Long-term efficacy following venous stenting for symptomatic iliofemoral venous obstruction: 5 year clinical and stent outcomes in a Southeast Asian population

Abstract

Objective

Endovenous stenting is the standard of care for symptomatic iliofemoral venous obstruction with good short-term patency. We present our long-term outcomes in a Southeast Asian cohort.

Methods

We conducted a single-center retrospective review of 76 patients (98 limbs) who underwent iliofemoral venous stenting between 2015 and 2020. Clinical data and stent outcomes were analyzed over a median follow-up of 65.3 months. Primary, assisted primary, and secondary patency were reviewed using Kaplan-Meier analysis. Stent-related complications, long-term clinical outcomes and factors associated with stent occlusion, were assessed.

Results

Of 98 limbs, 82.6% had nonthrombotic iliac vein lesions, 9.2% post-thrombotic syndrome, and 8.2% acute deep vein thrombosis. A 100% technical success rate was achieved. The 5-year cumulative primary, assisted primary, and secondary patency rates were 88.2%, 92.5%, and 98.9%, respectively. Primary patency for stenting for nonthrombotic iliac vein lesion, post-thrombotic syndrome, and deep vein thrombosis was 92.0%, 77.8%, and 62.5%, respectively. Loss of patency occurred mainly within the first year, largely from noncompliance and inadequate anticoagulation. Stent-related complications were rare (fracture 1.3%, migration 1.3%). The rate of stent occlusion (per limb) was 8.2%. At 5 years, ulcer-free rate was 82.7%, pain relief 91.8%, and sustained swelling relief 53.1%. No patient required major amputation, and all-cause mortality was unrelated to venous disease.

Conclusions

Dedicated venous stenting for symptomatic iliofemoral obstruction is safe and durable, with good long-term patency and clinical outcomes in this Southeast Asian cohort. Most reinterventions occurred early, highlighting the importance of procedural technique, inflow optimization, and adherence to antithrombotic therapy. These data support endovenous stenting as a long-term treatment strategy for iliofemoral venous obstruction.

Article Highlights.

• •Type of Research: Single-center, retrospective cohort study
• •Key Findings: Of 98 limbs treated with iliofemoral venous stenting for nonthrombotic iliac vein lesions, post-thrombotic syndrome or after acute deep vein thrombolysis, 5-year cumulative primary, assisted primary, and secondary patency rates were 88.2%, 92.5%, and 98.9%, respectively. Ulcer-free rate was 82.7%, pain relief 91.8%, and sustained swelling relief 53.1%.
• •Take Home Message: Dedicated venous stenting for symptomatic iliofemoral obstruction is safe and durable, with good long-term patency and clinical outcomes in this single-center Southeast Asian cohort with predominantly nonthrombotic iliac vein lesions. Stent occlusion was uncommon, typically occurred early within the first year and was most often related to anticoagulation noncompliance rather than technical or flow factors.

Venous disease is prevalent worldwide and imposes a considerable physical, psychological, and socioeconomic burden. Among its many causes, chronic iliofemoral venous obstruction is a key driver of symptom progression and disability, often manifesting as pain, swelling, recurrent cellulitis, or venous ulceration.1, 2, 3, 4

Anatomical obstruction of the abdominopelvic veins can arise from nonthrombotic iliac vein lesions (NIVLs), typically due to extrinsic compression such as May-Thurner syndrome, or from post-thrombotic changes after deep vein thrombosis (DVT). Both mechanisms disrupt venous architecture and compromise outflow. Because the iliac vein serves as the principal drainage pathway of the lower limb, persistent obstruction can result in severe, lifestyle-limiting symptoms. However, NIVL is frequently clinically silent or associated with only mild, nonspecific complaints, hence treatment decisions must be individualized to the clinical context rather than the anatomical finding alone.⁵

With the advent of the endovascular era in the early 1990s, minimally invasive interventions have now revolutionized the treatment of venous disease. The treatment paradigm has shifted away from open surgery or anticoagulation alone, and contemporary societal guidelines now advocate percutaneous endovenous stenting as the standard of care for patients with clinically significant femoroiliocaval deep venous obstructive disease. Deep venous stenting using dedicated venous stents have been shown to be safe with meta-analyses reporting 12-month patency rates of 83%, 90%, and 95% for primary patency, primary assisted patency, and secondary patency respectively and 36-month primary patency, primary-assisted patency, and secondary patency rates of 59%, 80%, and 86%, respectively.⁶

Despite these encouraging results, most data derive from Western populations. Asian patients differ in important aspects, including lower baseline incidence of venous thromboembolism and inherited thrombophilia,⁷ which may influence stent performance and long-term outcomes. Moreover, long-term follow-up remains limited.

We previously reported short term outcomes following deep venous stenting for iliofemoral vein pathologies in our institution in 2020.⁸ There was statistically significant reduction in Venous Clinical Severity Score (VCSS) and Villalta score, and cumulative primary and secondary patency rates at 24 months of 90% and 100%. In the present study, we independently evaluate the 5-year clinical and stent outcomes in a cohort with symptomatic iliofemoral venous obstruction treated at our center. We also analyze factors associated with stent occlusion, stent-related complications, and limb-related outcomes, to assess the long-term safety and effectiveness of venous stenting in an Asian population.

Methods

Study design

This is a single-center retrospective review of consecutive patients who had venous stents placed from April 2015 to January 2020 for symptomatic iliofemoral venous obstruction. Procedural details had been prospectively collected in a venous stenting database. Follow-up data was collected retrospectively from review of electronic medical records. A waiver from the institutional review board was obtained. Although the detailed methodology has been described previously,⁸ the current study represents an independent analysis focusing on long-term outcomes.

Setting and patients

The center is a 1800-bed tertiary teaching university hospital in Singapore, managing approximately 1000 chronic venous insufficiency-related admissions, 2000 outpatient multilayer compression bandage applications, and 250 superficial venous surgeries annually. Indications for iliac vein stent placement included patients with symptomatic chronic venous disease and iliofemoral venous outflow obstruction (total occlusion or ≥50% area stenosis) as determined by intravascular ultrasound (IVUS) examination of the iliac vein.^1,9 In patients with NIVL, stenting was restricted to those with moderate to severe symptoms clearly attributable to the venous lesion and was not performed in asymptomatic or minimally symptomatic individuals.¹⁰ Accordingly, patients with symptomatic, nonmalignancy-related NIVL or post-thrombotic syndrome (PTS) were included. Venous stents placed after deep vein thrombolysis or thrombectomy were also included.

Patient demographics including ethnicity, nature of disease (NIVL, PTS, or acute DVT), clinical manifestations, and preoperative Clinical, Etiology, Anatomy, Pathophysiology (CEAP) classification, VCSS and Villalta scores were obtained.11, 12, 13 Patients with acute DVT and PTS were co-managed with hematology or vascular medicine specialists. Thrombophilia testing, including tests for antiphospholipid antibodies, factor V Leiden, and antithrombin, was not routinely performed, unless clinically indicated as per American Society of Hematology guidelines.¹⁴ All patients with stent occlusion or recurrent DVT had thrombophilia screening tests performed.

Anatomical extent of disease was classified into five category types: (1) NIVL, (2) chronic venous obstruction (CVO) of iliac segment, (3) CVO of iliofemoral segment above common femoral vein (CFV) confluence, (4) CVO of iliofemoral segment extending into the FV or deep FV, and (5) CVO of iliofemoral segment involving both deep FV and FV, as per Jalaie et al.¹⁵ All patients had preoperative venous reflux duplex ultrasound examination of the lower limb to evaluate for patency and reflux of the deep and superficial veins. Additional cross-sectional imaging consisting of magnetic resonance venography or computed tomography venography was performed for all patients. Our center previously reported use of unenhanced magnetic resonance imaging using the relaxation-enhanced angiography without contrast and triggering protocol.¹⁶ IVUS examination, in addition to angiography in at least two projections, was used in all cases. If there were discrepancies, IVUS examination was the gold standard to evaluate the anatomical extent of disease. A ≥50% luminal area reduction was considered as significant.

Procedural details

All deep venous interventions were performed by consultant-grade vascular surgeons and interventional radiologists in a dedicated Angio suite. Venous access was obtained via the FV and/or the internal jugular vein. A 9F or 10F sheath was inserted followed by 5000 IU heparin with activated clotting time monitoring. All procedures were performed with IVUS examination using Visions PV .035 (Philips Medical Systems) in addition to venography to guide treatment decisions including maximal zone of compression, optimal landing zones, and stent sizing.^9,17,18 Predilation and postdilatation angioplasty was performed with noncompliant angioplasty balloons (Atlas/Atlas Gold, Becton, Dickinson and Company). Dedicated venous stents were placed, with the type of stent placed at the discretion of the proceduralist. Subcutaneous low-molecular-weight heparin 1 mg/kg was administered to all patients in recovery. Pneumatic compression was applied for all patients after the procedure in addition to full-length compression stockings.

Follow-up

After venous stenting, all patients were given novel oral anticoagulants for ≥3 months with additional single antiplatelet therapy (aspirin or clopidogrel) administered at the proceduralist's discretion. Stent patency was primarily assessed using ultrasound duplex examination at intervals of 2 weeks, 3 months, 6 months, and then annually. If high-risk stenting was performed for acute DVT or PTS, additional ultrasound examination was performed on post operative day 1. Computed tomography venography was additionally performed if patency was unclear on ultrasound examination or if there was concern of stent fracture or migration. Patients who were symptomatic with imaging findings of thrombosed stents or significant in-stent stenosis (>50%) were offered reintervention. Patients were de-escalated to single antiplatelet at 3 months after NIVL stenting if the surveillance duplex did not show in-stent stenosis. From 2020 onward, CYP2C19 genotyping was performed before prescribing long-term clopidogrel in recognition of diminished antiplatelet effect of clopidogrel in CYP2C19 poor and intermediate metabolizers which is present in 14% and 40% of Asians, respectively.¹⁹ Clopidogrel was the first-line single antiplatelet therapy for normal metabolizers, whereas poor or intermediate metabolizers received aspirin as an alternative in accordance with our center's protocol. Anticoagulation adherence was assessed during each clinical follow-up visit. Noncompliance was defined as failure to take medication as prescribed.

Anatomical outcome measurement

Long-term post-stenting patency was obtained, categorized into primary patency, primary assisted patency, and secondary patency as defined by the reporting standards of the Society for Vascular Surgery.²⁰ Primary patency was defined as stent patency in the absence of reintervention. Primary assisted patency was defined as recanalization of a nonoccluded stent through additional intervention. Secondary patency was defined as recanalization of an occluded stent through additional intervention.⁶ Occlusion and reintervention rates were obtained and subgroup analysis was performed to determine if stent manufacturer and type were significant for occlusion and/or reintervention. Potential contributors to stent occlusion included hematologic factors (eg, anticoagulation compliance or adequacy of dosing), flow factors (eg, inflow disease), and technical factors.^9,21,22 Primary contributing factors in cases of stent occlusion were analyzed retrospectively by two vascular surgeons. Other stent-related complications, such as fracture or migration, were also recorded.

Clinical outcome measurement

Pain relief and clinical features of chronic venous disease (edema improvement and the presence of venous ulcer) were assessed clinically at each follow-up visit.¹¹ Pain was evaluated based on patient-reported symptoms obtained through standard clinical questioning, whereas edema was determined by visual inspection and palpation for the degree of limb swelling. Venous leg ulcers were monitored through digital photographic documentation and standardized wound assessment.

Adverse limb events such as ipsilateral (to stent placement) or contralateral lower limb DVT, or lower extremity amputation, and mortality, were also audited.²³

Data analysis

Categorical variables were summarized as counts and percentages, and continuous variables as mean ± standard deviation or median ± interquartile range. Kaplan-Meier survival analysis was used to estimate primary, assisted primary, and secondary patency rates, as well as to analyze and compare resolution of pain, swelling, and ulcer after treatment. Cumulative survival curves were compared using the log-rank (χ²) test. Associations for both continuous and categorical variables were evaluated using Cox proportional hazards models, with results reported as hazard ratios with 95% confidence intervals. A P value of <.05 was considered significant. Variables with a P value of <.2 on univariate analysis were subsequently included in the multivariate analysis. Analyses were performed using jamovi (version 2.6) statistical software.²⁴

Results

Patient characteristics

A total of 76 patients with 98 limbs underwent iliac vein stenting between 2015 and 2020 (Table I), of whom bilateral disease was treated in 22 patients. Among limbs treated, left-sided disease (n = 47) was more common than right-sided disease (n = 7), which is in keeping with the typical distribution of NIVLs. The median follow-up time was 65.3 months, with most patients completing ≥5 years of follow-up. Over this period of time, 10 patients died and 7 were lost to follow-up.

Table I.

Patient characteristics

Study population
Patients	76
Limbs treated	98
Right only	7 (7.1%)
Left only	47 (48.0%)
Bilateral	22 (22.4%)
Follow-up, months	65.3 (16.4)
ASA classification
1	2 (2.6%)
2	45 (59.2%)
3	27 (35.5%)
Demographics
Male	42 (54.5%)
Female	34 (45.5%)
Ethnicity
Chinese	43 (56.6%)
Malay	8 (10.5%)
Indian	25 (32.9%)
Mean age, years	59.4 ± 11.5
Mean BMI, kg/m2	28.9 ± 6.4
Nature of disease
NIVL	81 (82.6%)
PTS	9 (9.2%)
DVT	8 (8.2%)
Jalaie classification of CVO (where applicable)
1	81
2	8
3	1
Preoperative CEAP
C1	0
C2	1 (1.0%)
C3	28 (28.6%)
C4	21 (21.4%)
C5	19 (19.4%)
C6	29 (29.6%)
Preoperative scores
VCSS	11 (5)
Villalta	13 (5)
Indication for surgerya
Leg pain	12 (12.2%)
Recurrent cellulitis	21 (21.4%)
Recalcitrant venous ulceration	29 (29.6%)
Leg edema	44 (44.9%)
In conjunction with DVT thrombectomy or fibrinolysis	7 (7.1%)

ASA, American Society of Anesthesiologists; BMI, body mass index; CVO, chronic venous obstruction; CEAP, Clinical, Etiology, Anatomy, Pathophysiology; DVT, deep vein thrombosis; NIVL, nonthrombotic iliac vein lesion; PTS, post-thrombotic syndrome; VCSS, Venous Clinical Severity Score.

Values are median (interquartile range) unless otherwise indicated.

^aIndications are not mutually exclusive.

The mean age of the study population was 59.4 years, and the cohort comprised of 43 Chinese (56.6%), 8 Malay (10.5%), and 25 Indian (32.9%) patients.

Common presenting complaints included persistent lower limb edema and recurrent cellulitis. Indications for stent placement per limb treated include leg pain (12.2%), recurrent cellulitis (21.4%), recalcitrant venous ulceration (29.6%), leg edema (44.9%), or performed in conjunction with DVT thrombectomy or fibrinolysis (7.1%). The median (interquartile range) preoperative VCSS and Villalta scores were 11 (5%) and 13 (5%) respectively.

Regarding the nature of disease, most patients (82.6%) had NIVL, nine patients had underlying PTS (9.2%) causing CVO, and eight had presented acutely with DVT (8.2%). CVO was further classified according to the Jalaie classification: type 1 (n = 81), type 2 (n = 8), and type 3 (n = 1). All but one patient had CEAP clinical score class ≥3 disease. CEAP class distribution follows as such: C2 (1.0%), C3 (28.6%), C4 (21.4%), C5 (19.4%), and C6 (29.6%). Of the NIVL patients, 1 (1.2%) had C2 disease, 20 (24.7%) had C3 disease, and 60 (74.1%) had C4-6 disease. The only patient with C2 disease had recurrent episodes of lower limb cellulitis and also underwent evaluation and stenting. There were no patients with detected thrombophilia in our cohort.

A total of 132 stents were placed across all limbs. The types of stents placed and number of stents placed per patient are as per Table II. A total of 70 limbs involved stenting confined to the iliac veins, 22 limbs with extension into the CFV and 5 patients with the stent(s) extending into the inferior vena cava (Table III). Superficial venous disease treatment was performed concomitantly in 12 limbs, and in 34 limbs it had been performed previously.

Table II.

Procedure details

Type of stent implanted
Becton Dickinson Venovo	41
Optimed Sinus-Venous	28
Optimed Sinus-Obliquus	2
Medtronic Abre	15
Zilver Vena	9
Others; nondedicated venous stents (Wallstent, Express LD)	3
No. of stents placed
1	45 (59.2%)
2	28 (36.8%)
3	3 (3.9%)
Superficial venous disease treatment
Concomitant	12
Previous	34

Table III.

Extent of stenting

Extent of stenting	Included extent(s)	Total
CIV and/or EIV only	CIV EIV CIV + EIV	70
Extension to CFV	EIV + CFV CIV + EIV + CFV IVC + CIV + EIV + CFV	22
Extension to IVC	IVC + CIV IVC + CIV + EIV IVC + CIV + EIV + CFV	5

CFV, Common femoral vein; CIV, common iliac vein; EIV, external iliac vein; IVC, inferior vena cava.

Anatomical outcome

There was 100% technical success in stent placement. Over the follow-up period, 13 of 76 patients (17.1%) required reintervention. Eight (61.5%) were for stent occlusion, three (23.1%) for symptomatic in-stent stenosis, one for stent fracture (7.7%), and one for additional stenting of critical stenosis (7.7%). Among the 13 patients, 2 (15.4%) were treated originally for PTS and 3 (23.1%) for DVT.

Loss of patency almost exclusively happened within the first year. Beyond the first year, two stents were occluded after 4.5 and 7.0 years, respectively. Stent occlusion rate was 8.2% at 8 out of 98 limbs. Noncompliance with anticoagulation accounted for 50% of occlusions, and inadequate dosing was responsible for 12.5%. In contrast, occlusions due to flow-related and technical factors were less frequent (Supplementary Table, online only).

Of these eight occlusions, six (75%) were salvaged successfully. Recanalization was not performed in two limbs with stent occlusion: one patient was not a candidate for further intervention owing to multiple comorbidities, and the other defaulted treatment. Of the 11 limbs that underwent reinterventions, 10 had one reintervention, while one required two reinterventions for symptomatic in-stent restenosis.

Overall primary, primary assisted, and secondary cumulative patency rates at 5 years were 88.2%, 92.5%, and 98.9% respectively (Fig, A).

Fig.

Kaplan-Meier analysis of cumulative primary (PP), assisted primary (PAP), and secondary (SP) patency after iliofemoral stenting. Numbers below the x-axis represent limbs at risk for each time interval. (A) All 98 limbs. (B) 81 limbs with nonthrombotic iliac vein lesion (NIVL). (C) Nine limbs with post-thrombotic syndrome (PTS). (D) Eight limbs with deep vein thrombosis (DVT).

The stented limbs with NIVL fared better as expected than those with thrombotic or post-thrombotic disease, with primary, primary-assisted, and secondary cumulative patency rates of 92.0%, 94.6%, and 98.7% (Fig, B-D). The χ² tests revealed a statistically significant difference in primary patency (P = .024) over 5 years among the three disease groups (NIVL, PTS, and DVT), but not for primary assisted (P = .087) or secondary (P = .82) patencies (Supplementary Fig, online only).

Significant stent-related complications were rare, with only one case each of fracture and migration (1.3% each). In-stent restenosis occurred in 10 patients (10/76 [13.1%]).

Factors associated with iliac vein stent occlusion

During the observation period, eight stents had occlusion. The proportional contribution of different factors to stent occlusion was analyzed (Table IV). PTS showed the strongest association with loss of patency (hazard ratio, 5.58; 95% confidence interval, 0.95-32.68; P = .057) although not statistically significant, likely reflecting the limited number of limbs treated for PTS in our cohort (9/98 [9.2%]). Age, extent of stent extension, or stent type did not appear to increase the risk of occlusion.

Table IV.

Hazard ratio for possible factors contributing to occlusion of inserted stents

Discrete variables	Occluded, frequency	Hazard ratio	P value
Operation side
Left	7
Right	1	1.53	.562
Nature of disease
NIVL	5
PTS	2	4.17	.089
DVT	1	2.10	.499
Stent extended to CFV
No	6
Yes	2	1.07	.935
Stent extended to IVC
No	8
Yes	0	0	0
Gender
Female	4
Male	4	1.34	.687
Concomitant superficial venous treatment
No	8
Yes	0	0	0
Previous superficial venous treatment
No	5
Yes	3	0.58	.510
Stent type
Becton Dickinson Venovo	3
Optimed Sinus-Venous	3	1.65	.543
Optimed Sinus-Obliquus	0	0	0
Medtronic Abre	1	2.40	.351
Others; nondedicated venous stents	1	2.12	.389

Continuous variables
Age (with increasing age)		1.01	.756
BMI		0.99	.871

Multivariate analysis
Nature of disease	NIVL
	PTS	5.58	.057
	DVT	2.74	.388
Sex (for MVA)	F
	M	1.27	.749
Limb	Left
	Right	1.6	.528
Age (for MVA)	Mean (SD)	1.02	.517

BMI, Body mass index; CFV, common femoral vein; DVT, deep vein thrombosis; IVC, inferior vena cava; MVA, multivariate analysis; NIVL, nonthrombotic iliac vein lesion; PTS, post-thrombotic syndrome; SD, standard deviation.

Clinical outcome

Target limb DVT developed in eight limbs (8.2%), and all were associated with ipsilateral stent occlusion. Contralateral limb DVT developed in two limbs (2.0%): one at 1 year post intervention in a patient treated for PTS, and the other at 3 years post intervention in a patient treated for NIVL. In both instances the stent had extended beyond the inferior vena cava bifurcation, resulting in bifurcation occlusion. There were no cases of limb amputation throughout follow-up.

At 5 years, 82.7% of limbs treated achieved freedom from ulcer, 91.8% pain relief, and 53.1% sustained swelling relief. Of 98 limbs, 29 had an active venous ulcer prestenting, of which 15 (51.7%) had healed completely over the course of follow-up. Ten patients (13.2%) died during follow-up; all deaths were unrelated to venous disease or stenting, and none were attributed to DVT. All long-term post-stenting outcomes are summarized in Table V.

Table V.

Outcomes

Stent-related complicationsa
Fracture	1 (1.3%)
Migration	1 (1.3%)
In-stent stenosis (not requiring reintervention)	10 (13.1%)
Reintervention (for stenosis, occlusion or fracture)	11 (14.5%)
Limb-related complicationsb
Target limb DVT/occlusion	8 (8.2%)
Contralateral limb DVT	2 (2.0%)
Major amputation	0 (0%)
5-year outcomesb
Freedom from ulcer	Yes = 81 (82.7%) No = 17 (17.3%)
Residual limb swelling	Yes = 46 (46.9%) No = 52 (53.1%)
Relief of pain	Yes = 90 (91.8%) No = 8 (8.2%)
Mortalitya	10 (13.2%) All unrelated to venous disease

DVT, Deep vein thrombosis.

^aPer patient.

^bPer limb treated.

Discussion

Our single-center series, using predominantly dedicated venous stents in NIVL patients, demonstrates satisfactory long-term clinical and anatomical outcomes after venous stenting for symptomatic iliofemoral venous obstruction. Stent-related complications were uncommon and ulcer healing rates were high on follow-up. The primary patency at 5 years was high at 92.0%, 77.8%, and 62.5% for the NIVL, PTS, and post-DVT stenting cohorts, respectively. A recent meta-analysis looked at 17 studies evaluating 2218 patients who had deep venous stenting using dedicated venous stents.⁶ At 36 months, the patency rates were 59%, 80%, and 86% for primary, primary assisted, and secondary patency, respectively. Our results are also comparable and improved when compared with other series,^2,4 which similarly show that stenting of the venous outflow obstruction of lower extremities can be performed with low risk, high long-term patency rates, low in-stent restenosis rates, and low re-intervention rates. Neglén et al³ also reported long-term outcomes in a group of 982 patients. The 72-month primary patency, primary-assisted patency, and secondary patency rates were 79%, 100%, and 100% for NIVL, respectively, and 57%, 80%, and 86% for PTS, respectively. Of note, cumulative rates of complete relief of pain and swelling were 62% and 32%, respectively, and ulcer healing was 58% at 5 years. Similarly, patients in our series had high rates of clinical success. Among those who underwent stenting in the context of preexisting or a history of venous ulceration (C5 or C6 disease) (n = 48), 65% (31/48) achieved ulcer-free status, indicating that the intervention provided durable benefit. Occlusion rates (8.2%) and reintervention rates (17.1%) were low, and stent-related malfunctions were rare with separate isolated cases of stent fracture and migration at 5 years, suggestive of improvement in procedural techniques and stent technologies. Despite our study cohort predominantly comprising NIVL patients, patency loss was generally lower than expected in all subtypes (NIVL, PTS, and DVT). The lower rates of patency loss observed among Asian patients in our study, relative to reports from Western cohorts, highlights an important area for future research to explore and validate these findings. As a general guide, population-wide incidence estimates for venous thromboembolism in Asia (13.8-19.9 per 100,000 people) are 15% to 20% of the levels recorded in Western countries (approximately 100 per 100,000 people).⁷ Unlike many predominantly female NIVL cohorts,²⁵ our series had more male patients, possibly reflecting the referral pattern at our institution rather than true epidemiologic differences.

Traditional risk factors for deep venous stenting occlusion include stents that cross the inguinal ligament, stents with poor inflow, early designed nondedicated venous stents, and PTS when compared with those treated for NIVL.¹ Analysis of our data showed that extent of stenting (CFV or iliocaval), gender, and stent type were not significant factors. Of note, PTS patients were 5.58 times more likely to have stent occlusion than NIVL patients, although this difference was not statistically significant (P = .057), highlighting the need to ensure heightened early surveillance.

Nonetheless, even when stenting is done for our PTS and DVT cohorts, long-term patency remained favorable. Although our numbers were small for these subgroups, patency loss rarely happened after 12 months. This finding is also similar to what the Erasmus group reported after evaluating long-term outcomes after stenting for DVT and PTS.²⁶ They concluded that reinterventions are typically performed in the first year and are potentially preventable by improving procedural techniques and patient selection. The same team concluded that occlusion is rare after 1 year of primary patency and that it may be possible to discharge selected patients from long-term surveillance. Although we do not advocate discharging our patients from long-term surveillance, there may be role of extending surveillance from annually to every 18 months for selected patients with no pathological findings on imaging, but more data are needed.

Although flow factors (particularly inflow disease) have been reported to account for the largest proportion of stent failures and occlusions,²⁷ hematological factors such as noncompliance with anticoagulation and inadequate dosing were more commonly implicated in our series. The absence of inherited thrombophilia in our cohort is consistent with the purported lower rates of inherited thrombophilias (factor V Leiden and prothrombin G20210A gene mutation) among Southeast Asians as compared with the West.²⁸ These findings suggest that focused interventions to improve patient adherence may have a significant impact on stent patency in this population.

This study has several limitations. As a retrospective, single-center analysis, the results may reflect local practice patterns and patient selection, which may limit generalizability. The majority of our cohort comprised patients with NIVL, and subgroup analyses of PTS and acute DVT were constrained by small numbers. Additionally, the relatively small number of stent occlusion events may have reduced the statistical power of the multivariate analysis. We did not have postintervention VCSS or Villalta scores for this cohort to compare with preintervention values, which limits the ability to quantify symptomatic improvement; these values were reported in our previous study with short-term follow-up of ≤24 months, demonstrating a 6- and 7-point mean reduction in VCSS and Villalta scores, respectively (P < .001) post stenting. Additionally, clinical outcomes in the present cohort were assessed based on clinical examination and patient-reported symptoms. Although these reflect meaningful clinical benefit, we acknowledge that the absence of validated scoring systems and disease-specific quality-of-life instruments represents a limitation of this study. Future prospective studies incorporating validated quality-of-life tools (eg, VEINES-QOL/Sym, CIVIQ-20) will be important to fully capture patient-reported outcomes.

Conclusions

Our study demonstrates that iliofemoral venous stenting achieves excellent 5-year patency and symptom relief in a contemporary Asian cohort, with outcomes comparable with those described in Western series. Stent occlusion was uncommon, typically occurred early, and was most often related to anticoagulation noncompliance rather than technical or flow factors. Importantly, dedicated venous stents were associated with very low complication rates, durable ulcer healing, and significant improvements in pain and swelling. These findings reinforce the role of endovenous stenting as the standard of care for clinically significant iliofemoral venous obstruction. Future research should focus on refining patient selection, optimizing inflow assessment, and developing evidence-based follow-up strategies to further enhance long-term outcomes.

Iliofemoral venous stenting is a safe and effective intervention with durable 5-year patency, low reintervention rates, and substantial clinical benefit. Most failures occur early and are potentially preventable with optimal technique, IVUS-guided stent placement, and strict adherence to anticoagulation. These results support the continued evolution of venous stenting as a cornerstone in the management of chronic iliofemoral venous obstruction.

The authors thank our previous colleagues Dr Sriram Narayanan, Dr Joseph Lo, and Dr Ivan Huang for their role in contributing patients to this series. Their efforts were helpful to the completion of this work.

Author contributions

Conception and design: YL, PP, EY

Analysis and interpretation: YL, PP, QH, MM, LQ, UP, GL, JK, YW, GC, GT, LC, ZL, EY

Data collection: YL, PP

Writing the article: YL, PP

Critical revision of the article: YL, PP, QH, MM, LQ, UP, GL, JK, YW, GC, GT, LC, ZL, EY

Final approval of the article: YL, PP, QH, MM, LQ, UP, GL, JK, YW, GC, GT, LC, ZL, EY

Statistical analysis: YL, EY

Obtained funding: Not applicable

Overall responsibility: EY

Funding

None.

Disclosures

None.

Appendix (online only)

Supplementary Fig (online only)

Kaplan-Meier analysis of (A) primary, (B) primary-assisted patency, and (C) secondary patency stratified by etiology. Differences were assessed using the log-rank χ² test. DVT, deep vein thrombosis; PTS, post-thrombotic syndrome; NIVL, nonthrombotic iliac vein lesion.

Supplementary Table (online only)

Etiologies contributing to occlusion of stents (n = 8)