ABSTRACT

PURPOSE

Malignant distal colorectal obstruction is associated with substantial morbidity and mortality when managed with emergency surgery. This study aimed to evaluate the technical and early clinical outcomes, safety, procedural radiation exposure, and stent-related reintervention of fluoroscopy-guided self-expandable metal stent (SEMS) placement.

METHODS

This retrospective single-center study included patients treated between January 2018 and December 2025. Bridge-to-surgery or palliative intent was established before stent placement and retrospectively abstracted from contemporaneous records. Primary end points were technical success and early clinical success within 48 hours. Secondary end points included adverse events within 30 days, stent-related reintervention, time to oral intake, total procedure time, fluoroscopy time, and dose-area product.

RESULTS

The cohort comprised 19 patients. Technical success was achieved in all patients [100%; 95% confidence interval (CI), 82.4%–100%], and early clinical success in 18 (94.7%; 95% CI, 74.0%–99.9%). Median time to oral intake was 3 days (interquartile range, 2–5 days). Median total procedure time was 38 minutes (range, 25–55 minutes). Median fluoroscopy time was 4.5 minutes (range, 3.8–8.0 minutes), and median dose-area product was 6.4 Gy·cm² (range, 5.1–8.8 Gy·cm²). One severe adverse event occurred (5.3%; 95% CI, 0.1%–26.0%): delayed perforation requiring surgery. No stent migration or 30-day mortality occurred. Stenting was performed with bridge-to-surgery intent in 5 patients and palliative intent in 14. All bridge-to-surgery patients underwent surgery after a median of 12 days; 3 underwent primary anastomosis and 2 required stoma creation. Among patients treated with palliative intent, 13 achieved early clinical success and 1 experienced clinical failure. One endoscopic reintervention for delayed stent dysfunction was performed at 150 days.

CONCLUSION

Fluoroscopy-guided SEMS placement achieved high technical and early clinical success in selected patients with distal malignant colorectal obstruction and may provide an effective interventional radiology–led decompressive option in experienced centers.

CLINICAL SIGNIFICANCE

Fluoroscopy-guided SEMS placement can provide timely decompression in selected patients with distal malignant colorectal obstruction within an interventional radiology–led multidisciplinary pathway, with endoscopic support available when needed.

Acute malignant colonic obstruction (MCO) is the initial presentation of colorectal cancer in approximately 8%–13% of cases and remains a time-sensitive condition in which prompt decompression is a major determinant of short-term outcomes.¹ Malignant distal colorectal obstruction may also arise from locally recurrent colorectal cancer or extracolonic malignancies through direct invasion, extrinsic compression, or peritoneal dissemination.^{2, 3} Emergency surgery in this setting is associated with substantial morbidity and mortality, particularly in patients who are physiologically vulnerable, and carries higher stoma rates, lower rates of primary anastomosis, and less frequent use of minimally invasive approaches. Meta-analyses further suggest that self-expandable metal stents (SEMSs) may improve short-term perioperative outcomes compared with urgent surgery, particularly by reducing postoperative morbidity and length of hospital stay.^{4, 5}

Since the first radiology-based reports in the late 1990s, colorectal SEMS placement has evolved from a predominantly palliative intervention into an established component of MCO management.^{6, 7} By restoring luminal patency, SEMS placement may convert an emergency presentation into a more favorable elective setting, facilitating physiologic optimization, completion of staging, and operative planning while increasing the likelihood of single-stage resection with primary anastomosis. Contemporary guidelines support SEMS use for palliation and, in selected patients, as a bridge to surgery, although uncertainties remain regarding patient selection, long-term oncologic outcomes, and optimal procedural strategy.^{1, 2}

Most available evidence derives from endoscopic stent placement performed with fluoroscopic support, whereas fluoroscopy-guided placement without routine endoscopic assistance remains less well characterized. Endoscopic support may be particularly useful for proximal lesions or markedly tortuous anatomy, whereas a transanal fluoroscopic approach may provide a more direct route to selected distal strictures. Fluoroscopy-guided SEMS placement should therefore be viewed as a complementary option within multidisciplinary care rather than as a competing alternative to endoscopy.^{2, 3}

Distal lesions nevertheless pose distinct technical challenges, including angulation, luminal distortion, patient discomfort, tenesmus, and concerns regarding stent anchoring. Radiology-based and contemporary fluoroscopy-guided series have reported high technical and clinical success in selected cases of malignant colorectal obstruction, including distal lesions.^7-10 Fluoroscopic placement may also provide a rescue option after failed endoscopic traversal.¹¹ Appropriate case selection should incorporate computed tomography (CT)–defined anatomy, the absence of perforation, the likelihood that focal decompression will restore clinically meaningful bowel passage, local expertise, and the availability of endoscopic support when needed.

At our institution, selected patients with distal malignant colorectal obstruction are managed using an interventional radiology–led fluoroscopic workflow, with endoscopic support available as needed. We evaluated this workflow in a contemporary single-center cohort that included primary colorectal tumors, malignant anastomotic recurrence, and extracolonic malignancies, focusing on technical and early clinical outcomes, adverse events, procedural radiation exposure, stent-related reintervention, and subsequent management after decompression.

Methods

Study design and patient selection

This retrospective single-center study was approved by the Hacettepe University Health Sciences Research Ethics Committee (decision number: SBA 26/067; date: January 20, 2026), and the requirement for individual informed consent was waived. All consecutive patients referred to the interventional radiology unit for fluoroscopy-guided stenting of distal malignant colorectal obstruction between January 2018 and December 2025 were screened for inclusion.

Distal obstruction was defined anatomically as a malignant stenosis involving the distal sigmoid colon, rectosigmoid junction, or rectum; no fixed distance from the anal verge was used as an eligibility threshold. Eligible patients had clinical symptoms of large-bowel obstruction and contrast-enhanced CT evidence of a dominant malignant stenosis with upstream colonic dilatation, without evidence of perforation. Stenting was considered when focal decompression of the index lesion was expected to restore clinically meaningful bowel passage.

Peritoneal carcinomatosis alone was not considered an exclusion criterion. Patients were not offered stenting when CT demonstrated peritoneal disease causing multilevel obstruction, including small-bowel involvement, because isolated treatment of the distal colorectal lesion was unlikely to provide adequate decompression. Patients with proximal colonic obstruction were outside the study’s prespecified scope. Treated patients were excluded from the analytic cohort when the indication was non-malignant or was not an obstructing colorectal lesion, including anastomotic leak, benign anastomotic stricture, tumor fistulization, and stenting after magnetic compression anastomosis in a patient with a colostomy and an obstructed anastomotic segment. The cohort derivation pathway is shown in Figure 1.

Multidisciplinary assessment and treatment intent

Most patients were referred by the general surgery service after clinical and contrast-enhanced CT evaluation; gastroenterology input was obtained when clinically feasible (Figure 2). Eligibility for fluoroscopy-guided stenting was determined jointly by interventional radiology, surgery, and oncology, considering lesion anatomy, disease extent, absence of perforation, the anticipated effectiveness of focal decompression, operability, and the planned oncologic pathway. Both endoscopic and fluoroscopic gastrointestinal stenting were available at our institution. For this distal cohort, fluoroscopy-guided placement was selected based on lesion location, CT-defined anatomy, procedural urgency, and local expertise. Endoscopic support was available if needed, but no combined procedure or same-session endoscopic assistance was required.

Bridge-to-surgery intent was established before stent placement in operable patients for whom elective resection after decompression was planned. Stenting was considered palliative when definitive resection was not planned because of unresectable or disseminated disease or the patient’s overall clinical condition. Treatment intent was retrospectively abstracted from contemporaneous medical records. Because urgent decompression was often required, preprocedural tumor board review was not mandatory; subsequent management was reassessed after decompression and discussed in a multidisciplinary setting when indicated.

Preprocedural assessment

All patients underwent clinical evaluation and contrast-enhanced CT before intervention to confirm the level and extent of obstruction, estimate stricture length, assess proximal bowel dilatation, and exclude overt perforation or diffuse multilevel disease. Baseline laboratory evaluation included hemoglobin, white blood cell count, and serum albumin. American Society of Anesthesiologists (ASA) class and Eastern Cooperative Oncology Group (ECOG) performance status were recorded as baseline clinical variables. Procedures were performed in the angiography suite under anesthesiology monitoring, generally with conscious sedation and escalation to moderate sedation when required by patient cooperation or procedural tolerance; prophylactic antibiotics were administered.

Fluoroscopy-guided stent placement

All procedures were performed by an interventional radiologist with at least 10 years of experience in non-vascular interventional radiology. Patients were positioned in the left lateral decubitus position for fluoroscopy-guided stent placement. After anal cannulation, the stenotic segment was approached with a hydrophilic guidewire (Radifocus; Terumo, Tokyo, Japan) and a 5-Fr Bern or vertebral catheter (Boston Scientific, Marlborough, MA, USA). Once the obstructing segment had been traversed, water-soluble iodinated contrast material was injected to confirm intraluminal positioning proximal to the lesion and to delineate stricture length and morphology. An Amplatz stiff guidewire (Boston Scientific, Marlborough, MA, USA) was then advanced across the lesion, and a 9- or 10-Fr sheath was advanced over it for stent deployment. All patients were treated with uncovered nitinol SEMSs (Niti-S D-type colonic stent; TaeWoong Medical, Gimpo, South Korea). Stent diameter was standardized to 22 or 24 mm, and stent length was selected according to fluoroscopic stricture measurements to ensure coverage beyond both lesion margins. Overlapping stents were used at the operator’s discretion for selected long-segment or multifocal strictures.

At the end of the procedure, fluoroscopy was used to confirm stent position, expansion, and restoration of luminal passage (Figure 3). No balloon dilation was performed before or after stent deployment. Low-pulse-rate fluoroscopy was used at 7.5 pulses/s, with collimation whenever feasible and without routine angiographic or digital subtraction acquisitions. Total procedure time was defined as the interval from initial transanal cannulation to final fluoroscopic confirmation of stent position and luminal passage and was retrospectively abstracted from the procedural records. Fluoroscopy time and dose-area product were recorded by the fluoroscopy system.

Postprocedural follow-up

Following stent placement, patients were monitored for abdominal distension, pain, passage of flatus and stool, and tolerance of oral intake. Abdominal radiographs were obtained during the initial hospitalization, and CT was performed selectively in patients with persistent or recurrent symptoms or suspected perforation, stent migration, or recurrent obstruction (Figure 4). Subsequent follow-up was not protocolized and was reconstructed from available inpatient records, outpatient surgical and oncologic assessments, and follow-up imaging. Follow-up ended at definitive surgery, death, stent-related reintervention, or the last documented clinical encounter.

Definitions and study end points

The primary end points were technical success and early clinical success. Technical success was defined as successful fluoroscopic deployment of the stent across the target lesion, with satisfactory coverage of the stenotic segment and restoration of luminal passage on contrast assessment. Early clinical success was defined as improvement in obstructive symptoms with passage of flatus and/or stool within 48 hours, without the need for urgent alternative decompression. Time to oral intake was evaluated separately as a secondary outcome.

Secondary end points included total procedure time, fluoroscopy time, dose-area product, time to oral intake, stent migration, adverse events within 30 days, 30-day mortality, and stent-related reintervention. Stent dysfunction was defined as recurrent obstructive symptoms attributable to the treated segment that prompted endoscopic, radiologic, or surgical reintervention. The stent-related reintervention-free interval was defined as the time from stent placement to the first intervention for documented stent dysfunction. In patients without stent-related reintervention, follow-up ended at the time of definitive surgery, death, or the last documented clinical assessment.

Migration was defined as radiographic displacement of the stent from its original position, regardless of clinical consequence. Adverse events were retrospectively classified according to the Society of Interventional Radiology adverse-event severity scale.¹² Routine endoscopic reassessment was not performed.

Statistical analysis

Continuous variables were summarized as median [interquartile range (IQR)] or range, as appropriate, and categorical variables as counts and percentages. Exact two-sided 95% confidence intervals (CIs) for proportional outcomes were calculated using the Clopper–Pearson method. Because of the small sample size and descriptive study design, no formal between-group hypothesis testing was performed. Stent-related reintervention and the corresponding interval were reported descriptively. Formal time-to-event analysis of stent patency was not performed because follow-up was non-protocolized and definitive surgery and death represented frequent competing events. Statistical analyses were performed using IBM SPSS Statistics, version 26.0 (IBM Corp., Armonk, NY, USA).

Results

Cohort derivation

During the study period, 37 patients were referred for possible fluoroscopy-guided colorectal stenting. Eight did not undergo stenting: 4 had peritoneal disease causing multilevel obstruction, including small-bowel involvement, and 4 had proximal colonic obstruction outside the prespecified scope of the study. Among the remaining 29 treated patients, 10 were excluded because the indication was not distal malignant colorectal obstruction, including anastomotic leak (n = 4), benign anastomotic stricture (n = 2), tumor fistulization (n = 3), and stenting after magnetic compression anastomosis in a patient with a colostomy and an obstructed anastomotic segment (n = 1). The final analytic cohort comprised 19 patients: 9 with primary colorectal malignancy, 3 with malignant anastomotic recurrence after rectal cancer surgery, and 7 with extracolonic malignancy involving the distal colorectum (Figure 1).

Baseline and lesion characteristics

The cohort included 19 patients (10 men and 9 women) with a median age of 67 years (IQR, 54–76.5 years); ASA class was II in 11 patients and III in 8 patients, and ECOG performance status was 0 in 6 patients, 1 in 9 patients, and 2 in 4 patients. Median hemoglobin was 11.4 g/dL (IQR, 10.0–12.3 g/dL), median white blood cell count was 8.8 × 10⁹/L (IQR, 4.9–13.6 × 10⁹/L), and median serum albumin was 3.2 g/dL (IQR, 2.7–3.7 g/dL); hypoalbuminemia was present in 13 of 18 patients with available values. Peritoneal metastases or carcinomatosis were present in 12 patients (63%). The etiology of obstruction was primary colorectal malignancy in 9 patients, malignant anastomotic recurrence in 3, and extracolonic malignancy in 7. The treated obstruction was located in the rectum (n = 10), rectosigmoid junction (n = 5), or sigmoid colon (n = 4). Median fluoroscopic stricture length was 6.5 cm (IQR, 5.0–7.5 cm), median distance from the anal verge was 12.5 cm (IQR, 7.0–16.5 cm), and median stent length was 10 cm (IQR, 9–12 cm) (Table 1).

Procedural and early clinical outcomes

Technical success was achieved in all 19 patients (100%; 95% CI, 82.4%–100%), and early clinical success was achieved in 18 patients (94.7%; 95% CI, 74.0%–99.9%). Median total procedure time was 38 minutes (range, 25–55 minutes). Median fluoroscopy time was 4.5 minutes (range, 3.8–8.0 minutes), and median dose-area product was 6.4 Gy·cm² (range, 5.1–8.8 Gy·cm²). Median time to oral intake was 3 days (IQR, 2–5 days). One severe adverse event occurred (1/19, 5.3%; 95% CI, 0.1%–26.0%): delayed perforation on postprocedural day 6 requiring surgery. No stent migration or 30-day mortality was observed (0/19 for each; 95% CI, 0%–17.6%) (Table 2).

Technical success was achieved in all etiologic subgroups. Early clinical success was achieved in 9 of 9 patients with primary colorectal malignancy, 2 of 3 with malignant anastomotic recurrence, and 7 of 7 with extracolonic malignancy. The only clinical failure and severe adverse event occurred in the malignant anastomotic recurrence group. In a patient with clinical failure, a retrospective review suggested that traction and fixed angulation due to desmoplastic peritoneal disease limited effective bowel passage despite restoration of luminal patency in the treated segment. Because of the small subgroup sizes, no formal between-group comparisons were performed (Table 3).

Treatment intent, subsequent management, and stent-related reintervention

Stenting was performed with bridge-to-surgery intent in 5 patients and palliative intent in 14. All 5 bridge-to-surgery patients subsequently underwent surgery at 4, 7, 12, 12, and 15 days after stent placement, corresponding to a median interval of 12 days (IQR, 7–12 days). Three underwent resection with primary anastomosis, and 2 required stoma creation. In both patients who required stoma creation, the operative course reflected additional intra-abdominal disease burden or multilevel obstruction rather than a technical failure at the index stent site. In 1 bridge-to-surgery patient, only partial symptomatic relief was documented due to a second obstructive process involving the transverse mesocolon rather than stent-site failure.

Among the 14 patients treated with palliative intent, 13 achieved early clinical success and continued palliative management; 1 experienced clinical failure and underwent surgery for delayed perforation. Among the 13 patients who achieved early clinical success and continued palliative management, median available follow-up was 35 days (IQR, 17–90 days). Among all 14 patients treated with palliative intent, 1 (7.1%) underwent endoscopic reintervention for delayed stent dysfunction at 150 days; this event occurred in a patient who had achieved early clinical success. No additional stent-related reintervention was documented before death or the last available clinical follow-up. Fluoroscopy-guided stent placement was successful in 1 patient after failed colonoscopic traversal (Table 4).

Discussion

This retrospective single-center study demonstrated high technical and early clinical success for fluoroscopy-guided SEMS placement in selected patients with distal malignant colorectal obstruction. This study adds contemporary outcome data on an interventional radiology–led fluoroscopic workflow in a distal cohort that included primary colorectal malignancy, malignant anastomotic recurrence, and extracolonic malignancy. The findings should be interpreted in the context of careful case selection and substantial operator experience.

Our overall outcomes compare favorably with those reported in previously fluoroscopy-guided series. Zeng et al.⁹ reported technical and clinical success rates of 97.5% and 96.6%, respectively, whereas Wan et al. reported corresponding rates of 98.5% and 95.5% in malignant rectal obstruction.¹⁰ In the present cohort, technical success was achieved in all three etiologic subgroups. Early clinical success was achieved in all patients with primary colorectal and extracolonic malignancy and in 2 of 3 patients with malignant anastomotic recurrence. The only clinical failure and severe adverse event occurred in the anastomotic recurrence group, although the small subgroup sizes preclude meaningful comparisons among etiologies.

Our cohort largely comprised patients with advanced metastatic disease, many of whom had peritoneal metastases or carcinomatosis and were not candidates for definitive surgery. In most patients, SEMS placement provided clinically meaningful decompression without the need for immediate emergency surgery, as reflected by early symptom relief and resumption of oral intake. By relieving obstruction and often preserving intestinal continuity, stenting may also facilitate subsequent oncologic and supportive care. In this context, SEMS may serve not only as a local decompressive measure but also as a practical component of palliative oncologic management.

Our institutional workflow consisted of prompt surgical referral; contrast-enhanced CT review; multidisciplinary assessment of whether focal distal decompression was likely to restore bowel passage; fluoroscopy-guided transanal stent placement by an interventional radiologist with at least 10 years of non-vascular interventional radiology experience under anesthesiology monitoring; and postprocedural reassessment by surgery and oncology. Endoscopic colorectal stenting at our institution is performed by the gastroenterology service through a separate clinical workflow. In contrast, the patients in the present cohort were referred to interventional radiology predominantly by general surgery, with gastroenterology consultation obtained when clinically appropriate. Endoscopic support was available if needed, but no combined procedure or same-session endoscopic assistance was required. Because a complete cross-service, procedure-specific registry was unavailable and referral pathways and case selection differed between services, a reliable contemporaneous comparison could not be performed. Accordingly, the present study was intended to describe our interventional radiology–led fluoroscopic workflow rather than to establish superiority over endoscopic techniques.

The distinction between technical and clinical success is particularly relevant in advanced or anatomically complex diseases. The only clinical failure occurred despite technically successful stent deployment in a patient with malignant anastomotic recurrence and peritoneal carcinomatosis. Retrospective review suggested that traction and angulation related to desmoplastic peritoneal disease limited effective passage despite restoration of luminal patency. This case illustrates that correct stent deployment does not invariably provide adequate decompression when fixed angulation or multifocal disease persists. Similarly, in a patient who subsequently underwent surgery, the index stent provided only partial symptomatic relief because of a second obstructive process involving the transverse mesocolon rather than failure at the treated segment. These observations underscore the importance of distinguishing focal technical success from the overall clinical effect of decompression in patients with extensive intra-abdominal disease.

SEMS placement may convert an emergency presentation into a more controlled operative pathway in appropriately selected patients. Pooled analyses have reported high technical and clinical success, favorable primary anastomosis rates, and low perioperative mortality in the bridge-to-surgery setting.⁵ More recent evidence suggests lower morbidity and stoma rates than emergency resection without a clear adverse effect on disease-free or overall survival in selected patients with left-sided MCO.¹³ The interval from stent placement to surgery may also be individualized after successful decompression.¹⁴ In our cohort, bridge-to-surgery intent was established before stent placement in five operable patients, all of whom subsequently underwent surgery after a median of 12 days. Three underwent primary anastomosis, whereas two required stoma creation due to additional intra-abdominal disease burden or multilevel obstruction.

Palliative stenting may relieve obstruction while avoiding emergency surgery and reducing stoma burden in patients without a planned definitive resection. Meta-analytic data suggest shorter hospitalization and lower early morbidity than surgery, although later reintervention is more frequent.^15-17 In the present cohort, 13 of 14 patients treated with palliative intent achieved early clinical success. Among the 13 patients who continued palliative management, 1 underwent endoscopic reintervention for delayed stent dysfunction at 150 days, and no additional stent-related reintervention was documented before death or the last available follow-up. These findings indicate effective short-term decompression in most patients but should not be interpreted as evidence of durable stent patency, given the median follow-up of 35 days, the non-protocolized assessment, and the frequent occurrence of death as a competing event. Although patient-reported quality of life was not assessed, symptom relief and resumption of oral intake provided pragmatic indicators of early clinical recovery.

Procedural radiation exposure was also evaluated. Median fluoroscopy time was 4.5 minutes (range, 3.8–8.0 minutes), and median dose-area product was 6.4 Gy·cm² (range, 5.1–8.8 Gy·cm²), within a fluoroscopy-only workflow using 7.5 pulses/s and no routine angiographic or digital subtraction acquisitions. Median total procedure time was 38 minutes (range, 25–55 minutes). These measurements are provided as procedural reference data rather than comparative benchmarks, because radiation exposure and procedure duration are influenced by equipment, patient habitus, lesion complexity, collimation, and institutional technique.

This study has several limitations. First, its retrospective, single-center design and small sample size limit precision and preclude meaningful statistical comparisons among etiologic subgroups. Second, the cohort was clinically heterogeneous and restricted to distal obstruction, which limits generalizability to proximal lesions. Third, procedures were performed by a single interventional radiologist with at least 10 years of non-vascular interventional radiology experience; therefore, the high technical success may not be directly reproducible in centers with less experience. Moreover, a complete cross-service, procedure-specific registry of endoscopy-assisted colorectal SEMS procedures was unavailable, precluding a reliable contemporaneous comparison; differences in referral pathways and case selection also introduced potential selection bias. A prospective study based on a shared procedure-specific registry would provide a more appropriate framework for comparing fluoroscopic and endoscopic workflows. Follow-up was not protocolized and was reconstructed from clinical records and available imaging, creating a risk of incomplete event ascertainment. The short follow-up in the palliative group and the frequent competing events of death and definitive surgery precluded formal time-to-event analysis of stent patency. Finally, quality of life was not formally assessed, and the clinical benefit of decompression was inferred from symptom improvement, passage of flatus or stool, and resumption of oral intake.

Fluoroscopy-guided SEMS placement achieved high technical and early clinical success in selected patients with distal malignant colorectal obstruction. Within an experienced, multidisciplinary setting, it may provide an effective interventional radiology–led option for bridge-to-surgery decompression and palliation. The small cohort, absence of a comparative group, and non-protocolized follow-up warrant cautious interpretation and further evaluation in larger studies.