Balancing Days at Home and Major Adverse Limb Events After Surgical Versus Endovascular Lower‐Extremity Revascularization in Chronic Kidney Disease Authors: Tara I. Chang, Gomathy Parvathinathan, Sai Liu, Jun Young Lee, Ali Etemadi, Shipra Arya, Margaret R. Stedman Published: 2026 ABSTRACT BackgroundThe risk–benefit trade‐offs after lower‐extremity revascularization among older patients with chronic kidney disease remain unclear. We compared how surgical versus endovascular lower‐extremity revascularization associates with clinical outcomes that may be valued differently by patients.MethodsWe selected adults aged ≥66 years with chronic kidney disease who underwent lower‐extremity revascularization from January 1, 2008, to December 31, 2017, with ≥12 months Medicare Parts A and B before the procedure. The primary outcome was days at home at 30, 180, and 365 days after revascularization. The secondary outcome was major adverse limb events (ie, lower‐extremity revascularization or major amputation). We report outcomes as the difference in survived days at home and days lost due to major adverse limb events by restricted mean time lost.ResultsAmong 87 446 patients, 15.3% underwent surgical and 84.7% underwent endovascular revascularization. Surgical revascularization was associated with fewer days at home at 30 days (difference, −8.4 [95% CI, −8.6 to −8.2]), 180 days (difference, −21.2 [95% CI, −22.1 to −20.4]) and 365 days (difference, −31.9 [95% CI, −33.5 to −30.6]) compared with endovascular. Surgical revascularization was associated with fewer days lost to major adverse limb events at 180 days (difference, −6.9 [95% CI, −7.6 to −5.8]) and at 365 days (difference, −17.3 [95% CI, −19.3 to −14.7]).ConclusionsIn this study of older patients with chronic kidney disease, surgical lower‐extremity revascularization was associated with fewer survived days at home, but also with fewer days lost to major adverse limb events compared with an endovascular approach. Considering different risk–benefit profiles when planning for lower‐extremity revascularization is essential to facilitate shared decision making. Nonstandard Abbreviations and Acronyms Best Endovascular Versus Best Surgical Therapy in Patients With Chronic Limb‐Threatening Ischemia major adverse limb events restricted mean time lost Clinical PerspectiveWhat Is New? Among older Medicare beneficiaries with chronic kidney disease undergoing lower‐extremity revascularization, surgical revascularization was associated with fewer survived days at home compared with endovascular revascularization across 30, 180, and 365 days after the procedure, but also with more days without major adverse limb events. What Are the Clinical Implications? In selecting a revascularization strategy for older patients with chronic kidney disease, clinicians should consider the differing values of outcomes such as maximizing time at home with an endovascular approach versus minimizing repeat revascularization risk with surgical revascularization, emphasizing the importance of discussing these competing priorities to support patient‐centered decision making. Clinical Perspective What Is New? Among older Medicare beneficiaries with chronic kidney disease undergoing lower‐extremity revascularization, surgical revascularization was associated with fewer survived days at home compared with endovascular revascularization across 30, 180, and 365 days after the procedure, but also with more days without major adverse limb events. Among older Medicare beneficiaries with chronic kidney disease undergoing lower‐extremity revascularization, surgical revascularization was associated with fewer survived days at home compared with endovascular revascularization across 30, 180, and 365 days after the procedure, but also with more days without major adverse limb events. Among older Medicare beneficiaries with chronic kidney disease undergoing lower‐extremity revascularization, surgical revascularization was associated with fewer survived days at home compared with endovascular revascularization across 30, 180, and 365 days after the procedure, but also with more days without major adverse limb events. What Are the Clinical Implications? In selecting a revascularization strategy for older patients with chronic kidney disease, clinicians should consider the differing values of outcomes such as maximizing time at home with an endovascular approach versus minimizing repeat revascularization risk with surgical revascularization, emphasizing the importance of discussing these competing priorities to support patient‐centered decision making. In selecting a revascularization strategy for older patients with chronic kidney disease, clinicians should consider the differing values of outcomes such as maximizing time at home with an endovascular approach versus minimizing repeat revascularization risk with surgical revascularization, emphasizing the importance of discussing these competing priorities to support patient‐centered decision making. In selecting a revascularization strategy for older patients with chronic kidney disease, clinicians should consider the differing values of outcomes such as maximizing time at home with an endovascular approach versus minimizing repeat revascularization risk with surgical revascularization, emphasizing the importance of discussing these competing priorities to support patient‐centered decision making. Peripheral artery disease (PAD) affects 200 million people worldwide and 9 million people in the United States. 1 Chronic kidney disease (CKD) affects 20 million people in the United States 2 and confers a markedly higher risk for PAD. 3 In the 2024 American College of Cardiology/American Heart Association guidelines on PAD management, 4 CKD was identified as a PAD‐related risk amplifier due to shared risk factors such as smoking, hypertension, hyperlipidemia, or diabetes, 5 , 6 , 7 , 8 and also the common underlying pathophysiology of chronic inflammation and oxidative stress. CKD may also confer additional independent risk factors such as disorders in bone and mineral metabolism or uremia. 9 , 10 , 11 , 12 One cornerstone of treatment for lower‐extremity PAD is revascularization to improve limb perfusion and prevent amputation. A recent randomized clinical trial of patients showed that patients randomized to a surgical revascularization strategy had lower rates of major adverse limb events (MALEs) and death compared with an endovascular strategy. 13 However, patient‐centered outcomes like health‐related quality of life did not differ meaningfully between groups. 14 The divergence between a “traditional” outcome like MALEs and patient‐centered outcomes highlights the importance of examining both types of outcomes in comparative effectiveness studies. Moreover, this divergence may also be more pronounced among patients with CKD because greater symptom burden, frailty, impaired wound healing, and lower physical resilience may limit improvements in health‐related quality of life, even when limb outcomes improve. 14 While health‐related quality of life and other patient‐centered outcomes can be difficult to ascertain, days at home is a validated, patient‐centered outcome that integrates information on patient morbidity, functional status, and recovery 15 , 16 , 17 , 18 that can be easily ascertained within large administrative data sets. We used data from a nationally representative cohort of Medicare beneficiaries to compare surgical versus endovascular lower‐limb revascularization among older patients with CKD. We hypothesized that a surgical approach would be associated with fewer days at home but better MALE outcomes. Methods Data Source and Cohort Creation We performed a retrospective, population‐based cohort study using Medicare claims data from 2008 to 2020, obtained from the Centers for Medicare and Medicaid Services. Data for this project were accessed using the Stanford Center for Population Health Sciences Data Core under a Data Use Agreement with the Centers for Medicare and Medicaid Services (RSCH‐2023‐59 722). The data used in this analysis are subject to stringent privacy and confidentiality standards. As such, researchers can request access to the data through the Centers for Medicare and Medicaid Services. Interested parties can refer to the Centers for Medicare and Medicaid Services data access guidelines (cms.gov) for further information on obtaining relevant data. This study was approved by a Stanford institutional review board committee (Protocol No. 70241), and a waiver of consent was granted. Specifically, we used the Medicare 20% Master Beneficiary Summary File; Medicare Provider Analysis and Review; and Outpatient, Carrier, and Hospice claims. 19 , 20 , 21 , 22 , 23 , 24 , 25 , 26 , 27 , 28 Beneficiaries were linked across data sets via unique identifiers in the Master Beneficiary Summary File. We extracted primary and secondary diagnosis and procedure codes and ascertained death using the Master Beneficiary Summary File and Medicare Provider Analysis and Review file. We included adults aged ≥66 years to ensure at least 12 months of continuous enrollment in Medicare Parts A and B and excluded beneficiaries with Part C coverage (Medicare Advantage). From this cohort, we identified patients who had a lower‐extremity revascularization between January 1, 2008, and December 31, 2017, to allow for at least 3 years of follow‐up after revascularization. We identified surgical or endovascular lower‐extremity revascularization using the International Classification of Diseases, Ninth Revision (ICD‐9) and Tenth Revision (ICD‐10) procedure codes, and Current Procedural Terminology codes (Tables S1–S5). 29 The index date was the date of the revascularization procedure. To identify patients with CKD, we required at least 2 outpatient CKD ICD‐9/10 diagnosis codes on separate days or 1 inpatient claim in the 1 year before revascularization (Table S6). For patients with multiple lower‐extremity revascularizations during the study period, we selected the first revascularization after a confirmed diagnosis of CKD as the index procedure. We excluded patients with end‐stage kidney disease before revascularization because patients receiving maintenance dialysis or with functioning kidney transplants differ significantly from patients with less severe CKD in physiology, hemodynamic stressors, inflammatory markers, and responsiveness to treatments. We also excluded patients with both surgical and endovascular revascularization on the same day and any patients whose index hospitalization exceeded 30 days. We also excluded beneficiaries residing outside the 50 United States or with “unknown” race (Figure 1). Cohort flow diagram of Medicare beneficiaries with chronic kidney disease who underwent lower‐extremity revascularization. Outcomes Primary Outcome Our primary outcome was survived days at home at 30, 180, and 365 days after the index revascularization. Days at home is a validated, patient‐centered outcome originating in the stroke literature. 15 , 16 , 17 Days at home were measured from the date of admission for inpatient revascularization or the procedure date for outpatient revascularization. First, we determined the number of institutional days as the sum of the days spent in the hospital, skilled nursing facility, or hospice facility within the follow‐up period. The follow‐up period was adjusted for insurance loss and death. We determined survived days at home by subtracting the total institutional days from the adjusted follow‐up time (Figure S1). Secondary Outcomes Our secondary outcome was MALEs, defined as a repeat lower‐extremity revascularization or an above‐ankle amputation. We categorized any amputations that occurred during a hospitalization for trauma, congenital or acquired deformities of the limbs, or malignancy affecting the limbs (identified using ICD‐9 or ICD‐10 diagnosis codes; Tables S1 and S5) 30 as non–PAD‐related and excluded these as end points. We counted repeat revascularization as an event only if it occurred >14 days after the index date; we counted any above‐ankle amputation as an event. We also examined each of the components of MALEs (amputation and repeat revascularization) separately at 30, 180, and 365 days after the index date. Covariates We ascertained demographic details, including age, sex, race and ethnicity, and geographic region of the beneficiary at the time of the index date. Using a 1‐year look‐back period before the index date, we ascertained the following comorbid conditions using billing claims codes: prior lower‐extremity revascularization, chronic limb‐threatening ischemia, and other comorbid conditions as defined by the Elixhauser algorithm (Table 1). 31 We estimated the Elixhauser comorbidities using the multimorbidity package in R (R Foundation for Statistical Computing, Vienna, Austria). 32 Of note, the Elixhauser algorithm definition of “renal failure” uses a narrower subset of diagnosis codes than the broader set of codes that we used to identify patients with CKD for the current analysis. Given potential secular trends in revascularization and other treatments, we included the calendar year of the index procedure in the models. Table 1Baseline Characteristics of Patients With Chronic Kidney Disease Undergoing Lower‐Extremity Endovascular or Surgical RevascularizationSurgical, n=13 418Endovascular, n=74 028SMD before IPTWSMD after IPTWAge, y, mean±SD77.6±7.3877.7±7.290.01270.0011Female sex5666 (42.2)35 931 (48.5)0.1270.0042Race and ethnicityAsian109 (0.8)720 (1.0)0.0170.0077Black1382 (10.3)8148 (11.0)0.02290.0071North American Native74 (0.6)437 (0.6)0.00520.0014Hispanic212 (1.6)1661 (2.2)0.04850.0027White11 523 (85.9)62 392 (84.3)0.04480.0085Other Race or Ethnicity118 (0.9)670 (0.9)0.00270.0037RegionMidwest3268 (24.4)17 817 (24.1)0.00670.0026Northeast2749 (20.5)11 490 (15.5)0.12950.0047South5521 (41.1)33 920 (45.8)0.09440.0027West1880 (14.0)10 801 (14.6)0.01660.0043Procedural year20081675 (12.5)7183 (9.7)0.08860.00320091496 (11.1)6868 (9.3)0.06180.001120101381 (10.3)6678 (9.0)0.0430.007820111394 (10.4)6906 (9.3)0.03560.00420121388 (10.3)7065 (9.5)0.02680.014520131252 (9.3)7360 (9.9)0.02070.015820141287 (9.6)7472 (10.1)0.01680.008520151202 (9.0)7714 (10.4)0.04940.004620161193 (8.9)8179 (11.0)0.07210.000920171150 (8.6)8603 (11.6)0.10140.0039Comorbid conditionsChronic limb‐threatening ischemia4768 (35.5)22 814 (30.8)0.10030.004Previous lower‐extremity revascularization1913 (14.3)7864 (10.6)0.11030.001Elixhauser algorithm–defined comorbid conditionsHypertension12 678 (94.5)70 601 (95.4)0.04040.0036Heart failure5747 (42.8)28 682 (38.7)0.08320.0046Valvular disease4691 (35.0)22 710 (30.7)0.09130.0055Diabetes6774 (50.5)42 341 (57.2)0.13490.0025Peripheral vascular disease11 070 (82.5)65 127 (88.0)0.15480.0121Pulmonary circulation disorder1503 (11.2)6928 (9.4)0.06070.0129Paralysis766 (5.7)3952 (5.3)0.01620.0002Other neurological disorders2879 (21.5)15 444 (20.9)0.01450.0039Chronic pulmonary disease5665 (42.2)27 543 (37.2)0.10260.0037Hypothyroidism2813 (21.0)16 551 (22.4)0.03380.0016Renal failure4914 (36.6)27 854 (37.6)0.02080.0011Liver disease499 (3.7)2764 (3.7)8.00E‐040.0039Chronic peptic ulcer disease140 (1.0)733 (1.0)0.00530.0004HIV and AIDS24 (0.2)97 (0.1)0.01220.0008Lymphoma235 (1.8)1122 (1.5)0.01860.002Metastatic cancer392 (2.9)1734 (2.3)0.03620.0046Solid tumor without metastasis1672 (12.5)8530 (11.5)0.02890.0037Rheumatoid arthritis/collagen vascular disease892 (6.6)5561 (7.5)0.03370.0013Coagulation deficiency1815 (13.5)7124 (9.6)0.12220.0067Obesity1700 (12.7)10 067 (13.6)0.02750.0002Weight loss1654 (12.3)7698 (10.4)0.06080.0007Fluid and electrolyte disorders5520 (41.1)27 403 (37.0)0.08450.0089Blood loss anemia794 (5.9)3417 (4.6)0.05830.0061Deficiency anemias6174 (46.0)32 213 (43.5)0.05030.0043Alcohol abuse172 (1.3)910 (1.2)0.00470.0006Drug abuse112 (0.8)601 (0.8)0.00250.0017Psychoses1040 (7.8)5627 (7.6)0.00560.0043Depression2011 (15.0)11 118 (15.0)<0.0010.0025All values are n (%) except where specified.IPTW indicates inverse probability of treatment weighting; and SMD, standardized mean differences. Baseline Characteristics of Patients With Chronic Kidney Disease Undergoing Lower‐Extremity Endovascular or Surgical Revascularization All values are n (%) except where specified. IPTW indicates inverse probability of treatment weighting; and SMD, standardized mean differences. Statistical Analysis We assessed group differences in baseline characteristics using standardized mean differences. 33 To address differences in baseline characteristics, we applied inverse probability of treatment weighting. We estimated each patient's propensity score (Table S7, Figure S2) for surgical versus endovascular revascularization using multivariable logistic regression, including all variables in Table 1. We calculated stabilized weights as the inverse of the estimated propensity score multiplied by the proportion treated. 34 We truncated the stabilized weights at the 99th percentile and reported the mean of the stabilized weights. We performed a Poisson model adjusting for the propensity score weights to estimate the incident rate per 100 person‐years. We used weighted fractional regression 35 to model the probability of survived days at home, our primary outcome. We predicted survived days at home by multiplying the predicted probability by the adjusted follow‐up period for each individual. We then calculated the marginal difference in days at home and estimated the 95% CIs from 100 bootstrap samples. For our secondary outcome, MALEs, we applied an inverse probability‐weighted adjusted estimator for the difference in cause‐specific restricted mean time lost (RMTL) between the surgical and endovascular procedure groups. 36 , 37 The RMTL is the difference between the specified follow‐up time (30, 180, or 365 days) and the restricted mean survival time (the area under the survival curve from time 0 to the specified follow‐up time). When there are no competing risks, the RMTL and the restricted mean survival time sum to the follow‐up time. We used the same weights as detailed above in the inverse probability of treatment weighting analysis, which included all the variables in Table 1. We calculated the weighted RMTL, the RMTL difference (surgical – endovascular), and 95% CI from 100 bootstrap samples. Negative RMTL differences indicate that the surgical group experienced fewer days lost to MALEs, while positive RMTL differences indicate that the endovascular group had fewer days lost to MALEs. We censored for death, insurance loss, or the end of follow‐up. We conducted several sensitivity analyses. First, we reduced the minimum gap time between repeat revascularization events from 14 to 3 days after the index date. Second, we treated death as a competing event for MALEs instead of as a censoring event. In this approach, the RMTL for death, RMTL for MALEs, and restricted mean survival time sum to the follow‐up time. We used the same inverse probability‐weighted adjustment to the model as described for the main analysis. Third, we assessed frailty using the Claims‐Based Frailty Index for each beneficiary using a 1‐year look‐back period. 38 , 39 We then repeated the analyses after stratifying the models by frailty score (nonfrail Claims‐Based Frailty Index ≤0.2, and frail Claims‐Based Frailty Index >0.2). 38 Finally, we repeated the analysis limiting the cohort to patients in whom the index revascularization year was from the most recent 5 years available (calendar years 2013–2017). All analyses were conducted in Redivis and the Carina environment using R version 4.4. within RStudio version 2024.12.1 (Posit Software, Boston, MA), and the Jupyter Notebooks (Jupyter Foundation) with the IRkernel. Results We identified 87 446 Medicare beneficiaries with CKD who underwent a lower‐extremity revascularization and met our inclusion and exclusion criteria (Figure 1). Among these patients, 13 418 (15.3%) underwent a surgical lower‐extremity revascularization and 74 028 (84.7%) underwent an endovascular revascularization. The mean age of the cohort was 77.6 years, 47.6% were women, and 84.5% were of the White race. The prevalence of chronic limb‐threatening ischemia was 35.5% in the patients undergoing surgical revascularization and 30.8% in the patients undergoing endovascular revascularization. Before applying inverse probability of treatment weighting, patients undergoing surgical revascularization were less likely to be women and have a higher prevalence of chronic limb‐threatening ischemia, previous lower‐extremity revascularization, chronic pulmonary disease, and coagulation deficiency and a lower prevalence of diabetes (Table 1). After inverse probability of treatment weighting, all variables were well balanced between the 2 treatment groups (standardized mean difference <0.1). In unadjusted analyses, more patients undergoing surgical compared with endovascular revascularization died at 30‐, 180‐, and 365‐day follow‐up, and these patients spent a smaller proportion of survived days at home due to more days spent in a hospital or nursing facility (Table 2, Figure S3). In adjusted analyses, patients who underwent surgical versus endovascular revascularization had fewer survived days at home at all time points: 30 days (−8.4 [95% CI, −8.6 to −8.2]), 180 days (−21.1 [95% CI, −22.0 to −20.4]), and 365 days (−31.9 [95% CI, −33.5 to −30.6]; Figure 2). Results were not materially changed in sensitivity analyses that stratified by frailty status or that restricted the cohort to index procedures occurring during calendar years 2013 to 2017 (Table S8). Table 2Unadjusted Days at Home, Hospital, or Nursing Facility in the Specified Time Period After Index Revascularization30 d180 d365 dSurgicalEndovascularSurgicalEndovascularSurgicalEndovascularDied, n (%)1268 (9.5)3075 (4.2)2738 (20.4)9403 (12.7)3643 (27.2)14 281 (19.3)Proportion of survived days at home, mean±SD50.2±39.080.0±33.671.4±33.985.3±26.975.1±33.186.3±25.5Days at home, mean±SD14.8±11.723.8±10.2122.2±64.9147.3±55.1250.8±136.0290.5±117.9Days in hospital, mean±SD9.1±8.23.7±6.515.1±17.38.04±14.418.4±21.411.1±18.6Days in nursing facility, mean±SD4.4±7.71.8±5.515.8±29.69.21±24.220.8±41.114.0±35.2Loss of insurance, n (%)30 (0.2)160 (0.2)193 (1.4)1157 (1.6)361 (2.7)2248 (3.0)The proportion of patients who died or lost their insurance is also provided. Unadjusted Days at Home, Hospital, or Nursing Facility in the Specified Time Period After Index Revascularization The proportion of patients who died or lost their insurance is also provided. Comparison of survived days at home and days lost to MALEs at the specified time points after surgical versus endovascular revascularization among patients with chronic kidney disease. Values shown are the difference in days (95% CI) and absolute number of survived days at home or days lost to MALE after surgical vs endovascular revascularization. For survived days at home, larger values are better, and negative differences indicate that endovascular revascularization is favored. For days lost to MALEs, smaller values are better, and negative differences indicate that surgical revascularization is favored. *P values all <0.001 for difference in days. MALE indicates major adverse limb event. In unadjusted analysis, patients who underwent surgical revascularization had a lower incidence of MALEs at 30, 180, and 365 days of follow‐up when compared with endovascular revascularization, a difference driven by fewer repeat revascularizations (Table 3). Of the 1966 patients with an index surgical revascularization who required a repeat revascularization at 1‐year follow‐up, 53.3% (n=1048) underwent another surgical revascularization. Among the 17 610 patients with an index endovascular revascularization who required a repeat revascularization at 1‐year follow‐up, 86.4% (n=15 213) underwent another endovascular revascularization. Above‐ankle amputation occurred in fewer patients than repeat revascularization overall but was more often seen among patients undergoing surgical versus endovascular revascularization. Table 3Unadjusted Number of Patients Experiencing a MALE and the Events per 100 Person‐Years at the Specified Follow‐Up Time PointsSurgical, N=13 418Endovascular, N=74 028n (%)Events per 100 person‐years (95% CI)n (%)Events per 100 person‐years (95% CI)30‐d follow‐upMALE655 (4.9)62.4 (57.6–67.4)4868 (6.6)84.0 (81.6–86.3)Repeat revascularization206 (1.5)20.7 (18.0–23.6)3653 (4.9)62.3 (60.3–64.3)Above‐ankle amputation449 (3.3)42.9 (39.0–47.0)1215 (1.6)21.8 (20.7–23.1)180‐d follow‐upMALE2077 (15.5)39.9 (38.2–41.6)15 616 (21.1)54.3 (53.4–55.1)Repeat revascularization1349 (10.1)26.8 (25.4–28.2)13 285 (17.9)45.7 (44.9–46.4)Above‐ankle amputation728 (5.4)15.3 (14.3–16.4)2331 (3.1)9.2 (8.8–9.5)365‐d follow‐upMALE2787 (20.8)29.8 (28.7–30.9)20 177 (27.3)39.2 (38.6–39.7)Repeat revascularization1966 (14.7)21.6 (20.6–22.5)17 610 (23.8)33.8 (33.3–34.3)Above‐ankle amputation821 (6.1)10.1 (9.5–10.7)2567 (3.5)6.0 (5.8–6.2)MALE indicates major adverse limb event. Unadjusted Number of Patients Experiencing a MALE and the Events per 100 Person‐Years at the Specified Follow‐Up Time Points MALE indicates major adverse limb event. In adjusted analyses, the number of days lost to MALEs at 30 days slightly favored endovascular revascularization, while the results at 180 and 365 days demonstrated fewer days lost to MALEs among the surgical revascularization group (Figure 2). However, for the specific components of MALEs, in adjusted analyses, patients undergoing surgical revascularization experienced fewer days lost to repeat revascularization at 30, 180, and 365 days but more days lost to above‐ankle amputation compared with patients undergoing endovascular revascularization (Figure 3). Comparison of days lost to repeat revascularization and above‐ankle amputation at the specified time points after surgical versus endovascular revascularization among patients with chronic kidney disease. Values shown are the difference in days (95% CI) and absolute number of days lost after surgical vs endovascular revascularization. Negative differences indicate that surgical revascularization is favored, while positive differences indicate that endovascular revascularization is favored. *P values all <0.001 for difference in days. Results were not materially changed in any of our sensitivity analyses that treated death as a competing risk for MALEs rather than a censoring event; stratified by frailty status on the basis of Claims‐Based Frailty Index; restricted to an index year between 2013 and 2017; and reduced gap time between repeat revascularizations from 14 to 3 days from the index date (Table S9). Discussion In our large, nationally representative sample of older adults with CKD undergoing lower‐extremity revascularization for PAD, we found that surgical revascularization was associated with fewer survived days spent at home (and, conversely, more time in a hospital or nursing facility after the procedure) at 30‐, 180‐, and 365‐day follow‐up compared with endovascular revascularization. In contrast, patients in the surgical group had fewer days lost to MALEs compared with endovascular revascularization during these same time points. The advantage in more MALE‐free days was driven primarily by fewer repeat revascularizations in the surgical group. While above‐ankle amputation was a comparatively less common outcome, patients undergoing surgical revascularization experienced more days lost to major amputation than patients undergoing endovascular revascularization. Our findings highlight the importance of having care teams carefully discuss these risk–benefit trade‐offs with patients and their families when making decisions about lower‐extremity revascularization strategies. In our analysis, 15.3% of older patients with CKD underwent surgical revascularization, and 84.7% underwent endovascular revascularization. This distribution is similar to what has been reported in prior studies of Medicare beneficiaries not limited to CKD (18.6% surgical versus endovascular) 40 and of Medicare beneficiaries with chronic limb‐threatening ischemia (17.5% surgical versus endovascular). 41 Few prior studies have compared the association of surgical versus endovascular revascularization on survived days at home, a validated patient‐centered outcome that directly correlates with health‐related quality of life. 17 , 42 A recent mixed‐methods study of US veterans after inpatient surgery 15 showed that patients viewed the home as a safe space for mental and emotional well‐being, interpreting the ability to return home as an important marker of recovery. A prior study of Medicare beneficiaries concluded that a gain in days at home of 24 days in 1 year is likely to be clinically meaningful in terms of functional status, self‐rated health, mood, and other patient‐centered outcomes. 42 In our study, the surgical versus endovascular group spent 24.3 fewer survived days at home in the first year after lower‐extremity revascularization, which thus likely represents a clinically meaningful outcome that warrants discussion with patients, enabling them to understand trade‐offs that extend beyond the risks of MALEs. Our findings on MALEs are generally consistent with the results from the recent BEST‐CLI (Best Endovascular Versus Best Surgical Therapy in Patients With Chronic Limb‐Threatening Ischemia) trial, which randomized patients with chronic limb‐threatening ischemia to surgical or endovascular revascularization. Among patients with a suitable single saphenous vein segment, surgical revascularization led to lower rates of MALEs or death compared with endovascular revascularization (hazard ratio [HR], 0.68 [95% CI, 0.59–0.79]). 13 Among the subgroup of patients with non–dialysis‐dependent CKD, the point estimate continued to favor surgery, but the confidence limits were wide (HR, 0.86 [95% CI, 0.60–1.23]), reflecting the relatively small number of patients with CKD included in this arm of BEST‐CLI (n=219). Similarly, our findings are consistent with the results from a post hoc analysis of BEST‐CLI that focused on the subgroup of patients with kidney failure receiving dialysis and found that endovascular revascularization was associated with a higher hazard of repeat revascularization (HR, 2.07 [95% CI, 1.06–4.05]). 43 Our findings favoring surgery for MALEs were largely driven by fewer days lost to repeat revascularization. In contrast, we found that surgical revascularization was associated with 0.4, 3.6, and 9.1 more days lost to major amputation at 30‐, 180‐, and 365‐day follow‐up, thereby favoring endovascular revascularization. In BEST‐CLI, surgical versus endovascular revascularization reduced the need for above‐ankle amputation (HR, 0.73 [95% CI, 0.54–0.98]) among patients with a suitable saphenous vein graft, but participants who required an alternative bypass conduit in the surgical group had numerically more above‐ankle amputations, although the confidence limits were wide and largely overlapping with the endovascular group. 13 Our study, which included 87 446 patients with non–dialysis‐dependent CKD, provides greater statistical power and precision and extends the findings from the randomized trial to a real‐world clinical setting, thereby complementing and building upon these results from BEST‐CLI. Our findings underscore the importance of factoring in comorbidities like CKD and anatomic factors such as the availability of adequate vein conduits when discussing treatment options, given their potential influence on clinical outcomes. Patients with CKD have been shown to have higher rates of MALEs when compared with more general PAD cohorts in prior studies. For example, in a population‐based observational study using Swedish registry data, the incidence of MALEs following lower‐limb revascularization was 18.7% over a mean follow‐up time of 3.2 years or about 5.8% per year. 44 The mean age of the cohort in that study was 74 years, and only 8.4% of the cohort had CKD. In contrast, the 1‐year MALE event rate in our study of patients with CKD with a mean age of 78 years was 35% to 49%. In a separate study using US administrative data, the mortality rate among patients with PAD and advanced CKD was 4 times higher than for patients with preserved kidney function. 45 The reasons for the higher risks of adverse clinical outcomes in CKD, even after adjusting for the higher prevalence of comorbid conditions such as diabetes and hypertension, is thought to stem from factors unique to CKD such as higher rates of inflammation, toxic uremic solutes, and vascular calcification. 46 Patients with CKD therefore require dedicated study to understand their specific risk–benefit profiles. Limitations Our study has several strengths. Our findings were robust across multiple sensitivity analyses, and our large cohort spans diverse geographic regions within the United States, enhancing the generalizability of our results. We examined survived days at home, a patient‐centered outcome that has not been extensively studied in PAD. Nonetheless, there are important limitations to acknowledge. First, as our analysis was based on Medicare claims data, we lacked access to laboratory parameters such as albuminuria or serum creatinine values, which precluded our ability to analyze the cohort by CKD stage. Second, information about medication use was not available. Third, we did not have data on vascular anatomy, wound characteristics, type of bypass conduit used, or other clinical parameters that impact MALE outcomes. The claims data also preclude identification of whether repeat revascularizations targeted the same vessel or a different one. Fourth, we excluded patients receiving maintenance dialysis or with functioning kidney transplants, so our results may not be applicable to those populations. Finally, although we were able to balance all observable baseline characteristics, residual confounding in the absence of randomization remains possible. Future randomized trials that focus specifically on the CKD population and compare outcomes after surgical versus endovascular lower‐extremity revascularization remain essential. Conclusions In summary, in this national study of older patients with CKD, surgical versus endovascular lower‐extremity revascularization was associated with fewer survived days at home but more days free from MALEs. However, endovascular revascularization was associated with fewer days lost to major amputation. These distinct risk–benefit profiles underscore the importance of individualized care and shared decision‐making when selecting a revascularization strategy for these high‐risk patients with CKD. Sources of Funding This work was funded by a grant from the National Institutes of Health/National Heart, Lung, and Blood Institute (R01HL151351) to T.I.C. The PHS Data Core is supported by a National Institutes of Health National Center for Advancing Translational Science Clinical and Translational Science Award (UL1TR003142) and from internal Stanford funding. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Disclosures T.I.C. reports serving as a consultant for AstraZeneca, Bayer, Novo Nordisk, Eli Lilly, Humacyte, Prokidney, Alexion, Tricida, and the George Clinical Institute; and has received salary support from CSL Behring through funds paid directly to her employer. None of these relationships are related to the current manuscript. S.A. reports receiving consulting fees from Gore Medical and Novartis pharmaceuticals. The remaining authors have no disclosures to report. Supporting information Tables S1–S9 Figures S1–S3 RECORD statement