Abstract
Study Design
Retrospective cohort study.
Objectives
This study aims to investigate the incidence and risk factors for 30- and 90-day reoperations following biportal endoscopic lumbar discectomy for lumbar disc herniations.
Methods
We retrospectively reviewed patients who underwent biportal endoscopic lumbar discectomy. We included cases of single-level, central, or paracentral lumbar disc herniations treated with biportal endoscopic lumbar discectomy. Patients were categorized according to whether reoperation occurred within 30 or 90 days after the initial surgery. Patient demographics, radiologic characteristics of disc herniations, and spinopelvic parameters were obtained.
Results
A total of 1510 patients were included in the study. The 30- and 90-day reoperation rates were 2.1% (31/1510) and 3.0% (46/1510), respectively. The most common cause of reoperation at both 30 and 90 days was recurrent disc herniation. In the multivariable logistic regression analysis, diabetes mellitus (DM) (OR 2.83; 95% CI 1.22-6.52; P = .015 for 30-day; OR 2.31; 95% CI 1.12-4.75; P = .023 for 90-day) and prolonged length of postoperative hospital stay (LOS) (OR 1.36; 95% CI 1.22-1.51; P < .001 for 30-day; OR 1.44; 95% CI 1.31-1.60; P < .001 for 90-day) were significantly associated with both 30- and 90-day reoperations.
Conclusion
The incidence of 30- and 90-day reoperations was 2.1% and 3.0%, respectively. Considering that prolonged LOS may be a consequence rather than a cause of reoperation, DM was the only significant independent risk factor associated with both 30- and 90-day reoperations.
Introduction
Lumbar disc herniation is a common spinal disorder that often requires surgical intervention when conservative treatment fails.1-3 In recent years, biportal endoscopic discectomy has emerged as a minimally invasive alternative to conventional techniques.4-9 Biportal endoscopic surgery utilizes two portals—one for instruments (working portal) and one for visualization (endoscopic portal)—to facilitate sufficient decompression with magnified, clear visualization and minimal tissue disruption (Figure 1). Previous studies have reported its clinical effectiveness, including favorable outcomes such as pain reduction, functional improvement, and shorter recovery times.10-13 Intraoperative endoscopic views of biportal endoscopic lumbar discectomy. (A) Identification of herniated disc fragments. (B) Removal of herniated disc fragments. (C) Complete decompression of the thecal sac and traversing nerve root
Despite its increasing adoption, evidence regarding short-term postoperative outcomes specific to biportal endoscopic lumbar discectomy remains limited. While 30- and 90-day reoperation or readmission rates have been studied in microscopic or full-endoscopic discectomy, corresponding studies for biportal endoscopic discectomy are scarce.14-20 Investigating these early outcomes is important, as early reoperations may reflect technical failure, recurrent disc herniation, or other procedure-related complications. Identifying risk factors associated with early reoperation can help improve postoperative management and patient counseling, especially for those with predisposing conditions.
However, to date, no study has reported the incidence and risk factors for early reoperation following biportal endoscopic lumbar discectomy. Although early outcomes of discectomy using endoscopic techniques have been reported, these studies were primarily based on full-endoscopic spine surgery.14,15 There may be differences in visualization compared to biportal techniques. In full-endoscopic spine surgery, angled endoscopes are commonly used, which may provide a different visual field compared to the 0-degree endoscope typically used in biportal techniques. Such differences in visualization may influence the incidence of incomplete decompression. Therefore, this study aims to investigate the incidence and risk factors for 30- and 90-day reoperations following biportal endoscopic lumbar discectomy for lumbar disc herniations.
Methods
The study was conducted in accordance with the Declaration of Helsinki and was approved by the Ethics Committee of our hospital (Institutional Review Board [IRB] No. 2024-W11). The requirement for informed consent was waived by the IRB due to the retrospective nature of the study. This study has been reported in line with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines. 21
Materials
We retrospectively reviewed patients who underwent biportal endoscopic lumbar discectomy between October 2018 and October 2024. All surgeries were performed by board-certified neurosurgeons who had completed 1-2 years of fellowship training in open spine surgery, followed by an additional 1-2 years of dedicated training in biportal endoscopic surgery prior to independent practice. Therefore, the potential impact of the learning curve on surgical outcomes was considered to be limited. The indications for surgery were as follows: (1) persistent back or leg pain caused by lumbar disc herniation that did not improve after 4-6 weeks of conservative treatment, including medication, physical therapy, or epidural steroid injections; (2) intractable pain despite conservative treatment; (3) motor deficits caused by disc herniation; and (4) cauda equina syndrome.
We included single-level, central or paracentral lumbar disc herniations that were treated using biportal endoscopic lumbar discectomy through an interlaminar approach in patients aged 18-80 years. Patients were categorized based on the occurrence of reoperation within 30 or 90 days following initial operation. We excluded patients with multi-level disc herniations, recurrent disc herniations, foraminal or extraforaminal disc herniations, or those treated with open lumbar discectomy, full-endoscopic discectomy, or percutaneous endoscopic lumbar discectomy (PELD). Disc herniations resulting from trauma or associated with segmental instability or infection were also excluded. Additionally, patients with less than 3 months of follow-up were not included in the study.
At our institution, routine postoperative magnetic resonance imaging (MRI) is obtained within 2 days after surgery to identify any residual disc material or incomplete decompression, and such cases were excluded to avoid misclassifying residual disc as true recurrent disc herniation.
Clinical Variables
Patient demographics were collected from the electronic medical records. The following variables were obtained: age, sex, body mass index (BMI), smoking status, and past medical history. Comorbidities were assessed using the age-adjusted Charlson Comorbidity Index (CCI). 22 Preoperative physical status was evaluated according to the American Society of Anesthesiologists (ASA) classification. 23 The length of postoperative hospital stay (LOS) was also recorded.
Radiographic Variables
We assessed the following radiographic variables on preoperative MRI: level of disc herniation; type (protrusion or extrusion); migration direction (superior or inferior); degree of disc degeneration; canal compromise caused by disc herniation; Modic changes of the endplate (Type I, II, or III); fat infiltration in the multifidus muscles; and the presence of vacuum phenomenon or disc calcification. The extent of disc migration was classified as low, high, or very high according to previously published criteria.24-26 Low-grade migration is defined as disc migration up to 3 mm below the upper pedicle (upward) or up to the middle of the lower pedicle (downward). High-grade migration extends beyond these reference points. Very high-grade migration reaches beyond the inferior margin of the pedicle in either direction. Disc degeneration was classified as Grade I to V according to Pfirrmann grading system. 27 Canal compromise was measured on the T2-weighted axial image showing the most severe compression: less than one-third was classified as “mild,” one-third to two-thirds as “moderate,” and greater than two-thirds as “severe”. 28 Fat infiltration of the multifidus muscle was classified as normal (0-10%), slight (10-50%), or severe (>50%) based on T1-weighted axial images. 29 Disc height (anterior and posterior) and the disc height index (DHI) were also measured. DHI was calculated as the ratio of the sum of anterior and posterior disc height to the sum of the superior and inferior disc depth (anteroposterior length).30,31 The presence of vacuum phenomenon or disc calcification was evaluated on preoperative computed tomography (CT) scans.
Statistical Analysis
Statistical analyses were conducted using R software (version 4.1.1; R Foundation for Statistical Computing). A P value of less than .05 was considered statistically significant, and all tests were two-tailed. Numerical variables were presented as mean ± standard deviation (SD) and compared between groups based on 30- and 90-day reoperation status using Student’s t-test or Welch’s t-test, depending on the results of Levene’s test for equality of variances. Categorical variables were presented as frequencies and percentages and compared between groups using the Chi-square or Fisher’s exact test.
To identify risk factors for 30- and 90-day reoperation, multivariable logistic regression analyses were performed using the backward stepwise selection method. The dependent variables were defined as reoperation within 30 or 90 days, while independent variables included clinical and radiographic parameters. Variables that produced extremely wide confidence intervals (CI) due to sparse data or quasi-complete separation were excluded from the analysis. To address potential multicollinearity, clinically overlapping variables and those with a variance inflation factor (VIF) greater than 10 were considered for removal from the multivariable model. We also compared incidence rates of reoperation between days 0-30 and 31-90 using Poisson regression with log(person-time) as an offset to estimate incidence rate ratio (IRR) and 95% CI.
Results
Demographics
Demographics of the Patients Undergoing Single-level Biportal Endoscopic Lumbar Discectomy
BMI, body mass index; HTN, hypertension; DM, diabetes mellitus, CCI, charlson comorbidity index; ASA, American Society of Anesthesiologists; LOS, length of postoperative hospital stay; *P < .05; **P < .01; ***P < .001; Bold means statistical significance.
Radiographic Characteristics
Radiologic Characteristics of the Patients
DHI, disc height index; PI, pelvic incidence; PT, pelvic tilt; SS, sacral slope; LL, lumbar lordosis, SD, standard deviation.
Reoperation Incidence and Causes
The incidence of reoperation within 30 days was 2.1% (31/1510), and within 90 days was 3.0% (46/1510). The incidence of reoperation within 30 days was significantly higher than 31-90 days reoperation (IRR 4.05; 95% CI 2.19-7.50; P < .001) (Figure 2). The most common cause of reoperation was recurrent disc herniation, accounting for 83.9% of 30-day reoperations and 84.8% of 90-day reoperations (Table 3). Reoperation incidence by postoperative period (0-30 vs 31-90 days). IRR, incidence rate ratio; CI, confidence interval; P < .001*** Causes of Reoperation Values are presented as number of patients (percentage of reoperation cases, percentage of total patients).
Risk Factors for 30–day Reoperation
Logistic Regression Analysis for 30-day Reoperation
BMI, body mass index; HTN, hypertension; DM, diabetes mellitus, CCI, charlson comorbidity index; ASA, american society of anesthesiologists; LOS, length of postoperative hospital stays, DHI, disc height index; PI, pelvic incidence; PT, pelvic tilt; SS, sacral slope; LL, lumbar lordosis; *P < .05; **P < .01; ***P < .001; Bold means statistical significance.
In the multivariable analysis, the presence of diabetes mellitus (DM) (OR 2.83; 95% CI 1.22-6.52; P = .015) and prolonged LOS (OR 1.36; 95% CI 1.22-1.51; P < .001) remained significantly associated with reoperations. In contrast, the presence of disc calcification was identified as a protective factor (OR 0.38; 95% CI 0.15-0.97; P = .043).
Risk Factors for 90–day Reoperation
Logistic Regression Analysis for 90-day Reoperation
BMI, body mass index; HTN, hypertension; DM, diabetes mellitus, CCI, charlson comorbidity index; ASA, american society of anesthesiologists; LOS, length of postoperative hospital stays, DHI, disc height index; PI, pelvic incidence; PT, pelvic tilt; SS, sacral slope; LL, lumbar lordosis; *P < .05; **P < .01; ***P < .001; Bold means statistical significance.
In the multivariable analysis, the presence of DM (OR 2.31; 95% CI 1.12-4.75; P = .023), and prolonged LOS (OR 1.44; 95% CI 1.31-1.60; P < .001) were significantly associated with reoperations.
Discussion
Key Findings
In this study of 1510 patients who underwent biportal endoscopic lumbar discectomy, the 30- and 90-day reoperation rates were 2.1% and 3.0%, respectively. The most common cause of reoperation was recurrent disc herniation, accounting for 83.9% of 30-day and 84.8% of 90-day reoperations. DM and prolonged LOS were significantly associated with an increased risk of reoperation, whereas disc calcification was associated with a lower risk.
Reoperation Incidence
Most reoperations occurred within the first 30 days postoperatively (2.1%), significantly more than during the 31- to 90-day period (1.0%) (IRR, 4.05; 95% CI, 2.19-7.50; P < .001) (Table 3, Figure 2). Therefore, the first postoperative month represents a critical period for postoperative management, and appropriate patient education should be emphasized during this time.
The incidence of reoperation in our study (2.1% at 30 days and 3.0% at 90 days) is consistent with the results of previous studies (1.7-5.4%).18,19,32-36 In a study comparing cohorts from the Spine Patient Outcomes Research Trial (SPORT) and the National Surgical Quality Improvement Program (NSQIP), the authors reported a 30-day reoperation rate of 2% in the NSQIP cohort and a 1-year reoperation rate of 7% in the SPORT cohort. 33 Bae et al. investigated risk factors for 90-day readmission following full-endoscopic transforaminal lumbar discectomy, and a sub-analysis showed a 90-day reoperation rate of 4.2% (65/1542). 14 A nationwide cohort study from Korea revealed reoperation rates of 4.1% (768/18 590) within 30 days and 5.4% (1001/18 590) within 90 days following surgery for lumbar disc herniation. 36 Another large population-based nationwide cohort study from Norway showed that 2.1% (735/34 639) of patients underwent reoperation within 90 days after the initial surgery. 32
Biportal endoscopic lumbar discectomy is essentially identical to open microscopic discectomy in terms of partial hemilaminectomy, ligamentum flavum removal, and discectomy, except that it is performed under continuous saline irrigation using an endoscope. Accordingly, biportal endoscopic lumbar discectomy would be expected to show reoperation rates similar to those of other conventional techniques, as demonstrated by our results.
Risk Factors
In logistic regression analysis, we found that age, the presence of hypertension and DM, age-adjusted CCI, ASA classification, LOS, and disc calcifications were significantly associated with 30- or 90-day reoperations (Tables 4 and 5). Risk factors, such as older age, hypertension, DM, age-adjusted CCI, and higher ASA classification, were consistent with findings from previous studies.19,22,23,33,37
Notably, DM was identified as an independent risk factor for both 30- and 90-day reoperations. This finding is supported by prior investigations of DM in the context of intervertebral disc degeneration.38-40 Laboratory studies consistently suggest that DM contributes to disc degeneration through altered biochemical pathways and chronic inflammatory responses. 39 Although clinical evidence remains limited, we postulate that DM may impair annular healing following discectomy, thereby increasing the risk of recurrence and subsequent reoperation. Accordingly, more cautious postoperative management may be warranted in patients with DM. In particular, stricter activity restrictions or extended use of bracing may be considered to minimize mechanical stress on the annular defect. In addition, closer postoperative monitoring and optimization of glycemic control may also be beneficial in reducing the risk of recurrence.
LOS was significantly longer in both the 30- and 90-day reoperation groups compared to the non-reoperation groups. Logistic regression analyses demonstrated that prolonged LOS was a significant factor associated with 30- and 90-day reoperations, although we could not find prior studies identifying prolonged LOS as a risk factor. At our institution, the most common cause of prolonged LOS was residual back or leg pain. Even when postoperative MRI confirmed complete removal of the disc herniation, persistent pain delaying discharge may represent a prodromal sign of recurrence. A small residual disc fragment within the disc space may herniate through the annular tear. Although such fragments may not be detectable on postoperative MRI, they can still cause residual back or leg pain. We postulate that these small residual fragments could serve as a nidus for recurrent disc herniation. However, caution is warranted when interpreting causality. Epidural hematomas typically occur during the initial hospital stay and often necessitate early reoperation. Similarly, early recurrence requiring reoperation during hospitalization may also contribute to a prolonged LOS. Therefore, prolonged hospitalization may not be a cause of reoperation, but rather a consequence of it.
Among the radiologic factors, disc calcification was the only independent protective factor for 30-day reoperation. Although no relevant studies have been identified, we hypothesize that calcified discs may reduce the risk of recurrence by acting as a mechanical barrier over the annular defect. Moreover, the presence of calcification may indicate a firmer disc consistency, which could further lower the likelihood of reherniation. In addition, calcified disc fragments may be less likely to migrate, which may further contribute to the reduced risk of recurrence.
Although some of the risk factors identified in our study align with previous reports, others did not. Meredith et al. reported that obesity was a strong and independent predictor of recurrent lumbar disc herniation. 41 However, BMI was not a significant risk factor in our analysis. Similarly, Pan et al. found that patients with recurrent herniation had significantly lower pelvic incidence (PI) and PI–lumbar lordosis (LL) mismatch. 42 In contrast, spinopelvic parameters, including PI, pelvic tilt (PT), sacral slope (SS), and PI–LL, were not significantly associated with reoperation in our study. BMI and spinopelvic parameters are known to reflect mechanical stress on the spine and intervertebral discs. Thus, higher BMI and lower PI or PI–LL mismatch have been proposed as potential risk factors for recurrent herniation.
A possible explanation for the discrepancy between our findings and prior literature is the surgical strategy employed at our institution. Except in cases of sequestrated herniation, we routinely perform aggressive internal discectomy to minimize the risk of residual loose fragments, which could lead to recurrence. 43 As a result, even under increased mechanical stress, the risk of recurrence and thus reoperation may be reduced due to the limited remaining disc material. In addition, our study focused on biportal endoscopic lumbar discectomy rather than open microscopic discectomy. Compared with open techniques, biportal endoscopy is associated with less muscle injury, which may help preserve postoperative segmental stability. 44 This difference may also contribute to the lack of a significant association of BMI and spinopelvic parameters with reoperation in our cohort.
Causes of Reoperations
(1) Recurrence
Recurrent disc herniation was the most common cause of reoperation in both the 30-day (83.9%, 1.7%) and 90-day (84.8%, 2.6%) groups in our study (Table 3). These findings are consistent with the 90-day incidence of recurrence reported in previous literature (2.1-4.6%).14,19 A common concern regarding biportal endoscopic discectomy is the potential for insufficient disc removal due to limited exposure. Nevertheless, the highly magnified endoscopic view and continuous fluid irrigation in biportal endoscopic discectomy enhance visualization, allowing for clearer identification of the ruptured annulus and herniated disc fragments. Furthermore, loose disc fragments that could cause recurrence can be more effectively removed, potentially resulting in less residual disc material. (2) Epidural hematoma
Postoperative hematoma was the second most common cause of early reoperation. At our institution, a Jackson-Pratt drain is routinely inserted for 1-2 days, which may contribute to the low incidence in our study (0.3%, 4/1510). This rate is consistent with incidence of symptomatic postoperative epidural hematoma in previous studies (0.1-0.2%).45-48 All epidural hematomas in our study occurred within the first 30 days postoperatively. Therefore, symptom recurrence or worsening in the early postoperative period should prompt consideration of an epidural hematoma. (3) Postoperative discal cyst
The pathophysiology of postoperative discal cyst is thought to involve fluid accumulation within a potential space that communicates with a residual annular defect.49,50 Kang et al. reported an incidence of 1.1% (15/1406) following full-endoscopic discectomy.
51
However, the incidence of postoperative discal cyst was only 0.2% (2/1510) within 90 days. This discrepancy may be due to cauterization of the annulus. At our institution, we routinely cauterize the annulus after medially retracting the thecal sac using a nerve root retractor to expose the annulus. Cauterization may induce shrinkage of the annulus and annular defect, thereby reducing the potential for fluid accumulation. (4) Infection
Infection was an extremely rare cause (0.1 %, 1/1510) of reoperation. Continuous saline irrigation during the operation may reduce the risk of postoperative infection. A meta-analysis of biportal endoscopic spine surgery reported infection rates ranging from 0% to 0.08%.52,53 Furthermore, an in vivo comparative study found that continuous saline irrigation during biportal endoscopic transforaminal lumbar interbody fusion (TLIF) significantly reduced intraoperative contamination compared to open TLIF (P = .029).
54
(5) Wound Complications
There were no cases of wound-related reoperation in our cohort. This finding is consistent with the report by Park et al., which found no wound dehiscence in the biportal group, compared to 10.6% (5/47) in the microscopic discectomy group (P = .018).
Limitations
This study has several limitations. First, this was a single-center retrospective study, which may limit the generalizability of the findings. In addition, we did not include a comparison group, such as microscopic discectomy, microendoscopic discectomy, or full-endoscopic discectomy, for comparison with biportal endoscopic discectomy. However, the relatively large cohort of 1510 patients provides strength to the analysis and may mitigate this limitation. Second, because reoperations were assessed only within 30 and 90 days postoperatively, long-term outcomes could not be evaluated. In particular, although aggressive internal discectomy may reduce early recurrence by minimizing residual disc fragments, it may involve potential trade-offs in the long term, including disc height loss, which can lead to back pain and foraminal stenosis. In addition, segmental instability may develop over time. Third, although we analyzed risk factors for reoperation, the heterogeneity of reoperation causes may limit the clarity and applicability of these findings. Nevertheless, this study includes the largest number of cases among published reports on biportal endoscopic lumbar discectomy to date. Furthermore, by limiting the analysis to single-level lumbar discectomy, we sought to minimize potential bias. Fourth, although DM was identified as a significant risk factor, data on glycemic control, such as HbA1c levels, were not available in this study. Therefore, we were unable to evaluate the impact of DM control status on reoperation risk.
Conclusion
The incidence of 30- and 90-day reoperations was 2.1% and 3.0%, respectively, which are comparable to those reported for other techniques. Most reoperations occurred within the first 30 days postoperatively (2.1%), significantly more than during the 31- to 90-day period (1.0%). Therefore, patient education regarding the risk of reoperation should be emphasized during the first 30 days after surgery.
Except for disc calcification, spinopelvic parameters and other radiologic findings were not associated with reoperation at either 30 or 90 days. Considering that prolonged LOS may be a consequence rather than a cause of reoperation, DM was the only independent risk factor associated with both 30- and 90-day reoperations.
Footnotes
Ethical Considerations
The study was conducted in accordance with the Declaration of Helsinki and was approved by the Ethics Committee of the Leon Wiltse Memorial Hospital (IRB No. 2024-W11) on September 25, 2025, with the need for written informed consent waived.
Consent to Participate
The requirement for informed consent was waived by the IRB due to the retrospective nature of the study (IRB No. 2024-W11).
Author Contributions
Sang Hyub Lee: Conceptualization, Data collection, Data analysis, Methodology, Investigation, Visualization, Writing - Original Draft, Writing - Review & Editing. Jae-Won Jang: Resources, Writing - Review & Editing, Supervision, Project administration. Hangyu Lee: Data collection, Writing - Review & Editing. Limjoon Yoon: Data collection, Writing - Review & Editing. Sejin Song: Data collection, Writing - Review & Editing. Dain Jeong: Data analysis. Junghan Seo: Resources, Writing - Review & Editing. Sang Youp Han: Resources, Writing - Review & Editing. Bang-Sang Hahn: Resources, Writing - Review & Editing. Jun Young Kim: Resources, Writing - Review & Editing. Jin Seop Hwang: Resources, Writing - Review & Editing. Dong-Geun Lee: Resources, Writing - Review & Editing. Jin-Sung Kim: Writing - Review & Editing. Dong Chan Lee: Resources, Writing - Review & Editing. Yong Eun Cho: Resources, Writing - Review & Editing. Choon Keun Park: Resources, Writing - Review & Editing.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Declaration of Conflicting Interests
The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Jin-Sung Kim is a member of the Global Spine Journal editorial board and an author of this article. However, he was not involved in the editorial review or decision-making process regarding this manuscript. The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data Availability Statement
The datasets generated during and/or analyzed during the current study are not publicly available.
Declaration
The authors have no personal, financial, or institutional interests regarding any of the drugs, materials, or devices described in this article.