Use of peripheral vascular intervention (PVI) for intermittent claudication (IC) continues to expand, but there is uncertainty whether baseline demographics, procedural techniques and outcomes differ by sex, race, and ethnicity. This study aimed to examine amputation and revascularization rates up to 4 years after femoropopliteal (FP) PVI for IC by sex, race, and ethnicity. Patients who underwent FP PVI for IC between 2016 and 2020 from the PINC AI Healthcare Database were analyzed. The primary outcome was any index limb amputation, assessed by Kaplan-Meier estimate. Secondary outcomes included index limb major amputation, repeat revascularization, and index limb repeat revascularization. Unadjusted and adjusted hazard ratios (HRs) were estimated using Cox proportional hazard regression models. This study included 19,324 patients with IC who underwent FP PVI, with 41.2% women, 15.6% Black patients, and 4.7% Hispanic patients. Women were less likely than men to be treated with atherectomy (45.1% vs 47.8%, p = 0.0003); Black patients were more likely than White patients to receive atherectomy (50.7% vs 44.9%, p <0.001), and Hispanic patients were less likely than non-Hispanic patients to receive atherectomy (41% vs 47%, p = 0.0004). Unadjusted rates of any amputation were similar in men and women (6.4% for each group, log-rank p = 0.842), higher in Black patients than in White patients (7.8% vs 6.1%, log-rank p = 0.007), and higher in Hispanic patients than in non-Hispanic patients (8.8% vs 6.3%, log-rank p = 0.031). After adjustment for baseline characteristics, Black race was associated with higher rates of repeat revascularization (adjusted HR 1.13, 95% confidence interval 1.04 to 1.22) and any FP revascularization (adjusted HR 1.10, 95% confidence interval 1.01 to 1.20). No statistical difference in amputation rate was observed among comparison groups. Women and men with IC had similar crude and adjusted amputation and revascularization outcomes after FP PVI. Black patients had higher repeat revascularization and any FP revascularization rates than did White patients. Black and Hispanic patients had higher crude amputation rates, but these differences were attenuated by adjustment for baseline characteristics. Black patients were more likely to receive atherectomy and had higher rates of any repeat revascularization and specifically FP revascularization. Further study is necessary to determine whether these patterns are related to disease-specific issues or practice-pattern differences among different populations.
Peripheral artery disease (PAD) is a prevalent disease that increases with age, affecting up to 15% of subjects aged >70 years in the United States, with known sex-based, racial, and ethnic differences. , Differences in outcomes after peripheral vascular intervention (PVI) exist and may be increasing in certain populations. , In short-term follow-up, women (combined intermittent claudication [IC] and chronic limb-threatening ischemia [CLTI]) experienced more periprocedural bleeding and in-hospital major amputations, and higher mortality rates than did men after PVI. For both Black and Hispanic patients, there were increased rates of limb amputation after revascularization compared with White patients. Whether these risks are due to disease severity at presentation, co-morbid risk factors, intensity of local vascular care, or differences in social determinants of health has not been well characterized.
Known outcome disparities after PVI by sex, race, and ethnicity are based on short-term and midterm studies of all-comers with PAD, including patients at greatest risk of limb loss presenting with CLTI. Less is known about those presenting in earlier stages of PAD with IC alone. Older data in women with IC after PVI from the Vascular Quality Initiative (VQI) registry show no differences in outcomes for reintervention, amputation, and survival at midterm follow-up. However, a study in Black and Hispanic patients with IC from the VQI registry identified these cohorts as having excess risk for major adverse limb events at 1-year follow-up. There are concerns around the overuse of invasive therapies for IC with concomitant underuse of guideline-directed medical therapies, with signals from real-world practice that PVI procedures in patients with IC may increase the risk of progression to CLTI and amputation compared with optimal medical management. , In light of this, identifying the presence of specific sex-, race-, and ethnicity-based differences associated with amputation risk is critical. This is compounded by femoropopliteal (FP) PVI being the most performed invasive therapy for lower extremity PAD, usually for the invasive management of IC.
With the proliferation of less invasive PVI modalities, there has been a shift to treating patients at earlier disease stages, for example, patients with IC and not CLTI, and in less acute settings, outpatient-based laboratories as opposed to hospital-based (OBLs). Despite these trends toward increased use, information regarding the longer-term outcomes after FP PVI for IC is incomplete. To address these gaps in knowledge, we sought to assess whether longer-term outcome rates of repeat revascularization and amputation in patients with IC who undergo FP PVI differ by sex, race, and ethnicity.
Methods
Data for this analysis were derived from the PINC AI Healthcare Data (formerly known as the Premier Healthcare Database, Premier Inc., Charlotte, North Carolina), which is an all-payer, geographically diverse United States database including >1,300 hospitals and 1.3 billion hospital encounters, populated with detailed patient and visit level data, including demographic information, date-stamped data from billed services (diagnostic and therapeutic), and information on hospital characteristics. Patients in this database can be tracked across inpatient and hospital-based outpatient encounters that occurred in the same hospital or hospital system over time. This study was deemed exempt from the Yale University Human Investigation Committee because it used deidentified data.
This study included patients aged ≥18 years with an International Classification of Diseases-Tenth Revision (ICD-10) diagnosis code or Current Procedural Terminology (CPT) code for IC discharged between January 1, 2016 and December 31, 2020. All ICD-10 and CPT codes selected by investigator consensus that were used to define the study population are included in Supplementary Table 1 . Exclusions included any patients who underwent interventions in other vascular beds, with a history of amputation, with previous lower extremity surgical bypass, and with CLTI. Included patients must have undergone a first instance of FP PVI as determined by ICD-10 procedure codes rather than CPT codes, given that laterality of PVI (i.e., left or right) was not discernible in some cases (codes in Supplementary Table 1 ). If a patient had multiple PVI procedures during the study period, the first was considered the index procedure. Study outcomes were compared in the following groups: women versus men, Black versus White, and Hispanic versus non-Hispanic.
The primary outcome was a composite of minor (below-the-ankle) and major (above-the-ankle) amputation on the index limb (ipsilateral to the index revascularization). Secondary outcomes included individual components of major amputation and rates of repeat revascularization (total and by index limb). The definition of “any amputation” or “any repeat revascularization” was not limited to the index limb, and index limb is related to the initial sidedness of the PVI procedure.
Continuous data were presented as means with SDs, based on the normality assumption for large sample size, and their means were compared using the Student’s t test. Categorical data were presented as frequencies and percentages and compared using a chi-square test or Fisher’s exact test. End of follow-up was defined as end of study period or patient death. Cumulative incidence rates of each outcome (e.g., amputation [both index limb major and all], repeat revascularization [both total and index limb]), were calculated using Kaplan-Meier, with the date of PVI as day 0. The log-rank test was used to compare the subgroups. Cox proportional hazard regression models were used to calculate hazard ratios (HRs) and associated 95% confidence intervals with adjustment for risk factors statistically or clinically associated with PAD and included chronic kidney disease (CKD), acute renal failure, congestive heart failure, diabetes, hypertension, hyperlipidemia, chronic obstructive pulmonary disease, smoking, coronary artery disease, cerebrovascular disease, and obesity. Analyses were performed using SAS software, version 9.4 (SAS Institute Inc., Cary, North Carolina).
Results
The study population included 19,324 patients with IC who underwent PVI in the FP segments: 41.2% were female; 13.9% were Black, and 4.7% were Hispanic. Women were older than men at index procedure ( Table 1 ) and were more likely to have congestive heart failure, diabetes, and obesity although less likely than men to have coronary artery disease and smoking history. Black patients were younger and more likely to have CKD, diabetes, smoking history, and obesity than were White patients ( Table 1 ). Hispanic patients were more likely to have CKD, diabetes, obesity, and coronary artery disease and less likely to smoke and have hypertension than were non-Hispanic patients ( Table 1 ).
| Female (n=7958) | Male (n=11364) | P-value | Black (n=2679) | White (n=14513) | P-value | Hispanic (n=907) | Not Hispanic (n=18417) | P-value | |
|---|---|---|---|---|---|---|---|---|---|
| Age (years +/- SD) | 70.3 ± 11.4 | 67.7 ± 10.1 | <.0001 | 64.9 ± 10.6 | 69.4 ± 10.6 | <.0001 | 68.2 ± 11.5 | 68.8 ± 10.7 | 0.395 |
| Sex | |||||||||
| Male | 1379 (51.5%) | 8786 (60.5%) | <.0001 | 538 (59.3%) | 10826 (58.8%) | 0.756 | |||
| Female | 1300 (48.5%) | 5727 (39.5%) | <.0001 | 369 (40.7%) | 7589 (41.2%) | 0.782 | |||
| Race | |||||||||
| Asian | 76 (1.0%) | 108 (1.0%) | 1.0000 | 3 (0.3%) | 181 (1.0%) | 0.051 | |||
| Black | 1300 (16.3%) | 1379 (12.1%) | <.0001 | 26 (2.9%) | 2653 (14.4%) | <.0001 | |||
| White | 5727 (72.0%) | 8786 (77.3%) | <.0001 | 592 (65.3%) | 13921 (75.6%) | <.0001 | |||
| Other | 727 (9.1%) | 897 (7.9%) | 0.0024 | 229 (25.2%) | 1396 (7.6%) | <.0001 | |||
| Unknown | 128 (1.6%) | 194 (1.7%) | 0.6079 | 57 (6.3%) | 266 (1.4%) | <.0001 | |||
| Hispanic | 369 (4.6%) | 538 (4.7%) | 0.7822 | 26 (1.0%) | 592 (4.1%) | <.0001 | |||
| CKD | 1526 (19.2%) | 2096 (18.4%) | 0.2029 | 649 (24.2%) | 2494 (17.2%) | <.0001 | 234 (25.8%) | 3388 (18.4%) | <.0001 |
| Acute Kidney Failure | 425 (5.3%) | 590 (5.2%) | 0.6467 | 164 (6.1%) | 741 (5.1%) | 0.034 | 81 (8.9%) | 934 (5.1%) | <.0001 |
| Congestive heart failure | 818 (10.3%) | 913 (8.0%) | <.0001 | 250 (9.3%) | 1264 (8.7%) | 0.299 | 92 (10.1%) | 1640 (8.9%) | 0.2109 |
| Diabetes | 3297 (41.4%) | 4,422 (38.9%) | 0.001 | 1225 (45.7%) | 5331 (36.7%) | <.0001 | 508 (56.0%) | 7212 (39.2%) | <.0001 |
| Hypertension | 4444 (55.8%) | 6306 (55.5%) | 0.638 | 1504 (56.1%) | 8061 (55.5%) | 0.582 | 471 (51.9%) | 10280 (55.8%) | 0.022 |
| Hyperlipidemia | 5140 (64.6%) | 7318 (64.4%) | 0.795 | 1719 (64.2%) | 9235 (63.6%) | 0.615 | 613 (67.6%) | 11847 (64.3%) | 0.047 |
| COPD | 1600 (20.1%) | 2136 (18.8%) | 0.024 | 393 (14.7%) | 3078 (21.2%) | <.0001 | 102 (11.2%) | 3634 (19.7%) | <.0001 |
| Any Nicotine | 2018 (25.4%) | 3491 (30.7%) | <.0001 | 882 (32.9%) | 4132 (28.5%) | <.0001 | 192 (21.2%) | 5317 (28.9%) | <.0001 |
| Smoking | 2015 (25.3%) | 3468 (30.5%) | <.0001 | 879 (32.8%) | 4109 (28.3%) | <.0001 | 192 (21.2%) | 5291 (28.7%) | <.0001 |
| CAD | 3518 (44.2%) | 5831 (51.3%) | <.0001 | 1149 (42.9%) | 7064 (48.7%) | <.0001 | 484 (53.4%) | 8867 (48.1%) | 0.002 |
| CVD | 790 (9.9%) | 1042 (9.2%) | 0.0807 | 244 (9.1%) | 1407 (9.7%) | 0.354 | 77 (8.5%) | 1755 (9.5%) | 0.324 |
| Obesity (BMI>30) | 1035 (13.0%) | 1081 (9.5%) | <.0001 | 353 (13.2%) | 1513 (10.4%) | <.0001 | 123 (13.6%) | 1994 (10.8%) | 0.012 |
| Acute Respiratory Failure | 201 (2.5%) | 187 (1.6%) | <.0001 | 40 (1.5%) | 319 (2.2%) | 0.018 | 27 (3.0%) | 361 (2.0%) | 0.039 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
