Procedure-Specific Postoperative Drainage Strategy in Transforaminal Full-Endoscopic Spine Surgery: A Quantitative Evaluation of Drain Output and Endoscopic Bleeding Findings

Abstract

Background

Postoperative drain use after transforaminal full-endoscopic spine surgery (TF-FESS) remains controversial, and procedure-specific criteria are not well defined. This study aimed to determine whether postoperative bleeding risk—quantified by postoperative drain output and intraoperative endoscopic bleeding findings—differs by TF-FESS procedure and whether these metrics can inform a procedure-specific drainage strategy.

Methods

We retrospectively reviewed 100 consecutive TF-FESS cases: 65 managed with postoperative drains, and 35 managed without drains after an institutional policy change (limited to full-endoscopic discectomy [FED] and lumbar foraminotomy). Procedures included FED, full-endoscopic lumbar foraminotomy (FELF), full-endoscopic ventral facetectomy (FEVF), transforaminal full-endoscopic lumbar undercutting laminectomy (TELUL), and pars crisscross decompression (pars crisscross). Postoperative drain output was analyzed in the drain group. Intraoperative endoscopic bleeding was assessed at 3 standardized steps: approach/foraminoplasty (step 1), procedure-specific decompression (step 2), and final hemostasis (step 3). Multivariable analysis identified predictors of higher drain output.

Results

Drain output differed significantly among procedures (analysis of variance, P < 0.001). FED and FELF showed low drain output, whereas FEVF, TELUL, and pars crisscross showed higher drain output. Invisible bleeding during step 2 was more frequent in TELUL and pars crisscross. On multivariable analysis, TELUL and FEVF independently predicted higher drain output. In the drain-free subgroup (n = 35; FED/FELF), no patient developed neurological deterioration or symptoms requiring emergent hematoma evacuation.

Conclusions

Bleeding risk in TF-FESS is strongly procedure dependent. Drain-free management was implemented only in a selected low-risk subgroup after a policy change; given the small sample size and the low baseline incidence of symptomatic hematoma, this cohort is not powered to assess safety outcomes, and drain-free observations should be interpreted as descriptive feasibility data. Postoperative drainage should be considered for advanced procedures involving epidural venous plexus manipulation and/or extensive cancellous bone exposure (FEVF/TELUL/pars crisscross). Given the nonrandomized, procedure-dependent drain policy, these findings are hypothesis generating.

Clinical Relevance

A procedure-specific, risk-stratified drainage approach may help optimize postoperative management and patient safety in TF-FESS.

Level of Evidence

3.

Introduction

Transforaminal full-endoscopic spine surgery (TF-FESS) has progressively expanded from percutaneous endoscopic discectomy,^1,2 now formally termed full-endoscopic discectomy (FED),³ to more advanced procedures, including full-endoscopic lumbar foraminotomy (FELF),^4–6 full-endoscopic ventral facetectomy (FEVF),⁷ transforaminal full-endoscopic lumbar undercutting laminectomy (TELUL),⁸ and transforaminal full-endoscopic pars crisscross decompression (pars crisscross).^9,10 These procedures now address a broad spectrum of lumbar degenerative disorders, ranging from foraminal stenosis to central canal compromise and pars-related pathology. Notable advantages of TF-FESS include its minimal invasiveness, which allows procedures to be performed under local anesthesia¹¹ with negligible intraoperative bleeding,¹² thereby facilitating early postoperative recovery.¹³

Postoperative drain placement is widely used in spine surgery to prevent epidural hematoma; however, strict criteria for its necessity in TF-FESS have not been established. Moreover, bleeding risks are thought to differ inherently depending on the surgical target: specifically, there may be differences between procedures addressing extra-canalicular pathology (such as FED and FELF) and those requiring intracanal manipulation or involving vascularly dense structures (such as FEVF, TELUL, and pars crisscross). However, studies quantitatively evaluating postoperative drain output or the safety of drain-free management in TF-FESS are scarce. This study aimed to determine procedure-specific indications for postoperative drainage in TF-FESS by quantitatively evaluating postoperative drain output and intraoperative endoscopic bleeding at each operative step. The safety of drain-free management introduced midstudy for FED and FELF was also assessed.

Materials and Methods

Study Design and Ethics Statement

The study protocol was reviewed and approved by the Institutional Review Board of our institution. Written informed consent for the collection and publication of clinical data and images was obtained from all enrolled patients. All procedures adhered to the ethical standards of the Declaration of Helsinki and complied with applicable laws and regulations in Japan.

This was a retrospective cohort study designed to quantitatively evaluate intraoperative and postoperative bleeding characteristics. It included 100 consecutive patients who underwent TF-FESS at our institution between December 2024 and November 2025. The cohort consisted of 69 male and 31 female patients, ranging in age from 19 to 93 years (mean 67.2 ± 17.3 years). All procedures were performed under local anesthesia and were restricted to unilateral, single-level decompression. Patients undergoing bilateral or multilevel procedures were excluded. Additionally, procedures for discogenic low back pain, specifically thermal annuloplasty¹⁴ and disc cleaning for Modic changes,^15,16 were excluded to focus on decompression surgeries. Debridement for spinal infection was also excluded. This study had 2 prespecified components: (1) procedure-specific quantitative analysis of postoperative drain output in cases managed with drains, and (2) descriptive safety surveillance for postoperative neurological deterioration in all cases, including drain-free FED/FELF cases introduced midstudy. Because drain placement was not randomized and became procedure-dependent after April 2025, comparisons between drained and drain-free groups are descriptive and cannot be interpreted as causal evidence regarding the safety of drain omission.

Drain Placement Policy

Drain-free management was applied only to FED and FELF during the latter period; therefore, comparisons between drained and drain-free groups are inherently confounded by procedure type and study period and should not be interpreted as causal evidence regarding drain omission. Accordingly, inferential analyses using drain output were restricted to the drain group, and drain-free outcomes are reported as descriptive feasibility observations.

During the first study period, from December 2024 to March 2025, postoperative drains were routinely placed for all TF-FESS procedures. In the subsequent period beginning in April 2025, based on the minimally invasive nature and procedural characteristics of these procedures, FED and FELF were managed without postoperative drains as a general policy. A total of 65 procedures were performed with drain placement and 35 without drains. Among the 65 cases managed with drains, the distribution of procedures was as follows: 20 FED, 12 FELF, 14 FEVF, 7 TELUL, and 12 pars crisscross procedures. The 35 drain-free cases consisted exclusively of FED and FELF, comprising 15 FED and 20 FELF procedures (Table 1).

Table 1.

Baseline characteristics of the entire cohort.

Characteristic	Total (n = 100)	Drain (n = 65)	Drain-Free (n = 35)	P
Age, y, mean ± SD	67.2 ± 17.3	65.7 ± 17.5	69.9 ± 16.9	0.246
Male sex, n (%)	69 (69.0)	47 (72.3)	22 (62.9)	0.334
Body mass index, mean ± SD	24.4 ± 3.3	24.8 ± 3.4	23.6 ± 3.0	0.073
Hypertension, n (%)	50 (50.0)	28 (43.1)	22 (62.9)	0.058
Antithrombotic medication, n (%)	24 (24.0)	12 (18.5)	12 (34.3)	0.076
Operation time, min, mean ± SD	87.2 ± 21.9	91.7 ± 22.5	78.8 ± 18.1	0.003
Side (right/left), n	62/38	39/26	23/12	0.320
Surgical level, n				0.304
L2/L3	2	1	1
L3/L4	12	10	2
L4/L5	41	28	13
L5/L6	2	2	0
L5/S1	43	24	19
Procedure, n				<0.001
FED	35	20	15
FELF	32	12	20
FEVF	14	14	0
TELUL	7	7	0
Pars crisscross	12	12	0

Abbreviations: FED, full-endoscopic discectomy; FELF, full-endoscopic lumbar foraminotomy; FEVF, full-endoscopic ventral facetectomy; Pars crisscross, pars crisscross decompression; TELUL, transforaminal endoscopic lumbar undercutting laminectomy.

Note: Descriptive summary by drain policy; direct group comparisons are confounded by procedure type and study period. P values are unadjusted and should be interpreted cautiously because drain-free management was applied only to FED/FELF after a policy change.

Surgical Technique

All procedures were performed under local anesthesia via a transforaminal approach and began with the approach and standardized foraminoplasty (step 1) involving partial resection of the superior articular process (SAP). The subsequent procedure-specific decompression (step 2) targeted distinct pathologies (Figure 1). FED involved standard discectomy using the outside-in technique. FELF extended decompression by further resecting the SAP and removing the ligamentum flavum to decompress the exiting nerve root. FEVF extended decompression to the lateral recess by resecting the ventral aspect of the facet joint. TELUL addressed central stenosis through undercutting laminectomy (ventral resection of the inferior articular process [IAP]) and partial flavectomy following total SAP removal. Pars crisscross targeted the ragged edge in isthmic spondylolisthesis by sequentially resecting the SAP, floating lamina, and the tip of the IAP. Throughout all surgeries, irrigation pressure was consistently maintained at 60 mmHg. Before wound closure, a gelatin-thrombin matrix hemostatic sealant (Floseal; Baxter International Inc., Deerfield, IL, USA) was routinely applied to the operative field.

Figure 1.

Radiological classification of transforaminal full-endoscopic spine surgery based on postoperative axial computed tomography images. (a) Full-endoscopic discectomy (FED): standard discectomy; bone resection is limited to the approach (foraminoplasty). Full-endoscopic lumbar foraminotomy (FELF): decompression is confined to the foramen, leaving the spinal canal intact. (b) Full-endoscopic ventral facetectomy (FEVF): lateral recess decompression via ventral facet resection. Transforaminal full-endoscopic lumbar undercutting laminectomy (TELUL): central decompression via total superior articular process removal and undercutting laminectomy. Pars crisscross decompression (pars crisscross): removal of the pars ragged edge and surrounding bony structures.

Assessment of Intraoperative Bleeding

Intraoperative bleeding was assessed in real time at each predefined operative step by a consensus of at least 3 spine surgeons present in the operating room. When initial impressions differed, the operative field was discussed, and the final classification was determined by group consensus; at least 1 evaluator was a board-certified spinal endoscopic surgery specialist accredited by the Japanese Orthopedic Association. The agreed classification for each step was documented intraoperatively in the operative record and extracted during chart review. During step 1 and step 2, the endoscopic field was classified as either clear or oozing, indicating that visualization was adequately maintained and minor bleeding could be promptly controlled, or invisible bleeding, defined as persistent obscuration of the endoscopic view that did not clear for at least 10 seconds despite standard hemostatic maneuvers. Step 3 assessed hemostasis at the end of surgery. A 4-grade scale was used: grade 0 indicated a completely clear operative field with no bleeding; grade 1 represented mild oozing not requiring additional hemostasis; grade 2 reflected identifiable bleeding that required further bipolar or hemostatic intervention; and grade 3 was defined as persistent obscuration of the endoscopic view due to bleeding. Surgery was concluded only after achieving grade 0 or grade 1 hemostasis.

Postoperative Management and Data Collection

In the drain placement group (65 cases), a 10-Fr suction catheter (UK Drain; Nipro Corporation, Osaka, Japan) connected to a 125 mL portable suction reservoir (Slim Drain Bag; Nipro Corporation) was placed at the lateral foraminal level adjacent to the foraminoplasty site, and its location was confirmed fluoroscopically at the end of surgery in all cases. The reservoir was used as a closed-suction system activated by manual compression and maintained by a 1-way valve mechanism. The suction drain was removed on the first postoperative day. To ensure data consistency, drain removal was performed at a standardized time to maintain a substantially uniform collection period across all cases. The total fluid volume collected from the end of surgery until removal was recorded as the postoperative drain output. Regarding safety assessment, all 100 patients—including the 35 drain-free cases—were closely monitored for postoperative complications. Specifically, clinical signs suggestive of symptomatic spinal epidural hematoma, such as intractable leg pain or neurological deficits (eg, motor palsy), were rigorously evaluated. Symptomatic hematoma was defined as a condition requiring emergency surgical evacuation.

Statistical Analysis

Postoperative drain output was compared among procedures using analysis of variance followed by Tukey’s post-hoc testing. Associations between drain output and bleeding characteristics at step 1 and step 2 were evaluated using Student’s t tests. Descriptive comparisons within the FED/FELF subgroup (drain vs drain free) were performed to explore potential differences in baseline characteristics, recognizing that allocation was policy based rather than randomized. Multivariable linear regression analysis was performed with postoperative drain output as the dependent variable and surgical procedure, intraoperative bleeding category at step 1 and step 2, hemostasis grade at step 3, body mass index (BMI), hypertension, antithrombotic medication use, and operative time as independent variables. Given the modest sample size in the drain group (n = 65), covariates were selected a priori based on clinical plausibility, and regression findings are interpreted as exploratory. Statistical significance was defined as P < 0.05. All statistical analyses were performed using EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan), which is a graphical user interface for R (The R Foundation for Statistical Computing, Vienna, Austria).

Results

Baseline characteristics of the entire cohort (n = 100) are summarized in Table 1. Because drain-free management was implemented only for FED and FELF after an institutional policy change, the drain and drain-free groups differed by procedure type and study period; therefore, formal baseline comparisons between these groups should be interpreted cautiously. Consistent with the inclusion of more complex procedures in the drain group, operative time was longer in the drain group (Table 1).

Among the 65 cases in which postoperative drains were used, the drain output ranged from 1 to 105 mL, with a mean of 32.1 ± 25.0 mL. Significant differences were observed in operative time across procedures, with pars crisscross requiring the longest duration (Table 2). When stratified by procedure, FED demonstrated the lowest drain output (15.6 ± 12.5 mL), followed by FELF (28.3 ± 24.5 mL). In contrast, procedures requiring extensive bony resection or intracanal manipulation showed higher drain outputs: FEVF averaged 36.0 ± 24.2 mL, TELUL 52.3 ± 24.6 mL, and pars crisscross 46.8 ± 26.8 mL. Analysis of variance confirmed a significant difference in drain output among procedures (P < 0.001; Figure 2).

Table 2.

Postoperative drain output and intraoperative bleeding characteristics according to TF-FESS procedure (drain group only).

Variable	FED (n = 20)	FELF (n = 12)	FEVF (n = 14)	TELUL (n = 7)	Pars Crisscross (n = 12)	P
Postoperative drain output, mL, mean ± SD	15.6 ± 12.5	28.3 ± 24.5	36.0 ± 24.2	52.3 ± 24.6	46.8 ± 26.8	<0.001
Range	1–50	1–66	1–105	3–77	1–77
Operation time, min, mean ± SD	79.9 ± 16.0	92.3 ± 16.5	81.4 ± 12.7	89.4 ± 14.7	124.2 ± 19.2	<0.001
Step 1 Invisible bleeding, n (%)	4 (20.0)	3 (25.0)	2 (14.3)	1 (14.3)	3 (25.0)	0.938
Step 2 Invisible bleeding, n (%)	6 (30.0)	6 (50.0)	5 (35.7)	4 (57.1)	9 (75.0)	0.129
Step 3 Hemostasis grade 1 (oozing), n (%)	10 (50.0)	7 (58.3)	5 (35.7)	4 (57.1)	9 (75.0)	0.372

Abbreviations: FED, full-endoscopic discectomy; FELF, full-endoscopic lumbar foraminotomy; FEVF, full-endoscopic ventral facetectomy; Pars crisscross, pars crisscross decompression; TELUL, transforaminal endoscopic lumbar undercutting laminectomy; TF-FESS, transforaminal full-endoscopic spine surgery.

Note: Values are presented as mean ± SD (range) or number (%). P values were calculated using analysis of variance, and post hoc comparisons were performed using Tukey’s test. Steps for intraoperative bleeding: step 1, approach and foraminoplasty; step 2, procedure-specific decompression; and step 3, end of surgery.

Figure 2.

Postoperative drain output according to transforaminal full-endoscopic spine surgery procedure (drain group only). A significant difference in drain output was observed among the procedures (analysis of variance [ANOVA], P < 0.001), with full-endoscopic ventral facetectomy (FEVF), transforaminal full-endoscopic lumbar undercutting laminectomy (TELUL), and pars crisscross decompression (pars crisscross) showing higher drain output than full-endoscopic discectomy (FED) and full-endoscopic lumbar foraminotomy (FELF).

The frequency of invisible bleeding during step 1 was consistently low across all procedures. Analysis of intraoperative bleeding characteristics revealed no significant relationship between step 1 bleeding and postoperative drain output. In step 2, where procedure-specific decompression is performed, the frequency of invisible bleeding varied considerably across procedures. Whereas FED exhibited invisible bleeding in 30% of cases, TELUL and pars crisscross procedures demonstrated substantially higher frequencies at 57% and 75%, respectively. Although this difference did not reach statistical significance (P = 0.129), cases exhibiting invisible bleeding during step 2 showed a trend toward higher postoperative drain outputs compared with clear/oozing cases (n = 30 vs n = 35; 38.3 ± 26.9 mL vs 26.7 ± 22.3 mL, P = 0.068).

At the end of surgery (step 3), all procedures achieved hemostasis, classified as either grade 0 or grade 1. Grade 1 cases showed higher drain outputs than grade 0 cases (n = 35 vs n = 30; 37.6 ± 26.1 mL vs 25.6 ± 22.4 mL), and although this difference approached statistical significance (P = 0.051), it did not strictly meet the threshold.

Multivariable linear regression analysis identified TELUL (P = 0.002) and FEVF (P = 0.003) as independent predictors of increased postoperative drain output, while pars crisscross demonstrated a borderline trend toward association (P = 0.075; Table 3). Notably, FELF was not a significant predictor of increased drain output (P = 0.198). Intraoperative bleeding parameters at step 1 (P = 0.707), step 2 (P = 0.323), and step 3 (P = 0.172), as well as hypertension (P = 0.178), antithrombotic medication use (P = 0.773), BMI (P = 0.606), and operative time (P = 0.537), were not independently associated with postoperative drain output.

Table 3.

Multivariable linear regression analysis for predictors of postoperative drain output.

Variable	β Coefficient (mL)	95% CI	P
Procedure
FELF vs FED	10.9	−5.9 to 27.6	0.198
FEVF vs FED	24.6	8.6 to 40.6	0.003
TELUL vs FED	33.3	13.2 to 53.3	0.002
Pars crisscross vs FED	20.6	−2.2 to 43.4	0.075
Intraoperative bleeding (step 1)
Invisible vs clear/oozing	2.8	−12.2 to 17.8	0.707
Intraoperative bleeding (step 2)
Invisible vs clear/oozing	6.2	−6.3 to 18.8	0.323
Intraoperative hemostasis (step 3)
Grade 1 vs grade 0	8.1	−3.6 to 19.7	0.172
Hypertension	−8.9	−22.0 to 4.2	0.178
Antithrombotic medication use	−2.4	−18.8 to 14.0	0.773
Body mass index	−0.5	−2.5 to 1.5	0.606
Operative time (per min)	0.12	−0.27 to 0.51	0.537

Abbreviations: FED, full-endoscopic discectomy; FELF, full-endoscopic lumbar foraminotomy; FEVF, full-endoscopic ventral facetectomy; Pars crisscross, pars crisscross decompression; TELUL, transforaminal endoscopic lumbar undercutting laminectomy.

Note: Dependent variable was postoperative drain output (mL). Study population was the drain group (n = 65). Reference procedure was FED. Model was forced-entry multivariable linear regression. R² = 0.377. Steps for intraoperative bleeding: step 1, approach and foraminoplasty; step 2, procedure-specific decompression; and step 3, end of sugery.

Regarding postoperative complications, no patients in either cohort (n = 100)developed symptoms suggestive of postoperative symptomatic hematoma or new-onset neurological deficits.

Discussion

This study demonstrated that postoperative bleeding in TF-FESS is strongly dependent on the surgical procedure, particularly on the specific anatomical structures targeted for decompression. FED and FELF involve only extra- to intraforaminal manipulation and rarely contact epidural venous structures, which is considered to result in consistently low postoperative drain outputs. These findings are consistent with previous reports showing that FESS creates minimal soft-tissue disruption and virtually no postoperative dead space, which contributes to its extremely low rates of symptomatic epidural hematoma (0.1%–1%) reported in large clinical series.^17–19 In our cohort, no drain-free FED/FELF case required emergent hematoma evacuation or developed new neurological deficits; however, because symptomatic hematoma is rare and drain-free management was introduced in a nonrandomized, procedure-dependent manner, these observations should be interpreted as feasibility data rather than definitive evidence of safety.

In establishing our study protocol, we referred to evidence from general spinal surgery regarding risk factors for postoperative hematoma. A recent comprehensive meta-analysis by Luo et al identified older age, obesity (high BMI), and hypertension as significant risk factors for spinal epidural hematoma.²⁰ Furthermore, regarding antithrombotic medication, Saitta et al reported that preoperative aspirin use and complicated hypertension significantly increased the risk of postoperative hematoma.²¹ Although findings on antithrombotics vary across studies, their potential impact on hemostasis remains a critical clinical concern. Based on these precedents, we incorporated these demographic and clinical variables—age, sex, BMI, hypertension, and antithrombotic use—into our analysis to evaluate their association with postoperative drain output within the drain group. Interestingly, our multivariable analysis revealed that these systemic factors were not independent predictors of postoperative drain output in this cohort. Instead, the specific surgical procedure emerged as the primary determinant of postoperative bleeding risk.

In contrast to low-risk procedures, TELUL, pars crisscross, and FEVF addressing the lateral recess require manipulation in anatomically vascular-rich zones. Specifically, FEVF and TELUL involve the lateral recess, a region densely populated by the epidural venous plexus. Furthermore, in pars crisscross for isthmic spondylolisthesis, the surgical target includes the pars defect, where hypertrophic synovium with significant neovascularization is frequently encountered. The necessity of resecting these highly vascularized tissues and exposing extensive cancellous bone surfaces creates an environment inherently prone to bleeding. This intrinsic hemorrhagic tendency directly correlates with our findings of higher frequencies of invisible bleeding during step 2 and greater postoperative drain outputs in these procedures. Although the hemostasis grade at the end of surgery (step 3) showed a borderline association with drain output in univariable analysis, it was not an independent predictor after multivariable adjustment, suggesting that final endoscopic appearance alone does not fully reflect cumulative intraoperative bleeding.

Our advocacy for a procedure-specific approach is further supported by high-level evidence regarding drain use in lumbar decompression. A systematic review and meta-analysis by Davidoff et al concluded that wound drainage offers no significant benefit in reducing the risk of symptomatic epidural hematoma or infection in non-instrumented lumbar decompression surgery.²² Consistent with this, drainage has been reported to be unnecessary for single-level decompression.²³ Furthermore, a large multicenter study revealed that symptomatic hematomas developed even in patients with drains, suggesting that drain placement does not guarantee the prevention of hematoma formation.²⁴ These findings align with our results for FED and FELF, supporting selective consideration of drain omission for these procedures, while recognizing that our drain-free subgroup was not powered to establish safety. However, we maintain a cautious stance for complex TF-FESS procedures such as TELUL, pars crisscross, and FEVF. Unlike open surgery, endoscopic hemostasis in vascular-rich zones can be technically challenging due to limited visualization. Therefore, despite the general evidence supporting drain-free decompression, the specific hemostatic risks associated with these advanced endoscopic techniques warrant the continued use of drains.

Taken together, the present results indicate that drain placement in TF-FESS should not be uniform but rather tailored to the specific surgical procedure. Selective drain omission can be considered for FED and FELF, in which postoperative drain output is consistently low in drained cases, and no emergent hematoma evacuation was observed in our drain-free experience; however, these observations are descriptive and do not establish safety. In contrast, postoperative drainage should be considered for TELUL, pars crisscross, and FEVF with lateral recess involvement, where exposure of the epidural venous plexus and cancellous bone may increase the risk of residual oozing and postoperative fluid collection. This procedure-specific approach provides a practical framework for endoscopic lumbar decompression while emphasizing the need for prospective validation.

Limitations

Limitations of this study include its retrospective, single-institution design and the nonrandomized, procedure-dependent drain policy that changed over time, which precludes causal inference regarding drain necessity. Postoperative drain output was available only in drained cases and is an imperfect surrogate for bleeding because it likely reflects a mixture of blood, residual irrigation fluid, and serous fluid; although irrigation pressure was standardized at 60 mmHg, variation in irrigation volume and operative duration may influence measured output. Postoperative imaging to quantify epidural hematoma volume was not routinely performed. Because symptomatic epidural hematoma is rare, the drain-free subgroup was not powered to establish safety, and drain-free observations should be interpreted as descriptive feasibility data. The multivariable model included multiple predictors relative to the drain-group sample size (n = 65), raising the possibility of overfitting; therefore, regression results should be interpreted cautiously. Intraoperative bleeding classification was semiquantitative; although assessments were determined by real-time multisurgeon consensus, formal inter-rater reliability testing was not performed. Finally, surgeon-related factors and variations in postoperative mobilization may influence bleeding and should be addressed in future prospective studies.

Conclusion

Postoperative bleeding risk in TF-FESS is not uniform but is strongly influenced by the specific surgical procedure and its anatomical target. Procedures largely confined to the foraminal zone (FED and FELF) showed low drain output in drained cases and no emergent hematoma evacuation in our selected drain-free experience; however, this cohort was not powered to evaluate safety outcomes, and drain-free observations should be interpreted as descriptive feasibility data. Conversely, advanced techniques involving epidural venous plexus manipulation and/or extensive cancellous bone exposure—particularly FEVF and TELUL, and potentially pars crisscross—were associated with higher drain output, supporting consideration of postoperative drainage for these procedures. We propose a procedure-specific, risk-stratified approach to drain use as a practical framework for endoscopic lumbar decompression, while emphasizing the need for prospective validation.