Abstract
Purpose
Atherectomy has emerged as an alternative to percutaneous transluminal angioplasty (PTA) for endovascular reopening. Despite increasing use of atherectomy (and higher cost of atherectomy catheters compared with balloon catheters), few studies have compared outcomes and costs with other reopening strategies.
Methods
We performed a retrospective cohort study involving all patients undergoing isolated femoropopliteal PTA ( n =69) or atherectomy ( n =92) at our institution from 1/2005 to 4/2006. The choice of reopening strategy was left to the treating physician, and no patients with relative contraindications to stent placement (specifically common femoral artery lesions) were included. Device and supply costs were calculated using the hospital resource-based accounting system, and other costs were calculated using the hospital micro-cost accounting system. Professional fees were calculated from the Medicare Fee Schedule.
Results
Baseline characteristics were generally well matched. There were no significant differences in complications (vascular complications, urgent repeat reopening, death, myocardial infarction, or stroke) between groups (PTA 8.7% vs. atherectomy 5.4%, P =.53). PTA required more balloons (2.0±0.8 vs. 0.7±1.0, P <.001) and stents (1.5±0.8 vs. 0.2±0.5, P <.001), but fewer atherectomy catheters (0.0±0.0 vs. 1.2±0.4, P <.001). Neither procedural supply costs (PTA $3137±1459 vs. atherectomy $3338±1505, P =.20) nor total costs differed between PTA and atherectomy patients ($10,945±4521 vs. $10,783±3857, P =.42).
Conclusions
Initial outcomes and costs are comparable for femoropopliteal PTA and atherectomy. The choice of reopening strategy should therefore be based on operator experience and anatomic suitability. Further studies are required to determine whether there are differences in long-term outcomes or costs between these approaches.
1
Introduction
Lower extremity peripheral arterial disease (PAD) is present in 10–20% of patients over age 55 . Initial treatment includes cardiovascular risk factor modification for all patients as well as medication, smoking cessation, and exercise therapy to control symptoms of intermittent claudication (IC) . For those unresponsive to conservative measures, reopening is indicated to relieve severe symptoms and for limb salvage . Current reopening options include percutaneous transluminal angioplasty (PTA) with or without adjunctive stenting, atherectomy using either rotational or directional cutting systems , and lower extremity bypass surgery.
Although PTA with stenting and bypass surgery are both highly effective therapies for aortoiliac disease, treatment of femoropopliteal disease is more problematic. Depending on patient and lesion characteristics, restenosis and reocclusion remain common after PTA (with or without stenting) and long-term (3–5 years) primary and secondary graft primary patency rates after bypass surgery range from 57% to 90%, respectively . In light of these limitations, there has been increasing interest in alternative reopening strategies. The Silverhawk device (ev3 Endovascular, Plymouth, MN, USA) is a directional atherectomy device commonly used for percutaneous endovascular reopening. Despite its increasing popularity, however, there are no studies comparing its short- or long-term outcomes with those of alternative approaches to femoropopliteal reopening. Moreover, its acquisition cost is approximately $1900 per device, and multiple devices may be required to complete a single procedure. Because of the cost of atherectomy devices and the possibility of higher procedural costs as compared with conventional endovascular reopening, we undertook the present study to evaluate in-hospital outcomes and costs of atherectomy compared to conventional PTA with bailout stenting for patients undergoing femoropopliteal reopening.
2
Methods
2.1
Patient population
This study included all patients undergoing isolated femoropopliteal PTA or atherectomy at a single center between 1/2005 and 4/2006. Patients with common femoral artery lesions were excluded due to the relative contraindication to stent placement. If both PTA and atherectomy were used in a single patient, the treatment group assignment was determined based on the intended initial form of definitive therapy determined by detailed chart review.
2.2
Data sources and definitions
Clinical data were obtained by detailed chart review using a standardized data collection instrument. Hypertension, hypercholesterolemia, diabetes mellitus, chronic renal insufficiency (defined as a serum creatinine level >1.5 mg/dl), coronary artery disease, and congestive heart failure were recorded as documented by the admitting physician. Rutherford class was obtained from the preprocedural history and physical or from the attending physician’s assessment at the time of arteriography. Smoking was defined as current tobacco use at the time of admission. Laboratory data were obtained from preprocedural laboratories—typically performed within 1 week of reopening. The last measurement of ankle brachial index (ABI), duplex imaging, or computed tomography angiograms (CTA) were reviewed and assessed for lesion severity. If maximal estimated lesion severity was provided, it was averaged for the group, and the most severe measurement was used in a composite comparison if multiple evaluations were available. Outside angiograms were included as an assessment of lesion severity only if there were no noninvasive evaluations available. Qualitative assessments of mild, moderate, or severe stenosis were recorded if there was no estimate of lesion severity provided. Noninvasive studies were considered normal with ABI of 0.9 or greater or reported stenosis of 0%, mildly abnormal with ABI of 0.70–0.89 or reported stenosis of 1–69%, moderately abnormal with ABI of 0.40–0.69 or a reported stenosis of 70–90%, and severe disease was defined as ABI under 0.40 or maximum estimated stenosis of >90%.
In-hospital outcomes were assessed from hospital records and discharge summaries. Lesion length, percent stenosis, and reference vessel diameter were obtained from review of diagnostic angiograms and procedural reports. Complications included vascular-related, urgent repeat reopening, death, myocardial infarction, or stroke. Vascular complications were defined as bleeding requiring transfusion or intervention to repair vascular damage (i.e., thrombin injection or surgical repair of pseudoaneurysm). Technical success was defined as successful endovascular treatment of the lesion with <40% residual stenosis (by visual estimate), and procedural success was defined as technical success with no in-hospital complications (as defined above). The protocol was approved by the local institutional review board.
Since our goal was to focus on complications and resource utilization that was directly attributable to the reopening procedures, we reviewed the hospital course for each patient and excluded those resources that were not directly attributable to the reopening procedures or their complications (e.g., preprocedure hospital days for another condition). Preprocedure hospital days were considered part of the reopening procedure if they were required to allow for hydration, laboratory assessment, and noninvasive testing. Postprocedure hospital days were considered procedure related if complications from the procedure led to testing or treatment prolonging the hospitalization.
2.3
Cost methodology
Total costs included supply costs, other procedural costs, non–procedure-related hospital costs, and physician fees assessed in constant 2005 US dollars. Procedure-related supply costs, including balloons, stents and atherectomy devices, contrast dye, and guide wires, were obtained directly from the hospital’s cardiovascular laboratory accounting system. Interventional suite costs (excluding relevant supply costs) were based on room costs to the hospital as well as direct labor and other nonlabor costs. Nonprocedural hospital costs included room and ancillary costs, such as histology, specimen procurement, and nutrition, in addition to specific department costs (laboratory, radiology, pharmacy). These costs were based on direct costs obtained from the hospital accounting system. Physician costs were derived by applying 2005 Medicare Relative Value Unit-based rates to current procedural terminology codes for each procedure.
2.3.1
Statistical analysis
Demographic and clinical data are presented as mean±S.D. for continuous variables and proportions for categorical variables. Data were compared by ANOVA for continuous variables and χ 2 or Fisher’s Exact Test, where appropriate, for categorical variables. Cost data are presented as mean±S.D. with selected data reported as median and interquartile range. P values for the difference in mean costs between treatment groups were obtained using bootstrap resampling . All P values are two-tailed, and P <.05 was considered significant. All analyses were performed using SAS version 9.1 (SAS Institute, Cary, NC, USA).
2
Methods
2.1
Patient population
This study included all patients undergoing isolated femoropopliteal PTA or atherectomy at a single center between 1/2005 and 4/2006. Patients with common femoral artery lesions were excluded due to the relative contraindication to stent placement. If both PTA and atherectomy were used in a single patient, the treatment group assignment was determined based on the intended initial form of definitive therapy determined by detailed chart review.
2.2
Data sources and definitions
Clinical data were obtained by detailed chart review using a standardized data collection instrument. Hypertension, hypercholesterolemia, diabetes mellitus, chronic renal insufficiency (defined as a serum creatinine level >1.5 mg/dl), coronary artery disease, and congestive heart failure were recorded as documented by the admitting physician. Rutherford class was obtained from the preprocedural history and physical or from the attending physician’s assessment at the time of arteriography. Smoking was defined as current tobacco use at the time of admission. Laboratory data were obtained from preprocedural laboratories—typically performed within 1 week of reopening. The last measurement of ankle brachial index (ABI), duplex imaging, or computed tomography angiograms (CTA) were reviewed and assessed for lesion severity. If maximal estimated lesion severity was provided, it was averaged for the group, and the most severe measurement was used in a composite comparison if multiple evaluations were available. Outside angiograms were included as an assessment of lesion severity only if there were no noninvasive evaluations available. Qualitative assessments of mild, moderate, or severe stenosis were recorded if there was no estimate of lesion severity provided. Noninvasive studies were considered normal with ABI of 0.9 or greater or reported stenosis of 0%, mildly abnormal with ABI of 0.70–0.89 or reported stenosis of 1–69%, moderately abnormal with ABI of 0.40–0.69 or a reported stenosis of 70–90%, and severe disease was defined as ABI under 0.40 or maximum estimated stenosis of >90%.
In-hospital outcomes were assessed from hospital records and discharge summaries. Lesion length, percent stenosis, and reference vessel diameter were obtained from review of diagnostic angiograms and procedural reports. Complications included vascular-related, urgent repeat reopening, death, myocardial infarction, or stroke. Vascular complications were defined as bleeding requiring transfusion or intervention to repair vascular damage (i.e., thrombin injection or surgical repair of pseudoaneurysm). Technical success was defined as successful endovascular treatment of the lesion with <40% residual stenosis (by visual estimate), and procedural success was defined as technical success with no in-hospital complications (as defined above). The protocol was approved by the local institutional review board.
Since our goal was to focus on complications and resource utilization that was directly attributable to the reopening procedures, we reviewed the hospital course for each patient and excluded those resources that were not directly attributable to the reopening procedures or their complications (e.g., preprocedure hospital days for another condition). Preprocedure hospital days were considered part of the reopening procedure if they were required to allow for hydration, laboratory assessment, and noninvasive testing. Postprocedure hospital days were considered procedure related if complications from the procedure led to testing or treatment prolonging the hospitalization.
2.3
Cost methodology
Total costs included supply costs, other procedural costs, non–procedure-related hospital costs, and physician fees assessed in constant 2005 US dollars. Procedure-related supply costs, including balloons, stents and atherectomy devices, contrast dye, and guide wires, were obtained directly from the hospital’s cardiovascular laboratory accounting system. Interventional suite costs (excluding relevant supply costs) were based on room costs to the hospital as well as direct labor and other nonlabor costs. Nonprocedural hospital costs included room and ancillary costs, such as histology, specimen procurement, and nutrition, in addition to specific department costs (laboratory, radiology, pharmacy). These costs were based on direct costs obtained from the hospital accounting system. Physician costs were derived by applying 2005 Medicare Relative Value Unit-based rates to current procedural terminology codes for each procedure.
2.3.1
Statistical analysis
Demographic and clinical data are presented as mean±S.D. for continuous variables and proportions for categorical variables. Data were compared by ANOVA for continuous variables and χ 2 or Fisher’s Exact Test, where appropriate, for categorical variables. Cost data are presented as mean±S.D. with selected data reported as median and interquartile range. P values for the difference in mean costs between treatment groups were obtained using bootstrap resampling . All P values are two-tailed, and P <.05 was considered significant. All analyses were performed using SAS version 9.1 (SAS Institute, Cary, NC, USA).
3
Results
3.1
Patient population
Between 1/2005 and 4/2006, a total of 161 eligible isolated femoropopliteal reopening procedures were performed. The intended treatment strategy was PTA in 69 patients and atherectomy in 92. Table 1 displays baseline demographic and clinical characteristics for the two groups. There were few differences between the PTA and atherectomy groups. Overall, the mean age was 68 years, 61% were male and 93% were Caucasian. There were similar rates of hypertension, hypercholesterolemia, diabetes mellitus, active smoking, chronic renal failure, and congestive heart failure. Patients undergoing PTA were more likely to have documented coronary artery disease (82.6% vs. 67.4%, P =.03) and also tended to have more baseline renal insufficiency (11.6% vs. 4.3%, P =.08). Over one-half of patients had undergone at least one prior lower extremity reopening procedure (52% PTA vs. 58% atherectomy, P =.49). The majority of these were prior endovascular reopening (76.4% among PTA patients and 94.2% among atherectomy patients). A minority of patients (14.7% PTA and 0 atherectomy) had both prior PTA and surgery, while 8.8% of PTA and 5.8% of atherectomy patients had undergone prior lower extremity bypass surgery ( P =.005 for difference in type of prior reopening between groups).
| PTA ( n =69) | ATH ( n =92) | P value | |
|---|---|---|---|
| Age, years | 67.3±22.0 | 68.6±13.6 | .65 |
| Male, % | 60.9 | 58.7 | .78 |
| Caucasian, % | 92.8 | 92.4 | .62 |
| Hypertension, % | 92.8 | 90.2 | .57 |
| Hyperlipidemia, % | 98.6 | 94.6 | .24 |
| Diabetes mellitus, % | 40.6 | 38.0 | .74 |
| Insulin-requiring, % | 18.8 | 16.5 | .70 |
| Tobacco use, % | 23.2 | 29.3 | .38 |
| Renal insufficiency, % | 11.6 | 4.3 | .08 |
| Serum creatinine, mg/dl | 1.4±1.1 | 1.2±0.8 | .36 |
| Coronary artery disease, % | 82.6 | 67.4 | .03 |
| Congestive heart failure, % | 18.8 | 10.9 | .15 |
| Prior LE revascularization, % | 52.2 | 57.6 | .49 |
| Bilateral PAD, % | 75.4 | 76.1 | .92 |
| Initial stenosis, % | 84.7±14.6 | 83.5±14.5 | .45 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
