processes, and supraspinous ligaments. This preservation is crucial for maintaining spinal stability, a finding consistent with previous research [29].

Furthermore, in this study, postoperative visual analog scale (VAS) scores for lumbar and leg pain significantly decreased in all three groups compared to preoperative values, indicating that each technique effectively alleviated pain. However, the most pronounced improvement was observed in the AUSS group. The UBE technique enables spinal canal decompression while preserving the medial branches of the dorsal rami without additional muscle traction, thereby reducing pain. While maintaining a minimally invasive approach, the AUSS technique further expands the operative field and decompression range, minimizing the removal of normal tissue and significantly reducing collateral damage.This technique may contribute to improved postoperative pain relief, particularly by minimizing collateral tissue damage. In contrast, the PEID technique may lead to transient nerve stimulation or microtrauma due to direct contact between the tubular retractor and the spinal nerve roots or adjacent structures, potentially causing postoperative pain.At the six-month follow-up, ODI and EQ-5D-5 L scores improved significantly in the AUSS group compared to the UBE and PEID groups. Previous studies have reported that recurrent back and lower limb pain occurs in up to 30% of patients following discectomy, primarily due to postoperative fibrosis [30]. This fibrosis replaces normal epidural tissue, resulting in nerve root compression, recurrent pain, and ischemia due to epidural traction. Preventing or mitigating epidural fibrosis is therefore crucial in reducing postoperative pain [31, 32]. The ligamentum flavum (LF) serves as a natural barrier that limits fibrosis [33]. The AUSS technique, with its flexible surgical angle and enhanced visualization, allows better preservation of the LF, an advantage that is challenging to achieve with other minimally invasive techniques.Regarding complications, two cases of dural tears and one case of transient sensory abnormality due to temporary nerve root injury were reported in the AUSS group. In the UBE group, one case of dural tear and two cases of transient sensory abnormalities were recorded, whereas the PEID group experienced three cases of dural tears and three cases of transient sensory abnormalities. Previous studies have identified dural tears as the most common complication in minimally invasive spine surgery, with an incidence rate of 1.9–5.8% [34], consistent with our findings. These tears typically occur at the cranial or caudal ends of the target segment, the intervertebral disc level, or the medial facet joint, with headache being the most common clinical manifestation. Risk factors include mechanical or radiofrequency-induced dural injury, intraspinal adhesions, large intervertebral disc fragments, and dural laxity [12, 35]. Thus, a thorough understanding of dural-spinal ligament anatomy and proper instrument handling is essential.The incidence of transient sensory abnormalities was approximately 0.14%, primarily manifesting as postoperative pain and numbness [36]. In the PEID group, nerve root injury was often attributed to nerve rotation and compression from the intravertebral tubular retractor. In UBE procedures, nerve root traction or displacement may be required to optimize the surgical field and facilitate decompression. Although excessive traction is avoided, minor nerve root traction or compression may still occur due to the high precision required, potentially causing transient postoperative numbness. These symptoms typically resolve within three months.The high flexibility and enhanced visualization of the AUSS technique may help reduce the incidence of such complications.C-reactive protein (CRP), an acute-phase protein synthesized by the liver, is widely used to assess inflammation, infection, and muscle damage due to its high sensitivity [37]. In the absence of infection, peak postoperative CRP levels are considered markers of tissue injury [38]. CRP levels rise rapidly postoperatively, peak on the third day, and gradually decline, typically returning to baseline within 10 to 14 days. In spinal surgery, peak CRP levels correlate with surgical type and duration, with higher values observed in fixation surgeries compared to single-segment decompression procedures [39].Serum creatine kinase (CK) levels are a widely recognized indicator of postoperative muscle injury, typically peaking on the first postoperative day. CK elevation is closely associated with the invasiveness of the procedure and mechanical stress on the paraspinal muscles during surgery [40–42]. In this study, CK levels were lower in the PEID group than in the UBE and AUSS groups. Across all three groups, CK levels peaked on the first postoperative day and subsequently declined to normal levels. Additionally, CRP levels were lower in the PEID group compared to the other two groups, as shown in Figure 7. This finding is attributed to the PEID technique’s use of sequential dilators and blunt obturators for muscle splitting, which preserves paraspinal muscle integrity.The AUSS technique employs a single-port, dual-channel system that allows simultaneous visualization and instrument operation through a 1.5 cm incision, reducing soft tissue dissection by 50% compared to the UBE system. Consequently, AUSS may result in reduced soft tissue trauma compared to UBE, based on incision length and dissection range.The UBE technique, utilizing a dual-port system, employs sequential dilators, bipolar radiofrequency probes, and continuous saline irrigation to create an operative space within the interlaminar gap without a working tube. It also incorporates minimal muscle dissection techniques. The extent of soft tissue damage is closely linked to intraoperative blood loss. In this study, blood loss was lowest in the PEID group. The AUSS technique, which offers a larger surgical field and greater maneuverability, overcomes the limitations of single-axis endoscopic techniques while minimizing tissue trauma. Additionally, its high-definition arthroscopic imaging system provides clear visualization of the spinal cord, neural structures, epidural venous plexus, and microvasculature. This enhanced visualization allows for effective decompression while improving hemostasis.In summary, all three minimally invasive spine surgery techniques demonstrated favorable clinical efficacy in treating lumbar spinal stenosis. However, surgical approach selection should be tailored to individual patient symptoms and clinical needs. This study, however, has several limitations. The AUSS technique, a refinement of the domestic UBE technique, is relatively new, and the six-month follow-up period is insufficient to assess long-term clinical outcomes. Additionally, this study was a single-center retrospective analysis with a limited sample size. The learning curve remains undetermined, and recurrence and reoperation rates could not be definitively assessed. Future studies should include large-scale, prospective, multicenter trials to compare the long-term clinical outcomes of AUSS with other endoscopic techniques.

Fig. 7.

A: Trends in C-reactive protein (CRP) changes at different time points among three surgical groups.BTrends in creatine kinase (CK) changes at different time points among three surgical groups

Fig. 8.

A case of lumbar spinal stenosis treated with AUSS surgery is presented. The patient, a 49-year-old female, had experienced low back pain with radiating pain in the lower limbs for 5 years, which had recently worsened, prompting her admission to the hospital. She was diagnosed with lumbar spinal stenosis. Figures A-H represent preoperative imaging, while Figures I-L show postoperative imaging.

Conclusion

Arthroscopic-assisted Uni-portal Spinal Surgery(AUSS), unilateral biportal endoscopic (UBE) technique, and percutaneous endoscopic interlaminar discectomy (PEID) are all effective minimally invasive approaches for lumbar spinal stenosis (LSS). AUSS offers shorter operative time and better preservation of anatomical structures. PEID minimizes intraoperative tissue damage and reduces inflammatory response, while UBE balances visualization with operational flexibility. All three techniques demonstrate good short-term clinical outcomes. The surgical approach should be tailored to each patient’s symptoms and individualized needs. Although AUSS demonstrated certain intraoperative advantages, this study did not show that AUSS is superior to UBE or PEID in clinical efficacy or complication control. As a novel technique, AUSS may improve postoperative pain and quality of life, providing a valuable addition to the minimally invasive treatment options for lumbar spinal stenosis.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

Shihao Zhou and xiaowan xu conceived and designed the trial.Tianluo Guo collected patient sampl.junhao sun analysed and interpreted the data.Jiancuo A the final version of the manuscript. All authors read and approved the final submitted manuscript.

Funding

This study was supported by Qinghai Provincial Science and Technology Program: Construction and Application of Personalized Surgical Strategies for Severe Congenital Scoliosis Based on Multidimensional Digital Technology (2023-SF-118).

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval

All procedures were conducted in accordance with the ethical standards of the institutional and/or national research committees. This study was approved by the Ethics Committee of Qinghai Red Cross Hospital (KY-2024-60). Informed consent was obtained from all patients.

Competing interests

The authors declare no competing interests.