Socioeconomic factors predict successful supervised exercise therapy completion.

OBJECTIVE: Supervised exercise therapy (SET) for patients with intermittent claudication (IC) can lower the risk of progression to chronic limb-threatening ischemia and amputation, while preserving and restoring functional status. Despite supporting evidence, it remains underutilized, and among those who initiate programs, attrition rates are extremely high. We hypothesize that socioeconomic factors may represent significant barriers to SET completion. METHODS: Patients with IC referred to SET at a multi-hospital, single-institution health care system (2018-2022) from a prospectively maintained database were retrospectively analyzed. Our primary endpoint was SET program completion and graduation, defined as completion of 36 sessions. Our secondary endpoints were vascular intervention within 1 year of referral and change in ankle-brachial index (ABI). Baseline demographics were assessed using standard statistical methods. Predictors of SET graduation were analyzed using multivariable logistic regression generating adjusted odds ratios (aORs) with 95% confidence intervals (CIs). Change in ABI was analyzed using t-test between subgroups. Reasons for attrition were tabulated. Patient Health Questionnaire-9 (PHQ-9), metabolic equivalent level, Vascular QOL, Duke Activity Status, and ABI were analyzed using paired t-tests across the entire cohort. RESULTS: Fifty-two patients met inclusion criteria: mean age 67.85 ± 10.69 years, 19 females (36.54%), mean baseline ABI of 0.77 ± 0.16. The co-pays for 100% of patients were fully covered by primary and secondary insurance plans. Twenty-one patients (40.38%) completed SET. On multivariable analysis, residence in a ZIP code with median household income <$47,000 (aOR, 0.10; 95% CI, 0.01-0.76; P = .03) and higher body mass index (aOR, 0.81; 95% CI, 0.67-0.99; P = .04) were significant barriers to SET graduation. There were no differences in ABI change or vascular intervention within 1 year between graduates and non-graduates. Non-graduates reported transportation challenges (25.00%), lack of motivation (20.83%), and illness/functional limitation (20.83%) as primary reasons for SET attrition. Metabolic Equivalent Level (P ≤ .01) and Duke Activity Status scores (P = .04) were significantly greater after participating in a SET program. CONCLUSIONS: Although SET participation improves lower extremity and functionality outcomes, only 40% of referred patients completed therapy in our cohort. Our findings suggest that both socioeconomic and functional factors influence the odds of completing SET programs, indicating a need for holistic pre-referral assessment to facilitate enhanced program accessibility for these populations.

Racial Disparities in Treatment Indications and Outcomes for Limb Ischemia.

BACKGROUND: Studies suggest that the Affordable Care Act (ACA) of 2014 has improved access to vascular care and vascular outcomes among patients suffering from peripheral arterial disease (PAD). We sought to examine the racial disparities that exist in patients with PAD who have undergone lower extremity bypass (LEB) or a peripheral vascular intervention (PVI) using the Vascular Quality initiative (VQI) database. METHODS: The VQI infrainguinal and PVI datasets were queried for patients receiving elective and urgent LEB or PVI between 2016, 2 years after ACA implementation, and in 2021. Patients undergoing interventions urgently/emergently or for aneurysm were excluded. The primary outcome was major adverse limb event (MALE-defined as any vascular reintervention or above-ankle amputation) free survival at 1 year. Standard statistical methods were utilized as appropriate. RESULTS: A total of 17,455 LEB and 87,475 PVIs were included in this analysis. Black persons present at a younger age when compared to non-Hispanic White persons (NHW) and are more likely to have diabetes, hypertension, end-stage renal disease (ESRD), and higher rates of prior amputation. Black persons are more likely to present with chronic limb-threatening ischemia (CLTI) rather than claudication, and in a more urgent setting. Postoperative outcomes show higher rates of major amputations among racial minorities, specifically Black persons for both LEB (1.8% vs. 0.8% P < 0.001) and PVI (20.8% vs. 16.8% P < 0.001). Black persons are at higher risk of 1-year MALE for LEB (36.7% vs. 29.9% P < 0.001) and PVI (31.0% vs. 21.7%; P < 0.001). Even after adjusting for confounding variables, Black persons have a higher risk of 1-year MALE for LEB, with an adjusted hazard ratio (aHR) of 1.15 (95% CI [1.05-1.26], P = 0.003) and PVI (aHR 1.18 95% CI [1.12-1.24], P < 0.001). Other major determinates of 1-year MALE on multivariate Cox regression included CLTI (LEB aHR 1.57 95% CI [1.43-1.72], P < 0.001; PVI aHR 2.29 95% CI [2.20-2.39], P < 0.001) and history of prior amputation (LEB aHR 1.35 95% CI [1.17-1.56], P < 0.001; PVI aHR 1.5 95% CI [1.4-1.6], P < 0.001). CONCLUSIONS: Compared to NHW persons, Black persons present with more advanced vascular disease regardless of the operative indication. Black persons are also at significantly higher risk of 1-year MALE. Despite some advances in more accessible care through the ACA of 2014, our observations suggest that Black persons still have significantly worse outcomes due to variety of variables that need further investigation.

Carotid endarterectomy remains cost-effective for the surgical management of carotid stenosis.

OBJECTIVE: Transcarotid arterial revascularization (TCAR) has gained popularity as an alternative to carotid endarterectomy (CEA) and transfemoral carotid artery stenting (TFCAS), potentially combining the benefits of a minimally invasive approach with a lower risk of procedural stroke compared with TFCAS. Emerging evidence shows TCAR to have excellent perioperative outcomes. However, the cost-effectiveness of TCAR is not well-understood. METHODS: Incorporating data from Carotid Revascularization Endarterectomy vs Stenting Trial (CREST), the Vascular Quality Initiative Surveillance Project, and local cost data, we compared the cost-effectiveness of these three treatment modalities (TFCAS, CEA, and TCAR) for both symptomatic and asymptomatic carotid stenosis using a Markov state-transition model to quantify lifetime costs in United States dollars and effectiveness in quality-adjusted life-years (QALYs). We accounted for perioperative stroke and myocardial infarction, as well as long-term risks of stroke and restenosis. Based on CREST, we assumed a start age of 69 years and a cost-effectiveness acceptability threshold of $100,000/QALYs gained. Sensitivity analyses were performed. RESULTS: In the base-case scenario, TCAR cost $160,642/QALY gained compared with CEA, greater than the frequently cited $100,000/QALY gained threshold. TFCAS was more expensive and less effective than other strategies, largely due to a greater periprocedural stroke risk. In one-way sensitivity analysis, if TCAR stroke risk was <0.9% (base-case risk, 1.4%), than it was economically favorable compared with CEA at its current procedural cost. Alternatively, if TCAR procedural costs were reduced by approximately $2000 (base-case cost, $15,182), it would also become economically favorable. In a probabilistic sensitivity analysis, varying all parameters simultaneously over distributions, CEA was favored in 80% of model iterations at $100,000/QALY, with TCAR favored in 19%. CONCLUSIONS: At current cost and outcomes, TCAR does not meet a traditional cost-effectiveness threshold to replace CEA as the primary treatment modality for carotid stenosis. TFCAS is the least cost-effective strategy for carotid revascularization. Given these observations, TCAR should be limited to select patients, specifically those at high physiologic and anatomic risk from CEA. However, TCAR can become cost-effective if its cost is reduced. Given the current outcomes and cost, CEA remains the most cost-effective treatment for carotid revascularization.

Female Gender is a Predictor of Lower Iliac Vein Stenting Patency Rates.

BACKGROUND: Iliac venous stenting (IVS) for thrombotic and nonthrombotic venous disease is increasingly used as evidence of the safety, efficacy and durability of these interventions increases. Female gender has been implicated as a predictor of failure in arterial endovascular interventions. We hypothesize that female gender could be predictive of patency rates of iliac vein stenting. METHODS: Consecutive patients who underwent IVS for thrombotic or nonthrombotic venous disease at our institution from 2007 until 2019 were identified and divided into groups based on gender. Operative notes, venograms, and the electronic health record were then queried to obtain operative details, co-morbid conditions, postoperative outcomes and stent patency. Study outcome was long term patency rate. The data was analyzed using chi-square, logistic regression, and Kaplan-Meier analysis as appropriate. RESULTS: A total of 200 consecutive patients (231 limbs) were identified in our retrospective analysis, with a mean age of 48.8 ± 17.3, and BMI of 31.6 ± 8.6. Of those, 119 (59.5%) patients, (131 [56.8%] limbs) were female. Comparisons between the gender groups revealed no difference in age, BMI, or preoperative comorbidities. There was no difference in type of venous disease between male (85% thrombotic, 15% nonthrombotic) and female (84% thrombotic, 16% nonthrombotic), P= 0.830. The male cohort was more likely to present with leg ulceration (17% vs. 4.6%, P = 0.002), and the female cohort was more likely to present with leg edema (98.5% vs. 93.0%, P= 0.03). The male cohort had a higher rate of caval (48% vs. 33.6%, P= 0.027) and infrainguinal stent extension. (11% vs. 6.9%, P= 0.02). Females had a higher rate of left sided stenting (80.9% vs. 66/0%, P= 0.010). There was no difference in the median stent diameter used between the cohorts. Primary patency at 5 years was significantly higher for the male cohort (94.1% vs. 74.4%, P= 0.01) On adjusted multivariable cox regression female gender was a predictor of loss of primary patency within 5 years (HR, 4.04; P= 0.007). CONCLUSIONS: In this single center retrospective analysis of IVS, male patients were found to have better primary stent patency compared to female.

Contemporary management of acute and chronic mesenteric ischemia: 10-year experience from a multihospital healthcare system.

OBJECTIVE: Endovascular and hybrid methods have been increasingly used to treat mesenteric ischemia. However, the long-term outcomes and risk of symptom recurrence remain unknown. The objective of the present study was to define the predictors of postoperative morbidity, mortality, and patency loss for acute mesenteric ischemia (AMI) and chronic mesenteric ischemia (CMI). METHODS: The inpatient and follow-up records for all patients who had undergone revascularization for AMI and CMI from 2010 to 2020 at a multicenter hospital system were reviewed. Patency and mortality were evaluated with Cox regression, visualized with Kaplan-Meier curves, and compared using log-rank testing. Patency was further evaluated using Fine-Gray regression with death as a competing risk. The postoperative major adverse events (MAE) and 30-day mortality were evaluated with logistic regression. RESULTS: A total of 407 patients were included, 148 with AMI and 259 with CMI. For the AMI group, the 30-day mortality was 31%. Open surgery was associated with lower rates of bowel resection (odds ratio [OR], 0.23; 95% confidence interval [CI], 0.13-0.61). The etiology of AMI also did not change the outcomes (OR, 1.30; 95% CI, 0.77-2.19). Adjusted analyses indicated that a history of diabetes (OR, 2.77; 95% CI, 1.37-5.61) and sepsis on presentation (OR, 2.32; 95% CI, 1.18-4.58) were independently associated with an increased risk of 30-day MAE. In the CMI group, open surgery and chronic kidney disease were associated with a higher incidence of MAE (OR, 3.03; 95% CI, 1.14-8.05; OR, 2.37; 95% CI, 1.31-4.31). In contrast, chronic kidney disease (OR, 3.02; 95% CI, 1.10-8.37) and inpatient status before revascularization (OR, 2.78; 95% CI, 1.01-7.61) were associated with increased 30-day mortality. For the CMI group, the endovascular cohort had experienced greater rates of symptom recurrence (29% vs 13%) with a faster onset (endovascular, 64 days; vs bypass, 338 days). CONCLUSIONS: AMI remains a morbid disease despite the evolving revascularization techniques. An open approach should remain the reference standard because it reduces the likelihood of bowel resection. For CMI, endovascular interventions have improved the postoperative morbidity but have also resulted in early symptom recurrence and reintervention. An endovascular-first approach should be the standard of care for CMI with close surveillance.

Cost-effectiveness analysis of immediate access arteriovenous grafts versus standard grafts for hemodialysis.

OBJECTIVE: Immediate-access arteriovenous grafts (IAAVGs), or early cannulation arteriovenous grafts (AVGs), are more expensive than standard grafts (sAVGs) but can be used immediately after placement, reducing the need for a tunneled dialysis catheter (TDC). We hypothesized that a decrease in TDC-related complications would make IAAVGs a cost-effective alternative to sAVGs. METHODS: We constructed a Markov state-transition model in which patients initially received either an IAAVG or an sAVG and a TDC until graft usability; patients were followed through multiple subsequent access procedures for a 60-month time horizon. The model simulated mortality and typical graft- and TDC-related complications, with parameter estimates including probabilities, costs, and utilities derived from previous literature. A key parameter was median time to TDC removal after graft placement, which was studied under both real-world (7 days for IAAVG and 70 days for sAVG) and ideal (no TDC placed with IAAVG and 1 month for sAVG) conditions. Costs were based on current Medicare reimbursement rates and reflect a payer perspective. Both microsimulation (10,000 trials) and probabilistic sensitivity analysis (10,000 samples) were performed. The willingness-to-pay threshold was set at $100,000 per quality-adjusted life-year (QALY). RESULTS: IAAVG placement is a dominant strategy under both real-world ($1201.16 less expensive and 0.03 QALY more effective) and ideal ($1457.97 less expensive and 0.03 QALY more effective) conditions. Under real-world parameters, the result was most sensitive to the time to TDC removal; IAAVGs are cost-effective if a TDC is maintained for ≥23 days after sAVG placement. The mean catheter time was lower with IAAVG (3.9 vs 8.7 months; P < .0001), as was the mean number of access-related infections (0.55 vs 0.74; P < .0001). Median survival in the model was 29 months. Overall mortality was similar between groups (76.3% vs 76.7% at 5 years; P = .33), but access-related mortality trended toward improvement with IAAVG (6.1% vs 6.8% at 5 years; P = .052). CONCLUSIONS: The Markov decision analysis model supported our hypothesis that IAAVGs come with added initial cost but are ultimately cost-saving and more effective. This apparent benefit is due to our prediction that a decreased number of catheter days per patient would lead to a decreased number of access-related infections.

Single- versus multiple-stage catheter-directed thrombolysis for acute iliofemoral deep venous thrombosis does not have an impact on iliac vein stent length or patency rates.

BACKGROUND: Incomplete venous thrombolysis and residual nonstented iliac vein disease are known predictors of recurrent deep venous thrombosis (DVT). Controversy exists as to whether the number of thrombolysis sessions affects total stent treatment length or stent patency. The goal of this study was to evaluate the outcomes of patients who underwent single vs multiple catheter-directed lysis sessions with regard to stent extent and patency. METHODS: Consecutive patients who underwent thrombolysis and stenting for acute iliofemoral DVT between 2007 and 2018 were identified and divided into two groups on the basis of the number of treatments performed (one vs multiple sessions). Operative notes and venograms were reviewed to determine the number of lytic sessions performed and stent information, including size, location, total number, and length treated. End points included total stent length, 30-day and long-term patency, and post-thrombotic syndrome (Villalta score ≥5). The χ2 comparisons, logistic regression, and survival analysis were used to determine outcomes. RESULTS: There were 79 patients who underwent lysis and stenting (6 bilateral interventions; mean age, 45.9 ± 17 years; 48 female). Ten patients (12 limbs) underwent single-stage treatment with pharmacomechanical thrombolysis, and the remaining 69 (73 limbs) had two to four operating room sessions combining pharmacomechanical and catheter-directed thrombolysis. Patients who underwent a single-stage procedure were older and more likely to have a malignant disease. These patients received less tissue plasminogen activator compared with the multiple-stage group (17.2 ± 2.2 mg vs 27.6 ± 11.6 mg; P = .008). Average stent length was 8.8 ± 5.2 cm for the single-stage group vs 9.2 ± 4.6 cm for the multiple-stage group (P = .764). Patients who underwent a single-stage procedure had no difference in average length of stay from that of patients who underwent multiple sessions (8.5 days vs 5.9 days; P = .269). The overall 30-day rethrombosis rate was 7.3%. Two-year patency was 72.2% and 74.7% for the single and multiple stages, respectively (P = .909). The major predictors for loss of primary patency were previous DVT (hazard ratio [HR], 5.99; P = .020) and incomplete lysis (HR, 5.39; P = .014) but not number of procedures (HR, 0.957; P = .966). The overall post-thrombotic syndrome rate was 28.4% at 5 years and was also not associated with the number of treatment sessions. CONCLUSIONS: Single- vs multiple-stage thrombolysis for DVT is not associated with a difference in extent of stent coverage. Patency rates remain high for iliac stenting irrespective of the number of lytic sessions, provided lysis is complete and the diseased segments are appropriately stented.

Persistently low inferior vena cava filter retrieval rates in a population-based cohort.

BACKGROUND: Practice patterns associated with inferior vena cava (IVC) filter placement have seen considerable variation in the last decade. We used a statewide administrative database to examine trends in IVC filter placement and retrieval in the general population. METHODS: We reviewed Florida state inpatient and ambulatory surgery databases from 2004 to 2014. International Classification of Diseases, Ninth Revision diagnosis and procedure codes and Current Procedural Terminology codes were searched for patients undergoing inpatient or outpatient IVC filter placement, and each patient was longitudinally tracked to the time of inpatient or outpatient filter retrieval. For inpatient filter placements, associated diagnoses were reviewed to identify indications for placement. Univariate and multivariate logistic regression models were constructed to identify factors associated with improved retrieval rates. RESULTS: During the 11-year period, 131,791 IVC filter placements were identified, with a 50% increase from 2004 to 2010 and a 24% decline from 2010 to 2014. Median age at filter placement was 71 years (interquartile range, 57-81 years). Mean follow-up after filter placement was 17.3 ± 25.5 months. Only 8637 filters (6.6%) were retrieved. The annual retrieval rate trended upward, from 3.4% in 2004 to 8.5% in 2013 (P < .001). Median filter dwell time was 96.5 days (interquartile range, 44-178 days). Diagnoses associated with filter placement included venous thromboembolism (75.9%), trauma (35.0%), hemorrhage (29.9%), malignant disease (29.4%), and stroke (5.1%). Retrieval rates were highest in younger patients (34.0% in patients younger than 20 years) and lowest in Medicare patients (2.5%). In a multivariate logistic regression model, Medicare was associated with decreased retrieval rates (odds ratio, 0.33; 95% confidence interval, 0.31-0.35; P < .001) after adjusting for age and associated diagnoses. Weaker risk factors included increased age, white race, and diagnoses of deep venous thrombosis, pulmonary embolism, and malignant disease. A trauma diagnosis was associated with improved retrieval. To further investigate the Medicare effect, a propensity score-matched model was created to better account for confounding effects. In this model, Medicare persisted as a risk factor for decreased filter retrieval (odds ratio, 0.43; 95% confidence interval, 0.40-0.46; P < .001). CONCLUSIONS: IVC filter placements, after a substantial increase between 2004 and 2010, have been declining since 2010. Retrieval rates in the general population are steadily improving but continue to lag behind those described in center-specific literature. Increased age and Medicare as the primary payer are the strongest risk factors for lack of filter retrieval. Widespread improvements on a national scale are needed to improve the appropriateness of filter placements and to enhance filter retrieval rates.

Implementation of drug-eluting stents for the treatment of femoropopliteal disease provides significant cost-to-system savings in a single-state outpatient simulation.

OBJECTIVES: Initial data on drug-eluting stents (DES) shows that they may increase the durability of endovascular treatment of superficial femoral artery disease compared with traditional bare-metal stents (BMS). Observed decreased target lesion revascularization (TLR) rates have potential for cost savings despite an increased initial cost. The purpose of this study was to run a simulation model of progressive transition from BMS to DES over 5 years evaluating the overall cost impact of that transition. METHODS: Florida State Ambulatory Databases were searched for all patients undergoing superficial femoral artery stenting in 2013 using Current Procedural Terminology codes 37226 and 37227. A simulation model was developed to estimate the impact of a progressive transition from BMS to DES over a 5-year horizon in this patient population. Cost estimates were determined from available cost charge ratio data. For the 5-year model, 2013 served as the initial year with each subsequent year based on the expected number of interventions per year. Up to one TLR per patient was assumed for the model. The 5-year TLR rates for DES and other parameter estimates were based on pooled data from the literature. Institutional data were used to estimate that up to 48% of superficial femoral artery lesions would fit the instructions for use for the Zilver PTX (Cook Medical, Bloomington, Ind), which is currently the only DES approved by the U.S. Food and Drug Administration for peripheral interventions. The net budget impact was expressed as the difference in total costs (primary stenting and reinterventions) for a scenario where BMS is progressively replaced by Zilver PTX compared with a scenario of BMS only. Multiple sensitivity analyses were performed on the base scenario. RESULTS: We identified 4107 peripheral interventions in the first year that fit our study. The overall cost for these procedures in Florida database was $51,362,142.00. In the base case scenario, DES was introduced slowly into the population at a rate of 8% per year up to 48% at the end of the model. This strategy resulted in an overall cost savings of $1,688,953.72 compared with the model with BMS alone. Sensitivity analyses including slower adoption of DES up to only 24% at 5 years, a 20% increase in TLR rates per year for the DES, and a 10% reduction in TLR rates per year for BMS still resulted in a net savings. As long as the additional cost of a DES compared with BMS is less than $677, the DES model remains less expensive. CONCLUSIONS: The adoption of DES in lieu of traditional BMS can lead to significant cost savings in a single state model over a short time horizon.

Survival and long-term cardiovascular outcomes after carotid endarterectomy in patients with chronic renal insufficiency.

BACKGROUND: Multiple studies have evaluated the perioperative outcomes of patients with chronic renal insufficiency (CRI) undergoing carotid endarterectomy (CEA), generally indicating worse survival and cardiovascular (CV) outcomes, although not consistently and with a paucity of long-term data. The present study addresses the perioperative and long-term impact of CRI on CV events and survival after CEA. METHODS: A cohort of consecutive patients treated with CEA between January 1, 2000, and December 31, 2008, was analyzed based on medical records and Social Security Death Index. Estimated glomerular filtration rate (GFR) was assessed at baseline. Renal function was used to divide patients into 3 groups: normal (GFR ≥ 60 mL/min/1.73 m(2)), moderate CRI (GFR, 30-59), and severe CRI (GFR <30). The end points were major coronary events, major cerebrovascular events (any stroke), noncardiac vascular interventions (aortic disease, carotid disease, and critical limb ischemia), and mortality. Subgroup analysis based on the presence of preoperative neurologic symptoms was also performed. Survival analysis and Cox regression models were used to assess the effect of baseline predictors. RESULTS: A total of 1,342 CEAs (mean age, 71.2 ± 9.2 years; 55.6% male; 35.3% symptomatic) were performed during the study period with a mean clinical follow-up of 57 months (median, 55; range, 0-155 months). Eight hundred sixty-eight (65%) patients had normal renal function, 414 (31%) had moderate CRI, and 60 (4%) had severe CRI (24 on dialysis). The combined 30-day stroke/death rates for the symptomatic and asymptomatic groups were 3.2% and 1.4% (normal renal function), 5.7% and 2.6% (moderate CRI), and 14.3% and 10.3% (severe CRI), respectively, with the differences being significant only for the severe-CRI group. At 5 years, the severe-CRI group experienced significantly more coronary events (36.9% vs. 16.3%, P < 0.001), more cerebrovascular events (21.6% vs. 6.3%, P < 0.001), and deaths (70.0% vs. 20.3%, P < 0.001), whereas the moderate-CRI group had no significantly different outcomes compared with the normal group, except for mortality (29.8% vs. 20.3%, P < 0.001). After adjusting for all risk factors, severe CRI remained predictive of coronary events (hazard ratio [HR], 2.21; 95% confidence interval [CI], 1.25-3.90; P = 0.007), cerebrovascular events (HR, 3.11; 95% CI, 1.44-6.74; P = 0.004), and mortality (HR, 4.36; 95% CI, 3.00-6.34; P < 0.001). Symptomatology at baseline was predictive of 5-year mortality (HR, 1.43; 95% CI, 1.14-1.81; P = 0.002). The need for noncardiac vascular interventions was equally distributed among all the groups. CONCLUSIONS: Severe but not moderate CRI is associated with poor perioperative outcomes and is an independent predictor of CV events and death at 5 years after CEA. The decision to perform CEA in symptomatic and asymptomatic patients with severe CRI should be individualized given the poor reported outcomes.

The impact of a systemwide policy for emergent off-hours venous duplex ultrasound studies.

BACKGROUND: We evaluated the impact of an after-hours policy regulating venous duplex ultrasound (VDU) for deep vein thrombosis (DVT) diagnosis on resource utilization and patient care. METHODS: On July 1, 2007, we altered the approach to emergent VDU of patients with potential DVT during off-hours (defined as 5:00 p.m. to 7:00 a.m. weekdays, after 3:30 p.m. Saturdays and Sundays). Instead of 24 hr access, we permitted a venous duplex study in the noninvasive vascular laboratory (NIVL) only after meeting set criteria developed collaboratively across services. In the emergency department (ED), we based all VDU requests on a preset modified Wells score (MWS) as determined by the ED physician. Those patients with MWS 0 or 1 and those above 1 who could receive empiric single-dose low-molecular weight heparin (LMWH) received next-morning imaging unless consultation with a vascular surgeon created an emergent imaging plan. In parallel, inpatient emergent VDU was permitted only after contact with an attending vascular surgeon and where empiric short-term anticoagulation could not occur safely. We tracked NIVL utilization, patient morbidity, sonographer retention, and satisfaction. RESULTS: The number of overall off-hours emergent VDUs decreased from 59 to 19/month after implementation. Testing was deferred in 52 ED patients: 15 stayed in the ED for testing in the morning and 37 were discharged to be tested the following day. Thirty-one of 37 patients returned for testing as outpatient follow-up. Twenty-eight received ED LMWH while awaiting testing. No adverse events were noted with the delay. The mean MWS for ED after-hours studies was 2.9+/-1.6 and that for deferred ER studies was 2.4+/-1.3 (p=0.005). Incidentally, overall off-hours inpatient and ED VDU requests decreased 64% with no clinical adverse events in the first year. The rate of overall positive studies done off-hours increased from 6.7% to 20% (p<0.0001). Sonographer satisfaction was maintained with regulation of call. CONCLUSION: Our collaborative approach allowed off-hour VDU utilization to decrease without any measurable negative care impact.