Clinical efficacy of intervertebral foraminoscopic debridement with catheter drainage for lumbar spinal epidural abscess Authors: Yinqiu He, Lijin Yang, Tianhao Liu, Xin Long, Xuefeng Feng, Zenglin Xiao, Zucai Pang Published: 2026 ABSTRACT AbstractObjectivesTo evaluate the clinical efficacy of percutaneous transforaminal endoscopic debridement with catheter drainage for the treatment of single-segment lumbar spinal epidural abscess (SEA).MethodsWe retrospectively analysed clinical data from 12 patients with nonspecific single-segment spinal epidural abscess who underwent percutaneous transforaminal endoscopy at our hospital between March 2019 and June 2023. Intraoperative lesion tissue was subjected to microbial culture and histopathological examination. Postoperative anti-infection treatment was given based on drug sensitivity and clinical response. Postoperative CT/MRI was performed to evaluate abscess clearance. WBC, CRP, and ESR levels were measured before and after surgery, and the degree of symptom relief was evaluated using the VAS, ODI, and MacNab criteria. Dynamic radiographs were obtained to evaluate spinal stability, and imaging data were reviewed to assess infection recurrence.ResultsAll patients successfully underwent surgery without postoperative complications. Various indicators improved significantly at 1 week, 1 month, and 3 months after surgery, as well as at the last follow-up, compared with preoperative levels (P < 0.05). Clinical symptoms improved significantly. According to MacNab criteria, the excellent and good rate was 91.7%. Two patients (16.7%) underwent lumbar fusion surgery due to new neurological symptoms in both legs after 3 months. In the remaining patients, infectious symptoms improved, and no recurrence occurred during the follow-up period.ConclusionsPercutaneous transforaminal endoscopic debridement with catheter drainage is a valuable and minimally invasive surgical approach for the treatment of SEA with preliminary feasibility and safety under certain conditions.Keywords: Spinal epidural abscess, Percutaneous transforaminal endoscopy, Minimally invasive technique Introduction Spinal epidural abscess (SEA) is a purulent infection caused by pathogenic bacteria entering the epidural space and represents a severe infection of the epidural cavity. Owing to the absence of valves in the venous plexus, which allows bidirectional systemic circulation, SEA often arises from haematogenous spread or diffusion from adjacent osteomyelitis or discitis. Although SEA is uncommon, its incidence is increasing [1]. Early diagnosis and prompt treatment are crucial, and patients typically require surgical decompression combined with long-term antibiotic therapy [2, 3]. However, even with surgical intervention, SEA has a high mortality rate, with some studies reporting a three-month mortality of up to 15% and a recurrence rate of up to 30% [4]. As SEA predominantly affects individuals over 60 years of age and those with compromised immunity or multiple comorbidities, safer alternatives are needed to reduce mortality, recurrence, long-term sequelae, and healthcare costs. In this study, we applied percutaneous transforaminal endoscopy to treat SEA, offering a safe surgical approach. The objectives of this study were to (i) introduce the application of percutaneous transforaminal endoscopic debridement and catheter drainage in SEA surgery; (ii) report the initial results; and (iii) evaluate its feasibility and safety. Methods Patients and data collection Our institution obtained written informed consent from all participants. We retrospectively collected data from patients with nonspecific SEA treated at our hospital between March 2019 and June 2023. The inclusion criteria were as follows: ① patients with no underlying cause or with nonspecific single-segment lumbar infection; ② patients with significant low back pain, with or without symptoms of nerve compression, and poor response to conservative treatment; ③ patients whose MRI findings indicated abscess formation in the intervertebral space or spinal canal, with compression of the dural sac and/or nerve roots; ④ patients who provided informed consent for the treatment plan; and ⑤ patients who were available for follow-up for more than 1 year. The exclusion criteria were as follows: ① multisegmented lumbar infection; ② severe spinal deformity, instability, or severe vertebral bone destruction; ③ large paravertebral abscess; ④ abscess extending over two segments; ⑤ tolerance to the prone position < 30 min; ⑥ significant coagulation dysfunction; and ⑦ incomplete follow-up data. Twelve SEA patients, aged 52 to 77 years (mean: 65.8 ± 7.3 years), met the inclusion criteria and were included in this study. The duration of symptoms ranged from 2 to 21 days (mean: 7 ± 6.35 days). Comorbidities included diabetes mellitus in 5 patients, end-stage renal disease in 4 patients, coronary heart disease in 3 patients, a history of spinal surgery (percutaneous vertebroplasty) in 2 patients, a history of cancer in 2 patients, and pulmonary infection in 1 patient. The affected segments were L1/2 in 1 patient, L2/3 in 1 patient, L3/4 in 2 patients, L4/5 in 5 patients, and L5/S1 in 3 patients (Table 1). All patients presented with low back pain and unilateral or bilateral nerve damage at admission. Eight patients (66.7%) had a fever, with a maximum temperature of 40.6 °C. Preoperative X-ray, CT, and MRI were performed to determine the extent of the abscess, the specific segment, degree of vertebral destruction, and spinal stability (defined as sagittal translation > 3 mm, or an angular change > 10°). Table 1Summary of demographic characteristics for patients treatedCase No.Clinical symptomsDiseased SegmentsComorbiditiesWBC (10^9/L)ESR (mm/h)CRP (mg/L)SDP (days)1AL5/S1Coronary atherosclerotic heart disease16.7248101.8222BL4/5Chronic renal insufficiency22.286090.7333AL5/S1Chronic renal failure16.9475128.6364AL4/5Diabetes mellitus, Lung cancer, Chronic cardiac insufficiency35.814797.3675BL2/3Hypertension, Diabetes mellitus9.888958.3856AL3/4Hypertension16.4668165.1177AL4/5Chronic renal failure, Hypertension, Diabetes mellitus16.6462256.98218AL1/2Hypertension, Coronary atherosclerotic heart disease12.5196137.6059BL4/5Spinal cord injury, Hypertension23.5970314.76210BL4/5Sepsis, Lobar pneumonia22.5747116.94211BL5/S1Breast malignancy, Chronic renal insufficiency, Hypertension10.616090.60512AL3/4Hypertension, Diabetes12.145352.0419Abbreviations: A Lumbar pain, B Lumbar pain with lower limb, WBC White blood cell count, CRPC-reactive protein, ESRErythrocyte sedimentation rate, SDP Symptom duration period Summary of demographic characteristics for patients treated Abbreviations: A Lumbar pain, B Lumbar pain with lower limb, WBC White blood cell count, CRPC-reactive protein, ESRErythrocyte sedimentation rate, SDP Symptom duration period Surgical method After successful intravenous anaesthesia, the patient was positioned prone, and the responsible segment was located under fluoroscopy to mark the puncture site. The surgical area was routinely disinfected, draped, and covered with 3 M film. Local infiltration anaesthesia with 1% lidocaine was administered from the puncture site to the superior articular process. Under fluoroscopic guidance, the puncture needle was inserted into the target location of the lesion gap. After confirming the appropriate puncture position using C-arm fluoroscopy, an 18G puncture needle with a 20 ml syringe was used to aspirate pus for examination. If the sample was not extracted, 5–10 ml of normal saline was injected and re-aspirated. A guidewire and dilating cannula were then introduced, and a trephine was used to partially reshape the articular process, expand the intervertebral foramen, and establish a working channel. A transforaminal endoscope was inserted to explore the spinal canal. If the spinal epidural abscess was in the intervertebral space, it could be seen flowing out. A portion of the lesion tissue was removed using nucleus pulposus forceps for pathological examination and microbial culture. Necrotic tissue and the abscess wall surrounding the lesion were removed, and the surrounding capsule was widely cleared. The area was explored sequentially from the head to tail, as well as within and outside the intervertebral foramen, to observe any residual necrotic tissue or abscess. The intervertebral space was continuously flushed with normal saline, and the nerve roots and dural sac were examined to ensure no significant compression, allowing the dural sac to pulsate freely. After decompression, a flushing and drainage tube was inserted through the working cannula into the lesion gap. The flushing system consisted of a disposable epidural steel wire catheter (Jiangsu Maichuang Medical Device Co., Ltd., Model F3-IV, specification: d; outer diameter: 1.0 mm) and a disposable drainage tube (Suzhou Tianping Huachang Medical Device Co., Ltd.; specification F12; outer diameter: 4.0 mm). A small incision was made 3 cm from the end of the negative-pressure drainage tube, and the disposable epidural steel wire catheter was threaded through the opening into the tip of the drainage tube. Water was tested and seen flowing out of the drainage tube opening. The connection was sealed with sterile adhesive film to prevent leakage. The incision was sutured and fixed, completing the surgery (Fig. 1). All procedures were performed by the same surgeon with more than 10 years of experience. A–C. Appearance of the flushing and drainage tube; D–E. Connect syringes and drainage tube; F. Appearance of the placed drainage tube Postoperative management Pathological examination and bacterial culture were performed on the lesion tissues, and pus samples were obtained during the operation. Sensitive antibiotics were promptly administered based on drug sensitivity test results; empirical broad-spectrum antibiotics were used if the cultures were negative. Postoperatively, the drainage status was closely monitored, and the drainage tube was kept unobstructed. If no drainage fluid was observed, blockage of the drainage tube orifice was considered, and normal saline was injected to flush the tube. The drainage tube was removed after no further drainage and two consecutive negative bacterial cultures. If there was no spinal instability after surgery and the patient could tolerate pain with improved fatigue symptoms, they were encouraged to wear a brace and stand and walk as soon as possible, while taking precautions to prevent the drainage tube from falling off. Intravenous antibiotics were administered for 4 weeks, followed by oral medication for an additional 3 months. Routine blood tests, inflammatory markers, and outpatient follow-ups were conducted to monitor recovery of postoperative symptoms and signs. Relevant imaging studies (X-ray, CT, MRI, or enhanced MRI) were reviewed to evaluate spinal mechanical stability and infection control. Efficacy evaluation General information for 12 patients and preoperative inflammatory markers, including WBC, CRP, and ESR, were recorded to assess patients’ general condition before surgery (Table 1). Clinical and radiological evaluations, including X-ray, CT, MRI, or enhanced MRI, were performed to assess local infection control and spinal mechanical stability. Operative time and intraoperative blood loss were recorded. Preoperative and postoperative Oswestry Disability Index (ODI), visual analogue scale (VAS), and Japanese Orthopaedic Association (JOA) scores were used to assess clinical outcomes, with comparisons presented in Table 2. The modified MacNab criteria were applied to assess efficacy at the final follow-up. Pathological results and positive bacterial culture findings were documented. Inflammatory markers, including WBC, CRP, and ESR, were compared before and after surgery to assess the patient’s physical status. Table 2Clinical outcomes of patients before surgery and one week after surgeryPre-operationPost-operationp-valueVAS7.58 ± 0.994.75 ± 0.62<0.001ODI(%)63.17 ± 7.0653.50 ± 8.74<0.001JOA10.75 ± 2.7717.75 ± 4.00<0.001Abbreviations: ODI Oswestry Disability Index, VAS Visual analogue scale, JOA Japanese Orthopaedic Association Clinical outcomes of patients before surgery and one week after surgery Abbreviations: ODI Oswestry Disability Index, VAS Visual analogue scale, JOA Japanese Orthopaedic Association Statistical methods Data analysis was performed using SPSS 27.0 statistical software (IBM Corporation, USA). Mauchly’s test of sphericity was conducted for VAS, ODI, JOA, WBC, CRP, and ESR values before and after surgery. One-way repeated-measures analysis of variance was used for indicators that met the sphericity assumption, while the Greenhouse–Geisser correction was applied when the assumption was violated. Post hoc comparisons were conducted using the least significant difference (LSD) method for pairwise comparisons at each time point. A P value < 0.05 was considered statistically significant. GraphPad Prism 10.0 was used for graphical presentation. Results The surgical outcomes are summarized in Table 3. All patients completed the procedure without any damage to the dural sac, nerve roots, blood vessels, or organs. Additionally, there were no complications, such as haematoma, sinus formation, or mixed infection. The operative time ranged from 75 to 125 minutes, with an average of 98.33±16.28 minutes. Intraoperative blood loss ranged from 10 to 40 mL, with a mean of 22.5±8.91 mL. The length of hospital stay ranged from 20 to 56 days, with an average of 28.92±10.74 days. All patients were followed up for more than one year, with a follow-up duration of 13 to 24 months (mean 17.58±3.83 months). There were no cases of infection recurrence or spread. However, two patients developed neurological symptoms in both legs three months after surgery, and CT re-examination showed intervertebral space collapse, secondary intervertebral foramina, and spinal stenosis. Therefore, they underwent secondary bone graft fusion and internal fixation. Table 3Surgical technique for the patients treatedCase No.Operation time (minutes)Blood loss (mL)Causative organismLength of hospitalisation(days)Follow-up (months)Drainage time(days)Drainage volume (mL)18525Streptococcus pasteurianus2216738212015Escherichia coli301584039530Staphylococcus aureus2614736410020Klebsiella pneumoniae subsp. pneumoniae2624942511015Negative241873969010Escherichia coli292283978020Staphylococcus haemolyticus232473787535Gemella morbillorum561894398540Staphylococcus aureus45168401011515Streptococcus agalactiae24146351110020Klebsiella pneumoniae subsp. pneumoniae22137381212525Negative2017736 Surgical technique for the patients treated Postoperatively, patients experienced significant relief of lumbar and lower-extremity symptoms, with varying degrees of improvement in lower-extremity numbness and weakness. No patients reported worsened neurological symptoms. Three patients who had presented with urinary dysfunction of different severities before surgery fully recovered during follow-up. If postoperative spinal stability was maintained and patients could tolerate pain and fatigue symptoms had improved, early ambulation with brace support was encouraged, taking care to protect the drainage tube from dislodging. At multiple postoperative follow-up points, VAS, JOA, and ODI scores were significantly improved compared with preoperative values (P < 0.05; Fig. 2). At the final follow-up, clinical efficacy was evaluated using the modified MacNab criteria, yielding an excellent and good rate of 91.7%, with 9 cases rated excellent and 2 cases as good. (a) Visual analogue scale (VAS) for low back pain; (b) Oswestry Disability Index (ODI), (c) Japanese Orthopaedic Association (JOA) scores. Box plots illustrate changes from the pre-surgery assessment to the last follow-up. Follow-up scores are presented for presurgery, 1 week post-surgery, and at 4, 12, 24, and the last follow-up. The y-axis represents the different outcome measures in each panel. VAS, ODI, and JOA scores showed statistically significant improvement compared to pre-surgical values (∗ P<0.05) Laboratory test results All patients demonstrated a gradual decrease in WBC, ESR, and CRP levels postoperatively. Compared with preoperative values, there were statistically significant differences at multiple postoperative follow-ups and at the final follow-up (P < 0.05; Fig. 3). A. White blood cells (WBC), B. C-reactive protein (CRP), C. Erythrocyte sedimentation rate (ESR). Line graphs illustrate changes from the pre-surgery assessment to the last follow-up. Follow-up scores include pre-surgery, 1 week post-surgery, and at 4, 12, 24 weeks, and the last week. The y-axis expresses different outcome measures in each panel. All findings are statistically significant compared to pre-surgery values (* P<0.05) Typical case A patient was admitted in 2021 with “low back pain accompanied by pain, numbness, and weakness in both lower limbs for one week”. On admission, they presented with a low-grade fever of 38.9 °C. Low back pain was pronounced and progressively worsened, making it difficult to turn around. They exhibited tenderness and percussion pain in the lumbosacral region, with restricted lumbar flexion, extension, and rotation. The muscle strength of their lower limbs was grade 3, and the VAS score was 7. Blood tests showed a WBC count of 16.46×109/L, HGB concentration of 83 g/L, ESR of 68.00 mm/h, CRP concentration of 165.11 mg/L, and PCT concentration of 2.700 ng/ml. The patient’s past medical history included chronic renal failure, haemodialysis, and type 2 diabetes mellitus. Imaging revealed an abscess in the L4–L5 spinal canal with significant nerve compression and infection of the L5/S1 intervertebral space. Emergency L4/5 percutaneous transforaminal endoscopic abscess debridement and drainage were performed. Based on drug sensitivity, the patient was treated with imipenem for 1 week, after which blood and drainage fluid cultures became negative (-). Due to recurrent vomiting and diarrhoea, considered adverse drug reactions, the treatment was switched to ceftriaxone—to which the pathogen showed similar susceptibility—for an additional three weeks. The patient was subsequently discharged and prescribed oral levofloxacin combined with cefuroxime for an additional 10 weeks. Follow-up examinations showed normal inflammatory markers and recovery of neurologic function, with restored independent mobility. See Fig. 4. Imaging data of typical cases A–D: CT and MRI indicate abscess formation in the spinal canal at L4–L5, with oedema in the L5 and S1 vertebral bodies; E: External view of the intervertebral foramen endoscope; F–G: Tissue adhesion and abscess formation observed under endoscopy; H–I: Placement of the drainage tube guided by a guidewire; J: Illustration of drainage tube placement; K–L: MRI one month after surgery shows reduced oedema in the L5 and S1 vertebral bodies and resolution of the abscesses in the spinal canal from L4 to L5; M–N: MRI three months after surgery indicates the disappearance of oedema in the L5 and S1 vertebral bodies; O–R: One-year postoperative review shows the absence of oedema in the L5 and S1 vertebral bodies, endplate sclerosis, and osteoproliferation around the vertebral bodies Discussion Spinal epidural abscess refers to a purulent infection of the fat tissue and venous plexus within the epidural space of the vertebral canal. It can progress from symptom onset to irreversible neurological deficits in just a few days. SEA is most common in individuals over 60 years of age, and abscesses tend to occur in patients with low immunity and multiple underlying comorbidities, such as diabetes, chronic hepatitis, uraemia, intravenous drug use, tuberculosis, and alcohol addiction [5]. Ghobrial et al. [6] identified previous revision surgery, body mass index, intravenous drug use, smoking, previous radiation therapy, diabetes, chronic systemic infection, and previous osteomyelitis as common risk factors. Huang et al. [7] reported risk factors, including older age (>70 years), diabetes, renal insufficiency, and MRSA infection. In our case series, 5 patients had diabetes, 4 had end-stage renal disease, 3 had coronary heart disease, 2 had a history of spinal surgery, 2 had a history of cancer, and 1 had a pulmonary infection. Lenga et al. [8] reported 2-year follow-up results of simple decompression and fusion surgery in 35 SEA patients aged 80 years and older. Among them, 18 underwent simple decompression, and 17 underwent fusion. The bacterial species identified in the two groups were Staphylococcus aureus (fusion group: 70.6%, decompression group: 44.4%), Escherichia coli (fusion group: 16.7%, decompression group: 11.8%), Enterococcus (fusion group: 11.1%, decompression group: 5.9%), Pseudomonas aeruginosa (fusion group: 5.6%, decompression group: 5.6%), and multiple pathogens (fusion group: 5.6%, decompression group: 5.6%). In the fusion group, 16.7% of pathogens were negative, compared with 5.9% in the decompression group. In our study, Staphylococcus aureus and Escherichia coli were the most frequently isolated microorganisms in cultures, consistent with previous research. Some studies suggest that anti-infective therapy alone is suitable only for patients in the early stages without, or with only mild, neurological symptoms, those who cannot tolerate surgery, or those who have been completely paralyzed for more than 72 hours. Patients with neurological damage should undergo surgical decompression as soon as possible [9, 10]. Suppiah et al. [11] compared the efficacy of conservative antibiotic treatment with surgery combined with antibiotic treatment for SEA. They concluded that patients without severe neurological impairment and with early bacterial culture confirmation of infection type could be treated with conservative antibiotic therapy. Kitov et al. [3] treated 34 SEA patients between 2009 and 2018. Nine patients (26.5%) underwent interlaminar decompression or hemilaminectomy, 25 (73.5%) underwent laminectomy, and 15 (44.1%) underwent posterior pedicle screw fixation. Overall, 67.6% had a good prognosis, 32.4% had a poor prognosis, and 8.8% died. They concluded that emergency surgery is the preferred treatment for SEA. Reihsaus et al. [12] conducted a meta-analysis of treatment methods for 915 patients with spinal epidural abscesses, with a surgical rate of 89% and 11% receiving drug therapy alone. Stratton et al. [13] analysed 12 studies involving 489 SEA patients who received drug therapy, and found that the failure rate of conservative treatment ranged from 29.3% to 35.1%. Similarly, Gardner et al. [10] included 14 studies and reported a conservative treatment failure rate of 29.40% for nonsurgical SEA patients. Common risk factors for treatment failure include acute neurological damage, diabetes, advanced age, and Staphylococcus aureus infection. The authors suggest that although there is increasing evidence supporting nonsurgical treatment for appropriate patients, surgical intervention will remain the primary treatment for SEA. Lyons et al. [1] included all SEA patients from a tertiary centre between April 2011 and May 2019. Among them, 108 patients underwent surgical treatment, while 105 initially received drug therapy, with 22 (21%) experiencing conservative treatment failure. Patients with failed conservative treatment had significantly higher CRP levels, longer symptom duration, and a higher incidence of concurrent nonspinal infections. Nine percent of patients had skip lesions, which were associated with significantly higher white blood cell counts, faster erythrocyte sedimentation rates, and higher rates of bacteraemia and nonspinal infections. These factors can help identify individuals at risk of conservative treatment failure, necessitating more clinical evaluation and consideration of active surgical intervention. Page et al. [14] developed a predictive scoring model to guide clinical decision-making for the diagnosis and treatment of SEA. They retrospectively analysed 159 SEA patients without neurological impairment, with 50 requiring delayed surgery and 109 cured through surgical intervention. The SEA predictive scoring incorporated factors such as active malignancy, discitis, microbiological identification, blood culture, and sex. A score of ≤2 was associated with a conservative treatment failure rate of <5%, whereas a score of ≥7 was associated with a failure rate of >95%. This predictive model for risk stratification can identify high-risk patients who require surgical intervention. Ghobrial et al. [6] explored the issue of surgical timing. Fifty-four patients underwent surgery within 24 hours of admission (with an average time from admission to incision of 11.2 hours), while 33 patients underwent surgery more than 24 hours after admission (average of 59 hours). There were no significant differences between the two groups in terms of spinal epidural abscess presentation, location, nerve damage grade, or risk factors. However, the conclusion indicated that surgery within 24 hours had a relative advantage in terms of patient neurological recovery grading at discharge. The surgical objective is to swiftly remove the lesion to achieve decompression and obtain reliable bacterial culture results from pus samples. For patients with single-segment abscesses who have poor physical condition, multiple comorbidities, and intolerance to open surgery, an increasing number of scholars are adopting percutaneous transforaminal endoscopic surgery. This approach allows procedures to be performed under local anaesthesia, enables the retrieval of sufficient diseased tissue and pus for examination, minimizes trauma and blood loss, facilitates early mobilization, and permits precise and safe removal of lesion tissue under direct endoscopic visualization. Proietti L [15] reported that decompression, aspiration of pus, and thorough irrigation of the epidural space to eliminate most lesions and relieve compression of the dural sac and nerves constitute the rationale for surgical necessity. Continuous irrigation with normal saline facilitates the timely removal of abscesses, necrotic tissue, and inflammatory materials. The findings of this study indicate significant reductions in postsurgical WBC, CRP, and ESR levels compared with presurgical levels. Additionally, patients demonstrated notable improvements in VAS scores for lumbar and leg pain, JOA scores, and ODI indices. According to MacNab criteria, the postoperative excellent and good rate reached 91.7%, which is generally consistent with previous reports. Liu Y [16] compared the efficacy of percutaneous endoscopic lumbar debridement with posterior lumbar interbody fusion in patients with pyogenic discitis. Compared with the fusion group, the endoscopic group had a shorter operative time, less intraoperative blood loss, and a shorter hospital stay (P<0.01). At the final follow-up, both groups showed significant improvements in ESR, CRP, and ASIA classification (P<0.001), with no significant differences between groups (P>0.05). This study revealed that endoscopic debridement offers advantages, including minimal surgical trauma, shorter operative time, and faster postoperative recovery. This study highlights areas for perioperative medical optimization. In our study, two patients underwent secondary bone graft fusion and internal fixation. Postoperative cultures for these patients were negative, and after initial empirical treatment with vancomycin combined with third-generation cephalosporins for 4 weeks, they were switched to oral third-generation cephalosporins. This outcome may be due to the long duration of symptoms, radiological risk factors before surgery, including obvious intervertebral disc collapse, a significant degree of vertebral body destruction, and partial impairment of preoperative stability. Subsequently, severe subsequent bone destruction, vertebral collapse, and worsened neurological symptoms occurred, and secondary internal fixation surgery was needed. However, we believe that delayed instability after surgery is not related to aggressive debridement. The surgery uses targeted puncture, does not require facet joint arthroplasty, and does not compromise vertebral body stability. Therefore, delayed instability is considered to be related to the preoperative condition. Chia S L [17] and other researchers have expressed skepticism about the routine placement of negative-pressure drainage tubes after debridement for spinal epidural abscess, emphasizing the critical importance of complete intraoperative abscess removal. Despite this, numerous scholars continue to advocate drainage tube placement following debridement to facilitate the drainage of residual pus and promote recovery. In our exploration of drainage tube placement methods, a drainage tube was initially placed through the percutaneous transforaminal endoscopic channel. However, postoperative drainage was often obstructed or even non-existent. We speculated that the drainage tube orifice might be blocked. We subsequently attempted to insert a puncture needle into the lesion site 2 cm from the foraminotomy incision and gradually expanded the tube under direct visualization until the maximum size was reached. A drainage tube (serving as an irrigation tube) was then placed through the expanded tube, while a second drainage tube (serving as a drainage tube) was placed through the percutaneous transforaminal endoscopic channel. After confirming correct placement of both tubes and successful drainage of fluid upon irrigation, the tubes were secured. Postoperatively, the irrigation tube was connected to normal saline for continuous irrigation, and the drainage tube was connected to negative-pressure drainage. However, issues such as obstructed irrigation, excessive negative pressure, and inadequate drainage persisted. During our follow-up attempts, we placed an epidural steel wire catheter within the drainage tube. This method allowed us to observe water flowing from the drainage tube orifice during irrigation via the epidural catheter. When postoperative drainage was obstructed, we often attributed it to a blocked drainage tube orifice. Repeated irrigation through the catheter orifice generally cleared the obstruction, allowing successful drainage to continue. Strengths and limitations of this study This study has several limitations. As a single-centre retrospective study with a small sample size, it may affect the generalizability of the results and introduce potential biases. The follow-up period for evaluating long-term efficacy was relatively short, with a maximum of only 24 months. Longer follow-up may be needed to validate the durable effects of this surgical approach. The authors believe that percutaneous transforaminal endoscopy is suitable for single-segment spinal epidural abscesses and yields satisfactory short-term clinical outcomes. Its minimally invasive nature provides surgical opportunities for patients with poor baseline conditions. This technique involves minimal intraoperative blood loss, does not affect spinal stability, allows early patient mobilization with brace support, and facilitates the removal of abscess contents through continuous saline irrigation. In addition, intraoperative visualization enables retrieval of a large number of tissue samples, thereby improving tissue culture positivity rates. The placement of irrigation and drainage tubes facilitates continuous postoperative irrigation and drainage. This approach offers a new method in clinical decision-making when faced with the dilemma of choosing between open surgery and conservative treatment. However, larger multicentre studies with expanded sample sizes and longer-term follow-up are needed to further validate these findings. Conclusions In patients with spinal epidural abscess, percutaneous endoscopic debridement with catheter drainage is feasible, with satisfactory initial results and no catastrophic complications. Thus, this method has preliminary feasibility and safety under certain conditions. Therefore, we believe that percutaneous endoscopic debridement with catheter drainage is worth trying. Larger multicentre studies with larger sample sizes and longer-term follow-up might be used as alternative surgical treatment options for patients with spinal epidural abscess. Acknowledgements The authors wish to thank the auxiliary diagnosis provided by staff in the Clinical Laboratory and Pathology Department, as well as the understanding and support from all participants. Clinical trial number Not applicable. Abbreviations Spinal epidural abscess Oswestry disability index Visual analogue scale Japanese orthopaedic association Least significant difference White blood cell count C-reactive protein Erythrocyte sedimentation rate Symptom duration period Lumbar pain Lumbar pain with lower limb Authors' contributions XL participated in data collection and analysis, YQH, THL contributed to drafting the manuscript. LJY contributed to revise the manuscript. YQH, LJY contributed equally to the manuscript. XFF, XL, LJY, ZLX and ZCW performed the surgery and contributed to the study design and conception, interpretation of data, and revision of the manuscript. All authors reviewed and approved the final manuscript. Funding The authors declare that no funds, grants, or other support were received. Data availability All relevant data are included in this article. Additional de-identified data are available from the corresponding author on reasonable request and with approval from the institutional research center/ethics committee. Declarations Ethics approval and consent to participate All procedures performed in studies involving human participants were in accordance with the ethical standards of the Clinical Trial Ethics Committee of the ShunDe Hospital, GuangZhou University of Chinese Medicine (Foshan, China; approval No. KY-2023105) and with the 1964 Helsinki Declaration and its later amendments, or comparable ethical standards. Informed consent was obtained from all participants. All methods were carried out in accordance with the relevant guidelines and regulations (Declaration of Helsinki). Consent for publication All participants gave written informed consent for their personal or clinical details along with any identifying images to be published in this study. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Yinqiu He and Lijin Yang contributed equally to this work. References Associated Data Data Availability Statement All relevant data are included in this article. Additional de-identified data are available from the corresponding author on reasonable request and with approval from the institutional research center/ethics committee.