Print this pageCost Per Day of Patency: Understanding the Impact of Patency and Reintervention in a Sustainable Model of Healthcare
Back to Annual MeetingBack to Program
Dorian J deFreitas*, Mark L Manwaring*, Clark M Kardys*, Frank M Parker, Charles S Powell, Michael C Stoner*
East Carolina University, Greenville, NC
Background: Peripheral arterial disease is prevalent in the United States with 8 to 12 million Americans carrying the diagnosis account for over 400,000 hospitalizations per year. America's aging population and the rising cost of health care has placed an untenable burden on society. Cost effectiveness in surgery is of critical importance if we are to develop a sustainable model of health care. Hospital costs for the treatment of peripheral vascular disease are primarily driven by device costs, length of hospital stay, and operating room utilization.
Open revascularization is the gold standard in the treatment of femorpopliteal occlusive disease, although many authors tout endovascular therapy as both a first-line and definitive therapy. Those who champion open revascularization highlight its excellent efficacy and durability. The obvious advantages of endovascular therapy include; low morbidity, ease of reintervention and lower cost of intervention. However, current cost models only take into account initial cost which is flawed and myopic, as it does not account for procedural durability and patient longevity. With this in mind we set out to review our contemporary experience with open and endovascular therapy in the treatment of femoropopliteal occlusive disease. The aim of this study is to utilize a new model of cost analysis that accounts for initial cost, patency and subsequent interventions and to determine if the initial cost savings of endovascular therapy is sustained over time.
Methods: A retrospective review of all open and endovascular treatments of femoropopliteal disease from 2003-2006. Demographic, clinical variables, initial and all subsequent costs were recorded as well as patency data according to SVS guidelines.
An amortized cost model was developed. Total cost was determined over time, using initial cost and all subsequent revascularization costs as the numerator. The denominator was the time from the index procedure until the time point when the intervention had failed and no other revascularization procedure was undertaken. Open revascularization after endovascular therapy was considered a failure of revascularization.
Cost per day of patency is expressed at any given time interval (t) as follows:
(SEE FORMULA)
Where pa = maximum number of days of assisted patency, Cost(i) = total hospital costs at time interval i days from index procedure (t=0). Through this cost model, cost per day of patency was assessed at any given time interval, from the index procedure.
Results: Over a three year period a total of 381 femoropopliteal segments were treated in 359 patients. There were a total of 183 femoropopliteal segment treated with open revascularization and 198 treated with endovascular therapy. Both treatment groups were well matched with no significant difference in clinical variables including: Rutherford category, age, gender, diabetes, and hypertension, end stage renal disease requiring dialysis tobacco, run off and post operative pharmacotherapy (Table 1).
| Open | Endovascular | P | |
| Age (years) | 63.9±0.9 | 66.0±0.9 | 0.10 |
| Male | 63.1% | 57.9% | 0.29 |
| Critical limb ischemia | 55.7% | 43.4% | 0.10 |
| Diabetes | 48.1% | 50.0% | 0.70 |
| ESRD | 13.1% | 9.1% | 0.80 |
| Creatinine > 1.5 mg/dL | 14.8% | 15.6% | 0.80 |
| ≤ 1 patent tibial artery | 49.1% | 53.0% | 0.45 |
| TASC C or D | N/A | 37.7% | N/A |
| Lesion length > 10 cm | N/A | 21.3% | N/A |
| Re-intervention rate | 22.0% | 18.0% | 0.54 |
Table 1. Sample demographic, clinical and anatomical characteristics of open and endovascular cases. TASC = Trans-Atlantic Inter-Society Consensus, ESRD = End stage renal disease.
Patency was assessed by SVS guidelines utilizing lifetable analysis. Primary assisted patency at twelve months was 78% for the open group and 66% for the endovascular group. (P=0.01, Figure 1). Primary assisted patency for patient with critical limb ischemia at twelve months was 58% in the open group and 47% in the endovascular group. Primary assisted patency for patients with claudication at twelve months was 90% in the open group versus 76% in the endovascular group.
(SEE FIGURE)
Figure 1. Kaplan-Meyer curve demonstrating probability of primary-assisted patency from time of index procedure in both open (red) and endovascular (blue) cases, all indications (P=0.002).
Initial cost of open therapy was significantly higher in all subgroups (Table 2) Using the model of amortized cost described above the cost per day of patency was calculated at one year. Despite the difference in initial cost, our model showed no statistically significant difference in amortized cost at one year between open and endovascular therapy regardless of indication (Table 3). Claudicants treated by simple angioplasty showed the lowest cost per day of patency ($26±14) although this did not reach statistical significance when compared to open therapy.
| Open - Initial cost ($) | Endovascular - Initial cost ($) | P | |
| All indications | 12, 389 +/- 408 | 6,739 +/- 260 | <0.001 |
| Critical limb ischemia | 13,277 +/- 598 | 7,176 +/- 309 | <0.001 |
| Claudicants | 11,042 +/- 468 | 6287 +/- 415 | <0.001 |
Table 2. Initial cost of open and endovascular femoropopliteal revascularization stratified by indication.Utilizing the model, subgroup analysis was undertaken to identify risk factors in which one therapy was more cost effective. While there were trends for cost saving for endovascular therapy in end-stage renal disease requiring dialysis, renal insufficiency (creatinine > 1.5 md/dL) and congestive heart failure, none of these results reached statistical significance. Because of their major impact on the cost model, early failures, which were defined as primary patency of less than 30 days, were also examined using multivariate logistic regression technique. End stage renal disease (ESRD) requiring dialysis was an independent risk factor for early failure in patients with critical limb ischemia for both open and endovascular therapy. (O.R. = 3.48, P = 0.048). No other variable was a significant correlate of early failure.
| Open ($/patient-day patency) | Endovascular ($/patient-day patency) | P | |
| All indications | 229 +/- 106 | 185 +/- 124 | 0.71 |
| Critical limb ischemia | 261 +/- 58 | 306 +/- 128 | 0.33 |
| Claudicants | 259+/- 189 | 86+/-52 | 0.31 |
Table 3. Cost per patient-day of patency at 12 months from index procedure, stratified by indication.
Discussion: This study provides a well matched group of patients treated for femoropopliteal occlusive disease. While this study is limited by its retrospective nature, and the inherent treatment biases that exist in any practice, it follows the literature in that the primary assisted patency of open revascularization is superior to that of endovascular therapy. The cost of open revascularization is also significantly higher than endovascular therapy, as has also been described. Our new model of cost per patient-day of patency gives a surrogate measure of a cost benefit ratio by adding in previously unconsidered factors such as overall patency and reintervention rates. Interestingly despite a two-fold difference in initial cost, the cost savings of endovascular therapy is not carried out over time. The demise of the cost benefit of endovascular therapy lies in its lower patency rates and need for subsequent reintervention. This is evident particularly in the critical limb ischemia cohort where at one year, endovascular therapy is clearly trending to become the more expensive modality. The 11% difference in patency at one year and the early failures of patency (<30 days) are the major influences on this high cost.
Claudicants treated by endovascular therapy have a noticeable trend toward cost savings compared to open therapy, despite a 14 % difference in patency at one year. In particular, those treated by simple angioplasty with no other adjunctive interventions showed the greatest savings. This difference is driven by less early failures in the endovascular limb and the high expense of reintervention in the open group. These data suggest that if one were to employ only simple angioplasty alone without adjunctive measures (i.e. atherectomy, cryoplasty, and stenting) that endovascular therapy is the most economically sustainable modality in the treatment of claudicants.
The cost per day of patency of endovascular therapy appears to favor those with ESRD, renal insufficiency and heart failure. Caution must be taken when considering these results because of low patient numbers and high error margins. The trend favoring endovascular therapy relates to high rates of early failures in both groups which would favor the modality with the lower initial cost. ESRD offers a complex problem as it has a high early failure rate in both open and endovascular therapy in patients with critical limb ischemia. Based on our results it is difficult to recommend any therapy based on our cost analysis. The poor cost benefit ratio encountered when treating ESRD patients with critical limb ischemia is prohibitive in all but a select few, and argues for conservative therapy and early primary amputation.
Conclusion: Open revascularization has a superior patency rate compared to endovascular therapy. The initial cost saving of endovascular therapy is not sustained over time. Patency appears to be more important than initial cost when determining overall cost to the health care system. Open revascularization shows a trend towards being more cost effective at one year in patients with critical limb ischemia than endovascular therapy. Endovascular therapy trends to be more cost effective than open therapy in patients with claudication. Using a model of initial cost is a flawed methodology of assessing a cost benefit ratio. Models for healthcare utilization in vascular surgery must to take into account initial cost, patency and reintervention rate if a sustainable growth model of cost is to be developed.
Back to Annual Meeting
Back to Program