Current outcomes of interventions to treat mixed arterial-venous ulcers

Abstract

Objective

To analyze outcomes of initial combined arterial and venous interventions (AVI) compared with isolated venous interventions (VI) alone in patients with mixed arterial-venous ulcer (MAVLU).

Methods

Between 2012 and 2022, all patients presenting with MAVLU, defined as having a venous ulcer with a Toe Brachial Index <0.7 or an absolute toe pressure <60 mmHg. Two groups were identified: VI and AVI. The primary outcomes assessed were ulcer healing at 12 weeks and freedom from ulcer recurrence within 1 year.

Results

303 patients (57% female; mean age 59 ± 11 years) presented with active MAVLU; 44% underwent VI, and 56% underwent AVI. The Wound Bed Score was more severe in the AVI group compared to the VI group. In addition to medical and wound care therapies used across the entire cohort, venous ablation with thermal or nonthermal modalities was performed in 93% of cases, sclerotherapy in 61%, stab phlebectomy in 39%, and iliac vein stenting in 4%. In the AVI group, 76% of patients underwent endovascular arterial intervention, while 24% had open bypass procedures. All arterial procedures were successful and resulted in an increased TBI. Ulcer healing rates at 12 weeks were 9% for the VI group and 28% for the AVI group (mean; p = .01). At 1 year, freedom from ulcer recurrence was higher in the AVI group (87 ± 3%) compared to the VI group (64 ± 3%; mean ± SEM p = .01).

Conclusion

Compared to isolated venous interventions, combined arterial and venous interventions accelerate initial ulcer healing and reduce ulcer recurrence in MAVLU.

Keywords

venous ulcers arterial disease interventions outcomes

Introduction

The development of venous ulceration remains a significant public health issue, with up to 1% of the population suffering from a venous leg ulcer (VLU).^1,2 Current treatment shows that healing rates in patients with a VLU and normal arterial perfusion range from 63% at 10 weeks retrospectively to 33% at 12 weeks prospectively.^3–5 Clinical trials have demonstrated that early venous intervention can enhance ulcer healing or prolong the time to recurrence.⁶ However, clinical trials on venous interventions have excluded patients with decreased arterial perfusion to ensure a more homogeneous population. Mixed arterial-venous disease can be present in up to 26% of patients with VLUs.⁷ Evidence of concomitant arterial insufficiency (ankle-brachial index; ABI of <0.8) affects healing metrics for VLUs.⁴ Limbs with a VLU and an ABI of <0.8 can take up to 19 weeks to heal, compared with 9 weeks for VLUs with an ABI of > 0.8.⁴ This may be due to a lack of compression therapy because of the fear of compounding arterial malperfusion or due to the underlying arterial disease. Several authors have shown that a tailored compression regimen in patients with diabetes and arterial disease is effective without resorting to an arterial intervention.^8,9 There is evidence to suggest that in the setting of a venous ulcer with an ABI of <0.5, the arterial disease should be addressed, which generally entails treating iliac and femoral artery disease.¹⁰ There are differing views on whether to adopt a venous-only approach or a staged approach with an arterial intervention followed by a venous intervention for patients with MAVLU.⁷ This study aims to analyze the outcomes of combined arterial and venous interventions compared to venous interventions alone among patients with MAVLU at a single center.

Patients and methods

Study design

This was a retrospective study. Between 2012 and 2022, all patients presenting with a de novo active VLU were surveyed. From this cohort, two groups were identified: those who had initial venous interventions (VI) only and those who had staged arterial and venous interventions (AVI). Arterial interventions were performed before venous interventions. Demographics, comorbidities, prior interventions, pathophysiology, pathology, Interventions, and adverse events were recorded. Consent was waived because patients were not directly studied, and the data collection was retrospective and de-identified. The study was approved by the institutional review board. The study conformed with the STrengthening the Reporting of OBservational studies in Epidemiology (STROBE) guidelines (https://www.strobe-statement.org/).

Setting

Patients were drawn from a single urban tertiary referral, level 1 trauma center in San Antonio associated with two urban wound care clinics.

Selection

Inclusion criteria were all patients with a de novo active VLU, in the presence of superficial truncal venous reflux (great saphenous vein - GSV; small saphenous vein - SSV and perforator - PERF) and without evidence of deep venous reflux on duplex imaging, and a Toe Brachial index (TBI) <0.7 or absolute toe pressure <60 mmHg. Patients with iliac venous obstructive disease and/or superficial reflux were also included. Criteria for superficial venous reflux were defined as retrograde flow duration >500 ms, and criteria for deep venous reflux were defined as retrograde flow duration >1000 ms in the femoral or popliteal segments. Patients with a history of deep venous thrombosis (DVT) were confirmed to have a normal deep venous system. Ulcers with atypical features were also excluded. Additionally, patients with recurrent venous ulcers were also excluded. Only one ulcer on the index leg per patient was included in this review to maintain the independence of observations and avoid bias based on patients presenting with bilateral ulcers.

Ulcer evaluation

The patients were clinically assessed using the Revised Venous Clinical Severity Score (rVCSS), which can be used to measure the disease severity of the venous disease.¹¹ Photographs of the wounds were obtained at the initial visit and at 4-weeks intervals/Prior to taking the digital images, the wound was cleansed with 0.9% saline solution. The lower limb with the ulcer was placed against a white background for the photograph. A ruler in cm was placed close to the ulcer and used as the reference object. The photograph was taken with a standard camera that had automatic white balance, high sharpness, and flash disabled. The camera was kept perpendicular to the wound (90°) and 50 cm away, moving closer or farther as necessary. Wound area measurements were obtained from the digital photographs using NIH ImageJ 1.53i software.¹² All wound tracings were subsequently verified by a blinded wound expert at the end of data collection. Based on these images, the ulcer was assessed by the Wound Bed Score¹³ and the Margolis Score.¹⁴ The Wound Bed Score (WBS) is based on nine criteria: Black Eschar (% wound area), Eczema/Dermatitis, Depth. Granulation tissue, Scarring, Color of wound bed (pink/red), Edema/Swelling, Resurfacing Epithelium, Exudate Amount. Each wound can have a maximum score of 16 (the best score), to a minimum score of 0 (the worst score) (see e-Appendix).¹³ The Margolis Score is based on two criteria: Each wound was assessed by Ulcer Area (cm²) and Ulcer Duration (weeks), and each variable was scored from 0 to 2 and summed together, with the larger number reflecting a harder-to-heal wound (see e-Appendix).¹⁴

Management

Diagnostic imaging

All patients presenting with a de novo venous ulcer underwent venous and arterial imaging. To determine the presence of venous disease, all patients underwent superficial and deep venous duplex ultrasound scanning to evaluate the limb for reflux. If there were abnormal findings in the common femoral vein or significant above-knee swelling, a concomitant iliac vein duplex was performed. To evaluate for arterial disease, all patients underwent an initial noninvasive evaluation on presentation, including ankle-brachial index (ABI) and/or toe-brachial index (TBI). In the presence of a low ABI or TBI, a vascular non-invasive or contrast study (duplex scan, arteriogram, or CT angiography) was obtained to define the anatomy.

Wound algorithm

Compression of the VLU was achieved with an Unna boot in accordance with current societal guidelines.^1,15 Compression protocols were not altered after arterial intervention. Standard and PCR-based wound culture data were obtained in most cases, and appropriate antimicrobial therapy was used only in the presence of a positive culture. Skin substitutes were used after 4 weeks of compression, and in cases where the wound was not progressing, and there was an absence of an active infection. A maximum of 10 placements was performed on each wound over a 6-month period.

Venous algorithm

Following identification of superficial venous incompetence, great saphenous and small saphenous ablation was performed where there was >500 ms of reflux and appropriate anatomic criteria. In the presence of residual perforator reflux, perforator ablation was performed where there was a reversal of flow from the deep system to the superficial system of >500 ms, and where there was a diameter of more than 3.5 mm.

Arterial algorithm

Following identification of occlusive lesions in the lower extremity, endovascular or open intervention was performed at the discretion of the individual attending to attain maximal revascularization.

• Endovascular Intervention: Following diagnostic angiography, intervention (angioplasty stenting or atherectomy) on femoral, tibial, and pedal vessels was performed whenever possible at the same setting with the patient receiving systemic heparin to ensure an activated clotting time (ACT) of 250 s, and completion angiography was performed to assess the final technical result. An attempt was made to intervene on as many vessels as possible during the first intervention. Attempts were made to secure two tibial vessel runoff to the foot with particular attention to the vessels associated with the appropriate angiosomes. Both antegrade and pedal access techniques were used during an endovascular intervention.^16–23

• Open Procedures: Common femoral endarterectomy was performed using established techniques and vessels were patched with a bovine pericardial patch. Vein bypass using standard techniques was employed in patients with a suitable target vessel, acceptable venous conduit, and good pedal runoff. Both reversed, in situ, and non-reversed configurations were performed under general anesthesia.^{16–18,21–23} Prosthetic bypass was used where the vein was considered unsuitable in diameter <3.0 mm or length

Follow-up

After discharge, patients attended wound care and podiatric surgery to manage their foot wounds as frequently as necessary to ensure wound healing and renewed ambulation. Patients were followed by the vascular clinic and underwent routine duplex ultrasound follow-ups at 1, 3, and every 6 months following their procedure using criteria previously described.²¹ During follow-up, angiography was only performed if non-invasive studies suggested restenosis/occlusion (positive duplex scan with a drop in toe-brachial index (TBI) of >0.1 and/or if the patient had recurrent symptoms (worsening ischemia and/or poor wound healing); angiography was performed as appropriate.^16,21,24,25

Date variables

The definition of the variables used in the study are shown in the e-appendix.

Definition of outcomes

The primary outcomes were initial VLU healing at 12 weeks and freedom from recurrence of VLU within 1 year. A VLU was defined as a full-thickness defect of skin, most frequently in the ankle region, that fails to heal spontaneously and is sustained by chronic venous disease, based on venous duplex ultrasound testing, Initial healing was defined as complete epithelialization of the VLU, with no scab formation or drainage, sustained for at least two consecutive assessments over 4 week period, Recurrence was defined as the return of the VLU after it was deemed as initially healed. The secondary outcomes were Amputation-free survival (AFS) and Freedom from modified major adverse limb events (MALE). Amputation-free survival is defined as the duration of time a patient remains alive without undergoing a major limb amputation. Modified major adverse limb events is defined as major amputation (amputation above the ankle) or any venous or arterial open or percutaneous reintervention on the index leg.

Statistical analysis

Measured values are reported as percentages or mean ± SD. Non-parametric ANOVA or χ² testing or Fisher exact test was used to analyze individual variables and significance compared to the VI group (Control group). Rates of initial Ulcer healing and Ulcer recurrence, Amputation-free survival, and freedom from modified MALE were calculated using life table analyses and presented as mean ± SEM. The log-rank test was used to determine differences between cohorts on life tables, with significance compared to the VI group. Univariate and multivariate analyses were performed to define factors impacting categorical outcomes. Univariate and multivariate Cox proportional hazard analysis was performed to determine factors affecting time-dependent outcomes. The models included all variables with a p = .05 in covariate analysis. A backward elimination procedure using a p = .2 was used to reduce each model. The results are expressed as an adjusted hazard ratio (HR) or an odds ratio (OR) as appropriate, with a corresponding set of 95th percent confidence intervals (CI). The VI group was used as the control group in these multivariable analyses. Analyses were performed using JMP software (version 9.0; SAS Institute, Cary, NC) and RStudio (version 1.2.5001; RStudio, Inc., Boston, MA).

Results

Patient population

Three hundred and three patients (57% female, age 59 ± 11 years, mean ± SD) presented with an active VLU and evidence of arterial insufficiency and were classified as MAVLU. These patients represented 27% of all VLUs that presented to the vascular surgery service. Of these 303 patients, 134 (44%) had initial venous interventions (VI), while 169 (56%) had initial concomitant arterial intervention followed by venous intervention (AVI). The majority of patients in both groups were female and Hispanic, reflecting the demographics of the catchment area (Table 1). The groups had no significant differences in either arterial or venous comorbidities (Table 2). There was a significantly increased presentation time in patients who underwent VI compared to those who underwent AVI (8 vs 14 days; Table 1). However, the time to obtain arterial testing was much longer in those who had undergone venous interventions (9 ± 5 vs 28 ± 4 days; mean ± SD; p = .001). An initial larger ulcer size and higher wound bed score correlated with reduced time to presentation and arterial testing (p = .02).

Table 1.

Demographics, venous pathology and ulcer characteristics.

Demographics	VI	AVI	p-value
Patients	134	169
Gender (Female)	82 (61%)	96 (57%)	.48
Caucasian	28 (21%)	32 (19%)	.97
African-American	1 (%)	2 (1%)
Hispanic	103 (77%)	134 (79%)
Other	1 (1%)	2 (1%)
Age
Average age (mean ± SD, years)	67 ± 9	65 ± 9	.06
Time frames (weeks)
Average time ulcer present before treatment	7 ± 2	5 ± 3	.001
Average time ulcer present before arterial testing	14 ± 3	8 ± 3	.001
Ulcer size (sq cm)
<5	60 (45%)	39 (23%)	.001
5–20	46 (34%)	61 (36%)
>20	28 (21%)	69 (41%)
Ulcer position
Medial	72 (54%)	76 (45%)	.24
Lateral	35 (26%)	42 (25%)
Circumference	27 (20%)	51 (30%)
Prior venous interventions
No intervention	118 (88%)	147 (87%)	.99
Thrombolysis/thrombectomy	3 (2%)	4 (2%)
Venous stenting	4 (3%)	6 (3%)
Phlebectomy	1 (1%)	2 (1%)
Endo-venous ablation	4 (3%)	6 (3%)
Sclerotherapy	3 (2%)	4 (2%)
Pathophysiology
Superficial system reflux alone	119 (89%)	139 (82%)	.09
Superficial and perforator reflux	15 (11%)	30 (18%)	.09
Venous scores (mean ± SD)
Revised venous clinical severity score (rVCSS)	14 ± 5	18 ± 5	.001
Wound bed score	16 ± 3	10 ± 5	.001
Margolis score	1.5 ± 0.1	1.9 ± 0.1	.001

Values are n (%) unless specified.

Table 2.

Comorbidities.

Co-morbidities	VI	AVI	p-value
Nicotine abuse	46 (34%)	59 (35%)	.86
Current nicotine abuse	17 (13%)	24 (14%)	.80
Coronary artery disease	43 (32%)	64 (38%)	.28
Congestive heart failure	60 (45%)	79 (47%)	.73
Hypertension	118 (88%)	145 (86%)	.61
Diabetes	122 (91%)	159 (94%)	.32
HB_Aic >7%	118 (88%)	150 (89%)	.79
Hyperlipidemia	119 (89%)	150 (89%)	.99
Venous-thromboembolism	11 (8%)	14 (8%)	.99
Current anticoagulation	101 (75%)	132 (78%)	.54
Metabolic syndrome	111 (83%)	140 (83%)	.99
Obesity	87 (65%)	120 (71%)	.26
Chronic renal insufficiency	12 (9%)	19 (11%)	.57
On dialysis	3 (2%)	5 (3%)	.59
Hypothyroidism	36 (27%)	47 (28%)	.84
Charlson comorbidity index
Charlson comorbidity index (mean ± SD)	6.6 ± 3.4	6.8 ± 3.9	0.63
Charlson comorbidity >4	105 (78%)	144 (85%)	0.12
Modified frailty index
Modified frailty index (mean ± SD)	0.46 ± 0.24	0.47 ± 0.22	.71
Considered to be frail	60 (45%)	74 (44%)	.86
Pre-operative ambulatory status
Ambulatory	43 (32%)	66 (39%)	.09
Ambulatory/homebound	60 (45%)	71 (42%)
Non-ambulatory/transfer	24 (18%)	17 (10%)
Non-ambulatory/bedridden	7 (5%)	15 (9%)

Values are n (%) unless specified.

Anatomic and hemodynamic presentation

The ulcerations in the AVI group were noted to be significantly larger than those in the VI group, but their locations on the leg were equivalent. Among the 303 patients, 100% had GSV incompetence, 33% had SSV incompetence, and 11% had perforator incompetence. Of the patients undergoing arterial intervention, 9% had significant iliac disease, 44% had superficial femoral arterial disease, 23% had tibial arterial disease, and 24% had multi-level disease. There were no differences in venous pathology between the groups, and prior venous or arterial interventions were equivalent (Table 1 and Appendix Table 1). The rVCSS, Wound Bed Score, and Margolis score were higher in the AVI group compared to the VI group (Table 1). Absolute toe pressures were significantly lower in the AVI group compared to the VI group on initial presentation (Appendix Table 1). When ABIs of the index limb were stratified across the entire cohort, 34% had an ABI <0.5, and 66% had an ABI of 0.5–0.8. In the AVI group, 47% had an ABI of <0.5 and 53% had an ABI of 0.5–0.8, and in the VI group, 18% had an ABI of <0.5 and 82% had an ABI of 0.5–0.8 (AVI vs VI; P = .001). The distribution by Trans-Atlantic Inter-Society Consensus (TASC)-II for iliac disease and by Global Limb Anatomic Staging System (GLASS) category for the femoro-popliteal and infra-popliteal segments in the AVI group are shown in Appendix Table 1.

Interventions

Management of the MAVLU included arterial intervention in the AVI group and venous ablation of the superficial veins, serial compression, and advanced wound care in both VI and AVI groups. Medical therapy was completed in 97% of the patients and consisted of oral trental, multivitamins, and aspirin (Table 3). There was no difference between the VI and AVI groups. Within the group of 303 patients, 94% underwent consistent compression therapy (Table 3). Wound debridement with chemical agents, with or without additional sharp debridement, was noted in 89% of the patients. Antibiotics were selected and administered based on deep wound cultures in 88% of the patients. Skin substitutes were applied in 30% of cases (Table 3). There was no significant difference between the two groups in the use of these advanced wound care modalities.

Table 3.

Venous and Ulcer interventions.

Medical	VI	AVI	p-value
Medical therapy	131 (98%)	164 (97%)	.58
Compression	129 (96%)	156 (97%)	.14
Local
Debridement	122 (91%)	146 (86%)	.17
Antibiotics	115 (86%)	151 (89%)	.33
Skin substitutes	44 (33%)	47 (28%)	.38
Venous intervention
Vein ablation	121 (91%)	160 (95%)	.20
Sclerotherapy	84 (63%)	101 (60%)	.59
Phlebectomy	55 (41%)	63 (28%)	.50
Iliac vein stenting	4 (3%)	7 (4%)	.64

Values are n (%) unless specified.

Venous interventions were performed in the majority of patients with superficial and perforator incompetence. Venous ablation using either thermal or nonthermal modalities was performed in 93% of the cases, sclerotherapy in 61%, stab phlebectomy in 39%, and iliac vein stenting in 4%. The groups were equivalent regarding venous interventions (Table 3).

Of those undergoing arterial intervention followed by venous interventions, 76% of patients underwent endovascular intervention, and the remainder underwent bypass surgery (all were femoropopliteal bypasses). Endovascular interventions were performed in 76% of the patients: 26% had an iliac intervention, 24% had a femoropopliteal intervention, and 15% had a proximal tibial intervention. Multilevel endovascular interventions were performed in 11% of the patients. Open procedures were performed in 24%: 12% were femoropopliteal bypasses (87% of vein and 13% with prosthetic), while the remainder (15%) were common femoral artery endarterectomies. Hybrid procedures consisted of common femoral artery endarterectomies with ipsilateral iliac stenting in 5%. There was an average 0.20 ± 0.13 mmHg increase in TBI after arterial intervention, with 77% of the patients achieving an increase of >0.1 in TBI at 1 month. Following arterial intervention in the AVI group, 7% had an ABI of <0.5, 11% had an ABI of 0.5–0.8, and 82% had an ABI >0.8. The median time between arterial intervention was 3 weeks (range 2–6 weeks) for endovascular interventions and 9 weeks (range 5–12 weeks for open surgical interventions. Median follow-up was 38 months (15–41 months), with 11% of patients lost to follow-up.

Early outcomes

There were no perioperative deaths in either group. At 90 days, both groups had a 1% 90-days mortality (Table 4). Overall, 90-days morbidity was significantly higher in the AVI group compared to the VI group. Systematic and local complications were not significantly different; target vessel complications (artery or vein) were significantly different between the groups, with more issues occurring in the arterial group (Table 4). Of the arterial Society for Vascular Surgery objective performance goals, 30-days MACE, 30-days modified MALE, and 30-days amputations were equivalent and not significantly different (Table 4). The majority of the endovascular and venous interventions were performed as outpatient or 23-h observation episodes of care. In those requiring admission (observation or inpatient), the length of stay was significantly longer in the AVI group than in the VI group, due to an open surgical bypass. The majority of the patients were discharged home after interventions. Readmissions were similar between the groups (Table 4).

Table 4.

Immediate outcomes.

Outcomes	VI	AVI	p-value
90-days mortality	1 (1%)	2 (1%)	.99
90-days morbidity	11 (8%)	22 (13%)	.16
Systemic	1 (1%)	5 (3%)	.23
Vessel	1 (1%)	15 (9%)	.02
Local	1 (1%)	2 (1%)	.99
30-days objective performance goals
30-days MACE	1 (1%)	2 (1%)	.62
30-days modified MALE	3 (2%)	10 (6%)	.09
30-days major amputations	1 (1%)	2 (1%)	.99
Hospital metrics
Length of stay, median (IQR)	0 (0–1)	2 (0–7)	.001
Readmissions	4 (3%)	8 (5%)	.38
Disposition
Home	70 (52%)	100 (59%)	.34
Rehabilitation facility	34 (25%)	37 (22%)
Skilled nursing facility	23 (17%)	19 (11%)
Hospital	8 (6%)	14 (8%)

Values are n (%) unless specified.

MACE: major adverse cardiovascular events; MALE: major adverse limb events.

Crossover late interventions in VI group

In the VI group, 31% of patients with a primary VLU had an unplanned arterial intervention after their venous intervention during follow-up to facilitate wound healing. The median time to arterial intervention was 11 weeks (range 7–16 weeks). Indications for the crossover were slow healing, prompting arterial studies, and early recurrence or the development of arterial symptoms. These cases were considered major adverse limb events and censored from the initial healing and VLU recurrence analysis of the VI group. By multivariate analysis, a longer time before presentation (p = .04), wound size >20 sq cm (p = .02), increasing wound bed scores (p = .01), a Charlson Comorbidity Index >4 (p = .01) and ABI <0.5 (p = .01) were associated with crossover.

Late outcomes

Initial ulcer healing was 9 ± 2% (mean ± SEM) and 29 ± 2% at 12 weeks (p = .001) and 46 ± 4% and 65 ± 4% at 24 weeks for VI and AVI, respectively (p = .03; Figure 1(A)). By Cox proportional hazard analysis, wound size >20 sq cm (p = .03), wound bed severity (p = .03), Charlson Comorbidity Index >4 (p = .02), presence of frailty (p = .02), increasing wound bed scores (p = .01), and venous intervention alone (p = .01) were associated with poorer initial ulcer healing. At 3 years, freedom from ulcer recurrence was 64 ± 3% (mean ± SEM) and 87 ± 3% at 1 year (p = .01) and 32 ± 5% and 58 ± 5% for VI and AVI, respectively (p = .02; Figure 1(B)). By Cox proportional hazards analysis, a Charlson Comorbidity Index greater than 4 (p = .02), the presence of frailty (p = .02), and the occurrence of an arterial MALE or venous intervention alone (p = .01) were associated with ulcer recurrence. In the AVI group, limb-based patency was 68 ± 4% (mean ± SEM) at 3 years. By Cox proportional hazard analysis, lower limb-based patency was associated with diabetes (p = .04), chronic renal insufficiency (p = .03), Charlson Comorbidity Index >4 (p = .04), and presence of frailty (p = .02). Freedom from modified MALE was 61 ± 5% (mean ± SEM) and 72 ± 5% for VI and AVI, respectively (p = .04; Figure 2(a)). Events that contributed to the loss of freedom from modified MALE were endovascular reintervention, arterial intervention after a venous intervention, and major amputation. In Cox proportional hazard analysis, no factor indicated a need for intervention for MALE. Amputation-free survival was 80 ± 6% and 63 ± 5% for VI and AVI groups at 3 years, respectively (p = .01; Figure 2(a)). By Cox proportional hazard analysis, a Charlson Comorbidity Index greater than 4 (p = .02), the presence of frailty (p = .02), and ulcer recurrence (p = .03) were associated with lower amputation-free survival.

Figure 1.

The life table curves for time to initial healing (A), and freedom from recurrence (B) are shown for the two intervention groups: Venous intervention - VI (•) and arterial and venous intervention - AVI (○). Data are the mean ± standard error of the mean (SEM) and number at risk shown in the table where appropriate. No error bars are displayed if the standard error of the mean is >10%, and the data set terminates if the number at risk is <10. The log-rank test was used to determine differences between groups on life tables.

Figure 2.

The life table curves for freedom from MALE (A), and amputation free survival (B) are shown for the two intervention groups: venous intervention - VI (•) and arterial and venous intervention - AVI (○). Data are the mean ± standard error of the mean (SEM) and number at risk shown in the table where appropriate. No error bars are displayed if the standard error of the mean is >10%, and the data set terminates if the number at risk is <10. The log-rank test was used to determine differences between groups on life tables.

Discussion

In this current single-center retrospective study, approximately 27% of patients presenting with a VLU had evidence of arterial insufficiency as determined by TBI. In those patients with MAVLU, those who underwent VI had similar venous pathophysiology on duplex imaging as the AVI group. There was a significant increase in presentation time before intervention in patients who underwent VI compared to those who underwent AVI. Furthermore, the time to obtain arterial testing was also longer in those who underwent VI. The initial ulcer sizes were larger, and the wound bed scores were more severe in the AVI group than in the VI group. An initial larger ulcer size and higher wound bed scores correlated with reduced time to presentation and earlier arterial testing. The TBI was significantly lower in the AVI group compared to the VI group. Both groups had a similar treatment profile for their venous disease, and those initially treated for arterial disease predominantly had iliac and femoropopliteal interventions. Patients in the AVI group had an accelerated initial wound healing rate compared to the VI group and a lower VLU recurrence rate. However, those undergoing AVI had a higher MALE rate and a lower AFS in the longer term compared to those who underwent VI, which was related to the severity of the underlying arterial disease.

Marin et al.,²⁶ compared clinical presentation and ulcer characteristics between patients presenting with MAVLU and those with a VLU. In their report, the authors did find that the surface areas of the MAVLU and the VLU were similar, but the MAVLU were considered to be deeper and have a greater eschar burden. This would support our data that wound bed scores are worse in MAVLU treated by AVI. Patients with MAVLU tended to present more rapidly to medical attention, usually having a shorter ulcer duration at initial assessment.²⁶ Furthermore, in another study on chronic leg ulcers with arterial insufficiency, the only risk factor associated with slower wound closure was initial wound size on presentation,²⁷ The current study expands on these findings and shows that in a population of patients with a MAVLU, those who undergo an arterial intervention followed by a venous intervention present faster, have more rapid arterial testing, and have larger, more complex wounds compared to those MAVLU that undergo venous interventions alone. It shows that there is a correlation between the size and complexity of the wound and the pattern of clinical interventions performed.

The mainstay of care for a VLU is compression therapy and local wound care. Multiple reports have demonstrated that initial healing of the VLU with intact arterial perfusion can be achieved in approximately 10 weeks. In addition to compression and wound care for VLU, there is sufficient evidence to show that the interruption or ablation of incompetent superficial veins will result in improved healing of VLUs and lead to increased intervals for ulcer recurrence.^28–31 In a scoping review of clinical practice guidelines for venous leg ulcer (VLU) management, all recommend below-knee compression to improve healing outcomes after calculating the ankle-brachial index to rule out significant arterial disease. However, there is a lack of consensus and a lack of clear guidance in relation to the specific ABI range where standard compression therapy can be used and where modified compression can be used.⁸ A recent systematic review suggested that modified compression therapy (MCT) has a role in selected cases of MAVLU in those with moderate PVD (ABI 0.5–0.8)³² There is a consensus that an ankle-brachial index of less than 0.5 or an ankle pressure of <60 mmHg, necessitates revascularization first, and compression therapy is contraindicated prior to that.³³ In the current study, patients were offered standard compression or modified compression based on their presenting TBI during the efforts to obtain initial healing of the VLU in the MAVLU population. After a systematic review, Alagha et al. found that there were no direct comparisons of a strategy of venous interventions first (either modified compression, surgery, or ablation) compared to arterial intervention first in patients with MAVLU.³⁴ In addition, there is little data on the use of therapies stratified by the baseline ABI of 0.5 to 0.85 and ABI <0.5.³⁴

Several reports have demonstrated that arterial revascularization can expedite MAVLU healing and improve quality of life.^10,35 Open surgical revascularization heals about 60% of MAVLU patients, although subsequent venous intervention may be needed in a staged manner.³⁶ Karkos and associates support the use of combined arterial and venous interventions in a small study by demonstrating in their series that there was a higher healing rate with concomitant revascularization and venous intervention.³⁷ Early reports suggested that in patients with MAVLU and moderate arterial disease (ABI 0.5–0.8), revascularization should be reserved for those who failed conservative management with supervised modified compression therapy (ulcers increasing in size, new ulcers developing, and increased pain).^10,33,38 The progress in endovascular techniques and outcomes has shifted this approach to a more aggressive interventional strategy. Lantis JC and colleagues³⁵ performed only percutaneous transluminal angioplasty (PTA) for 27 patients with MAVLU and ABI <0.7, subsequently followed by medical management, including compression therapy initiated 6–10 days after the endovascular intervention. No venous interventions were performed. This revascularization strategy, combined with medical therapy, resulted in an initial healing rate of 75% at 10 weeks, which increased to 100% at 22 weeks after the endovascular intervention. Georgopoulos et al. reported that successful revascularization in MAVLU patients, as demonstrated by normalization of ABI post-operatively, was followed by compression therapy. The arterial intervention combined with the best medical management, including compression, significantly shortened healing time compared to those patients who received compression-only (16.6 ± 2.6 weeks vs 24.7 ± 3.2 weeks, p < .001).¹⁰ In the current study, arterial interventions, when coupled with a venous intervention, did accelerate both initial VLU healing and recurrence. Importantly, those patients in the AVI group experienced higher MALE rates and lower amputation-free survival than those in the VI group. This likely reflects a greater arterial burden in the patients than in those in the VI group as was indicated by the lower hemodynamic parameters at presentation.

This study is the first to report crossover events in the MAVLU patients. Among patients with a primary VLU who underwent VI initially, 31% underwent an unplanned arterial intervention during follow-up to facilitate wound healing. The crossover was associated with a longer time before the presentation of VLU, which may indicate poorer access to healthcare and health insurance or poorer health literacy. Patients also had more comorbidities as determined by Charlson comorbidity scores >4. Furthermore, an initial lower ABI (<0.5) was associated with a crossover event. Upon presentation, the size and severity of the VLU were also important findings consistent with the findings of Marin et al.,²⁶ and led to crossover from VI group to a delayed arterial intervention.

This study has limitations. It is retrospective; the treatment protocols were not driven by a standard protocol but rather by individual attending decision-making, which may have created inherent bias and may not make the two treatment groups fully comparable. It is also from a specific region with a distinct population, and thus, the results may not be generalizable. It is also important to note that both scientific information and technology evolved over the study period.

In conclusion, approximately 27% of patients presenting with active venous leg ulcers (VLUs) demonstrate arterial insufficiency. On presentation, MAVLUs are larger and have more severe wound severity. Initial ulcer healing and freedom from ulcer recurrence are superior in AVI relative to VI. Up to one-third of patients will crossover during follow-up to receive arterial interventions in the VI group. In the presence of impaired arterial perfusion, combined venous and arterial interventions accelerate wound healing and decrease ulcer recurrence in MAVLU.

Supplemental material

Supplemental Material - Current outcomes of interventions to treat mixed arterial-venous ulcers

Supplemental Material for Current outcomes of interventions to treat mixed arterial-venous ulcers by Joseph P. Hart, Mona Li and Mark G. Davies in Phlebology.

Footnotes

Author’s note

Presented at the International Venous Congress (Miami, FL; April 18 - 20, 2024).

ORCID iDs

Joseph P. Hart

Mona Li

Mark G. Davies

Author contributions

Study design: all authors. Obtain data: all authors. Statistical analysis: all authors. Data interpretation: all authors. Manuscript draft: all authors. Critical revision: all authors.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Guarantor

Mark G Davies.

Supplemental material

Supplemental material for this article is available online.

Appendix

Appendix Table 1.

Hemodynamics and GLASS disease grading.

Hemodynamics	VI	AVI	p-value
Toe pressure (mmHg) (mean±SD)	47 ± 8	32 ± 8	.001
Prior arterial intervention
No intervention	129 (96%)	162 (96%)	.99
Aorto-iliac	3 (2%)	3 (2%)
Femoro-popliteal	1 (1%)	2 (1%)
Tibial	1 (1%)	2 (1%)
TASC II iliac grading
TASC II A/B		123 (73%)
TASC II C/D		46 (27%)
Femoro-popliteal (FP) disease grading
FP Grade 0		24 (14%)
FP Grade 1		49 (29%)
FP Grade 2		66 (39%)
FP Grade 3		22 (13%)
FP Grade 4		6 (5%)
Infra-popliteal (IP) disease grading
IP Grade 0		93 (55%)
IP Grade 1		49 (29%)
IP Grade 2		15 (9%)
IP Grade 3		10 (6%)
IP Grade 4		2 (1%)
GLASS staging
Stage 1		123 (72%)
Stage II		35 (21%)
Stage II		11 (7%)

Values are n (%) unless specified.

TASC II: Trans Atlantic Societal Consensus II; GLASS: Global Limb Anatomic Staging System.

References

O’Donnell

Jr. Passman

Marston

, et al. Management of venous leg ulcers: clinical practice guidelines of the Society for Vascular Surgery ® and the American Venous Forum. J Vasc Surg 2014; 60(2 Suppl): 3s–59s. https://doi.org/10.1016/j.jvs.2014.04.049

Gloviczki

Lawrence

Wasan

, et al. The 2023 Society for Vascular Surgery, American Venous Forum, and American Vein and Lymphatic Society clinical practice guidelines for the management of varicose veins of the lower extremities. Part II: endorsed by the society of interventional radiology and the society for vascular medicine. J Vasc Surg - Venous Lymphat Disord 2024; 12(1): 101670. https://doi.org/10.37923/phle.2024.22.s.s1

Nelson

Iglesias

Cullum

, et al. Randomized clinical trial of four-layer and short-stretch compression bandages for venous leg ulcers (VenUS I). Br J Surg 2004; 91(10): 1292–1299. https://doi.org/10.1002/bjs.4754

Marston

Carlin

Passman

, et al. Healing rates and cost efficacy of outpatient compression treatment for leg ulcers associated with venous insufficiency. J Vasc Surg 1999; 30(3): 491–498. https://doi.org/10.1016/s0741-5214(99)70076-5

Cai

Hitchman

Mohamed

, et al. Endovenous ablation for venous leg ulcers. Cochrane Database Syst Rev 2023; 7(7): Cd009494. https://doi.org/10.1002/14651858.cd009494.pub3

Howard

Kothari

, et al. The role of superficial venous surgery in the management of venous ulcers: a systematic review. Eur J Vasc Endovasc Surg 2008; 36(4): 458–465. https://doi.org/10.1016/j.ejvs.2008.06.013

Hedayati

Carson

Chi

, et al. Management of mixed arterial venous lower extremity ulceration: a review. Vasc Med 2015; 20(5): 479–486. https://doi.org/10.1177/1358863X15594683

Weller

Team

Ivory

, et al. ABPI reporting and compression recommendations in global clinical practice guidelines on venous leg ulcer management: a scoping review. Int Wound J 2019; 16(2): 406–419. https://doi.org/10.1111/iwj.13048

Mosti

Cavezzi

Bastiani

, et al. Compression therapy is not contraindicated in diabetic patients with venous or mixed leg ulcer. J Clin Med 2020; 9(11): 3709. https://doi.org/10.3390/jcm9113709

10.

Georgopoulos

Kouvelos

Koutsoumpelis

, et al. The effect of revascularization procedures on healing of mixed arterial and venous leg ulcers. Int Angiol 2013; 32(4): 368–374.

11.

Vasquez

Rabe

McLafferty

, et al. Revision of the venous clinical severity score: venous outcomes consensus statement: special communication of the American Venous Forum ad hoc outcomes working group. J Vasc Surg 2010; 52(5): 1387–1396. https://doi.org/10.1016/j.jvs.2010.06.161

12.

Schneider

Rasband

Eliceiri

. NIH image to ImageJ: 25 years of image analysis. Nat Methods 2012; 9(7): 671–675. https://doi.org/10.1038/nmeth.2089

13.

Falanga

Saap

Ozonoff

. Wound bed score and its correlation with healing of chronic wounds. Dermatol Ther 2006; 19(6): 383–390. https://doi.org/10.1111/j.1529-8019.2006.00096.x

14.

Kantor

Margolis

. A multicentre study of percentage change in venous leg ulcer area as a prognostic index of healing at 24 weeks. Br J Dermatol 2000; 142(5): 960–964. https://doi.org/10.1046/j.1365-2133.2000.03478.x

15.

De Maeseneer

Kakkos

Aherne

, et al. Editor’s choice–European Society for Vascular Surgery (ESVS) 2022 clinical practice guidelines on the management of chronic venous disease of the lower limbs. Eur J Vasc Endovasc Surg 2022; 63(2): 184–267. https://doi.org/10.1016/j.ejvs.2021.12.024

16.

Baer-Bositis

Hicks

Haidar

, et al. Outcomes of re-intervention for recurrent symptomatic disease after tibial endovascular intervention. J Vasc Surg 2018; 68(3): 811–821. https://doi.org/10.1016/j.jvs.2017.11.096

17.

Smolock

Anaya-Ayala

Kaufman

, et al. Current efficacy of open and endovascular interventions for advanced superficial femoral artery occlusive disease. J Vasc Surg 2013; 58(5): 1267–1275. https://doi.org/10.1016/j.jvs.2013.02.252

18.

Davies

Bismuth

Saad

, et al. Outcomes of interventions for recurrent disease after endoluminal intervention for superficial femoral artery disease. J Vasc Surg 2010; 52(2): 331–339. https://doi.org/10.1016/j.jvs.2010.02.278

19.

Cheun

Jayakumar

Ferrer

, et al. Implications of early failure of isolated endovascular tibial interventions. J Vasc Surg 2020; 72(1): 233–240. https://doi.org/10.1016/j.jvs.2019.11.035

20.

Cheun

Jayakumar

Sideman

, et al. Outcomes of isolated inframalleolar interventions for chronic limb-threatening ischemia in diabetic patients. J Vasc Surg 2020; 71(5): 1644–1652. https://doi.org/10.1016/j.jvs.2019.07.094

21.

Cheun

Davies

. Influence of heel ulceration on limb salvage after open and endovascular intervention for limb-threatening critical ischemia. J Vasc Surg 2018; 68(2): e21. https://doi.org/10.1016/j.jvs.2018.05.073

22.

Hart

Davies

. Transitions of frailty after lower extremity interventions for chronic limb-threatening ischemia. J Vasc Surg 2025; 81(3): 730–742.

23.

Davies

Hart

. The value of restaging WIfI after initial vascular and podiatric intervention. Ann Vasc Surg 2024; 100: 262–263. https://doi.org/10.1016/j.avsg.2023.12.035

24.

Smolock

Anaya-Ayala

Bismuth

, et al. Impact of metabolic syndrome on the outcomes of superficial femoral artery interventions. J Vasc Surg 2012; 55(4): 985–993. https://doi.org/10.1016/j.jvs.2011.10.109

25.

Cheun

Hart

Davies

. The value of restaging WIfI (wound, ischemia, and foot infection) after initial vascular and podiatric intervention. Ann Vasc Surg 2025; 111: 319–330. https://doi.org/10.1016/j.avsg.2024.11.005

26.

Marin

Woo

. Clinical characteristics of mixed arteriovenous leg ulcers: a descriptive study. J Wound, Ostomy Cont Nurs 2017; 44(1): 41–47. https://doi.org/10.1097/WON.0000000000000294

27.

Marston

Davies

Armstrong

, et al. Natural history of limbs with arterial insufficiency and chronic ulceration treated without revascularization. J Vasc Surg 2006; 44(1): 108–114. https://doi.org/10.1016/j.jvs.2006.03.026

28.

Barwell

Davies

Deacon

, et al. Comparison of surgery and compression with compression alone in chronic venous ulceration (ESCHAR) study: randomised controlled trial. Lancet 2004; 363(9424): 1854–1859. https://doi.org/10.1016/S0140-6736(04)16353-8

29.

Harlander-Locke

Lawrence

Jimenez

, et al. Combined treatment with compression therapy and ablation of incompetent superficial and perforating veins reduces ulcer recurrence in patients with CEAP 5 venous disease. J Vasc Surg 2012; 55(2): 446–450. https://doi.org/10.1016/j.jvs.2011.08.009

30.

van Gent

Hop

van Praag

, et al. Conservative versus surgical treatment of venous leg ulcers: a prospective, randomized, multicenter trial. J Vasc Surg 2006; 44(3): 563–571. https://doi.org/10.1016/j.jvs.2006.04.053

31.

Gohel

Heatley

Liu

, et al. A randomized trial of early endovenous ablation in venous ulceration. N Engl J Med 2018; 378(22): 2105–2114. https://doi.org/10.1056/nejmoa1801214

32.

Lim

SLX

Chung

Holloway

, et al. Modified compression therapy in mixed arterial-venous leg ulcers: an integrative review. Int Wound J 2021; 18(6): 822–842. https://doi.org/10.1111/iwj.13585

33.

Ghauri

Nyamekye

Grabs

, et al. The diagnosis and management of mixed arterial/venous leg ulcers in community-based clinics. Eur J Vasc Endovasc Surg 1998; 16(4): 350–355. https://doi.org/10.1016/s1078-5884(98)80056-7

34.

Alagha

Lowery

, et al. “Veins first” versus “artery first” approach for management of mixed arterial venous leg ulcers (MAVLU): systematic review and meta-analysis. Phlebology 2024; 40: 2683555241282118.

35.

Lantis

2nd Boone

Lee

, et al. The effect of percutaneous intervention on wound healing in patients with mixed arterial venous disease. Ann Vasc Surg 2011; 25(1): 79–86. https://doi.org/10.1016/j.avsg.2010.09.006

36.

Alagha

Alfatih

Westby

, et al. Review of mixed arterial venous leg ulcers (MAVLU) disease in contemporary practice. Vasc Endovasc Surg 2024; 58(7): 747–751. https://doi.org/10.1177/15385744241264336

37.

Karkos

Holbrook

Makris

, et al. Simultaneous superficial venous surgery and on-table subintimal angioplasty in the treatment of venous leg ulcers with arterial compromise. J Endovasc Ther 2005; 12(6): 739–745. https://doi.org/10.1583/05-1614.1

38.

Humphreys

Stewart

Gohel

, et al. Management of mixed arterial and venous leg ulcers. Br J Surg 2007; 94(9): 1104–1107. https://doi.org/10.1002/bjs.5757

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.64 MB