Comparison of clinical outcomes between percutaneous endoscopic interlaminar discectomy and arthroscopy-assisted unilateral spinal surgery for lumbar disc herniation: A retrospective study

Abstract

Background:

The growing emphasis on enhanced recovery and reduced morbidity has accelerated the clinical adoption of endoscopic and percutaneous techniques to treat lumbar disc herniation (LDH). Our study aimed to compare the clinical outcomes, safety profiles, and learning curves between the percutaneous endoscopic interlaminar discectomy (PEID) and the arthroscopy-assisted unilateral spinal surgery (AUSS) in patients with LDH.

Methods:

A retrospective analysis was conducted on 60 patients who underwent either PEID (n = 30) or AUSS (n = 30). Patients were matched based on age, sex, and surgical level to improve the comparison. The primary outcomes were assessed using a Visual Analog Scale (VAS), Oswestry Disability Index (ODI), and modified MacNab criteria. The secondary outcomes including postoperative analgesic consumption, operation time, length of stay, surgical complications, postoperative recurrence, and anesthesia-related adverse events.

Results:

Both PEID and AUSS groups demonstrated significant postoperative improvements in VAS and ODI (P < 0.05). There were no significant differences between the two groups in VAS, ODI, modified MacNab criteria, postoperative analgesic consumption, surgical complications, postoperative recurrence, and anesthesia-related adverse events (P > 0.05). The AUSS group had significantly decreased operation time and increased length of stay (P < 0.05).

Conclusion:

Both the PEID and the AUSS groups show significant alleviation of pain and improvement of motor function after operation, few surgical complications and few postoperative recurrence, without intergroup difference. In addition, the AUSS group had significantly decreased operation time and significantly increased length of stay.

INTRODUCTION

Lumbar disc herniation (LDH) remains one of the leading causes of low back pain and radiculopathy worldwide, significantly affecting patient mobility and quality of life, while placing a heavy economic burden on healthcare systems.[1] Historically, open discectomy has served as the gold standard for surgical management.[2] However, the associated tissue trauma, prolonged hospitalization, and relatively high complication rates have led to a paradigm shift toward minimally invasive spine surgery (MISS).[3] The growing emphasis on enhanced recovery and reduced morbidity has accelerated the clinical adoption of endoscopic and percutaneous techniques.[4,5]

Percutaneous endoscopic interlaminar discectomy (PEID) was first introduced by Professor Ruetten and has since emerged as a core technique in MISS.[6] By utilizing a translaminar endoscopic approach, PEID enables direct visualization and targeted decompression with minimal soft tissue disruption. Clinical studies have highlighted advantages such as reduced operative blood loss, shorter hospital stay, and faster functional recovery.[5,7] Nevertheless, limitations including a narrow surgical field and high technical demands necessitate continued refinement and integration of adjunct technologies.

In recent decades, arthroscopy-assisted unilateral spinal surgery (AUSS) has become a new trend in MISS.[8] This technology represents an innovative fusion of spinal endoscopy with arthroscopic guidance. The hybrid approach improves intraoperative visualization, expands the scope of decompression, and simplifies anatomical navigation. AUSS is considered easier to adopt due to its familiar instrumentation and improved ergonomics, potentially shortening the learning curve.[9] However, literature evaluating its clinical efficacy, long-term outcomes, and procedural learning trajectories remains sparse, limiting evidence-based standardization.

Therefore, we conducted this retrospective cohort study and aimed to systematically compare the clinical outcomes, safety profiles, and learning curves between PEID and AUSS in patients with LDH. In addition, we assess the effects of thoracolumbar interfascial plane block combined with intravenous-inhalational anesthesia, a novel multimodal analgesic strategy, on postoperative pain management and anesthesia-related complications, hoping to provide more evidence-based information for spine surgeons.

MATERIALS AND METHODS

Patient case selection

This study received ethical approval from the clinical research ethics committee of our hospital and was conducted in accordance with the good clinical practice guidelines and the principles outlined in the Helsinki Declaration. Since this is a case retrospective study and we didn’t have direct contact with the participants, informed consent was not obtained from the participants. Patients diagnosed with LDH that treated with PEID or AUSS were retrospectively analyzed (since Jan 1, 2023, to Jan 1, 2025). Inclusion criteria were as follows: (1) the diagnostic criteria referred to the LDH clinical guideline of NASS Evidence-Based Guideline Development Committee;[10] (2) single-level herniation treated with PEID or AUSS; and (3) complete 1-year follow-up data. Exclusion criteria including: (1) incomplete follow-up data; (2) multi-level surgery; (3) patients with a history of previous spinal surgery; and (4) patients with concomitant conditions, including trauma, tumors, or infections. After that, we conducted a case–control matching process based on age, sex, and surgical level to improve the comparison between cases that underwent PEID and AUSS. We required the matched cases to have the same sex and surgical level, and to have an age difference within 10 years.

Surgical procedure

Percutaneous endoscopic interlaminar discectomy

The surgical procedure of PEID refers to professor Ruetten’s method.[6] Perform thoracolumbar interfascial plane block combined with intravenous-inhalational anesthesia and place the patient in a prone position, with hip and knee flexed. After fluoroscopy, mark the puncture point at a distance of 1.0 cm from the posterior midline to the affected side according to the corresponding interlaminar space. Puncture with an 18-G needle toward the inner side of the facet joint and the interlaminar space. When the needle reaches the ligamentum flavum, insert a guide wire, remove the needle, make a 7–8 mm skin incision along the puncture point, and gradually expand with an expansion sleeve. After sequential expansion, install a working sleeve and connect it to the endoscope and flushing system. Expand the hole on the ligamentum flavum and gently rotate the conical working sleeve into the spinal canal through the ligamentum flavum fissure under direct visualization, revealing fragmented nucleus pulposus and annulus fibrosus tissue. Use endoscopic nucleus pulposus forceps to remove the protruding nucleus pulposus until the posterior longitudinal ligament and surrounding adipose tissue of the nerve root are clearly visible. Use a bipolar radiofrequency electrotome to ablate and stop bleeding of residual nucleus pulposus tissue, and perform the annulus fibroplasty. The sign of sufficient decompression is that the nerve roots and dural sac can be seen fluctuating with breathing under the microscope. After careful exploration to confirm that there is no residual protrusion of the nucleus pulposus, and then remove the working sleeve followed by wound closure [Figure 1].

Figure 1.

Percutaneous endoscopic interlaminar discectomy for a 52-year male patient with L5/S1 disc herniation. (a and b) Preoperative MRI image showing compression of dura and right S1 nerve root by disc herniation; (c) Confirm the location of working cannula during surgery; (d) S1 nerve root (A) was compressed by prolapsed and sequestrated nucleus pulposus (B), with high tension of the nerve root; (e) The nerve root was decompressed (A) after discectomy. Annuloplasty had been made. Annulus fibrosus and posterior longitudinal ligament shrink (short black arrow)

Arthroscopy-assisted unilateral spinal surgery

The surgical procedure of AUSS refers to professor Song’s method.[8] Under thoracolumbar interfascial plane block combined with intravenous-inhalational anesthesia, the patient is positioned prone with C-arm fluoroscopy employed for intraoperative localization. Surgical landmarks are established at the lower third of the interpedicular line and the superior outer pedicle margins bilaterally. Percutaneous pedicle access is achieved using fluoroscopic guidance. Bilateral guide wires are positioned and secured after verifying needle placement through imaging. A 1.5 cm longitudinal incision is created at the designated surgical level. Through sequential dilation techniques, subcutaneous tissues are expanded to facilitate arthroscopic access (30° scope). Use a radio frequency probe (3.8 mm, 90°) for soft tissue dissection to expose critical anatomical landmarks, including the ligamentum flavum, V-point, facet articulations, and vertebral lamina margins. Surgical decompression involved sequential resection of inferior and superior articular processes, followed by thorough central canal and lateral recess expansion until adequate nerve root mobilization is achieved. Thoroughly remove the protruding nucleus pulposus and perform fibrous ring suturing. The procedure concludes with surgical site irrigation, drain placement, and layered wound closure following hemostasis verification [Figure 2].

Figure 2.

Arthroscopy-assisted unilateral spinal surgery for a 65-year female patient with L4/5 disc herniation. (a and b) Preoperative MRI image showing compression of dura and left L5 nerve root by disc herniation; (c) Confirm the location of high-speed burr during surgery; (d) L5 nerve root (A) was compressed by prolapsed and sequestrated nucleus pulposus (B), with high tension of the nerve root; (e) The nerve root was decompressed (A) after discectomy. The fibrous ring suturing had been made (short black arrow)

Outcome measures

The primary outcome measures including visual analogue scale (VAS)[11] pain scores (before operation, postoperative 1 day, postoperative 3 days, postoperative 1 month and postoperative 1 year), Oswestry Disability Index (ODI)[12] (before operation, postoperative 1 day, postoperative 1 month and postoperative 1 year) and modified MacNab criteria[13] (postoperative 1 year).

The secondary outcome measures including postoperative analgesic consumption within 3 days, operation time, length of stay, surgical complications (infection, hematoma, and nerve or spinal cord injury), postoperative recurrence, and anesthesia-related adverse events (dizziness and/or headache, nausea and/or vomiting, rash and/or itching, and respiratory depression).

Statistical analysis

Analyses were performed using SPSS 24.0 (IBM, Armonk, NY, USA). Categorical variables were presented as absolute numbers (percentages) and evaluated through Fisher’s exact test. Continuous variables were assessed for normality: the data that obeys or approximately obeys the normal distribution were reported as mean ± standard deviation and analyzed with independent t-tests, whereas nonnormally distributed data were expressed as median (interquartile range) and tested using the Mann–Whitney U-test. Meanwhile, P < 0.05 were considered statistically significant for all the above.

RESULTS

Based on inclusion/exclusion criteria, 125 patients were initially enrolled, with 83 underwent PEID and 42 underwent AUSS. Following case-control matching, a total of 60 patients were included ultimately, with 30 in the PEID group and 30 in the AUSS group. No significant differences existed in age, sex, or surgical segment between the two groups, as well as the VAS scores and ODI before operation [Tables 1 and 2].

Table 1.

General information of patients

Variable	PEID group (n=30), n (%)	AUSS group (n=30), n (%)	P
Sex
Male	14 (46.7)	14 (46.7)	1.000
Female	16 (53.3)	16 (53.3)
Age (year), mean±SD	59.0±10.6	61.3±12.1	0.697
Surgical level
L3/4	3 (10.0)	3 (10.0)	1.000
L4/5	11 (36.7)	11 (36.7)
L5/S1	16 (53.3)	16 (53.3)

SD - Standard deviation; PEID - Percutaneous endoscopic interlaminar discectomy; AUSS - Assisted unilateral spinal surgery

Table 2.

Comparison of primary outcome measures between the two groups

Variable	PEID group, median (P25–P75)	AUSS group, median (P25–P75)	Statistical value (U)	P
Before operation
Low back VAS	2.0 (2.0–4.0)	3.0 (2.0–4.0)	412.0	0.342
Lower extremity VAS	6.0 (6.0–7.0)	6.0 (6.0–7.0)	450.0	0.865
ODI	62.0 (60.0–66.7)	62.0 (58.0–66.7)	440.0	0.725
Postoperative 1 day
Low back VAS	2.0 (2.0–3.0)*	2.0 (2.0–3.0)*	415.5	0.372
Lower extremity VAS	3.0 (3.0–4.0)*	3.0 (2.0–3.0)*	420.0	0.452
ODI	31.1 (28.9–34.0)*	31.1 (28.9–35.6)*	420.5	0.458
Postoperative 3 days
Low back VAS	2.0 (2.0–3.0)*	2.0 (2.0–3.0)*	398.0	0.221
Lower extremity VAS	3.0 (2.0–4.0)*	3.0 (2.0–3.0)*	405.5	0.275
Postoperative 1 month
Low back VAS	1.0 (1.0–2.0)*	1.0 (1.0–2.0)*	432.0	0.635
Lower extremity VAS	2.0 (2.0–3.0)*	1.0 (1.0–2.0)*	398.5	0.221
ODI	20.0 (14.0–28.9)*	17.8 (14.0–20.0)*	412.0	0.342
Postoperative 1 year
Low back VAS	1.0 (0.0–2.0)*	1.0 (0.0–1.0)*	412.0	0.342
Lower extremity VAS	1.0 (1.0–2.0)*	1.0 (0.0–1.0)*	398.5	0.221
ODI	14.0 (13.3–17.8)*	14.0 (12.0–15.6)*	412.0	0.342
MacNab (excellent, good), n (%)	23 (76.7)	26 (86.7)	436.5	0.718

*Clinical outcomes showed significant difference before and after surgery (P<0.05). ODI - Oswestry Disability Index; VAS - Visual Analog Scale; PEID - Percutaneous endoscopic interlaminar discectomy; AUSS - Assisted unilateral spinal surgery

Primary outcome measures

The VAS (low back VAS and lower extremity VAS) scores and ODI at different postoperative time points were significantly decreased compared with before the operation in both groups (P < 0.05), which indicated significant alleviation of pain and improvement of motor function. There were no significant differences between the two groups in the VAS (low back VAS and lower extremity VAS) scores, ODI, and modified MacNab criteria at all postoperative time points (P > 0.05) [Table 2].

Secondary outcome measures

There were no significant differences in the postoperative analgesic consumption (Celecoxib), surgical complications (infection, hematoma, and nerve or spinal cord injury), postoperative recurrence, and anesthesia-related adverse events (dizziness and/or headache, nausea and/or vomiting, rash and/or itching, and respiratory depression) between the two groups (P > 0.05). Compared to the PEID group, the AUSS group had significantly decreased operation time (P < 0.05). However, the AUSS group had a significantly increased length of stay (P < 0.05) [Table 3].

Table 3.

Comparison of secondary outcome measures between the two groups

Variable	PEID group, median (P25–P75) or n (%)	AUSS Group, median (P25–P75) or n (%)	Statistical value (U)	P
Celecoxib consumption (tablet, 0.2 g)	6 (4,8)	5 (2,6)	350.0	0.090
Operation time (min)	130 (100–168)	100 (65–125)	280.5	<0.001*
Length of stay (day)	4 (3–0 4)	5 (3–6)	268.0	<0.001*
Surgical complications
Infection	0	0	-	1.000
Hematoma	0	0	-	1.000
Nerve or spinal cord injury	2 (6.7)	1 (3.3)	-	1.000
Postoperative recurrence	2 (6.7)	1 (3.3)	-	1.000
Anesthesia-related adverse events
Dizziness and/or headache	5 (16.7)	3 (10.0)	-	0.527
Nausea and/or vomiting	3 (10.0)	4 (13.3)	-	1.000
Rash and/or itching	0	0	-	1.000
Respiratory depression	0	0	-	1.000

*Clinical outcomes showed significant difference between the two groups (P<0.05). PEID - Percutaneous endoscopic interlaminar discectomy; AUSS - Assisted unilateral spinal surgery

DISCUSSION

The evolution of MISS continues to refine treatment paradigms for LDH, prioritizing reduced tissue trauma, faster recovery, and improved clinical outcomes. PEID and AUSS represent two distinct yet advanced MISS approaches. The comparative analysis between PEID and AUSS for treating LDH in this study reveals critical insights into their clinical efficacy, safety, and operation characteristics. Our findings align with-yet also refine-current understandings of MISS, offering nuanced perspectives for clinical decision-making.

Our results indicate significant and equivalent reductions in low back VAS, lower extremity VAS, and ODI for both PEID and AUSS groups compared to preoperative baselines, persisting across postoperative time points. This aligns with the fundamental goal of LDH surgery: effective neural decompression. The equivalence in the primary outcomes, including VAS, ODI, and modified MacNab criteria, strongly suggests that both techniques achieve comparable levels of pain relief and functional restoration in the hands of skilled surgeons.

The anatomical trajectory and endoscopic fluoroscopy of PEID and AUSS are very similar to traditional posterior laminectomy and discectomy. Both techniques require the removal of a small section of the lamina after gradual stripping of the muscle layers to establish a pathway between the intervertebral spaces.[14] They can avoid the removal of surrounding normal tissues during surgery, which greatly reduces the degree of damage to the surrounding tissues. Meanwhile, they can decompress the spinal canal and protect the medial branch of the dorsal branch of the spinal nerve without additional stretching of the relevant muscles, thus improving the patient’s pain.[5,15] The results also show significant improvement in lumbar spine function in both groups, with ODI measurements significantly lower than preoperative values. Consistent with the previous studies, this indicates that both techniques have an improvement effect on physical dysfunction and cause less damage to paraspinal muscles, which is beneficial for improving spinal stability.[16,17] In addition, both techniques can relieve patients’ pain and enable them to better maintain the movement of the spine after surgery, which is beneficial for rapid recovery and improvement of postoperative functional impairment. The equivalence in modified MacNab criteria further reinforces that the patient’s overall experience and perceived success of the surgery are similar between groups. This holistic measure incorporates not just pain and function but also return to work and daily activities, suggesting both techniques meet patient expectations comparably well.

Neither group experienced infection or hematoma. It is worth mentioning that theoretically, the tube rotation technology of single-axis endoscopy of PEID is more likely to cause iatrogenic nerve injury due to nerve tension. While during the procedure of AUSS, the nerves can be protected through nerve hooks or can be pulled apart after neurolysis, which reduces the risk of nerve injury. However, the incidence of nerve injury in both groups is low, and there is no significant difference between the two groups. This may be because the surgeon of this study has extensive experience and is sufficiently skillful in performing both types of surgeries, especially careful operation to protect the nerves.

The PEID group requires significantly longer operation time (median 130 vs. 100 min, P < 0.001) compared to the AUSS group. We think that this may be attributed to the following points: (1) PEID usually requires precise percutaneous puncture and stepwise dilation, relying on intraoperative fluoroscopy to repeatedly confirm the puncture path. AUSS reduces the need for layer-by-layer separation of surrounding tissues through single hole operation, combined with the wide-angle field of view and flexible instruments of arthroscopy. The assistance of arthroscopy may allow for more direct exposure of the target area and shorten the time required to establish a surgical pathway and to perform surgical procedures; (2) the space for endoscopic operation of PEID is limited, and instruments often need to be used alternately (such as placing the endoscope first and then switching to nucleus pulposus forceps), and narrow channels can easily cause instrument interference. AUSS adopts a multi-channel single-hole design, allowing instruments (such as light sources, attractants, and operating pliers) to enter simultaneously, reducing the frequency of instrument switching. The high-definition field of view of arthroscopy may also reduce the number of adjustments to the endoscopic lens; (3) the flushing of PEID endoscopy may be inefficient, especially when there is significant bleeding that requires repeated flushing or the use of electrocoagulation for hemostasis. The liquid infusion system of AUSS arthroscopy (such as saline flushing) can continuously maintain clear surgical field, reduce visual field interruption caused by bleeding, and avoid frequent hemostasis operations; (4) PEID technique is a difficult procedure with a steep learning curve. It has high requirements for puncture accuracy and endoscopic operation, and beginners need more time to locate and adapt to operations in narrow spaces. For AUSS, if the surgeon is familiar with arthroscopic techniques, the operation may be more proficient; meanwhile, the single hole design may also simplify steps and reduce learning difficulty.[15,18] However, PEID facilitates earlier discharge (median 4 vs. 5 days, P < 0.001). This result is contrary to the recent research.[19] The reason may be due to the smaller incision of PEID (7 mm vs. 15 mm), which results in relatively faster postoperative recovery. Meanwhile, since a drainage tube is placed during the operation, postoperative attention should be paid to volume of drainage and removal of the drainage tube, which may prolong the length of stay. The above results suggest that PEID is more beneficial to postoperative recovery speed, whereas AUSS optimizes intraoperative efficiency.

There is no significant difference in the incidence of postoperative recurrence between the two groups. For these two techniques, although the surgical approach and procedure are different, the scope of decompression and the impact on spinal stability are similar. Therefore, we believe that the postoperative recurrence depends more on patient factors (age, disc degeneration severity) and surgical completeness than the specific MISS technique.

In addition, this study introduces the thoracolumbar interfascial plane block combined with intravenous-inhalational anesthesia. The results show that the postoperative Celecoxib consumption within 3 days (1200 mg in PEID, 1000 mg in AUSS) is essentially consistent with the previous studies.[20,21] Moreover, the incidence of dizziness/headache and nausea/vomiting is 13.% (8/60) and 11.7% (7/60), respectively, which is lower than that under general anesthesia.[22] No serious anesthesia-related adverse events, such as respiratory depression, occurred. These findings indicate the good effectiveness and safety of the thoracolumbar interfascial plane block combined with intravenous-inhalational anesthesia.

The current study has several limitations. First, as a retrospective analysis, patient grouping was not conducted randomly. Spine surgeons decide on surgical procedures based on their clinical judgment, taking into account factors such as imaging features, severity of symptoms, and overall patient condition. Therefore, selection bias may affect the research results. Second, the relatively small sample size. Further expanding our sample population will reduce measuring deviation and eliminate coincidence as much as possible, or even obtain a definite subversive result. Third, the absence of standardized criteria for surgery selection may also affect the universality of the results. It is necessary to conduct prospective randomized controlled trials in the future to further validate the conclusions. Finally, the subjects included in this study all belong to a single center, which relatively limiting the research region. Therefore, in future clinical work, high-quality, multicenter, large sample, and wide case scope studies should be conducted to provide spine surgeons with the best evidence-based information.

CONCLUSION

Both the PEID and the AUSS groups show significant alleviation of pain and improvement of motor function after operation, few surgical complications and few postoperative recurrence, without intergroup difference. In addition, the AUSS group had significantly decreased operation time and significantly increased length of stay. However, larger sample and multicenter studies should be performed to improve the accuracy and reliability of the findings, which can provide spine surgeons with the best evidence-based information.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.