Abstract
Background:
The choice of incident amputation level can have a profound effect on clinical outcomes. Amputations at the transmetatarsal (TM) or transtibial (TT) levels result in greater preservation of function and mobility, whereas transfemoral (TF) amputations typically result in a greater adverse impact. Prior investigations have explored racial/ethnic and regional variation in incident amputation level. This study overcomes some of the methodological limitations seen in prior research through the use of a large national, multiyear veteran sample and by including only those who have undergone an incident amputation.
Objectives:
(1) Determine if there are national/regional differences in the frequency of incident TF amputation compared with TM and TT amputation, (2) Determine if race/ethnicity and geographic region are associated with incident TF amputation level, and (3) Determine if racial/ethnic disparities of incident TF amputation differ by the presence of diabetes or prior revascularization.
Design:
Retrospective cohort study of veterans undergoing an incident dysvascular lower extremity amputation.
Setting:
One hundred ten Veterans Affairs (VA) Medical Centers.
Participants:
Seven thousand two hundred ninety-six Veterans undergoing incident unilateral dysvascular lower extremity amputation identified in the Veterans Affairs Surgical Quality Improvement Program (VASQIP) database (2005–2014).
Interventions:
Not applicable.
Main Outcome Measure:
Incident amputation level.
Results:
The White, Black, and Hispanic risk for an incident TF amputation was 31% (n = 1356), 35% (n = 810), and 46% (n = 293), respectively. In the Continental region, Blacks who had not had a prior revascularization were more likely to undergo a TF amputation compared to Whites both with and without diabetes (odds ratio [OR] = 1.4; 95% confidence interval [CI], 1.1, 1.9 and OR = 1.5; 95% CI, 1.1, 2.1, respectively). In the Southeast region, Hispanics compared with Whites were at increased odds of undergoing a TF amputation, irrespective of a diabetes or a prior revascularization (ORs ≥ 2.9).
Conclusions:
Racial and ethnic disparities exist in choice of proximal compared with distal amputation in specific VA geographic regions.
INTRODUCTION
Health and health care disparities are the differences in the quality of health outcomes or access to care that exist between population groups. They may include differences because of race, ethnicity, sexual orientation, gender identity, socioeconomic status, gender, disability status, and region.1,2 Although some health care disparities in the United States have improved, according to the 2018 Agency for Healthcare Research and Quality report on Healthcare Quality and Disparities; there are still persistent disparities with Blacks and Hispanics having poorer outcomes.1
Research on health care disparities in patients undergoing dysvascular amputations have primarily focused on amputation risk.3-11 People of color, older patients, and those who are homeless, uninsured, use public insurance or have lower socioeconomic status are all demographic factors that result in an increased risk for lower extremity amputation.3-7 Blacks and Hispanics have an increased risk of both incident and repeat amputations, compared to non-Hispanic Whites.3,8-11 Similarly, health care disparities have been reported in level of amputation, with proximal amputations occurring more frequently in Blacks, women, those with low income, or older age, and those who use public insurance.3,12-15 Prior research studies evaluating the impact of race and ethnicity on level of amputation have had a number of methodological limitations including inadequate control of comorbid medical and functional factors.3,5-7 Additionally, prior research included primary and secondary amputations, rather than limiting the cohort to incident amputation.3,6,7,13,14 Surgical revisions and subsequent higher level ipsilateral amputations or contralateral amputations are common in this population.16 Not controlling for factors associated with transfemoral (TF) amputation and race and ensuring risk estimates are based on incident amputations can confound results leading to spurious estimates, including racial/ethnic disparities.
The evaluation of disparities in amputation level selection in the dysvascular amputee population is important because amputation level can profoundly affect patient mobility, function, and ability to return to prior levels of activity.17,18 Compared to transtibial (TT) amputation, TF level amputation results in a reduction in the use of a prosthesis, greater limitations in walking, and a greater likelihood of being nonambulatory with a prosthesis.19,20 Together these adverse impacts on mobility result in a reduced likelihood of living independently, compared to more distal amputations18,21 as well as a reduced quality of life,19,20 and an increased mortality rate, both acutely and at 1, 3, and 5 years.22,23
Diabetes and peripheral artery disease (PAD) play an important role in modifying the need for proximal compared with distal lower extremity amputation. Patients with diabetes but without PAD are more likely to require a more distal major amputation,24 whereas patients requiring revascularization, which is a surrogate for large vessel vascular disease, are more likely to require a more proximal lower extremity amputation.25 Racial/ethnic factors, as well as regional disparities in health and health care in patients with diabetes and PAD can also influence the overall risk of amputation and the choice of amputation level. The inception cohort in these prior studies often did not specify whether the amputation was the incident amputation or a secondary/tertiary amputation. This, along with a lack of adequate control of comorbidities, may account for some of the observed differences in amputation level distribution. Because of the profound effect of amputation level on function, there is a need to further define and quantify potential regional and racial/ethnic disparities in amputation level selection. Therefore, the objectives of this investigation were to (1) determine if there are national and regional differences in the choice of incident amputation level, (2) determine if race/ethnicity is associated with incident amputation level nationally and by geographic region, and (3) determine if racial/ethnic disparities in incident amputation level by geographic region differ by the presence of diabetes and a history of prior revascularization. Variables deemed potential confounders and/or effect modifiers of the association are included in an attempt to estimate the most precise associations and to determine if the odds of a TF amputation depend on the presence of other factors such as geographic region, diabetes, and prior revascularization.
METHODS
Study sample
Patients in this study were identified using the Veterans Affairs (VA) Surgical Quality Improvement Program (VASQIP) national data set, which was developed to monitor surgical care quality in the VA health system. VASQIP data are abstracted by nurse data managers from all 110 VA Medical Centers inpatient surgical programs. Data are limited to the first 36 cases performed during an 8-day cycle. At lower volume VA hospitals, data are collected from all abstraction-eligible cases performed during that time period. This method ensures representative samples from each VA hospital’s cases are captured and guarantees the quality and reliability of collected data. 26,27 The VASQIP was used to identify presurgical risk factors, which were supplemented with additional risk factors from the VA Corporate Data Warehouse (CDW). Methods for using the CDW have been reported previously.22,28
Study criteria
Patients undergoing an incident unilateral transmetatarsal (TM), TT, or TF amputation between October 1, 2004 and December 31, 2014 secondary to diabetes and/or PAD and were 40 years of age or older were included in the study. The amputations were presumed to be related to diabetes and/or PAD if the patient was greater than 40 years of age based on prior VASQIP studies23 and had a diagnosis code of diabetes and/or PAD. Previous analyses of VA amputation operations have established that less than 2% of all amputations are related to trauma or lower extremity cancer.29 The inclusion of TM amputation is important because of the increasing trend to attempt limb salvage and their association with high reamputation rates.28,30,31 To confirm the definitive level, we presumed that patients undergoing a guillotine procedure at the TT (Current Procedural Terminology [CPT] 27881, International Classification of Diseases [ICD] 9 84.13) and TF (CPT 27592) levels would have a closure procedure performed within 3 weeks of the guillotine procedure; therefore, we searched forward 3 weeks for the next procedure code to classify the incident amputation level. If the next subsequent procedure occurred greater than 3 weeks after the initial procedure, we presumed the initial guillotine code was an error; we accepted the initial guillotine procedure as a definitive level of amputation and any subsequent procedure as a reamputation. For those coded as guillotine procedures, without a subsequent closure procedure from VASQIP or CDW, VA Informatics and Computing Infrastructure chart annotation services were used to identify the definitive closure level and laterality. ICD Ninth Revision (ICD-9) procedure codes and CPT codes used for establishing amputation level and ICD-9 diagnosis codes used to define risk factors have been reported previously.22,28 Participants were excluded if they had specific preoperative diagnoses of coma, paraplegia, quadriplegia, disseminated cancer, or tumor of the central nervous system or were ventilator dependent. Participants were also excluded for the following circumstances: if they had a body mass index <15 or > 52 kg/m2 because these values are clinically implausible or represent very sick patients who are likely candidates for TF amputation only (Figure 1). This study was approved by the local facility institutional review board.
FIGURE 1.
Strobe diagram depicting total numbers acquired by Veterans Affairs Surgical Quality Improvement Program (VASQIP) and total number excluded to achieve final analysis cohort (N = 7296)
Potential confounders and effect modifiers
We carefully considered several potential confounders and effect modifiers of the association between race and incident TF amputation based on evidence from the literature and clinical experience. This led to the a priori selection of nine possible variables in addition to race/ethnicity, which was our primary exposure of interest. These included demographic factors (age, gender, geographic region), comorbidities (diabetes and renal failure), current smoking status, level of functional dependence, and vascular factors (prior revascularization and abnormal ankle brachial index [ABI] defined as an ABI < .9). These potential confounders or effect modifiers were extracted from the VASQIP and CDW data sources (Table 1). We included any revascularization (open, endo, or a combination) that occurred within the previous 12 months of the incident amputation as a potential effect modifier. We combined these into “any revascularization procedure.” We ensured that all factors preceded the incident amputation. When both data sources included a given risk factor, the VASQIP data were used as the primary data source. When VASQIP data were missing, the most proximate CDW value within 3 months before the date of incident amputation was used. Patients with diabetes (requiring oral agents or insulin) were identified from VASQIP and additional patients were added if one of the ICD-9 codes for diabetes (249.7, 250.7, 785.4, 443.81, 785.4, 249.8, 250.8, and 707.1, .11, .12, .13, .14, .15, .19) existed in the CDW. VA regions were organized according to a five-region national framework implemented in 2015 and include the North Atlantic (ME, VT, NY, NH, MA, CT, RI, NJ, MD, DE, District of Columbia, PA, VA, WV, NC); Southeast (KY, TN, AL, GA, SC, FL); Midwest (ND, SD, NE, KS, MO, IA, MN, WI, IL, MI, IN, OH); Continental (MT, WY, UT, CO, AR, OK, TX, MS, LA); and Pacific (WA, OR, CA, ID, NV, AZ, NM). Regions, diabetes, and prior revascularizations were evaluated as effect modifiers with the a priori hypothesis that these factors may influence the risk for TF amputation and that their presence or absence may be disproportionately distributed among races. Race and ethnicity (primary exposure of interest) were sorted into three categories: Black, Hispanic, or White. Within the databases used, Hispanic does not categorize race and refers to descendents of/or persons from Cuba, Puerto Rico, Central and South America, and Spain.
TABLE 1.
Baseline factors (potential confounders and effect modifiers) by race/ethnicity
| Risk factors | White (N = 4356) % or Mean (SD) |
Black (N = 2302) % or Mean (SD) |
Hispanic (N = 638) % or Mean (SD) |
|---|---|---|---|
| Incident amputation level | |||
| Transmetatarsal | 940 (22) | 475 (21) | 120 (19) |
| Transtibial | 2060 (47) | 1017 (44) | 225 (35) |
| Transfemoral | 1356 (31) | 810 (35) | 293 (46) |
| Demographics | |||
| Age (years) | 67.6 (10.1) | 67.2 (10.9) | 71.5 (11.5) |
| Male | 4315 (99) | 2277 (99) | 633 (99) |
| Geographic region | |||
| Pacific | 1033 (24) | 762 (33) | 70 (11) |
| Midwest | 910 (21) | 567 (25) | 340 (53) |
| Continental | 1022 (23) | 399 (17) | 19 (3) |
| North Atlantic | 655 (15) | 369 (16) | 79 (12) |
| Southeast | 756 (17) | 205 (9) | 130 (20) |
| Comorbidities | |||
| Diabetes | 3185 (73) | 1693 (74) | 559 (88) |
| Kidney failurea | 324 (7) | 338 (15) | 0 (0) |
| Health factors | |||
| Smoker within one year pre-op | 1856 (43) | 910 (40) | 127 (20) |
| Physical function | |||
| Independent | 2377 (55) | 1195 (52) | 229 (36) |
| Partially dependent | 1602 (37) | 816 (35) | 252 (40) |
| Totally dependent | 377 (9) | 291 (13) | 157 (25) |
| Vascular/limb status | |||
| Prior revascularization | 917 (21) | 407 (18) | 66 (10) |
| Abnormal ABI (<.9) | 2269 (52) | 1147 (50) | 140 (22) |
Abbreviations: ABI, ankle brachial index; eGFR, estimated glomerular filtration rate; SD, standard deviation.
eGFR <15 (mL/min/1.73 m2).
Defining incident amputation
To ensure that included participants had undergone an incident unilateral amputation and did not have a history of prior amputation (which is critical to answer our study objectives), eligible patients were linked to the VA CDW and a 5-year look back strategy (defined in prior publications) was employed.22,28 The presence of any diagnostic or procedure code during this interval that was related to amputation or its treatment resulted in the exclusion of those participants. Participants with a prior toe or ray amputation were not excluded from the cohort.
Outcomes
The primary outcome of this study was incident dysvascular amputation level. We were interested in estimating the odds of TF amputation compared to a more distal amputation (ie, incident TT and TM amputation were pooled into one category).
Statistical methods
There were few missing data for race/ethnicity and the nine risk factors and therefore a “complete case” analysis (n = 7296; 98% of those eligible) was used, as opposed to other approaches, such as multiple imputation of missing values. For descriptive statistics, we reported the frequency counts and percentages of the outcome (incident TF and TM + TTl) and the nine potential confounders by race/ethnicity and the distribution of the outcome by region. To evaluate potential effect modification, eight potential interactions with race/ethnicity, based on a priori hypotheses and support from the literature, were considered to evaluate possibly differential effects of race/ethnicity on risk of TF amputation in subgroups. The a priori potential interactions included Black and Hispanic race by Continental and Southeast regions (n = 4), by diabetes (n = 2), and by prior revascularization (n = 2). They were confirmed during our exploratory data analysis where differences in incident amputation level by race, region, diabetes, and revascularization were observed. To estimate the odds of incident TF amputation compared to a more distal amputation comparing Black and Hispanic race to white, while controlling for potential confounders and considering possible effective modification, we performed a multivariable logistic regression including all main effects and interaction terms in the model. Interpretation of main effects from a multivariable logistic regression model in the presence of several interaction terms is complex and challenging and not necessarily clinically meaningful because the odds ratios apply to factors that are not present in the interactions. For example, when the interaction of race and revascularization is included in the model by a specific region (eg, Southeast), then the odds ratio associated with a specific race and prior revascularization applies to other regions other than the Southeast. To provide the reader with more interpretable results, the odds ratios with 95% confidence intervals (CIs) comparing Blacks and Hispanics to Whites in each region stratified by diabetes and prior revascularization were calculated from the linear combination of multiple interaction coefficients from the multivariable model using the lincom command in Stata. Analyses were performed using Stata software, version 15.0 (Stata Corp., College Station, Texas, USA).
RESULTS
The “complete case” cohort included 7296 individuals with incident dysvascular amputation, (1535 [21%] TM amputees, 3302 [45%] TT amputees, and 2459 [34%] TF amputees [Figure 1]). The distributions of the nine potential confounders and effect modifiers by race/ethnicity are summarized in Table 1. The remaining summaries and analyses will combine the TM and TT amputees because the analysis estimated the risk of TF amputation (compared to a more distal amputation).
National and regional differences in the choice of incident amputation level independent of race/ethnicity
Among patients with diabetes and PAD who require a major amputation, the overall risk of incident TF amputation in the national VA cohort was 34%. This risk differed by geographic region. The risk of an incident TF amputation was 31% (n = 572), 42% (n = 769), 29% (n = 416), 37% (n = 412), and 27% (n = 290) in the North Atlantic, Southeast, Midwest, Continental, and Pacific regions, Table 2.
TABLE 2.
Risk of incident level of amputation stratified by region (independent of race/ethnicity)
| Amputation level | Pacific n (%) |
Midwest n (%) |
Continental n (%) |
North Atlantic n (%) |
Southeast n (%) |
|---|---|---|---|---|---|
| Transmetatarsal/Transtibial | 801 (73) | 1004 (71) | 910 (63) | 1293 (69) | 1553 (58) |
| Transfemoral | 290 (27) | 416 (29) | 412 (37) | 572 (31) | 769 (42) |
Association of race/ethnicity with incident amputation level nationally and by geographic region
Nationally, among those undergoing a major amputation for PAD or diabetes, the White, Black, and Hispanic risk for an incident TF amputation was 31% (n = 1356), 35% (n = 810), 46% (n = 293), respectively (Table 3). Greater disparities are seen by region with Blacks experiencing the greatest risk in the Continental region and Hispanics in the Southeast region. By comparison, risks are almost identical in the Pacific region and Hispanics appear to have a lower risk of TF amputation in the Midwest and North Atlantic region; however, the number of Hispanic patients, especially in the Midwest, is very small, which may limit the strength of this finding. These represent crude risks not adjusted for potential confounders and without considering differential effects by region, diabetes, or revascularization.
TABLE 3.
Risk of incident transfemoral amputation by race/ethnicity across Veterans Affairs (nationally) and stratified by region
| White (N = 4356) n (%) |
Black (N = 2302) n (%) |
Hispanic (N = 638) n (%) |
|
|---|---|---|---|
| Across Veterans Affairs | |||
| Transmetatarsal/Transtibial | 3000 (69) | 1492 (65) | 518 (54) |
| Transfemoral | 1356 (31) | 810 (35) | 293 (46) |
| Pacific | |||
| Transmetatarsal/Transtibial | 563 (73) | 153 (73) | 87 (75) |
| Transfemoral | 202 (27) | 55 (27) | 33 (25) |
| Midwest | |||
| Transmetatarsal/Transtibial | 723 (72) | 865 (66) | 16 (84) |
| Transfemoral | 279 (28) | 134 (34) | 3 (16) |
| Continental | |||
| Transmetatarsal/Transtibial | 424 (65) | 212 (57) | 55 (70) |
| Transfemoral | 231 (35) | 157 (43) | 24 (30) |
| North Atlantic | |||
| Transmetatarsal/Transtibial | 720 (70) | 517 (68) | 56 (80) |
| Transfemoral | 313 (30) | 245 (32) | 40 (20) |
| Southeast | |||
| Transmetatarsal/Transtibial | 579 (64) | 348 (61) | 121 (36) |
| Transfemoral | 331 (36) | 219 (39) | 219 (64) |
Racial/ethnic disparities in incident amputation by geographic region in the presence of diabetes and a history of prior revascularization
To summarize the effect of race/ethnicity on incident TF amputation, accounting for the linear combination of interaction term coefficients from the multivariable model, we displayed the odds ratios and 95% CIs in a table comparing the risk of Black and Hispanic versus White in the Continental and Southeast, stratified by the presence/absence of diabetes and prior revascularization (Table 4).
TABLE 4.
Odds ratio (OR) and 95% confidence intervals (CIs) comparing Black and Hispanic versus White risk of incident transfemoral amputation in the Continental and Southeast regions stratified by the presence or absence of diabetes and prior revascularization (significant odds ratios are in bold)
| Stratified by diabetes and revascularization | Continental OR | 95% CI | p Value | Southeast OR | 95% CI | p Value |
|---|---|---|---|---|---|---|
| Black versus White | ||||||
| No diabetes without revascularization | 1.5 | 1.1, 2.1 | .02 | 1.2 | .92, 1.7 | .15 |
| Diabetes without revascularization | 1.4 | 1.1, 1.9 | .02 | 1.2 | .91, 1.5 | .22 |
| No diabetes with revascularization | 1.0 | .67, 1.5 | .93 | .83 | .57, 1.2 | .34 |
| Diabetes with revascularization | .96 | .64, 1.4 | .83 | .78 | .54, 1.1 | .19 |
| Hispanic versus White | ||||||
| No diabetes without revascularization | 1.6 | .79, 3.1 | .20 | 3.2 | 1.9, 5.3 | <.001 |
| Diabetes without revascularization | 1.5 | .62, 3.6 | .37 | 3.0 | 1.4, 6.4 | .004 |
| No diabetes with revascularization | 1.5 | .57, 4.0 | .41 | 3.0 | 1.2, 7.5 | .02 |
| Diabetes with revascularization | 1.4 | .46, 4.5 | .53 | 2.9 | .99, 8.5 | .05 |
Comparing the odds of incident TF amputation for Blacks versus Whites, significantly higher risks for Blacks were identified in the Continental region but only when Blacks (with or without diabetes) did not undergo a prior revascularization. For Blacks who underwent a prior revascularization, there were no significant differences. Specifically, Blacks who did not undergo revascularization in the Continental region were at an increased odds of TF amputation compared to Whites (odds ratio [OR] = 1.4; 95% CI, 1.1, 1.9 and OR = 1.5; 95% CI, 1.1, 2.1 in those with and without diabetes, respectively). The odds were slightly higher in the Southeast region as well, but these differences were not statistically significant (Table 4).
Comparing the risk odds of incident TF amputation for Hispanics versus Whites, significantly higher risks for Hispanics were identified in the Southeast region. In contrast to Blacks in the Continental region, Hispanics in the southeast region were at an increased risk in all scenarios (ie, whether they had a prior revascularization or diabetes). Specifically, Hispanics were at an increased odds of TF amputation compared to Whites (ORs ≥ 2.9 and all effects were statistically significant except those with diabetes and a prior revascularization, p = .05), Table 4. The odds were slightly higher in the Continental region as well, but these differences were not statistically significant (Table 4).
DISCUSSION
The primary goal of this study was to determine if race/ethnicity is associated with incident TF amputation level in the U.S. veteran population. Because these associations can be confounded by several factors that may be associated with race or risk of TF amputation or be modified by the presence of other factors (ie, geographic region, diabetes, prior revascularization), we carefully selected these factors a priori based on the prior literature and clinical experience. We identified significant racial and ethnic differences in incident amputation level selection in specific subgroups of patients. In the Continental region, Blacks without revascularization, but both with and without diabetes, were at greater odds of an incident TF amputation compared to Whites. This was not the case if they had a revascularization procedure at some time in the year before the amputation where their odds were not elevated. In the Southeast region, Hispanics were at greater odds of a TF amputation compared to Whites, independent of the presence/absence of diabetes or a prior revascularization. These identified disparities in incident amputation level persisted after adjusting for other factors that might influence amputation risk. The results from the multivariable model were consistent with what was observed in the crude risks described in Table 3. Risks were highest in Blacks in the Continental region and highest in Hispanics in the Southeast region. Including the interactions with prior revascularization and diabetes further illustrated that, at least in Blacks, prior revascularization was not associated with an elevated odds of incident TF amputation. The identified differences in likelihood of TF amputation for Blacks and Hispanics are important because of the adverse impact of TF amputation compared with a more distal TT/TM amputation on key outcomes. They are more likely to experience a reduced prosthetic use, greater limitations in ambulation, loss of independent living, and a reduced quality of life.8-13
In the Continental region, Blacks were at increased odds of TF amputation compared to Whites, but only if they had not had a prior revascularization. This finding suggests that the greater use of revascularization, which is often performed to salvage a more distal amputation, may modify the risk for a higher level amputation. However, it is uncertain whether the availability of revascularization in regions where Black populations predominate is reduced or whether disease severity in Blacks is greater and the arterial lesions may not be revascularizable. Revascularization rates have been shown to differ both regionally and by racial characteristics, with Black patients less likely to receive a revascularization procedure.32-34 One well-designed study found that even in hospitals with the greatest capacity for vascular interventions, Black patients were less likely to receive revascularization and more likely to undergo amputation than White patients.35 This observation suggests that this difference may not be related to an access issue. Others have concluded that access to care likely does play a role because regions with a relatively greater Black population perform fewer vascular interventions overall.36 This finding might suggest that the extent of underlying disease may therefore contribute to the observed differences in revascularization outcomes. It is further supported by other studies that have shown that Black patients present for medical care later, having already developed gangrene, which increases the likelihood that an amputation may be necessary.35
In the Southeast region, Hispanics were at increased odds of TF amputation compared to Whites regardless of diabetes or a prior revascularization. It is interesting that Hispanic patients in our study do not see the same benefit of revascularization as Blacks. In our study population, Hispanic patients were less likely to have an abnormal ABI. Additionally, they were more likely to have diabetes, which is less associated with large vessel disease and more with nonocclusive small vessel disease due to circulatory dysfunction.37,38 The Hispanic patients in our study may have presented more often with a neuropathic diabetic foot wound, not able to be treated with revascularization. Prior research supports this, as Hispanic patients have been found to present for medical care later, having already developed gangrene, requiring more emergent care, with less likelihood of success with a revascularization.34,35,39 Within the VA, Hispanic patients have also been found to have fewer ambulatory care visits and less preventative care, which may further explain why they are presenting with more advanced disease.40,41
Study limitations
Although overcoming a number of methodologic limitations seen in prior research assessing regional variation in amputation level and racial/ethnic factors in amputation level selection, this study does have a number of limitations. Importantly, although our analysis did identify health disparities for Black and Hispanic patients, we are unable to determine the underlying reason(s). This study has some limitations to the generalizability of the findings. Although it included only veterans, this population is largely similar to the broader national population requiring dysvascular amputations. The number of women is small as are the number of individuals from racial/ethnic groups beyond Black and Hispanic populations. Additionally, the designation of Hispanic ethnicity does not specify race and represents a heterogenous population. Separate subgroup analysis by race and ethnicity was not possible. Finally, because the preamputation factors available for analysis were derived from an existing database, we were unable to measure or include other preoperative factors that may influence risk of a higher-level amputation and therefore unable to control for all possible confounding factors when estimating the risk of race/ethnicity on TF amputation.
Conclusions
Our study presents evidence that racial and ethnic factors in specific geographic regions are associated with more proximal amputations. The reasons for these differences in amputation level in some VA regions are unknown but follow-up studies to identify modifiable causes are important, so that we may target interventions that address the observed disparities. These include patient specific factors but may also reveal system, hospital, or provider practices that can be addressed to allow more equity in care.
ACKNOWLEDGMENTS
This work was supported by the US Department of Veterans Affairs, Office of Research and Development, Rehabilitation Research and Development Merit Review Award O1474-R and Center Grant 1I50RX002357-01. The funding sources were not involved in the study design or collection, analysis, and interpretation of data or the decision to submit the manuscript for publication. The contents do not represent the views of the US Department of Veterans Affairs or the United States Government.
Footnotes
CONFLICT OF INTEREST
The authors declare that there is no conflict of interest.
REFERENCES
- 1.National Healthcare Disparities Reports. Agency for Healthcare Research and Quality, Rockville, MD. 2019. doi: 10.4135/9781483346663.n402 [DOI] [Google Scholar]
- 2.Healthy People 2020. Washington, DC: U.S. Department of Health and Human Services, Office of Disease Prevention and Health Promotion; 2020. Accessed January 25, 2020. https://www.healthypeople.gov/ [Google Scholar]
- 3.Dillingham TR, Pezzin LE, MacKenzie EJ. Racial differences in the incidence of limb loss secondary to peripheral vascular disease: a population-based study. Arch Phys Med Rehabil. 2002;83(9):1252–1257. 10.1053/apmr.2002.34805 [DOI] [PubMed] [Google Scholar]
- 4.Lefebvre KM, Lavery LA. Disparities in amputations in minorities. Clin Orthop Relat Res. 2011;469(7):1941–1950. 10.1007/s11999-011-1842-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Arya S, Binney Z, Khakharia A, et al. Race and socioeconomic status independently affect risk of major amputation in peripheral artery disease. J Am Heart Assoc. 2018;7(2):e007425. 10.1161/JAHA.117.007425 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Skrepnek GH, Mills JL, Armstrong DG. A diabetic emergency one million feet long: disparities and burdens of illness among diabetic foot ulcer cases within emergency departments in the United States, 2006-2010. PLoS One. 2015;10(8):1–15. 10.1371/journal.pone.0134914 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Stevens CD, Schriger DL, Raffetto B, Davis AC, Zingmond D, Roby DH. Geographic clustering of diabetic lower-extremity amputations in low-income regions of California. Health Aff. 2014;33(8):1383–1390. 10.1377/hlthaff.2014.0148 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Collins TC, Johnson M, Henderson W, Khuri SF, Daley J. Lower extremity nontraumatic amputation among veterans with peripheral arterial disease: is race an independent factor? Med Care. 2002;40(suppl 1):106–116. 10.1097/00005650-200201001-00012 [DOI] [PubMed] [Google Scholar]
- 9.Rucker-Whitaker C, Feinglass J, Pearce WH. Explaining racial variation in lower extremity amputation: a 5-year retrospective claims data and medical record review at an urban teaching hospital. Arch Surg. 2003;138(12):1347–1351. 10.1001/archsurg.138.12.1347 [DOI] [PubMed] [Google Scholar]
- 10.Feinglass J, Rucker-Whitaker C, Lindquist L, McCarthy WJ, Pearce WH. Racial differences in primary and repeat lower extremity amputation: results from a multihospital study. J Vasc Surg. 2005;41(5):823–829. 10.1016/j.jvs.2005.01.040 [DOI] [PubMed] [Google Scholar]
- 11.Nguyen LL, Hevelone N, Rogers SO, et al. Disparity in outcomes of surgical revascularization for limb salvage: race and gender are synergistic determinants of vein graft failure and limb loss. Circulation. 2009;119(1):123–130. 10.1161/CIRCULATIONAHA.108.810341 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Van Houtum WH, Lavery LA, Armstrong DG. Risk factors for above-knee amputations in diabetes mellitus. South Med J. 1998;91(7):643–648. 10.1097/00007611-199807000-00007 [DOI] [PubMed] [Google Scholar]
- 13.Lefebvre KM, Metraux S. Disparities in level of amputation among minorities: implications for improved preventative care. J Natl Med Assoc. 2009;101(7):649–655. 10.1016/S0027-9684(15)30973-1 [DOI] [PubMed] [Google Scholar]
- 14.Lefebvre KM, Chevan J. Sex disparities in level of amputation. Arch Phys Med Rehabil. 2011;92(1):118–124. 10.1016/j.apmr.2010.10.005 [DOI] [PubMed] [Google Scholar]
- 15.Lefebvre KM, Chevan J. The persistence of gender and racial disparities in vascular lower extremity amputation: an examination of HCUP-NIS data (2002-2011). Vasc Med. 2015;20(1):51–59. 10.1177/1358863X14565373 [DOI] [PubMed] [Google Scholar]
- 16.Izumi Y, Satterfield K, Lee S, Harkless L. Risk of Reamputation in diabetic patients. Diabetes Care. 2006;29:566–570. 10.2337/diacare.29.03.06.dc05-1992 [DOI] [PubMed] [Google Scholar]
- 17.Peters EJ, Childs MR, Wunderlich RP, Harkless LB, Armstrong DG, Lavery LA. Functional status of persons with diabetes-related lower-extremity amputations. Diabetes Care. 2001;24(10):1799–1804. 10.2337/diacare.24.10.1799 [DOI] [PubMed] [Google Scholar]
- 18.Norvell DC, Turner AP, Williams RM, Hakimi KN, Czerniecki JM. Defining successful mobility after lower extremity amputation for complications of peripheral vascular disease and diabetes. J Vasc Surg. 2011;54(2):412–419. 10.1016/j.jvs.2011.01.046 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Sinha R, Van Den Heuvel WJA, Arokiasamy P. Factors affecting quality of life in lower limb amputees. Prosthet Orthot Int. 2011;35(1):90–96. 10.1177/0309364610397087 [DOI] [PubMed] [Google Scholar]
- 20.Nehler MR, Coll JR, Hiatt WR, et al. Functional outcome in a contemporary series of major lower extremity amputations. J Vasc Surg. 2003;38(1):7–14. 10.1016/s0741-5214(03)00092-2 [DOI] [PubMed] [Google Scholar]
- 21.Dillingham TR, Pezzin LE, MacKenzie EJ. Discharge destination after dysvascular lower-limb amputations. Arch Phys Med Rehabil. 2003;84(11):1662–1668. 10.1053/S0003-9993(03)00291-0 [DOI] [PubMed] [Google Scholar]
- 22.Norvell DC, Thompson ML, Boyko EJ, et al. Mortality prediction following non-traumatic amputation of the lower extremity. Br J Surg. 2019;106(7):879–888. 10.1002/bjs.11124 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Feinglass J, Pearce WH, Martin GJ, et al. Postoperative and amputation-free survival outcomes after femorodistal bypass grafting surgery: findings from the Department of Veterans Affairs National Surgical Quality Improvement Program. J Vasc Surg. 2001;34(2):283–290. 10.1067/mva.2001.116807 [DOI] [PubMed] [Google Scholar]
- 24.Davie-Smith F, Paul L, Nicholls N, Stuart WP, Kennon B. The impact of gender, level of amputation and diabetes on prosthetic fit rates following major lower extremity amputation. Prosthet Orthot Int. 2017;41(1):19–25. 10.1177/0309364616628341 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Goodney PP, Tarulli M, Faerber AE, Schanzer A, Zwolak RM. Fifteen-year trends in lower limb amputation, revascularization, and preventive measures among medicare patients. JAMA Surg. 2015;150(1):84–86. 10.1001/jamasurg.2014.1007 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Davis CL, Pierce JR, Henderson W, et al. Assessment of the reliability of data collected for the department of Veterans affairs National Surgical Quality Improvement Program. J Am Coll Surg. 2007;204(4):550–560. 10.1016/j.jamcollsurg.2007.01.012 [DOI] [PubMed] [Google Scholar]
- 27.Henderson WG, Daley J. Design and statistical methodology of the National Surgical Quality Improvement Program: why is it what it is? Am J Surg. 2009;198(suppl 5):S19–S27. 10.1016/j.amjsurg.2009.07.025 [DOI] [PubMed] [Google Scholar]
- 28.Czerniecki JM, Thompson ML, Littman AJ, et al. Predicting reamputation risk in patients undergoing lower extremity amputation due to the complications of peripheral artery disease and/or diabetes. Br J Surg. 2019;106(8):1026–1034. 10.1002/bjs.11160 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Mayfield JA, Reiber GE, Maynard C, Czerniecki JM, Caps MT, Sangeorzan BJ. Trends in lower limb amputation in the veterans health administration, 1989-1998. J Rehabil Res Dev. 2000;37(1):23–30. [PubMed] [Google Scholar]
- 30.Landry GJ, Silverman DA, Liem TK. Predictors of healing and functional outcome following transmetatarsal amputations. Arch Surg. 2011;146(9):1005–1009. [DOI] [PubMed] [Google Scholar]
- 31.Thorud JC, Plemmons B, Buckley CJ, Shibuya N. Mortality after nontraumatic major amputation among patients with diabetes and peripheral vascular disease: a systematic review. J Foot Ankle Surg. 2016;55(3):591–599. [DOI] [PubMed] [Google Scholar]
- 32.Albrektsen SB, Henriksen BM, Holstein PE. Minor amputations on the feet after revascularization for gangrene. A consecutive series of 95 limbs. Acta Orthop Scand. 1997;68(3):291–293. 10.3109/17453679708996704 [DOI] [PubMed] [Google Scholar]
- 33.Kota SK, Kota SK, Meher LK, Sahoo S, Mohapatra S, Modi KD. Surgical revascularization techniques for diabetic foot. J Cardiovasc Dis Res. 2013;4(2):79–83. 10.1016/j.jcdr.2012.10.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Regenbogen SE, Gawande AA, Lipsitz SR, Greenberg CC, Jha AK. Do differences in hospital and surgeon quality explain racial disparities in lower-extremity vascular amputations? Ann Surg. 2009;250(3):424–430. 10.1097/SLA.0b013e3181b41d53 [DOI] [PubMed] [Google Scholar]
- 35.Durazzo TS, Frencher S, Gusberg R. Influence of race on the management of lower extremity ischemia: revascularization vs amputation. JAMA Surg. 2013;148(7):617–623. 10.1001/jamasurg.2013.1436 [DOI] [PubMed] [Google Scholar]
- 36.Goodney PP, Holman K, Henke PK, et al. Regional intensity of vascular care and lower extremity amputation rates. J Vasc Surg. 2013;57(6):1471–1480.e3. 10.1016/j.jvs.2012.11.068 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Akbari CM, LoGerfo FW. Diabetes and peripheral vascular disease. J Vasc Surg. 1999;30(2):373–384. 10.1016/S0741-5214(99)70154-0 [DOI] [PubMed] [Google Scholar]
- 38.Aboyans V, Criqui MH, Denenberg JO, Knoke JD, Ridker PM, Fronek A. Risk factors for progression of peripheral arterial disease in large and small vessels. Circulation. 2006;113(22):2623–2629. 10.1161/CIRCULATIONAHA.105.608679 [DOI] [PubMed] [Google Scholar]
- 39.Mustapha JA, Fisher BT, Rizzo JA, et al. Explaining racial disparities in amputation rates for the treatment of peripheral artery disease (PAD) using decomposition methods. J Racial Ethn Heal Disparities. 2017;4(5):784–795. 10.1007/s40615-016-0261-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Saha S, Freeman M, Toure J, Tippens KM, Weeks C, Ibrahim S. Racial and ethnic disparities in the VA health care system: a systematic review. J Gen Intern Med. 2008;23(5):654–671. 10.1007/s11606-008-0521-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Washington DL, Villa V, Brown A, Damron-Rodriguez J, Harada N. Racial/ethnic variations in veterans’ ambulatory care use. Am J Public Health. 2005;95(12):2231–2237. 10.2105/AJPH.2004.043570 [DOI] [PMC free article] [PubMed] [Google Scholar]