Abstract
Objectives:
There have been no data to support the use of tibial interventions in the treatment of claudication to date. We aimed to characterize the practice patterns surrounding tibial peripheral vascular interventions (PVI) in patients with claudication in the United States.
Methods:
Using 100% Medicare fee-for-service claims from 2017 to 2019, we conducted a retrospective analysis of all patients undergoing an index PVI for claudication. Patients who had any previous PVI, acute limb ischemia, or chronic limb-threatening ischemia in the preceding 12 months were excluded. The primary outcome was the receipt or delivery of tibial revascularization during index PVI for claudication, defined as tibial PVI with or without concomitant femoropopliteal PVI. Univariable comparisons and a multivariable hierarchical logistic regression were used to assess patient and physician characteristics associated with the use of tibial PVI for claudication.
Results:
Of 59,930 Medicare patients who underwent index PVI for claudication between 2017 and 2019, 16,594 (27.7%) received a tibial PVI (38.5% isolated tibial PVI and 61.5% tibial PVI with concomitant femoropopliteal PVI). Of 1,542 physicians included in the analysis, the median physician-level tibial PVI rate was 20.0% (IQR 9.1%−37.5%). Hierarchical logistic regression suggested that patient-level characteristics associated with tibial PVI for claudication included male sex (adjusted odds ratio [aOR] 1.23), increasing age (aOR1.30–1.96), Black race (aOR 1.47), Hispanic ethnicity (aOR 1.86), diabetes (aOR 1.36), no history of hypertension (aOR 1.12), and never-smoking (aOR 1.64; all, P<0.05). Physician-level characteristics associated with tibial PVI for claudication included early-career status (aOR 2.97), practice location in the West (aOR 1.75), high-volume PVI practice (aOR 1.87), majority practice in an ASC/OBL setting (aOR 2.37), and physician specialty; vascular surgeons had significantly lower odds of performing tibial PVI compared to radiologists (aOR 2.98) and cardiologists (aOR 1.67; all, P<0.05). Average Medicare reimbursement per patient was dramatically higher for physicians performing high rates of tibial PVI (quartile 4 vs. quartile 1–3: $12,023.96 vs. $692.31 per patient, P<0.001).
Conclusions:
Tibial PVI for claudication are performed more commonly by non-vascular surgeons, in high-volume practices, and in high-reimbursement settings. This reveals a critical need to reevaluate the indications, education, and reimbursement policies surrounding these procedures.
Keywords: intermittent claudication, peripheral artery disease, peripheral vascular interventions, endovascular surgery, clinical guidelines, Centers for Medicare and Medicaid
Table of Contents Summary
More than a quarter of the 59,930 Medicare patients who underwent a peripheral vascular intervention for intermittent claudication underwent a tibial peripheral vascular intervention. Our findings reveal a critical need to standardize trainee education and clinical practice guidelines across specialties, and to reevaluate Medicare reimbursement policies surrounding the treatment of claudication.
Introduction:
The Society for Vascular Surgery (SVS) guidelines recommend initiating a trial of smoking cessation, risk factor modification, and supervised exercise therapy as first-line therapy for the treatment of intermittent claudication1. After a period of 3 to 6 months, patients who continue to experience severe lifestyle-limiting claudication symptoms can be considered for revascularization. These recommendations are consistent across the SVS guidelines for lower extremity disease1 and the American Heart Association (AHA)/American College of Cardiology (ACC) guidelines for claudication2.
For patients with claudication who are considered for revascularization following conservative management, peripheral vascular interventions (PVI) have been shown to be most effective for aortoiliac occlusive disease, followed by femoropopliteal disease2, 3. For isolated infrapopliteal artery disease in patients with claudication, the AHA/ACC guidelines consider the clinical usefulness of PVI to be “unknown”2 due to a lack of evidence, while the SVS guidelines explicitly state, “Isolated infrapopliteal interventions are not recommended for patients with [intermittent claudication]”1. This is due to there being no clear relationship between the presence of tibial disease and claudication symptoms1, and due to the anatomic durability of tibial PVI being demonstrably poor relative to more proximal interventions4.
There are significant variations in the practice patterns and technologies used in PVI for claudication. We have previously explored the use of femoropopliteal atherectomy for treating claudication and observed marked differences in its use according to practice location and physician specialty5. Given the variation in professional guidelines surrounding the efficacy of tibial interventions for claudication1, 2, we aimed to characterize the practice patterns surrounding tibial PVI in patients with claudication in the United States.
Methods:
Study Population
We used 100% Medicare fee-for-service claims from January 1, 2017 to December 31, 2019 to identify all patients who underwent an index PVI for claudication. Claudication was defined using International Classification of Diseases (ICD) codes (Supplemental Table 1). All patients with any prior PVI (n=31,116), acute limb ischemia (n=51,404), or chronic limb-threatening ischemia (n=174,205) in the preceding 12 months were excluded from analysis (total excluded n=219,723). We also excluded patients without a diagnosis of claudication (n=1,614), patients without 12 months of enrollment at the time of the index PVI (n=3,251), and patients missing any demographic information (n=14). Study approval was obtained from the Johns Hopkins Medicine Institutional Review Board.
Patient Characteristics
We used the Medicare Master Beneficiary Summary File6 to identify patient demographic characteristics including age, sex, race, and ZIP code, which was mapped to the Federal Information Processing Standard (FIPS) code using the sashelp.zipcode file (SAS Institute, Cary, North Carolina), and then the core-based statistical area (CBSA) code using the National Bureau of Economics Research’s CBSA to FIPS County Crosswalk7. These linkages were used to classify the population density of residence for each patient, with an urban area defined as having ≥50,000 people and a rural area defined as population <50,000 people with no urban core5. We used median household income as a metric for socioeconomic status, characterized into quartiles per the Agency for Healthcare Research and Quality’s 2018 estimates8.
We reviewed inpatient, outpatient, and carrier claims data to identify patients who underwent index PVI and identified those with claudication diagnosed in the preceding 12 months as the indication for the procedure. We defined comorbidities including diabetes, hypertension, end-stage renal disease, and ever-smoking based on a single inpatient diagnosis claim or at least two outpatient diagnosis claims more than 30 days apart.
Physician Characteristics
We used National Provider Identifier numbers to link physicians with the PVI procedures that they performed. We calculated the rate of index tibial PVI for claudication for all physicians who performed >10 index PVI procedures during the study period, in accordance with our data use agreement with the Centers for Medicare and Medicaid Services. Each physician’s rate of index tibial PVI was calculated using the number of tibial PVI performed for claudication (either isolated tibial PVI or tibial PVI with a concomitant femoropopliteal PVI) during the study period by a given physician and the total number of PVI performed for claudication during the study period by that same physician as the denominator. We evaluated the national distribution of index tibial PVI rates using a histogram and then classified physicians into quartiles based on their individual tibial PVI rates.
We used the Medicare Data on Provider Practice and Specialty6 and the Physician Compare National Downloadable File9 to identify physician demographics including sex, years since graduation from medical school, primary specialty, census region of practice location, and population density of practice location. Physician specialty is based on self-report to the Centers for Medicare & Medicaid Services and does not necessarily represent board certification. We calculated the overall percentage of services delivered in a freestanding ambulatory surgery center (ASC) or office-based laboratory (OBL) using summary statistics in the Medicare Data on Provider Practice and Specialty6. We also calculated the average Medicare-allowed payment (in USD) for index PVI procedures performed for claudication from 2017 to 2019 for each physician.
Outcome
The main outcome of the study was defined as any tibial PVI (either as an isolated intervention or with a concomitant femoropopliteal PVI) performed during a patient’s index intervention for claudication. Femoropopliteal and tibial PVI were identified using CPT codes 37224–37227 and 37228–37235, respectively (Supplemental Table 2). Per CPT coding, procedures performed on the tibioperoneal trunk, anterior tibial artery, posterior tibial artery, and/or peroneal artery fall into the same category, which we classified as “tibial PVI.” Only the first limb treated for any given patient was included in the analysis.
Statistical Analysis
We described patient and physician characteristics using count (percentage) and mean ± standard deviation or median (interquartile range), as appropriate. We used Chi-squared tests for categorical variables or Mood’s median tests for continuous variables to assess differences in baseline characteristics of patients undergoing tibial PVI versus non-tibial PVI for claudication, and to compare characteristics of physicians performing the highest rates of tibial PVI (quartile 4) versus lower rates of tibial PVI (quartiles 1–3) for claudication.
Univariable logistic regression was used to assess the association of patient and physician characteristics with tibial PVI for claudication. A multivariable hierarchical logistic regression model including a random intercept for physician to account for patient clustering by physician was subsequently used to identify patient- and physician-level characteristics associated with the use of tibial PVI. Patient covariates (age, sex, ethnicity/race, smoking history, and other comorbid conditions) were assessed in the first level of the model, and physician covariates (sex, primary specialty, years in practice, location of practice, volume of PVI performed, and proportion of practice in an ASC/OBL setting) were assessed in the second level of the model. All covariates were chosen a priori based on our previous work on this topic10.
Statistical analyses were performed using SAS Enterprise version 7.1 (SAS Institute, Cary, North Carolina). All results were deemed statistically significant at P < 0.05.
Results:
Patient Cohort
Of 59,930 Medicare patients who underwent index PVI for claudication between 2017 and 2019, 27.7% (n=16,594) received a tibial PVI. Among patients who received a tibial PVI, 38.5% (n=6,394) received an isolated tibial PVI and 61.5% (n=10,200) received a tibial PVI with a concomitant femoropopliteal PVI (Table 1). Of 17,081 tibial interventions performed in 16,594 patients, atherectomy was the most common (N=9,916, 59.8%), followed by angioplasty (N=6,122, 36.9%), stenting (N=542, 3.3%), and stenting with atherectomy (N=501, 3.0%). Additionally, 29.6% (n=4,908) of patients undergoing a tibial PVI received a multi-vessel tibial intervention, with atherectomy being the most common procedure performed on additional tibial vessels. Due to multiple procedure types on multiple vessels, the sum of tibial interventions totals to greater than 100%.
Table 1.
Characteristics of Medicare patients undergoing a tibial intervention vs. non-tibial intervention as their index intervention for claudication, 2017–2019
| Patient characteristics | Non-tibial PVI (N=43,336) | Tibial PVI (N=16,594) | P-value | |
|---|---|---|---|---|
| Age (years) | Median (IQR) | 73.63 (68.66, 79.20) | 75.33 (69.34, 81.61) | <0.001 |
| ≤64 | 4,703 (10.85) | 1,730 (10.43) | ||
| 65–74 | 19,927 (45.98) | 6,309 (38.02) | ||
| 75–84 | 14,954 (34.51) | 6,202 (37.37) | ||
| ≥85 | 3,752 (8.66) | 2,353 (14.18) | ||
| Sex | Male | 25,539 (58.93) | 9,825 (59.21) | 0.54 |
| Female | 17,797 (41.07) | 6,769 (40.79) | ||
| Race | White | 36,164 (83.45) | 11,658 (7025) | <0.001 |
| Black | 5,045 (11.64) | 2,707 (16.31) | ||
| Asian | 361 (0.83) | 281 (1.69) | ||
| Hispanic | 680 (1.57) | 1,338 (8.06) | ||
| Other or Unknown | 1,086 (83.45) | 610 (3.68) | ||
| Comorbidities | ESRD | 10,720 (24.74) | 4,882 (29.42) | <0.001 |
| Diabetes | 19,883 (45.88) | 9,206 (55.48) | <0.001 | |
| Hypertension | 38,685 (89.27) | 14,702 (88.60) | 0.02 | |
| Ever Smoking | 15,144 (34.95) | 3,606 (21.73) | <0.001 | |
| Population Density of Residence | Urban | 33,836 (78.08) | 13,449 (81.05) | <0.001 |
| Rural | 9,500 (21.60) | 3,145 (18.95) | ||
| Census Region of Residence | Midwest | 9,361 (21.60) | 2,761 (16.64) | <0.001 |
| Northeast | 6,102 (14.08) | 1,532 (9.23) | ||
| South | 21,862 (50.45) | 7,810 (47.07) | ||
| West | 5,974 (13.79) | 4,447 (26.80) | ||
| Other | 37 (0.09) | 44 (0.27) | ||
| Socioeconomic Status | 1st Quartile ($0–45999) | 12,510 (28.87) | 5,934 (35.76) | <0.001 |
| 2nd Quartile ($46000–60999) | 13,869 (32.00) | 5,166 (31.13) | ||
| 3rd Quartile ($61000–81999) | 9,826 (22.67) | 3,290 (19.83) | ||
| 4th Quartile ($82000+) | 6,987 (16.12) | 2,135 (12.87) | ||
| Unknown | 144 (0.33) | 69 (0.42) | ||
PVI is peripheral vascular intervention. IQR is interquartile range. ESRD is end-stage renal disease.
The median age of patients who underwent tibial PVI (75.3 years, IQR 69.3, 81.6 years) was higher than that of patients who underwent non-tibial PVI (73.6 years, IQR 68.7, 79.2 years) (P<0.001). Patients undergoing tibial PVI were more frequently of non-Hispanic Black or Hispanic race/ethnicity, had end-stage renal disease and diabetes, and were less frequently smokers compared to patients who did not receive a tibial PVI (all, P>0.05; Table 1). There were marked discrepancies in geographic residence between groups, with patients living in urban areas and in the Western region of the US receiving significantly more tibial PVI (both, P<0.001; Table 1).
Physician Cohort
Of the 1,542 physicians who performed >10 PVI during the study period included in this study, the mean tibial PVI rate was 26.2 ± 22.6% and the median tibial PVI rate was 20.0% (IQR 9.1%, 37.5%) (Figure 1). The fewest number of PVI procedures performed by a single physician during the study period was 11 and the most was 550. Physicians with higher rates of tibial PVI (i.e., quartile 4, or tibial PVI rate ≥37.5%) more frequently practiced in the Western region, had a primary specialty of radiology, had a high-volume practice performing PVI for claudication, and delivered a high overall percentage of services in an ASC or OBL (all, P<0.001; Table 2). There were no significant differences in physician sex, years since medical school graduation, or population density of practice location between the two groups (all, P>0.05; Table 2). The average Medicare reimbursement per patient was dramatically higher for physicians performing high rates of tibial PVI compared to physicians performing lower rates of tibial PVI (quartile 4 v quartile 1–3: $12,023.96 vs. $692.31 per patient, P<0.001).
Figure 1.
National distribution of physicians by their tibial peripheral vascular intervention rate for claudication among Medicare beneficiaries from 2017–2019 (N=1,542)
Table 2.
Characteristics of physicians* performing index PVI for claudication stratified by tibial PVI rate (< 37.5% vs. ≥ 37.5%)
| Physician Characteristics | Index Tibial PVI Rate <37.5% (N=1,152) | Index Tibial PVI Rate ≥37.5% (N=390) | P-value | |
|---|---|---|---|---|
| Sex | Male | 1,125 (97.66) | 381 (97.69) | 0.97 |
| Female | 27 (2.34) | 9 (2.31) | ||
| Years since medical school graduation | Median (IQR) | 24 (18, 31) | 23 (17, 32) | 0.28 |
| ≤10 years | 25 (2.17) | 20 (5.13) | ||
| 11–20 years | 357 (30.99) | 135 (34.62) | ||
| 21–30 years | 434 (37.67) | 109 (27.95) | ||
| ≥31 years | 318 (27.60) | 118 (30.26) | ||
| Unknown | 18 (1.56) | 8 (2.05) | ||
| Census region of practice location | Midwest | 233 (20.23) | 55 (14.10) | <0.001 |
| Northeast | 150 (13.02) | 40 (10.26) | ||
| South | 645 (55.99) | 209 (53.59) | ||
| West | 124 (10.76) | 85 (21.79) | ||
| Other | 0 | 1 (0.26) | ||
| Population density of practice location | Urban | 1,066 (92.53) | 355 (91.03) | 0.34 |
| Rural | 86 (7.47) | 35 (8.97) | ||
| Primary specialty | Vascular surgery | 468 (40.63) | 89 (22.82) | <0.001 |
| Cardiology | 554 (48.09) | 217 (55.64) | ||
| Radiology | 51 (4.43) | 55 (14.10) | ||
| Other | 79 (6.86) | 29 (7.44) | ||
| Number of patients treated with PVI during the study period | Median (IQR) | 17 (13, 25) | 22 (15, 39) | <0.001 |
| 11–15 | 486 (42.19) | 106 (27.18) | ||
| 16–24 | 368 (31.94) | 102 (26.15) | ||
| 23–550 | 298 (25.87) | 182 (46.67) | ||
| Overall percentage of services delivered in ASC or OBL | Median (IQR) | 59.24 (32.96, 78.37) | 79.18 (52.80, 94.47) | <0.001 |
| 0%−38% | 330 (28.65) | 57 (14.62) | ||
| 39%−64% | 305 (26.48) | 72 (18.46) | ||
| 65%−83% | 299 (25.95) | 88 (22.56) | ||
| 84%−100% | 218 (18.92) | 173 (44.36) | ||
| Average Medicare reimbursement per patient ($USD) | Median (IQR) | 692.31 (562.61, 9,401.54) | 12,023.96 (5,513.30, 14,430.02) | <0.001 |
Included physicians who treated > 10 patients with index peripheral vascular intervention during the study period.
PVI is peripheral vascular intervention. IQR is interquartile range. ESRD is end-stage renal disease. ASC is ambulatory surgical center. OBL is office-based laboratory. USD is United States Dollars.
Patient and Physician Characteristics Associated with Tibial PVI
After adjusting for patient and physician characteristics in a hierarchical logistic regression model, patient-level characteristics associated with tibial PVI for claudication included male sex (adjusted odds ratio [aOR] 1.23, 95% CI 1.16, 1.30), increasing age (aOR1.30–1.96, 95% CI 1.22, 2.15), Black race (aOR 1.47, 95% CI 1.35, 1.60), Hispanic ethnicity (aOR 1.86, 95% CI 1.58, 2.20), and diabetes (aOR 1.36, 95% 1.29, 1.44). Patients with hypertension (aOR 0.89, 95% 0.81, 0.97), with a history of smoking (aOR 0.61, 95% 0.57, 0.65), and with increasing median household income (quartile 4 vs. quartile 1, aOR 0.88, 95% CI 0.88–0.98) were less likely to receive a tibial PVI (Table 3).
Table 3.
Hierarchical logistic regression model (OR, 95% CI) assessing patient- and physician-level characteristics associated with tibial PVI for claudication
| Patient-level characteristics | Adjusted Odds Ratio (95% CI) | |
|---|---|---|
| Age (years) | ≤64 | 1.01 (0.92, 1.12) |
| 65–74 | Ref | |
| 75–84 | 1.30 (1.22, 1.38) | |
| ≥85 | 1.96 (1.80, 2.15) | |
| Sex | Male | Ref |
| Female | 0.81 (0.77, 0.86) | |
| Race | White | Ref |
| Black | 1.47 (1.35, 1.60) | |
| Asian | 1.02 (0.79, 1.32) | |
| Hispanic | 1.86 (1.58, 2.20) | |
| Other or unknown | 1.23 (1.04, 1.45) | |
| ESRD | Yes | 1.10 (1.03, 1.17) |
| No | Ref | |
| Diabetes | Yes | 1.36 (1.29, 1.44) |
| No | Ref | |
| Hypertension | Yes | 0.89 (0.81, 0.97) |
| No | Ref | |
| Ever-Smoking | Yes | 0.61 (0.57, 0.65) |
| No | Ref | |
| Median Household Income ($USD) | 1st Quartile ($0–45999) | Ref |
| 2nd Quartile ($46000–60999) | 0.94 (0.88, 1.01) | |
| 3rd Quartile ($61000–81999) | 0.90 (0.82, 0.97) | |
| 4th Quartile ($82000+) | 0.88 (0.80, 0.98) | |
| Unknown | 0.24 (0.81, 1.90) | |
| Physician-level characteristics | ||
| Sex | Male | Ref |
| Female | 1.18 (0.77, 1.81) | |
| Years since medical school graduation | ≤10 years | 2.97 (2.02, 4.39) |
| 11–20 years | 1.21 (1.03, 1.43) | |
| 21–30 years | 0.79 (0.67, 0.92) | |
| ≥31 years | Ref | |
| Census region of practice location | Midwest | 1.01 (0.85, 1.20) |
| Northeast | 0.84 (0.68, 1.03) | |
| South | Ref | |
| West | 1.79 (1.48, 2.16) | |
| Population density of practice location | Urban | Ref |
| Rural | 1.30 (1.03, 1.66) | |
| Primary specialty | Vascular surgery | Ref |
| Cardiology | 1.66 (1.44, 1.91) | |
| Radiology | 2.94 (2.26, 3.82) | |
| Other | 1.37 (1.05, 1.78) | |
| Number of patients treated with PVI during the study period | 11–15 | Ref |
| 16–24 | 1.18 (1.01, 1.39) | |
| 25–550 | 1.86 (1.59, 2.18) | |
| Overall percentage of services delivered in ASC or OBL | 0%–38% | Ref |
| 39%–64% | 1.32 (1.10, 1.59) | |
| 65%–83% | 1.39 (1.15, 1.68) | |
| 84%–100% | 2.38 (1.97, 2.87) | |
OR is odds ratio. CI is confidence interval. PVI is peripheral vascular intervention. IQR is interquartile range. ESRD is end-stage renal disease. ASC is ambulatory surgical center. OBL is office-based laboratory.
Physician-level characteristics associated with tibial PVI for claudication included early-career status (≤10 years since medical school graduation: aOR 2.97, 95% CI 2.02, 4.39), practice location in the West (vs. South, aOR 1.79, 95% CI 1.48, 2.16), high-volume PVI practice (aOR 1.86, 95% CI 1.59, 2.18), and increasing proportion of practice in an ASC/OBL setting (aOR 1.32 – 2.38, Table 3). Physician specialty was also significantly associated with tibial PVI for claudication: radiologists (aOR 2.94, 95% CI 2.26, 3.82), cardiologists (aOR 1.66, 95% 1.44, 1.91), and other specialties (aOR 1.37, 95% CI 1.05–1.78) were significantly more likely to perform a tibial PVI for claudication compared to vascular surgeons.
Discussion:
Our findings show that a large proportion of Medicare beneficiaries undergo tibial PVI for claudication. Since 2015, the SVS guidelines for lower extremity peripheral artery disease have recommended against the use of tibial interventions for claudication, and specifically isolated tibial interventions1. Despite this, 27.7% of all Medicare patients undergoing PVI for claudication received a tibial intervention, and 38.5% of those tibial PVI were performed in isolation, without a concomitant femoropopliteal PVI. Furthermore, as many as 29.6% of patients undergoing a tibial PVI received a multi-tibial intervention. We found that most of these procedures were performed by radiologists and cardiologists rather than vascular surgeons. Our study is one of the first to characterize the practice patterns surrounding tibial interventions for claudication using national data.
Overall, there is a paucity of data summarizing practice patterns around tibial PVI for claudication. Data from the Vascular Study Group of New England between 2003 and 2018 suggested that tibial interventions were performed in 5.7% of cases for claudication, and that isolated tibial interventions were performed in 1.7% of patients11. Data from the Vascular Quality Initiative during the same timeframe suggested that tibial interventions were performed in approximately 11% of PVI for claudication12. Both of these tibial PVI frequencies are much lower than the 27.7% that we report. However, the previous analyses were limited to institutions that participate in quality registries. The Vascular Quality Initiative has grown substantially over time to include 938 medical centers in the United States13, but is still limited to voluntarily participating groups. Our analysis has a broader scope in that it captures 100% Medicare fee-for-service beneficiaries treated by any physician in the US, and in a more contemporary timeframe, during which endovascular interventions have become much more popular14. Notably, both prior analyses showed that patients undergoing tibial interventions for claudication had a higher risk of major amputation compared to patients undergoing more proximal interventions11, 12.
We found a few patient variables associated with tibial PVI for claudication. Patients of Black and Hispanic race/ethnicity had higher tibial PVI rates compared to White patients, as did patients with diabetes. In contrast, patients with a history of smoking had lower rates of tibial PVI. The racial/ethnic disparities we report are consistent with our previous analysis of PVI for claudication, which showed a higher rate of PVI for claudication in general for Black patients15. The associations of diabetes, end-stage renal disease, and never smoking with tibial PVI is consistent with the general patterns of disease, as diabetes and end-stage renal disease more frequently affect the infrapopliteal arteries in peripheral artery disease16, whereas smoking typically results in femoropopliteal atherosclerosis17.
We also identified several notable physician characteristics associated with tibial PVI. Specifically, physicians with high-volume PVI practices, those practicing in outpatient settings, and non-vascular surgeons were more likely to perform tibial PVI for claudication. The specialty differences we observed may reflect differing opinions across professional societies. In April 2022, the SVS released new appropriate use criteria (AUC) for the management of intermittent claudication18. These AUC determined that infrapopliteal interventions have greater risks than benefits for intermittent claudication, and should not be performed in isolation or downstream from femoropopliteal revascularization18. This was unanimously agreed upon by a multidisciplinary panel of physicians, including cardiologists nominated by the American College of Cardiology, radiologists nominated by the Society of Interventional Radiology, and vascular surgeons in the SVS18. Preceding this, the ACC, AHA, Society for Cardiovascular Angiography and Interventions (SCAI), Society of Interventional Radiology (SIR), and Society of Vascular Medicine (SVM) released AUC for peripheral artery interventions in 2018. The previous AUC used a different tone than the most recent AUC by the SVS, declaring that infrapopliteal PVI for claudication “may be appropriate” in special circumstances19. Of note, the SVS was not included in the 2018 committee, and did not formally endorse those criteria. Thus, the recently published AUC by the SVS explicitly emphasized the importance of multidisciplinary representation, drawing attention to the fact that there has been controversy surrounding the management of claudication. It will be important to evaluate if the practice patterns surrounding the use of tibial PVI for claudication will change in the coming years, following publication of the recent multispecialty AUC.
The association of high-volume PVI practices and outpatient-based practices with tibial PVI for claudication were also notable. Physicians performing as many as 82–100% of their total PVI cases in the ASC/OBL setting had greater odds of performing tibial PVI than physicians who performed most of their cases in a hospital-based setting. There was a clear incremental association of ASC/OBL practice volume with tibial PVI rate, and the average Medicare reimbursement per patient was dramatically higher for physicians performing a high volume of tibial PVI than for physicians performing lower volumes of tibial PVI. While indications for tibial PVI cannot be elucidated from our analysis, these findings suggest a possible financial incentive related to tibial interventions for claudication. We have similarly shown high rates of atherectomy for claudication in outpatient versus hospital-based settings that are closely associated with reimbursement5. Another study using Medicare claims data also found higher rates of PVI being performed in outpatient versus in-hospital settings, also believed to be associated with Medicare reimbursement policies20. Although the differences in total reimbursement can, in part, be explained by added facility fees that are reimbursed in the ASC/OBL setting but not the hospital-based setting, this does not explain the disproportionately high rate of PVI being performed in the ASC/OBL setting overall. Taken together, there may be an opportunity for a quality improvement initiative that directly addresses the use of low-value procedures in high reimbursement settings that are potentially not indicated.
Finally, we found that physicians earlier in their career had nearly three-fold greater odds of performing tibial PVI than their more senior counterparts. This could potentially be explained by endovascular interventions becoming increasingly common with time14, 21, and/or the expansion of PVI to other (non-vascular) specialty training. Given the widespread use of PVI to treat claudication across a range of specialties, a common professional stance is critical to standardize the use of rapidly evolving endovascular technologies and their applications and indications for use.
Our study has a number of limitations. The analysis is limited to the Medicare population and may not reflect practice patterns surrounding tibial PVI for claudication in other patient populations. Our patient-level findings are also limited to the degree of detail that can be determined from CPT codes, such that the exact tibial vessel upon which the physician performed the procedure is unknown, as well as any additional details about anatomic pattern or severity of disease. For the same reason, we also could not ascertain the clinical reasoning that each physician had in performing tibial PVI. As noted above, the 2018 AUC by the AHA/ACC/SCAI/SIR/SVM deemed tibial interventions for claudication appropriate in special circumstances19, the details of which cannot be elucidated from claims data. Importantly, we focused on investigating practice patterns surrounding this procedure, rather than subsequent outcomes. Prior registry studies on the topic have reported higher rates of major amputation among patients who undergo tibial versus femoropopliteal interventions for claudication11, 12, but a national study to investigate the long-term outcomes associated with tibial PVI for claudication is still needed to fully appreciate the implications of this study. Finally, because of the clear practice variance by specialty, future research exploring the National Radiology Data Registry for Interventional Radiology (NRDR) and National Cardiovascular Data Registry Peripheral Vascular Intervention Registry™ (PVI Registry™) is also warranted.
Conclusion:
Tibial PVI is performed at high rates among Medicare patients with claudication. There are apparent practice discrepancies by specialty that likely reflect differing expert opinions on the utility of tibial PVI for claudication1, 2, 19 which have varied substantially until the recent publication of the 2022 SVS AUC18. Notably, we observed a higher rate of tibial PVI among physicians with high-volume PVI practices and majority practice in ASC/OBL settings. Overall, our data show widely varied practice patterns in the endovascular treatment of claudication, suggesting a critical need for evidence-based education, guidelines, and reimbursement policies surrounding these practices.
Supplementary Material
ARTICLE HIGHLIGHTS.
Type of Research:
A retrospective analysis of 100% of Medicare fee-for-service claims collected between 2017 and 2019.
Key Findings:
Among 59,930 Medicare patients who underwent a peripheral vascular intervention (PVI) for claudication between 2017 and 2019, 16,594 (27.7%) underwent a tibial PVI. After adjusting for patient and physician characteristics and practice factors, physicians who did not identify as vascular surgeons, performed a high volume of PVI for claudication, and conducted most of their cases in high-reimbursement settings during the study period were significantly more likely to perform high rates of tibial PVI.
Take Home Message:
Tibial PVI is performed in more than a quarter of patients with claudication, most frequently by non-vascular surgeons in high-volume, high-reimbursement settings. This reveals a critical need to reevaluate the indications, education, and reimbursement policies surrounding these procedures.
Disclosure:
CWH is funded by a grant from the National Institute of Health (NIH)/National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK), K23DK124515. MSC receives financial contributions from Abbott Vascular (DSMB) and is funded by a research grant from Profusa, Inc. The remaining authors have no competing financial interests.
Footnotes
Publisher's Disclaimer: This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Presented at the Vascular & Endovascular Surgery Society (VESS) Spring Meeting on June 15, 2022, in Boston, MA.
References
- 1.Society for Vascular Surgery Lower Extremity Guidelines Writing G, Conte MS, Pomposelli FB, Clair DG, Geraghty PJ, McKinsey JF, et al. Society for Vascular Surgery practice guidelines for atherosclerotic occlusive disease of the lower extremities: management of asymptomatic disease and claudication. J Vasc Surg. 2015;61(3 Suppl):2S–41S. [DOI] [PubMed] [Google Scholar]
- 2.Gerhard-Herman MD, Gornik HL, Barrett C, Barshes NR, Corriere MA, Drachman DE, et al. 2016 AHA/ACC Guideline on the Management of Patients With Lower Extremity Peripheral Artery Disease: Executive Summary: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. J Am Coll Cardiol. 2017;69(11):1465–508. [DOI] [PubMed] [Google Scholar]
- 3.Hossain P, Kokkinidis DG, Armstrong EJ. How To Assess a Claudication and When To Intervene. Curr Cardiol Rep. 2019;21(12):138. [DOI] [PubMed] [Google Scholar]
- 4.Bath J, Lawrence PF, Neal D, Zhao Y, Smith JB, Beck AW, et al. Endovascular interventions for claudication do not meet minimum standards for the Society for Vascular Surgery efficacy guidelines. Journal of Vascular Surgery. 2021;73(5):1693–700.e3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Hicks CW, Holscher CM, Wang P, Dun C, Abularrage CJ, Black JH 3rd, et al. Use of Atherectomy During Index Peripheral Vascular Interventions. JACC Cardiovasc Interv. 2021;14(6):678–88. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.ResDAC. Master Beneficiary Summary File (MBSF) Base. In: Center RDA, editor. 2019. [Google Scholar]
- 7.National Bureau of Economic Research. Census Core-Based Statistical Area (CBSA) to Federal Information Processing Series (FIPS) County Crosswalk.
- 8.Agency for Healthcare Research and Quality. 2018 (XLSX, 38MB). 2018. [DOI] [PubMed]
- 9.Centers for Medicare & Medicaid Services (CMS). National Downloadable File.
- 10.Hicks CW, Holscher CM, Wang P, Black JH 3rd, Abularrage CJ, Makary MA. Overuse of early peripheral vascular interventions for claudication. J Vasc Surg. 2020;71(1):121–30 e1. [DOI] [PubMed] [Google Scholar]
- 11.Levin SR, Farber A, Cheng TW, Arinze N, Jones DW, Rybin D, et al. Patients undergoing interventions for claudication experience low perioperative morbidity but are at risk for worsening functional status and limb loss. J Vasc Surg. 2020;72(1):241–49. [DOI] [PubMed] [Google Scholar]
- 12.Mullins CH, Novak Z, Axley JC, Sutzko DC, Spangler EL, Pearce BJ, et al. Prevalence and Outcomes of Endovascular Infrapopliteal Interventions for Intermittent Claudication. Ann Vasc Surg. 2021;70:79–86. [DOI] [PubMed] [Google Scholar]
- 13.Society for Vascular Surgery Vascular Quality Initative. VQI Participating Centers - May 2022. In: VQI-Participants-May-2022, editor. 2022. p. Latest Monthly Updates. [Google Scholar]
- 14.Goodney PP, Beck AW, Nagle J, Welch HG, Zwolak RM. National trends in lower extremity bypass surgery, endovascular interventions, and major amputations. J Vasc Surg. 2009;50(1):54–60. [DOI] [PubMed] [Google Scholar]
- 15.Hicks CW, Wang P, Bruhn WE, Abularrage CJ, Lum YW, Perler BA, et al. Race and socioeconomic differences associated with endovascular peripheral vascular interventions for newly diagnosed claudication. J Vasc Surg. 2020;72(2):611–21 e5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Mustapha JA, Diaz-Sandoval LJ, Saab F. Infrapopliteal calcification patterns in critical limb ischemia: diagnostic, pathologic and therapeutic implications in the search for the endovascular holy grail. J Cardiovasc Surg (Torino). 2017;58(3):383–401. [DOI] [PubMed] [Google Scholar]
- 17.Thiruvoipati T, Kielhorn CE, Armstrong EJ. Peripheral artery disease in patients with diabetes: Epidemiology, mechanisms, and outcomes. World J Diabetes. 2015;6(7):961–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Woo K, Siracuse JJ, Klingbeil K, Kraiss LW, Osborne N, Singh N, et al. Society for Vascular Surgery Appropriate Use Criteria for Management of Intermittent Claudication. J Vasc Surg. 2022. [DOI] [PubMed] [Google Scholar]
- 19.Bailey SR, Beckman JA, Dao TD, Misra S, Sobieszczyk PS, White CJ, et al. ACC/AHA/SCAI/SIR/SVM 2018 Appropriate Use Criteria for Peripheral Artery Intervention. Journal of the American College of Cardiology. 2019;73(2):214–37. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Turley RS, Mi X, Qualls LG, Vemulapalli S, Peterson ED, Patel MR, et al. The Effect of Clinical Care Location on Clinical Outcomes After Peripheral Vascular Intervention in Medicare Beneficiaries. JACC Cardiovasc Interv. 2017;10(11):1161–71. [DOI] [PubMed] [Google Scholar]
- 21.Thukkani AK, Kinlay S. Endovascular intervention for peripheral artery disease. Circ Res. 2015;116(9):1599–613. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.