Author + information
- Received April 6, 2016
- Revision received July 6, 2016
- Accepted August 11, 2016
- Published online November 14, 2016.
- Subhash Banerjee, MDa,b,∗ (, )
- Haekyung Jeon-Slaughter, PhDa,b,
- Shirling Tsai, MDa,b,
- Atif Mohammad, MDa,
- Mazin Foteh, MDc,
- Mazen Abu-Fadel, MDd,
- Osvaldo S. Gigliotti, MDe,
- Ian Cawich, MDf,
- Gerardo Rodriguez, MD, PhDf,
- Dharam Kumbhani, MDa,
- Tayo Addo, MDa,
- Michael Luna, MDa,
- Tony S. Das, MDg,
- Anand Prasad, MDh,
- Ehrin J. Armstrong, MDi,
- Nicolas W. Shammas, MDj and
- Emmanouil S. Brilakis, MD, PhDa,b
- aUniversity of Texas Southwestern Medical Center, Dallas, Texas
- bVA North Texas Health Care System, Dallas, Texas
- cCardiothoracic and Vascular Surgeons, Austin, Texas
- dOklahoma University Health Science Center, Oklahoma City, Oklahoma
- eSeton Medical Center, Austin, Texas
- fArkansas Heart Hospital, Little Rock, Arkansas
- gWalnut Hill Medical Center, Dallas, Texas
- hUniversity of Texas at San Antonio, San Antonio, Texas
- iDenver VA Medical Center, Denver, Colorado
- jMidwest Cardiovascular Research Foundation, Davenport, Iowa
- ↵∗Reprint requests and correspondence:
Dr. Subhash Banerjee, Division of Cardiology, Department of Internal Medicine, University of Texas Southwestern Medical Center and VA North Texas Health Care System, 4500 S. Lancaster Road, 111a, Dallas, Texas 75216.
Objectives The aim of this study was to assess actual procedural costs and outcomes comparing wire-catheter and dedicated chronic total occlusion (CTO) device strategies to cross peripheral artery CTOs.
Background Peripheral artery CTO interventions are frequently performed, but there are limited data on actual procedural costs and outcomes comparing wire-catheter and dedicated CTO devices.
Methods The XLPAD (Excellence in Peripheral Artery Disease Intervention) registry (NCT01904851) was accessed to retrospectively compare cost and 30-day and 12-month outcomes of wire-catheter and crossing device strategies for treatment of infrainguinal peripheral artery CTO.
Results Of all 3,234 treated lesions, 42% (n = 1,362) were CTOs in 1,006 unique patients. Wire-catheter approaches were used in 82% of CTOs, whereas dedicated CTO devices were used in 18% (p < 0.0001). CTO crossing device use was associated with significantly higher technical success (74% vs. 65%; p < 0.0001) and mean procedure cost ($7,800.09 vs. $4,973.24; p < 0.0001). Because 12-month repeat revascularization (11.3% vs. 17.2%; p = 0.02) and amputation rates (2.8% vs. 8.5%; p = 0.002) in the CTO crossing device arm were lower compared with the wire-catheter group, the net cost for an initial CTO crossing device strategy was $423.80 per procedure.
Conclusions An initial wire-catheter approach to cross a peripheral artery CTO is most frequently adopted. The use of dedicated CTO crossing devices provides significantly higher technical success and lower reintervention and amputation rates, at a net cost of $423.80 per procedure at 12 months.
Nearly one-half of all lower extremity percutaneous interventions performed in patients with symptomatic peripheral artery disease (PAD) involve chronic total occlusions (CTOs), most commonly in the superficial femoral artery (1). Interventions for CTO of the superficial femoral artery, popliteal artery, or tibial artery are technically more challenging and are associated with more periprocedural complications and lower rates of procedural success (2,3). Despite these challenges, the ability to successfully treat CTOs is nearly obligatory for an endovascular specialist. CTO interventions can be associated with high procedural cost by virtue of selecting CTO crossing and/or re-entry devices over lower cost guidewires and support catheters. Despite the current awareness of health care cost, there are very limited data on actual procedural costs and outcomes comparing the 2 crossing strategies.
We accessed the XLPAD (Excellence in Peripheral Artery Disease Intervention) registry (NCT01904851) to compare cost and procedural outcomes of wire-catheter and crossing device strategies for treatment of infrainguinal peripheral artery CTO in patients with symptomatic PAD.
The XLPAD registry catalogues procedural information and outcomes of consecutive infrainguinal peripheral artery interventions performed at 11 centers in the United States between January 2005 and October 2015. Details of this registry have been previously published (4). All procedural images are adjudicated by the Veterans Affairs (VA) North Texas angiography and ultrasound core laboratory. Baseline features, demographics, and laboratory and medical therapy data are recorded in the registry. We retrospectively analyzed the cost and procedural outcomes of wire-catheter and crossing device strategies for treatment of infrainguinal peripheral artery CTO.
Cost variables of CTO crossing include the cost of guidewire(s), support catheter(s), and/or specialized CTO crossing and re-entry devices in case of subintimal passage. Micro-costs of devices used to cross each CTO lesion were calculated. Indicators for each device used during crossing were created, and average retail price per device was then multiplied by the number of devices used to derive the total cost of a crossing strategy. When multiple lesions were treated per procedure, the average cost of crossing strategy per lesion was used to calculate procedural cost. Total procedural cost is inclusive of micro-costs of all devices, other listed equipment (balloons, stents, atherectomy) periprocedural medication(s), contrast used during each procedure, and labor cost (Online Table 1 for detailed items included for cost calculation). The cost estimates are based on VA procurement and human resource estimates.
CTO crossing strategies
Crossing strategies were categorized into 2 groups by choice of primary crossing device at the beginning of the procedure, wire-catheter and crossing device groups, and subsequently into 3 groups by secondary device choice: 1) wire-catheter only; 2) initial wire-catheter followed by crossing device use (wire-catheter first plus crossing device); and 3) crossing device. Technical success was defined as crossing the CTO and placement of a guidewire or crossing device in the distal true lumen confirmed by either angiography or intravascular ultrasound. Primary technical success underscores accessing the distal true lumen with the primary crossing strategy, whereas provisional technical success identifies the additional use of a crossing or re-entry device; subintimal re-entry is defined by the use of re-entry device or as reported by the operator (1). Procedural success is defined as successful treatment of the CTO lesion with ≤30% angiographic residual stenosis (Figure 1).
Outcomes and procedural characteristics
Outcomes include binary variables of lesion crossing described as technical success or failure and procedural success or failure. Procedural data include duration of the procedure, fluoroscopy time, and periprocedural complications. We also examined patient adverse events after procedures at 30 days and 12 months: all-cause death, nonfatal myocardial infarction, peripheral artery stent or vessel thrombosis, need for any endovascular or surgical revascularization, or target limb major amputation.
We conducted cost-benefit analysis to examine the clinical benefits of using specialized crossing devices to cross peripheral artery CTO lesions initially, compared with an initial wire-catheter approach. Costs used for this cost-benefit analysis were micro-costs of endovascular intervention procedures and 1-year patency measured by numbers of repeated revascularization procedures performed as a unit of effectiveness for each of the crossing strategies. Unlike other cost-effectiveness methods, cost-benefit analysis actually calculates monetary values of the benefits of both crossing devices and wire catheters and compares them to total costs of both groups using a net cost. A net cost was calculated by subtracting the difference in benefits (Δ benefits) from the difference in costs (Δ costs) between the crossing device and wire-catheter groups (net costs = Δ costs − Δ benefits). Benefits were calculated from the avoidance of opportunity costs associated with repeated endovascular and surgical interventions and amputation within a year of the index endovascular intervention. Opportunity costs for the use of CTO crossing device as a primary crossing strategy compared with wire catheter were calculated as a difference in estimated numbers of interventions needed in 12 months per 100 procedures times the average cost of interventions (repeat endovascular or surgical revascularization and amputation). Estimated numbers of repeat endovascular or surgical revascularization and amputation per procedure were calculated by a logistic model per primary strategy as follows:where bi values are estimates (1).
Descriptive statistics are reported, and Cochran-Mantel-Haenszel statistics are used for overall associations of categorical characteristics and outcomes (success and failure) within groups of primary and secondary crossing strategies. |Standard residual| >3 is set as a criterion for a true association. Analysis of variance is used to test the overall difference in continuous variables, cost of CTO crossing, procedural cost, and duration of procedure between crossing strategies, and Tukey-Kramer adjusted p values are reported for post hoc comparisons among the 3 crossing strategy groups. Skew (<2) and kurtosis (<3) were checked to ascertain the normality of these variables for analysis of variance. Thirteen percent of study samples (12.9%) had missing data on duration of procedure, and 29.8% had missing data on CTO lesion length. To effectively handle missing data, the multiple imputation technique is used to impute missing data of duration of procedure under the missing-at-random assumption. Mean and bootstrap 95% confidence intervals of net cost per group were estimated. The bootstrap method with 1,000 resampling technique was used to construct 95% confidence intervals (5). A p value <0.05 is set for statistical significance.
Of a total 2,429 procedures available from the XLPAD registry, 3,234 lesions were treated with an average of 1.33 and a maximum of 5 lesions per procedure. Multiple lesions (2 or more) were treated in 611 procedures. Of these, 3 lesions were treated in 165 procedures, 4 lesions were treated in 25 procedures, and 5 lesions were treated in 4 procedures. Among the 3,234 lesions, 42% were CTOs (n = 1,362). We therefore selected 1,362 CTO lesions in 1,200 procedures to examine crossing and procedural success and associated actual procedural costs in 1,006 unique patients.
Patient baseline characteristics are described in Table 1. There were no significant group differences in baseline patient or lesion characteristics, except age, Rutherford category, and reference vessel diameters. The wire-catheter group included older patients and those with more severe symptoms and smaller reference vessel diameters compared with the crossing device group. Superficial femoral artery CTO crossing was negatively associated with selecting a wire-catheter strategy initially.
An overwhelming number of operators opted for a primary wire-catheter approach to cross CTO lesions (82% [n = 1,097]) compared with a primary CTO device strategy (18% [n = 248]) (p < 0.0001) (Figure 1).
While the wire-catheter group had a significantly lower technical success rate compared with the crossing device group (65% vs. 74%; p < 0.0001), the need for subintimal re-entry (27% vs. 26%; p = 0.71) and procedural success rates (88% vs. 91%; p = 0.23) (Figure 2) were similar between the wire-catheter and crossing device groups. Within the wire-catheter group, procedural success was higher with the provisional use of a re-entry device (n = 220) compared with a crossing device (n = 168; 89% vs. 76%; p = 0.0004) (Online Table 2).
The cost associated with the entire procedure for the wire-catheter group (mean $4,973.24 ± $3,476.90) was significantly lower than that of the crossing device group (mean $7,800.09 ± $3,809.6) (t = −11.36, p < 0.0001) (Figure 3). Provisional use of crossing or re-entry devices following an initial wire-catheter approach was associated with a mean cost of $5,493.8 ± $3,523.6, significantly lower than the crossing device group (p < 0.0001) (Figure 4).
Figure 5 shows proportions of itemized procedural costs. The proportion of the cost associated with CTO crossing is substantially different between wire-catheter and crossing device groups, respectively (12% vs. 40%; t = −21.61, p < 0.0001).
We found no significant differences in procedure duration, fluoroscopy time, and volume of contrast use between primary wire-catheter and crossing device groups (Table 2). When considering the secondary device choices after failure of primary wire-catheter strategy, the crossing device group had significantly longer procedure duration (mean 157.42 ± 58.69 minutes) than the re-entry device group (mean 136.46 ± 55.25 minutes; t = 3.62, p = 0.0003).
The crossing device group used a significantly greater numbers of balloons (mean 1.60 ± 1.49) and stents (mean 1.33 ± 1.16) during the procedure compared with the wire-catheter group (balloons used: mean 1.44 ± 0.87, t = 2.42, p = 0.0157; stents used: mean 1.07 ± 1.31, t = 2.79, p = 0.0054).
The average differences in technical and procedural success rates between the primary wire-catheter and crossing device groups were 9.16% and 2.62%, respectively. The difference in total cost for 100 procedures was estimated as $282,685 between the crossing device and wire-catheter groups. Differences in total benefits for 100 procedures evaluated at 30 days and 12 months were $4,737.90 and $265,902.47, respectively. Total benefits of each crossing strategy were calculated by multiplying the average cost for each repeat revascularization or amputation procedure by an estimated reduction in the numbers of such procedures (Table 3). Therefore, net costs evaluated at 30 days between the crossing device and wire-catheter groups were $2,122.40 per procedure and $423.80 per procedure at 12 months. Sensitivity analysis of the cost estimation was conducted as well, and results are presented in Online Table 3.
There were no statistically significant differences in rates of death, nonfatal myocardial infarction, or peripheral artery stent or vessel thrombosis within 30 days (Figure 6) or at 12 months (Figure 7) post-procedure between the 2 primary crossing strategies. However, there was a significantly higher rate of amputation in the wire-catheter group at 30 days and a significantly higher rate of endovascular reintervention and amputation at 12 months (Figures 6 and 7). Twelve-month death and amputation rates were significantly higher with provisional use of re-entry device compared with crossing device following an initial wire-catheter approach (death: 1.2% vs. 5.5%, p = 0.026; amputation: 1.8% vs 8.2%, p = 0.006) (Online Table 2).
Subgroup analyses conducted for target vessels above and below the knee were similar (Online Tables 4 and 5).
This analysis presents a comprehensive assessment of cost and clinical outcomes for wire-catheter and crossing device strategies for the treatment of infrainguinal peripheral artery CTO in patients with symptomatic PAD. We conclude that a wire-catheter crossing strategy is initially used in the vast majority of cases. Compared with a dedicated CTO crossing device strategy, which is associated with significantly higher procedural cost, the primary wire-catheter approach provides a lower rate of patency, albeit with a similar overall procedural success rate.
These data from the XLPAD registry provide an in-depth cost estimate associated with infrainguinal peripheral artery CTO interventions. These procedures are frequently performed and are associated with significant direct cost related to the selection of an initial crossing strategy, adjunctive devices, medication use, procedure duration, and catheterization laboratory staff effort. This comprehensive accounting is made possible by the detailed procedure and device data capture nested within the registry. The ability to cross an infrainguinal peripheral artery CTO and access the true lumen is significantly higher with dedicated CTO crossing devices compared with a wire-catheter strategy (1). Although patient and lesion characteristics were balanced between the primary crossing groups, the role of a selection bias based on operator expertise and device availability cannot be neglected. However, all participating sites include experienced endovascular operators and have contributed both wire-catheter and CTO device cases to the registry. On the basis of the data presented, reference vessel diameter is likely to have affected the primary crossing strategy. The mean vessel diameter in the wire-catheter group is significantly smaller than in the crossing device arm, an observation confirmed by a primary wire-catheter approach selected more frequently for below-the-knee CTO crossings and significantly greater in patients with advanced symptoms (Rutherford categories IV and V). Given the recent advent and adoption of smaller caliber CTO crossing devices, some with adjunctive imaging capability, this trend could see significant changes.
The need for subintimal re-entry is overall similar for both the primary crossing strategies. However, compared with a 20% need for subintimal re-entry in the primary wire-catheter group, use of re-entry escalates to 53% following failure of the primary guidewire approach. Such procedures add incrementally to procedure duration and cost. Importantly, these data do not reflect key differences in either lesion length or the presence of chronically occluded in-stent restenotic lesions between the wire-catheter and crossing device groups. Use of re-entry devices in the primary crossing device arm is 26%. The initial cost advantage associated with using a wire-catheter approach over a primary crossing device is lost with the provisional use of re-entry or crossing devices. Future analysis from the XLPAD registry is planned to elucidate key predictors of wire-catheter crossing failure and primary CTO device success.
There is an ongoing debate among health economists regarding the use of cost-benefit or cost-effectiveness analyses to evaluate the benefits of medical interventions (6). Cost-benefit analysis was formerly considered the gold standard (7), as it provides monetary values for a unit of effectiveness of a medical intervention (8).
Our cost-benefit analysis results indicate that the estimated cost for the crossing device group compared with wire-catheter is greater than estimated benefits associated with lower repeat revascularization and amputation, thus accounting for a positive net cost. It is important to emphasize that at 12 months, this incremental net cost is marginal. This could be partly accounted for by the higher itemized component cost with substantial (≥5%) cost contribution, except for atherectomy, in the crossing device arm. We believe that given the larger calibers of vessels treated in the CTO device group, there is significantly greater use of balloons and stents. Operators can use these data to make individual decisions on selection of a CTO crossing strategy. However, the net cost for the crossing device group compared with the wire-catheter group significantly reduced to $423.80 when evaluated by 12-month repeat revascularization or amputation from the net cost of $2,122.40 evaluated by 30-day repeat revascularization or amputation. This is because of significant increase in rates of endovascular and surgical reintervention and amputation at 12 months from 30 days in the wire-catheter group compared with the crossing device group. This could be attributed to older age, advanced patient symptoms, smaller reference vessel diameter, and lower stent use in the wire-catheter group. Similarly, higher death and amputation rates in the provisional re-entry device arm compared with provisional use of crossing device following an initial wire-catheter approach may also be related to a significantly higher proportion of patients with critical limb ischemia in the re-entry device arm.
These data regarding direct procedural costs are complementary to overall hospitalization cost data published on the basis of administrative third-party payer databases (9). The overall impact of cost-saving strategies on clinical outcomes can be gleaned by being informed of hospitalization, follow-up, and direct procedure costs and outcomes (10). Toward this objective, these data from the XLPAD registry provide a granular breakdown of direct procedure costs, operator preferences, and associated clinical outcomes seldom gleaned from third-party aggregate cost analyses.
Key limitations of our analysis include selection bias associated with an observational registry, lack of actual crossing time that could provide an assessment of crossing efficiency, and accounting of overall hospitalization and follow-up costs. VA cost estimates were used, but major device and human resource professional costs are overall similar to those of participating non-VA institutions. Furthermore, a standard cost model was required for analysis. Additional analyses to study differences in health system costs are planned.
Despite these limitations, we hope that this cost and outcomes analysis from the XLPAD registry will be informative to an interdisciplinary group of providers caring for patients with PAD and to health systems in making treatment decisions that are based on composite assessment of both cost and patient outcomes. There exists variance in cost estimation of 2 competing primary crossing strategies due to model specifications in estimating reintervention and amputation. We also conducted sensitivity analysis of alternative models without incorporating CTO lesion length. The net cost of CTO crossing device as a primary strategy compared with wire catheter according to the alternative model was estimated as $167.80 at 12 months and $2,779.50 at 30 days (Online Table 3). Except for CTO length, the model did not converge with the inclusion of other anatomic and procedural characteristics.
Treatment of infrainguinal CTO is an important component of endovascular treatment of PAD. A wire-catheter approach to cross peripheral artery CTO is most often the first choice of operators. The use of dedicated CTO devices provides significantly smaller numbers of repeat revascularization and amputation, at an additional incurred cost.
WHAT IS KNOWN? Health care cost is an important consideration in selecting devices for peripheral artery CTO intervention, which is frequently encountered in clinical practice.
WHAT IS NEW? We compared device cost and clinical outcomes side by side and provide a measure of cost-effectiveness derived from the actual number of devices and human resources used during clinically indicated peripheral artery CTO interventions.
WHAT IS NEXT? Future prospective studies capturing the cost of peripheral artery interventions are needed to add to our knowledge base regarding cost-effective device interventions and outcomes. This report from the XLPAD registry is an important first step in this direction.
The authors acknowledge the contributions of M. Ishti Ali, MD, to the XLPAD registry. We also acknowledge the support of the University of Texas Southwestern Medical Center in establishing and managing the RedCap database software used in the XLPAD registry (Academic Information Systems National Institutes of Health grant UL1-RR024982).
For supplemental tables, please see the online version of this article.
Dr. Banerjee has received research and educational grants from Boston Scientific and Merck; and consulting and speaking honoraria from Medtronic, Cardiovascular Systems, Inc., and Gore. Dr. Mohammad has received speaking honoraria from Medicure. Dr. Addo has received speaking honoraria from AstraZeneca and Merck. Dr. Cawich has received a consulting fee from Avinger. Dr. Armstrong has received consulting and speaking honoraria from Abbott Vascular, Boston Scientific, Cardiovascular Systems, Medtronic, Merck, and Spectranetics. Dr. Das has received consulting and speaking honoraria from Abbott Vascular, Cordis, Cardinal, and Boston Scientific. Dr. Gigliotti has received research funding from Abbott Vascular, Medtronic, Bard, Janssen, and Bristol-Myers Squibb; and speaking honoraria from AstraZeneca and Terumo. Dr. Prasad has received research funding from Osprey and Medtronic; and speaking honoraria from St. Jude Medical, AstraZeneca, and Gilead. Dr. Shammas has received educational and research funding from Boston Scientific, Janssen, Gilead, Novartis, and Trainer Covidien. Dr. Brilakis has received research grants from Infraredx, Boston Scientific; and consulting honoraria and speaking fees from Abbott Vascular, Asahi, Cardinal Health, Elsevier, GE Healthcare, and St. Jude Medical. Dr. Brilakis’s spouse is an employee of Medtronic. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- chronic total occlusion
- peripheral artery disease
- Veterans Affairs
- Received April 6, 2016.
- Revision received July 6, 2016.
- Accepted August 11, 2016.
- American College of Cardiology Foundation
- Banerjee S.,
- Sarode K.,
- Patel A.,
- et al.
- Banerjee S.,
- Sarode K.,
- Mohammad A.,
- et al.
- Carr J.G.
- Jones W.S.,
- Mi X.,
- Qualls L.G.,
- et al.