Balloon-assisted endoscopic stricturotomy for small bowel fibrostenotic strictures: A comparison of Crohn’s disease-related and non-Crohn’s disease strictures

Abstract

BACKGROUND

Small-bowel fibrostenotic strictures often cause obstructive symptoms and require repeated interventions. Endoscopic stricturotomy (ESt) is a minimally invasive treatment option; however, whether reintervention-free survival differs by stricture etiology and which baseline factors influence reintervention risk remain uncertain.

AIM

To compare the reintervention-free survival after ESt between Crohn’s disease (CD)-related and non-CD small-bowel strictures and to evaluate the prognostic value of inflammatory biomarkers.

METHODS

We retrospectively reviewed 57 patients with small bowel strictures who underwent balloon-assisted enteroscopy-guided ESt at our center between April 2018 and April 2025. The primary outcome was reintervention-free survival. Kaplan-Meier survival analysis together with Cox proportional hazards modeling was performed to explore factors related to reintervention. Composite inflammatory markers, such as the neutrophil-to-lymphocyte ratio (NLR), were evaluated.

RESULTS

Reintervention-free survival was significantly higher in patients with non-CD strictures than in those with CD-related strictures. At 1 year, the reintervention-free survival rate was 82.2% vs 54.5% (log-rank P = 0.036; HR = 0.37, 95%CI: 0.15-0.94). In parsimonious multivariable analysis, non-CD etiology and baseline NLR remained independently associated with reintervention. Patients with a NLR < 2.4 had significantly higher reintervention-free survival than those with elevated NLR (P = 0.022).

CONCLUSION

ESt provides more favorable long-term outcomes in non-CD small-bowel strictures. A NLR < 2.4 predicts reduced reintervention risk and may be used for preprocedural risk stratification.

Key Words: Balloon-assisted enteroscopy; Endoscopic stricturotomy; Small bowel stricture; Crohn’s disease; Neutrophil-to-lymphocyte ratio; Fibrostenotic stricture; Reintervention

Core Tip: Endoscopic stricturotomy is an established minimally invasive treatment for small-bowel fibrostenotic strictures, but whether outcomes differ by stricture etiology is not well defined. In this retrospective study with long-term follow-up, non-Crohn’s disease strictures had more durable reintervention-free survival than Crohn’s disease-related strictures. A lower neutrophil-to-lymphocyte ratio was associated with a lower risk of reintervention, supporting its use as a simple preprocedural risk marker.

INTRODUCTION

Small bowel strictures are a significant cause of intestinal obstruction. Common clinical manifestations include recurrent abdominal pain, abdominal distension, vomiting, and signs of malnutrition[1], all of which severely impair quality of life and often require surgical intervention[2]. Advances in abdominal imaging, particularly computed tomography (CT) and magnetic resonance (MR) enterography, along with the increasing use of balloon-assisted enteroscopy (BAE), have significantly improved the detection of small bowel strictures[3,4].

Epidemiologic studies show that up to 70% of patients with Crohn’s disease (CD) develop stricturing complications during the disease course, with small bowel involvement occurring in approximately 40%-60% of these cases[5,6]. Benign small bowel strictures unrelated to CD (non-CD strictures), account for approximately 30% of all small-bowel strictures and arise from diverse etiologies, including post-tuberculous fibrosis, nonsteroidal anti-inflammatory drug (NSAID)-induced small bowel strictures (diaphragm disease), radiation enteritis, cryptogenic multifocal ulcerous stenosing enteritis (CMUSE), chronic ischemic enteritis, and other causes[7,8]. Although CD-related and non-CD strictures may appear similar on endoscopic evaluation, typically presenting as luminal narrowing with mucosal scarring, their underlying pathophysiology differs significantly[9]. CD-related strictures result from chronic transmural inflammation, during which pro-fibrotic cytokines, such as tumor necrosis factor alpha (TNF-α), interleukin-21, and transforming growth factor-beta (TGF-β), promote abnormal tissue remodeling and progressive fibrosis, ultimately leading to persistent luminal narrowing[10,11]. In contrast, non-CD strictures typically result from localized injury-repair disturbances, such as ischemia-reperfusion injury, NSAID- or radiation-induced mucosal damage, and postoperative fibrosis and generally lack a sustained inflammatory drive.

Although biologic therapies, such as anti-TNF-α agents, are effective in suppressing active mucosal inflammation, they have limited capacity to reverse established deeplayer fibrosis[12-14]. Consequently, surgery remains a critical treatment option, particularly for long fibrotic strictures[15]. However, surgical intervention is associated with high postoperative recurrence rates and an increased risk of short-bowel syndrome. Endoscopic therapy has emerged as valuable minimally invasive option for managing small-bowel strictures[16]. The two primary techniques are endoscopic balloon dilation (EBD) and endoscopic stricturotomy (ESt)[17]. EBD is widely used, most studies report technical success rates above 90%, and for short segment strictures, the procedure is both effective and safe[18-20]. However, restenosis remains common, often requiring repeated dilations[21]. In contrast, ESt involves direct incision of the fibrotic ring, allowing for more complete release of constricting tissue and potentially prolonged luminal patency, while maintaining a low rate of adverse events[22]. A recent study on inflammatory bowel disease (IBD)-related strictures reported a technical success rate of 100% for ESt, with only 15.3% of patients requiring surgery during a median follow-up period of 0.9 years[23]. These findings highlight the potential of ESt as an effective minimally invasive treatment option for patients with deep small-bowel strictures.

Several studies have compared the efficacy and safety of ESt with surgery or EBD[22-24]. However, long-term outcomes after ESt stratified by stricture etiology remain unclear, particularly for deep small bowel strictures treated using BAE. CD-related and non-CD strictures occur as a result of distinct inflammatory and fibrotic processes[15]. Therefore, long-term patency and reintervention risk after ESt may differ, although these differences have not been adequately explored[12]. In addition, the baseline systemic inflammatory status may affect post-procedural outcomes; however, the prognostic value of routinely available inflammatory markers in this setting is not well defined. Therefore, the aim of this study was to compare long-term outcomes after ESt between CD-related and non-CD small bowel strictures and explore clinically relevant predictors of reintervention.

MATERIALS AND METHODS

Study design and population

This single-center, retrospective study was conducted at a tertiary referral center where a substantial proportion of patients present with complex small bowel strictures. Consecutive patients undergoing BAE-based ESt for small-bowel strictures between April 2018 and April 2025 were included. A total of 57 patients met the prespecified eligibility criteria and were included in the final analysis (Figure 1). Eligible patients had symptomatic small-bowel strictures confirmed by BAE or CT/MR enterography and subsequently underwent ESt for benign-appearing, fibrosis-dominant stenoses. Patients were excluded if a malignant stricture was identified after ESt, emergency surgery was required due to procedure-related complications, or follow-up data was incomplete. Screening details and reasons for exclusion are shown in Figure 1.

Figure 1 Flowchart of patient selection and grouping. CD: Crohn’s disease.

Grouping strategy

Patients were categorized into two etiologic groups. The CD-related stricture group (n = 32) included patients who met contemporary diagnostic criteria for CD. In this group, strictures were considered fibrosis-dominant lesions arising from transmural inflammation. The non-CD group (n = 25) included patients in whom CD had been excluded. In these patients, strictures were attributed to other benign causes, including post-tuberculous fibrosis (n = 6), radiation enteritis (n = 6), chronic ischemic enteritis (n = 7), NSAID-induced diaphragm disease (n = 4), and CMUSE (n = 2).

ESt procedure

ESt was performed by four experienced endoscopists using a balloon-assisted enteroscope, primarily the Fujifilm EN-580T system (Fujifilm, Tokyo, Japan). Preprocedural evaluation included CT or MR enterography to characterize the strictures. All patients fasted and underwent standard bowel preparation. Procedures were performed under general anesthesia in most cases, with intravenous sedation used in selected cases. Antispasmodic agents were administered when intestinal peristalsis interfered with scope advancement.

At the stricture site, a Hook knife (KD-620UR; Olympus, Tokyo, Japan) was used to create longitudinal or radial incisions through the fibrotic tissue. Incisions were extended in a controlled manner until adequate luminal opening was achieved, permitting passage of the enteroscope through the stricture while preserving the muscular layer. Any intraprocedural bleeding or perforation was immediately managed using standard endoscopic techniques (Figure 2).

Figure 2 Representative techniques of endoscopic stricturotomy for small-bowel strictures. A: Narrow tubular stenosis in the ileum; B: Layered incision performed parallel to the bowel axis; C: Luminal patency restored after longitudinal incision; D: Circumferential narrowing of the ileum; E: Multiple radial incisions performed perpendicular to the stricture plane; F: Luminal patency restored after radial incision.

Definitions

Stricture severity during BAE was categorized as mild, moderate, or severe. Severe strictures were defined as those in which the enteroscope could not be passed through with a reasonable amount of pressure. When complete small-bowel examination was not feasible, CT or MR imaging was used to evaluate the number and distribution of strictures. In patients with multiple strictures, the index lesion was determined based on stricture severity and/or the presence of prestenotic dilation. A staged treatment approach was employed, with the most severe or symptomatic stricture treated first. For other strictures, endoscopic treatment was typically performed at 3-month intervals, based on clinical evaluation and symptom recurrence.

Age at onset was defined as the age at which stricture-related symptoms first occurred for patients with, and those without CD. Blood samples for inflammatory markers were obtained within 3 days before ESt. Composite inflammatory indices, including the neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), and C-reactive protein-to-lymphocyte ratio (CLR), were defined as ratios derived from routine blood count parameters and serum C-reactive protein levels. Intraprocedural bleeding was defined as bleeding occurring during or immediately after the procedure. Clinically significant bleeding was defined as bleeding associated with hemodynamic instability[22]. Intraprocedural perforation was defined as a full-thickness defect identified in real time by the endoscopist.

Outcomes

The primary outcome was reintervention-free survival, defined as the time from the first ESt to the recurrence of obstructive symptoms that necessitated subsequent endoscopic or surgical intervention, typically supported by imaging (CT or MR enterography) findings and/or endoscopic evidence of restenosis. Cumulative reintervention-free survival rates at 6 months, 1 year, and 2 years were compared between the two groups. Secondary outcomes included technical success, one-year surgery-free survival following ESt, and procedure-related adverse events. Technical success was defined as successful passage of the enteroscope through the treated stricture[25]. Procedure-related adverse events included intraprocedural or early postprocedural bleeding and perforation.

Follow-up

Follow-up was conducted through outpatient visits and telephone interviews at 3-, 6-, 12-, and 24-months after the first ESt, and continued until the occurrence of reintervention or the last follow-up. Incomplete follow-up was defined as loss to follow-up before 6 months without documented reintervention; patients meeting this criterion were excluded during the screening phase (Figure 1). The median follow-up duration was 40.9 months (range: 20.9-52.3 months), with the final follow-up conducted on July 30, 2025. According to this definition, no patients were lost to follow-up after inclusion, and complete data were available for all variables included in the survival analyses.

Statistical analysis

Statistical analyses were performed using SPSS version 27.0 and R version 4.4.3. Normally distributed continuous variables were expressed as mean ± SD. Non-normally distributed variables were expressed as median (interquartile range). Group comparisons were performed using the Student’s t test or the Mann-Whitney U test. Categorical variables were summarized as n (%) and compared using the χ² test or Fisher’s exact test, as appropriate.

Reintervention-free survival was estimated using the Kaplan-Meier method, and survival curves were compared between groups using the log-rank test. Univariable Cox proportional hazards regression was primarily applied to identify potential predictors of reintervention. Considering the limited number of reintervention events, multivariable Cox regression was performed using a parsimonious modeling approach to reduce the risk of overfitting. Variables were selected on the basis of clinical relevance and the results of univariable analyses. Composite inflammatory indices with potential collinearity were not entered simultaneously into the same multivariable model. The proportional hazards assumption was evaluated using standard graphical diagnostics, including log-minus-log survival plots and Schoenfeld residuals, and no major violations were identified. The optimal cutoff value for the NLR was determined using maximally selected log-rank statistics (maxstat package in R). HRs with 95%CIs were reported. A two-sided P value < 0.05 was considered statistically significant.

RESULTS

Baseline characteristics

A total of 57 patients who underwent ESt were included, comprising 32 patients with CD-related strictures and 25 patients with non-CD strictures. Baseline clinical and stricture characteristics are summarized in Table 1 and Supplementary Table 1. The two groups were comparable with respect to body mass index, sex distribution, disease duration, stricture characteristics (number, length, diameter, location, and prestenotic dilation), history of smoking, completeness of small bowel evaluation with BAE, and the presence of severe stenosis.

Table 1 Baseline characteristics of patients undergoing endoscopic stricturotomy.

Variable	Total (n = 57)	CD group (n = 32)	Non-CD group (n = 25)	Statistic	P value
Age at onset (year)	33.0 (26.0-48.0)	28.0 (23.0-36.0)	45.0 (32.0-55.0)	Z = -2.86	0.004
Stricture diameter (mm)	4 (3-5)	4 (3-5)	4 (3-4)	Z = -0.73	0.467
Sex				χ2 = 0.26	0.607
Male	39 (68.4)	21 (65.6)	18 (72.0)
Female	18 (31.6)	11 (34.4)	7 (28.0)
Prior abdominal surgery				χ2 = 4.40	0.036
No	36 (63.2)	24 (75.0)	12 (48.0)
Yes	21 (36.8)	8 (25.0)	13 (52.0)
Stricture location				-	0.772
Ileum	40 (70.2)	22 (68.8)	18 (72.0)
Jejunum	10 (17.6)	5 (15.6)	5 (20.0)
Ileum + jejunum	7 (12.3)	5 (15.6)	2 (8.0)
Severe stenosis				-	1.000
No	2 (3.5)	1 (3.1)	1 (4.0)
Yes	55 (96.5)	31 (96.9)	24 (96.0)
Stricture length (cm)				χ2 = 0.07	0.799
≤ 2	42 (73.7)	18 (72.0)	24 (75.0)
> 2	15 (26.3)	7 (28.0)	8 (25.0)
Prior biologic therapy
No	46 (80.7)	21 (65.6)	25 (100.0)	χ2 = 8.56	0.003
Yes	11 (19.3)	11 (34.4)	0 (0.0)

The data are presented as median (interquartile range) or n (%). Group comparisons were performed using Student’s t test, Mann-Whitney U test, χ² test, or Fisher’s exact test, as appropriate. A dash (-) in the statistics column indicates that Fisher’s exact test was used. Age at onset refers to the age at onset of stricture-related symptoms. Prior biologic therapy was applicable only to patients with Crohn’s disease and was therefore not considered a candidate variable for multivariable regression. CD: Crohn’s disease.

Significant differences were observed in age at onset and surgical history. Individuals in the non-CD group had a higher age at disease onset compared with those in the CD group (median 45.0 years vs 28.0 years, P = 0.004). In addition, a higher proportion of patients in the non-CD group had a history of prior abdominal surgery compared with the CD group (52.0% vs 25.0%, P = 0.036). Among patients in the CD group, 11 (34.4%) had received biologic therapy prior to ESt, including ustekinumab (n = 4), infliximab (n = 4), and adalimumab (n = 3). Other medical treatments included 5-aminosalicylates (31.3%), corticosteroids (6.3%), azathioprine (3.1%), thalidomide (9.4%), and enteral nutrition (15.6%).

Survival analysis

During a median follow-up of 40.9 months, a total of 24 reintervention events were observed in the study cohort. Kaplan-Meier analysis demonstrated that reintervention-free survival differed significantly between etiologic groups (Figure 3A). Patients with non-CD strictures had a higher reintervention-free survival than those with CD-related strictures (log-rank P = 0.036). The reintervention-free survival rates at 6 months, 1 year, and 2 years differed significantly between the CD and non-CD groups. In the CD group, the rates were 74.9%, 54.5%, and 43.6%, respectively, while in the non-CD group, the corresponding rates were 91.8%, 82.2%, and 76.3% (Supplementary Figure 1).

Figure 3 Reintervention-free survival following endoscopic stricturotomy in patients with small-bowel strictures. A: Kaplan-Meier curves comparing reintervention-free survival between patients with Crohn’s disease (CD)-related and non-CD small-bowel strictures; B: Kaplan-Meier curves stratified by neutrophil-to-lymphocyte ratio (NLR). Patients were categorized as low NLR (< 2.4) or high NLR (≥ 2.4). Shaded areas represent 95%CIs, and plus signs indicate censored observations. Survival curves were compared using the log-rank test. CD: Crohn’s disease; NLR: Neutrophil-to-lymphocyte ratio.

Cox proportional hazards regression analysis

In univariable Cox proportional hazards analysis (Table 2), non-CD etiology was linked to a significantly lower likelihood of reintervention (HR = 0.37, 95%CI: 0.15-0.94; P = 0.036). Among inflammatory markers, a higher NLR correlated with an elevated risk of reintervention (HR = 1.63, 95%CI: 1.14-2.32; P = 0.007). The CLR similarly showed an association with reintervention in univariable analysis (HR = 1.03, 95%CI: 1.01-1.05; P = 0.019). Other baseline variables, including age at onset, severe stenosis, stricture length, incision pattern, PLR, and erythrocyte sedimentation rate, were not significantly related to reintervention (all P > 0.05).

Table 2 Univariable Cox proportional hazards regression for reintervention.

Variable	Univariable analysis
	HR (95%CI)	P value
Etiology
CD (reference)	1.00
Non-CD	0.37 (0.15-0.94)	0.036
Severe stenosis
No (reference)	1.00
Yes	1.35 (0.18-10.02)	0.772
Stricture length (cm)
≤ 2 (reference)	1.00
> 2	0.67 (0.25-1.80)	0.430
Incision pattern
Radial (reference)	1.00
Longitudinal	0.43 (0.18-1.05)	0.065
Age at onset	0.98 (0.95-1.01)	0.168
NLR	1.63 (1.14-2.32)	0.007
PLR	1.00 (1.00-1.01)	0.458
CLR	1.03 (1.01-1.05)	0.019
ESR	0.96 (0.76-1.20)	0.721

Cox proportional hazards model. HRs reflect the risk associated with a 1-unit increase. A two-sided P value < 0.05 was considered statistically significant. Age at onset refers to the age at onset of stricture-related symptoms. CD: Crohn’s disease; NLR: Neutrophil-to-lymphocyte ratio; PLR: Platelet-to-lymphocyte ratio; CLR: C-reactive protein-to-lymphocyte ratio; ESR: Erythrocyte sedimentation rate.

In the parsimonious multivariable Cox proportional hazards model (Table 3), non-CD etiology and NLR remained independently associated with reintervention. Non-CD etiology was associated with a reduced risk of reintervention (adjusted HR = 0.29, 95%CI: 0.10-0.84; P = 0.023). In contrast, higher NLR was associated with an increased risk of reintervention (adjusted HR = 1.92, 95%CI: 1.28-2.88; P = 0.002). The incision pattern was not independently associated with reintervention after adjustment.

Table 3 Multivariable Cox proportional hazards regression for reintervention.

Variable	Multivariable analysis
	HR (95%CI)	P value
Etiology
CD (reference)	1.00
Non-CD	0.29 (0.10-0.84)	0.023
Incision pattern
Radial (reference)	1.00
Longitudinal	0.73 (0.28-1.79)	0.463
NLR	1.92 (1.28-2.88)	0.002

A parsimonious multivariable Cox proportional hazards model was constructed based on clinical relevance and univariable screening. Composite inflammatory indices were not simultaneously included due to potential collinearity. The proportional hazards assumption was assessed graphically. CD: Crohn’s disease; NLR: Neutrophil-to-lymphocyte ratio.

NLR optimal cutoff and stratified analysis

Using maximally selected log-rank statistics, the optimal prognostic cutoff for the NLR was 2.4 (Supplementary Figure 2). Patients were stratified into a low-NLR group (NLR < 2.4; n = 41) and a high-NLR group (NLR ≥ 2.4; n = 16). Kaplan-Meier analysis showed higher reintervention-free survival in the low-NLR group than in the high-NLR group (log-rank P = 0.022; Figure 3B).

Two Cox proportional hazards models were then fitted using the dichotomized NLR variable (Supplementary Table 2). In the unadjusted model, low NLR was associated with a lower risk of reintervention (HR = 0.40, 95%CI: 0.18-0.90; P = 0.027). After adjustment for selected clinically relevant non-inflammatory covariates, including stricture etiology and incision pattern, low NLR remained independently associated with a reduced risk of reintervention (adjusted HR = 0.34, 95%CI: 0.14-0.76; P = 0.009).

Efficacy and safety

A total of 114 BAE-based ESt procedures were conducted in 57 patients with fibrotic small bowel strictures associated with obstructive symptoms. ESt was performed to relieve luminal patency and enable a comprehensive small bowel endoscopic evaluation. The overall technical success rate was 91.2% (104/114). Luminal patency was achieved in 82.5% of patients (47/57). Technical success was comparable between groups. It was achieved in 81.3% of patients in the CD group (26/32) and 84.0% in the non-CD group (21/25; P > 0.05).

Technical failure occurred in 10 patients. Six failures, including three cases involving angulated strictures and three cases in which the scope could not be advanced despite adequate incision, occurred in the CD group. In the non-CD group, there were four failures, including one case involving adhesion-related angulation and three involving persistent non-traversability after incision. In an exploratory descriptive analysis, patients who experienced technical failure appeared to have slightly higher baseline NLRs than did those who achieved technical success (median NLR: 1.81 vs 1.75), although the distributions largely overlapped between the two groups (Supplementary Table 3).

No clinically significant intraprocedural bleeding was observed (0/57). Intraprocedural perforation occurred in 3 patients (5.3%), including two cases in the CD group and one case in the non-CD group. All perforations were closed endoscopically using clips, and no patient required emergency surgery. At 1 year, surgery-free survival did not differ between groups (72.5% in the CD group vs 85.9% in the non-CD group; P > 0.05).

DISCUSSION

ESt is increasingly used for the treatment of small-bowel fibrostenotic strictures[22-24,26]. By preserving bowel length, ESt avoids resection and, compared with EBD, has been associated with more durable luminal patency[23,27,28]. In this etiology-based analysis, reintervention-free survival after ESt was higher in patients with non-CD strictures than in those with CD-related strictures. This difference persisted across both univariable and parsimonious multivariable analyses. This finding supported a clinically meaningful difference in long-term outcomes between these two entities. Moreover, the baseline inflammatory status appeared to further stratify prognosis, as a NLR of < 2.4 was independently associated with a lower risk of reintervention and represents a readily available, noninvasive parameter for preprocedural risk assessment. Because inflammatory indices such as NLR and CLR are derived from related laboratory parameters, only NLR was included in the multivariable model to reflect the baseline inflammatory status.

Most existing studies have compared ESt with EBD or surgery, but long-term outcomes stratified by stricture etiology remain limited[21,23,24,29]. Lan et al[24] reported that, in patients with CD, ESt achieved surgery-free survival comparable to ileocecal resection and was associated with fewer complications, supporting its role as a bowel-preserving strategy. In anastomotic strictures, ESt was associated with higher technical success and improved clinical and endoscopic outcomes compared with EBD. It also reduced subsequent surgery (9.5% vs 33.5%, P = 0.03)[23]. However, whether etiology influences long-term prognosis has not been well studied[25]. Our findings indicate that long-term outcomes after ESt differ according to stricture etiology, with a higher risk of reintervention for CD-related strictures than for non-CD strictures.

In this study, patients with CD-related strictures had a significantly higher risk of reintervention than those with non-CD strictures. This difference is likely related to distinct pathological features between the two entities. CD-related strictures typically arise in the setting of chronic transmural inflammation and are characterized by inflammation driven, fixed fibrosis[30]. Persistent inflammatory activity promotes extracellular matrix deposition while limiting its degradation, resulting in the progressive fixation of the stricture[31]. Additionally, CD-specific changes in the intestinal microenvironment, such as creeping fat infiltration and microbial dysbiosis, may further contribute to the irreversibility of stricture formation[32]. Within this pathological context, ESt can effectively relieve mechanical luminal narrowing but is unlikely to reverse the underlying inflammation-fibrosis process[12,28]. Consequently, CD-related strictures may be more susceptible to restenosis and subsequent reintervention. This tendency was also observed in the technical failures in our cohort. Beyond differences in long-term reintervention risk, technical feasibility also varied across stricture phenotypes in our cohort. Among the 10 cases of technical failure, strictures were predominantly fixed or angulated, a morphology for which mucosal or submucosal incision alone was insufficient to achieve adequate luminal release. The baseline inflammatory status also appeared to differ between patients with technical success and those with failure, as reflected by NLR values; however, given the descriptive nature of these observations and the limited number of technical failures, this finding should be interpreted with caution.

From a technical perspective, these findings suggest that ESt may be better suited for short, concentric strictures predominantly involving the mucosal or submucosal layers. In contrast, angulated strictures or strictures with dense fibrotic tethering may have limited technical success, and early surgical consultation should be considered. Notably, these observations are intended to inform procedural decision-making rather than serve as determinants of the study’s primary outcome.

Regarding inflammatory markers, a baseline NLR < 2.4 was an independent predictor of remaining free from reintervention after ESt. NLR reflects systemic inflammatory burden and is consistently associated with IBD activity[33,34]. A meta-analysis including 2185 patients with IBD demonstrated significantly higher NLR in active CD compared with healthy controls (SMD = 3.53) and with patients in remission (SMD = 1.91; P < 0.05), reflecting neutrophil activation and lymphocyte depletion[33]. Langley et al[34] similarly reported good discriminative performance of NLR for detecting inflammatory activity. Using maximally selected log-rank statistics, we identified an NLR cutoff of 2.4 as the optimal threshold for predicting reintervention risk. Patients with NLR < 2.4 had significantly higher reintervention-free survival compared to those with elevated NLR. This association remained robust when NLR was independently examined in multivariable analysis after adjustment for selected non-inflammatory covariates.

This finding is consistent with known mechanisms: Increased systemic inflammation promotes Th2/Th17 polarization and activates the TGF-β pathway, which accelerates extracellular matrix deposition, tissue remodeling, and predisposes to restenosis, even after successful mechanical dilation[32,35]. Accordingly, NLR is more appropriately interpreted as a surrogate marker of the inflammation-fibrosis axis rather than a direct mechanistic driver of restenosis. It supports its use in postoperative risk stratification and may help guide the need for intensified anti-inflammatory therapy in CD-related strictures.

This study has several limitations. First, because ESt is primarily performed in tertiary centers, selection and referral bias may have influenced the study population. Second, the single-center retrospective design and modest sample size (n = 57) limited the statistical power of the study, particularly for subgroup analyses, and constrained our ability to fully characterize etiological heterogeneity among non-CD strictures. In addition, residual confounding related to prior or concomitant medical therapy—particularly exposure to biologics among patients with CD—may not have been fully accounted for in this retrospective analysis. Detailed CD-specific characteristics, including inflammatory activity, as well as post-procedural treatment adjustments were not systematically incorporated into the multivariable models. Moreover, only the baseline NLR was evaluated without assessment of longitudinal changes following ESt. Third, reintervention was primarily defined by symptom recurrence prompting further endoscopic or surgical management. Although imaging and/or endoscopic confirmation of restenosis was obtained in most cases, some degree of subjectivity in outcome assessment cannot be entirely excluded.

Despite these limitations, the present study provides clinically relevant real-world data on long-term outcomes after ESt across different stricture etiologies. Future multicenter, prospective studies with larger cohorts and longer follow-up durations are warranted to reduce referral bias, address the heterogeneity among non-CD strictures, and refine risk prediction for reintervention. These findings may assist in patient selection for ESt. Considering stricture etiology and baseline inflammatory status during preprocedural assessment may help inform decision-making and better anticipate long-term outcomes.

CONCLUSION

Patients with non-CD small bowel strictures have more favorable outcomes after ESt compared to those with CD-related strictures. A baseline NLR < 2.4 was associated with a lower risk of reintervention, serving as a simple and practical biomarker for preprocedural risk stratification.

ACKNOWLEDGEMENTS

We would like to thank all participants in this study.