Abstract
Objective:
Patients with chronic kidney disease (CKD) are considered a high-risk population, and the optimal approach to the treatment of carotid disease remains unclear. Thus, we compared outcomes following carotid revascularization for patients with CKD by operative approach of carotid endarterectomy (CEA), transfemoral carotid artery stenting (TFCAS), and transcarotid arterial revascularization (TCAR).
Methods:
The Vascular Quality Initiative was analyzed for patients undergoing carotid revascularizations (CEA, TFCAS, and TCAR) from 2016 to 2021. Patients with normal renal function (estimated glomular filtration rate >90 mL/min/1.72 m2) were excluded. Asymptomatic and symptomatic carotid stenosis were assessed separately. Preoperative demographics, operative details, and outcomes of 30-day mortality, stroke, myocardial infarction (MI), and composite variable of stroke/death were compared. Multivariable analysis adjusted for differences in groups, including CKD stage.
Results:
A total of 90,343 patients with CKD underwent revascularization (CEA, n = 66,870; TCAR, n = 13,459; and TFCAS, n = 10,014; asymptomatic, 63%; symptomatic, 37%). Composite 30-day mortality/stroke rates were: asymptomatic: CEA, 1.4%; TCAR, 1.2%; TFCAS, 1.8%; and symptomatic: CEA, 2.7%; TCAR, 2.3%; TFCAS, 3.7%. In adjusted analysis, TCAR had lower 30-day mortality compared with CEA (asymptomatic: adjusted odds ratio [aOR], 0.4; 95% confidence interval [CI], 0.3–0.7; symptomatic: aOR, 0.5; 95% CI, 0.3–0.7), and no difference in stroke, MI, or the composite outcome of stroke/death in both symptom cohorts. TCAR had lower risk of other cardiac complications compared with CEA in asymptomatic patients (aOR, 0.7; 95% CI, 0.6–0.9) and had similar risk in symptomatic patients. Compared with TFCAS, TCAR patients had lower 30-day mortality (asymptomatic: aOR, 0.5; 95% CI, 0.2–0.95; symptomatic: aOR, 0.3; 95% CI, 0.2–0.4), stroke (symptomatic: aOR, 0.7; 95% CI, 0.5–0.97), and stroke/death (asymptomatic: aOR, 0.7; 95% CI, 0.5–0.97; symptomatic: aOR, 0.6; 95% CI, 0.4–0.7), but no differences in MI or other cardiac complications. Patients treated with TFCAS had higher 30-day mortality (aOR, 1.8; 95% CI, 1.2–2.5) and stroke risk (aOR, 1.3; 95% CI, 1.02–1.7) in symptomatic patients compared with CEA. There were no differences in MI or other cardiac complications.
Conclusions:
Among patients with CKD, TCAR and CEA showed rates of stroke/death less than 2% for asymptomatic patients and less than 3% for symptomatic patients. Given the increased risk of major morbidity and mortality, TFCAS should not be performed in patients with CKD who are otherwise anatomic candidates for TCAR or CEA.
Keywords: Carotid disease, Carotid stenosis, Carotid stenting, Chronic kidney disease, End-stage renal disease, Transcarotid artery revascularization
Carotid artery stenosis is a well-recognized cause of cerebral ischemia. Stroke is the second leading cause of death worldwide, and approximately 15% of strokes are secondary to carotid artery disease.1,2 Currently, carotid endarterectomy (CEA) remains the dominant method for carotid revascularization for non-high-risk patients. Historically, transfemoral carotid artery stenting (TFCAS) was considered the procedure of choice for those with unfavorable anatomy, radiation, or high-risk patients due to advanced comorbidities.3 However, studies have shown that TFCAS has been associated with an increased risk of both intraoperative and long-term stroke.4 Transcarotid artery revascularization (TCAR) with temporary flow reversal has emerged as an alternative that has shown similar stroke and death rates compared with CEA and lower stroke rates compared with TFCAS.5,6
In patients with chronic kidney disease (CKD), a group historically at higher risk during carotid interventions, recent findings suggest that TCAR can be safely performed even in those with end-stage renal disease on dialysis, without increasing the risk in stroke, myocardial infarction (MI), or the composite risk of stroke/death/MI compared with non-dialysis patients.7 However, comprehensive studies comparing outcomes of CEA, TFCAS, and TCAR in patients with CKD are lacking, and current guidelines offer limited direction.8 This study aims to evaluate the effectiveness of carotid revascularization techniques in the CKD population, seeking to identify the most favorable approach.
METHODS
Data source.
This study analyzed prospectively collected data from the Society of Vascular Surgery Vascular Quality Initiative (VQI) database. The VQI is a national quality improvement initiative containing patient-and procedure-specific data from more than 4000 participating centers across the United States, Canada, and Singapore.9 The study was approved by the VQI research advisory committee and deemed exempt by the University of Washington Institutional Review Board secondary to the de-identified nature of the registry.
Patient cohort.
This study included patients 18 years and older with CKD from 2016 to 2021. CKD stages were classified according to Kidney Disease: Improving Global Outcomes (KDIGO) guidelines as follows: stage 2 (glomular filtration rate [GFR], 60–89 mL/min/1.72 m2), stage 3 (GFR, 30–59 mL/min/1.72 m2), stage 4 (GFR, 15– 29 mL/min/1.72 m2), or stage 5 (GFR <15 mL/min/1.72 m2). Dialysis patients included those on hemodialysis and peritoneal dialysis. Patients classified as CKD stage 5 were analyzed separately if they were receiving dialysis treatment prior to revascularization. Patients with a GFR greater than 89 mL/min/1.72 m2 (CKD stage 1), indicating normal or near normal renal function, were excluded from this study. Estimated GFR was calculated using the 2021 CKD-EPI equation.10
Patients were categorized based on their symptom status for the procedure and then further stratified according to the operative approach used. Symptomatic carotid stenosis was defined as the presence of ipsilateral cortical or ocular symptoms indicative of likely transient ischemic attack (TIA) or stroke prior to the intervention. Operative approaches compared included CEA, TCAR, and TFCAS. TFCAS was defined as open cutdown to the femoral artery with femoral artery access or direct percutaneous femoral artery access through the skin for stent placement or angioplasty. Patients treated with direct carotid stenting without flow reversal (n = 4633) or upper extremity stenting via the brachial or radial artery (n = 960) were excluded.
Clinical and outcome variables.
Demographic data, comorbidities, and perioperative medication were stratified by symptom-status cohorts and then compared by approach type. In the VQI dataset, TIA is defined as focal neurologic deficits lasting less than 24 hours, whereas stroke is defined as symptoms lasting for 24 or longer hours. A history of coronary artery disease (CAD) included history of asymptomatic CAD, MI within 6 months of procedure, and unstable angina. A history of congestive heart failure (CHF) was defined as those with New York Heart Association Class I to IV. A history of chronic obstructive pulmonary disease (COPD) includes progressive lung diseases such as emphysema, chronic bronchitis, asbestosis, and non-reversible asthma with or without use of home oxygen. A history of diabetes mellitus included those patients with and/or without insulin use. A history of smoking was defined as either having current smoking use within the last month, or prior use defined as greater than or equal to 1 month prior to the procedure. Operative details were also compared. Variables included history of prior CEA or stenting, prior neck radiation, high-risk anatomy, American Society of Anesthesiologists (ASA) class, anesthesia type, and procedure time. High-risk anatomy was defined by prior neck surgery, neck stoma, contralateral laryngeal nerve palsy, common carotid artery lesions below the clavicle, or a combination thereof.
Primary outcomes evaluated included 30-day mortality, in-hospital stroke, MI, and a composite outcome of either 30-day mortality or stroke. Stroke was characterized as any ischemic or hemorrhagic event, whether ipsilateral, contralateral, cortical, retinal, or vertebrobasilar, identified postoperatively and prior to discharge. Secondary outcomes included other cardiac complications (new-onset CHF or dysrhythmias), prolonged length of stay defined as greater than or equal to 2 days, return to the operating room during the same admission, cranial nerve injury, reperfusion injury, access site complications, and hemodynamic instability requiring vasoactive support. Perioperative medication was also compared at admission and at discharge.
Statistical analyses.
Categorical variables were compared using the Pearson χ2 analyses; continuous variables were assessed for normalcy of distribution and compared using Kruskal-Wallis rank sum tests. Multivariable logistic regressions were completed adjusting for age, sex, CKD stage, anesthesia type, presence of dialysis, and history of CHF, diabetes, and COPD. Survival analysis was completed using Kaplan-Meier estimates, log-rank sum test, and adjusted Cox proportional hazards regression. Statistical significance was accepted at a P value of < .05. Analyses were performed using R statistical analysis software (RStudio, PBC).
RESULTS
Demographics and comorbidities.
Of 90,343 patients with CKD undergoing revascularization (CEA, n = 66,870; TCAR, n = 13,459; TFCAS, n = 10,014), 63% were asymptomatic and 37% symptomatic (Table I). Among symptomatic patients, the median age was 76 years for TCAR, compared with 74 and 72 years for CEA and TFCAS, respectively (P < .001). Among asymptomatic patients, sex differed (male: CEA, 59%; TCAR, 62%; TFCAS, 62%; P < .001). No difference was seen among symptomatic patients.
Table I.
Demographics, history, and preoperative medication by operative approach
| Asymptomatic (n = 57,037) | Symptomatic (n = 33,306) | |||||||
|---|---|---|---|---|---|---|---|---|
| CEA (n = 44,977) |
TCAR (n = 8151) |
TFCAS (n = 3909) |
P value | CEA (n = 21,893) |
TCAR (n = 5308) |
TFCAS (n = 6105) |
P value | |
| Age, years | 73 (12) | 76 (11) | 73 (11) | < .001 | 74 (13) | 76 (13) | 72 (14) | < .001 |
| Sex | < .001 | .08 | ||||||
| Male | 26,344 (59) | 5065 (62) | 2437 (62) | 13,274 (61) | 3263 (61) | 3793 (62) | ||
| Female | 18,633 (41) | 3086 (38) | 1472 (38) | 8619 (39) | 2045 (39) | 2312 (38) | ||
| Race | < .001 | < .001 | ||||||
| White | 40,460 (90) | 7451 (91) | 3551 (91) | 19,315 (88) | 4736 (89) | 5367 (88) | ||
| Black or African American | 2168 (4.8) | 376 (4.6) | 180 (4.6) | 1325 (6.1) | 320 (6.0) | 450 (7.4) | ||
| Asian | 500 (1.1) | 76 (0.9) | 38 (1.0) | 230 (1.1) | 54 (1.0) | 61 (1.0) | ||
| American Indian or Alaskan Native | 113 (0.3) | 31 (0.4) | 9 (0.2) | 62 (0.3) | 24 (0.5) | 18 (0.3) | ||
| Native Hawaiian or other Pacific Islander | 51 (0.1) | 8 (<0.1) | 4 (0.1) | 14 (<0.1) | 4 (<0.1) | 2 (<0.1) | ||
| More than 1 race/other | 1685 (3.7) | 209 (2.6) | 127 (3.2) | 947 (4.3) | 170 (3.2) | 207 (3.4) | ||
| Prior stroke | 3144 (7.0) | 1254 (15) | 688 (18) | < .001 | 12,379 (57) | 2856 (54) | 3277 (54) | < .001 |
| Time from stroke to procedure date, weeks | 2.4 (12.4) | 24.6 (213.3) | 28.3 (211.6) | < .001 | ||||
| Prior TIA | 1975 (4.4) | 749 (9.2) | 380 (9.7) | < .001 | 10,308 (47) | 2413 (45) | 2374 (39) | < .001 |
| Time from TIA to procedure date, weeks | 3.4 (9.7) | 4.0 (17.1) | 3.6 (16.3) | < .001 | ||||
| Comorbidities | ||||||||
| Hypertension | 41,337 (92) | 7528 (92) | 3578 (92) | .24 | 19,770 (90) | 4911 (93) | 5488 (90) | < .001 |
| CAD | 13,719 (31) | 4609 (57) | 2086 (53) | < .001 | 5437 (25) | 2607 (49) | 2803 (46) | < .001 |
| CHF | 6225 (14) | 1505 (18) | 771 (20) | < .001 | 2865 (13) | 1015 (19) | 1136 (19) | < .001 |
| Diabetes | 17,443 (39) | 3120 (38) | 1633 (42) | < .001 | 8126 (37) | 2103 (40) | 2433 (40) | < .001 |
| COPD | 10,517 (23) | 2083 (26) | 1067 (27) | < .001 | 4885 (22) | 1347 (25) | 1616 (26) | < .001 |
| Smoking | 32,672 (73) | 5926 (73) | 2900 (74) | .11 | 15,320 (70) | 3705 (70) | 4323 (71) | .39 |
| Preoperative medicationa | ||||||||
| Aspirinb | 37,781 (84) | 7315 (90) | 3478 (89) | < .001 | 18,450 (84) | 4749 (89) | 5168 (85) | < .001 |
| P2Y12 inhibitorc | 14,905 (33) | 7152 (88) | 3197 (82) | < .001 | 9553 (44) | 4600 (87) | 4558 (75) | < .001 |
| Dual anti-platelet | 12,415 (28) | 6531 (80) | 2902 (74) | < .001 | 8182 (37) | 4235 (80) | 4156 (68) | < .001 |
| Anticoagulationd | 5521 (12) | 1233 (15) | 536 (14) | < .001 | 3089 (14) | 823 (16) | 918 (15) | .02 |
| Statin | 38,095 (85) | 7305 (90) | 3370 (86) | < .001 | 18,872 (86) | 4757 (90) | 5024 (82) | < .001 |
| Beta-blockere | 26,646 (59) | 4939 (61) | 2368 (61) | .03 | 11,865 (54) | 3063 (58) | 3249 (53) | < .001 |
CAD, Coronary artery disease; CEA, carotid endarterectomy; CHF, congestive heart failure; COPD, chronic obstructive pulmonary disease; TCAR, transcarotid artery revascularization; TFCAS, transfemoral carotid artery stenting; TIA, transient ischemic attack.
Data are presented as number (%) or median (interquartile range).
Boldface P values indicate statistical significance.
Taken within 36 hours of procedure.
Includes drugs containing ASA, such as aggrenox.
Includes clopidogrel, prasugrel, ticlopidine, ticagrelor, aggrenox, or other or PAR1 inhibitor.
Includes vitamin K antagonists, dabigatran, rivaroxaban, apixaban, edoxaban and other; taken within 30 days of procedure.
Includes chronic (>30 days) use; use less than 30 days prior to procedure and day-of procedure.
Asymptomatic patients differed in rates of CAD (CEA, 31%; TCAR, 57%; TFCAS, 53%; P < .001) CHF (CEA, 14%; TCAR, 18%; TFCAS, 20%; P < .001), and COPD (CEA, 23%; TCAR, 26%; TFCAS, 27%; P < .001). Rates of diabetes also differed between approach groups (CEA, 39%; TCAR, 38%; TFCAS, 42%; P < .001). Hypertension and a history of smoking were prevalent across all approaches without significant differences. Prior history of stroke and TIA rates also differed in the asymptomatic cohort (Stroke: CEA, 7%; TCAR, 15%; TFCAS, 18%; P < .001; TIA: CEA, 4.4%; TCAR, 9.2%; TFCAS, 9.7%; P < .001).
In the symptomatic cohort, rates of CAD differed (CEA, 25%; TCAR, 49%; TFCAS, 46%; P < .001) between groups. Differences in CHF (CEA, 13%; TCAR, 19%; TFCAS, 19%; P < .001), diabetes (CEA, 37%; TCAR, 40%; TFCAS, 40%; P < .001), and COPD (CEA, 22%; TCAR, 25%; TFCAS, 26%; P < .001) were also present. Unlike the asymptomatic cohort, the symptomatic cohort had differences in rates of hypertension (CEA, 90%; TCAR, 93%; TFCAS, 90%; P < .001). History of smoking did not differ by operative approach. Prior stroke and TIA rates also differed (stroke: CEA, 57%; TCAR, 54%; TFCAS, 54%; P < .001; TIA: CEA, 47%; TCAR, 45%; TFCAS, 39%; P < .001).
The majority of patients in this study were in CKD stage 2 (asymptomatic: CEA, 62%; TCAR, 59%; TFCAS, 59%; symptomatic: CEA, 66%; TCAR, 62%; TFCAS, 62%) (Table II). In the asymptomatic group, median baseline GFR differed significantly between approach groups (CEA, 65; TCAR, 63; TFCAS, 64 mL/min/1.72 m2; P < .001). Differences were also seen in median baseline GFR among the symptomatic group (CEA, 67; TCAR, 65; TFCAS, 65 mL/min/1.72 m2; P < .001). Dialysis prevalence was similar across treatments in the asymptomatic group, but different in the symptomatic group (CEA, 1.3%; TCAR, 2.0%; TFCAS, 2.2%; P < .001).
Table II.
Univariate analysis of baseline glomular filtration rate (GFR), chronic kidney disease (CKD) stage, and dialysis
| Asymptomatic (n = 57,037) | Symptomatic (n = 33,306) | |||||||
|---|---|---|---|---|---|---|---|---|
| CEA (n = 44,977) |
TCAR (n = 8151) |
TFCAS (n = 3909) |
P value | CEA (n = 21,893) |
TCAR (n = 5308) |
TFCAS (n = 6105) |
P value | |
| Baseline GFR, mL/min/1.72 m2 |
65 (26) | 63 (26) | 64 (26) | < .001 | 67 (25) | 65 (25) | 65 (26) | < .001 |
| CKD stage (GFR, mL/min/1.72 m2) |
< .001 | <0.001 | ||||||
| 2 (60–89) | 27,901 (62) | 4769 (59) | 2312 (59) | 14,382 (66) | 3266 (62) | 3793 (62) | ||
| 3a/b (30–59) | 14,943 (33) | 3025 (37) | 1414 (36) | 6666 (30) | 1785 (34) | 2009 (33) | ||
| 4 (15–29) | 1242 (2.8) | 189 (2.3) | 105 (2.7) | 496 (2.3) | 136 (2.6) | 152 (2.5) | ||
| 5 (<15)a | 891 (2.0) | 168 (2.1) | 78 (2.0) | 349 (1.6) | 121 (2.3) | 151 (2.5) | ||
| Dialysis | 678 (1.5) | 143 (1.8) | 63 (1.6) | .24 | 287 (1.3) | 107 (2.0) | 132 (2.2) | < .001 |
CEA, Carotid endarterectomy; TCAR, transcarotid artery revascularization; TFCAS, transfemoral carotid artery stenting. Data are presented as number (%) or median (interquartile range).
Boldface P values indicate statistical significance.
Excludes dialysis patients.
Use of preoperative antiplatelet, anticoagulation, statin, and beta-blocker use also differed among groups in the asymptomatic cohort (ASA, P2Y12 inhibitor, dual anti-platelet, anticoagulation, statin; P < .001; beta-blocker; P = .03) (Table I). The use of preoperative medication also differed among groups in the symptomatic cohort (ASA, P2Y12, dual anti-platelet, statin, beta-blocker; P < .001; anticoagulation; P = .02). These differences were also present at discharge for both groups, with the exception of aspirin in the symptomatic cohort (P = .61) (Table III).
Table III.
Discharge medications
| Asymptomatic (n = 57,037) | Symptomatic (n = 33,306) | |||||||
|---|---|---|---|---|---|---|---|---|
| CEA (n = 44,977) |
TCAR (n = 8151) |
TFCAS (n = 3909) |
P value | CEA (n = 21,893) |
TCAR (n = 5308) |
TFCAS (n = 6105) |
P value | |
| Aspirin | 40,806 (91) | 7470 (92) | 3574 (91) | .02 | 19,886 (91) | 4844 (91) | 5544 (91) | .61 |
| P2Y12 inhibitor | 18,688 (42) | 7849 (96) | 3749 (96) | < .001 | 10,077 (46) | 5090 (96) | 5723 (94) | < .001 |
| Dual antiplatelet therapy | 16,218 (36) | 7215 (89) | 3440 (88) | < .001 | 8809 (40) | 4659 (88) | 5247 (86) | < .001 |
| Anticoagulation | 6002 (13) | 1240 (15) | 559 (14) | < .001 | 3302 (15) | 870 (16) | 912 (15) | .04 |
| Statin | 39,992 (89) | 7498 (92) | 3506 (90) | < .001 | 20,095 (92) | 4956 (93) | 5534 (91) | < .001 |
| Beta-blocker | 25,849 (57) | 4562 (56) | 2159 (55) | .002 | 11,120 (51) | 2782 (52) | 2998 (49) | .002 |
CEA, Carotid endarterectomy; TCAR, transcarotid artery revascularization; TFCAS, transfemoral carotid artery stenting. Data are presented as number (%).
Boldface P values indicate statistical significance.
Operative details.
Differences were observed among both symptom status cohorts across all measured operative details (Table IV). Notable differences included those seen in prior CEA/CAS, neck radiation, high-risk anatomy, and anesthesia type. Differences were observed in both cohorts in terms of prior CEA/CAS (asymptomatic: CEA, 17%; TCAR, 33%; TFCAS, 42%; P < .001; symptomatic: CEA, 9.7%; TCAR, 19%; TFCAS, 24%; P < .001). Frequency of prior neck radiation also differed (asymptomatic: CEA, 1.2%; TCAR, 5.2%; TFCAS, 9.3%; P < .001; symptomatic: CEA, 1.3%; TCAR, 4.8%; TFCAS, 7.2%; P < .001). This was also seen for high-risk anatomy (asymptomatic: CEA, 4.2%; TCAR, 44%; TFCAS, 45%; P < .001; symptomatic: CEA, 3.4%; TCAR, 40%; TFCAS, 37%; P < .001). Anesthesia type also differed (asymptomatic [general]: CEA, 92%; TCAR, 84%; TFCAS, 23%; P < .001; symptomatic [general]: CEA, 94%; TCAR, 84%; TFCAS, 26%; P < .001).
Table IV.
Univariate analysis of operative details and prior surgical/radiation history
| n (%) | Asymptomatic (n = 57,037) | Symptomatic (n = 33,306) | ||||||
|---|---|---|---|---|---|---|---|---|
| CEA (n = 44,977) |
TCAR (n = 8151) |
TFCAS (n = 3909) |
P value | CEA (n = 21,893) |
TCAR (n = 5308) |
TFCAS (n = 6105) |
P value | |
| Prior CEA/carotid stent | 7860 (17) | 2729 (33) | 1631 (42) | < .001 | 2128 (9.7) | 1000 (19) | 1477 (24) | < .001 |
| Prior neck radiation | 558 (1.2) | 422 (5.2) | 363 (9.3) | < .001 | 285 (1.3) | 254 (4.8) | 441 (7.2) | < .001 |
| High-risk anatomya | 1887 (4.2) | 3590 (44) | 1778 (45) | < .001 | 743 (3.4) | 2138 (40) | 2283 (37) | < .001 |
| ASA class | < .001 | < .001 | ||||||
| 2 | 1692 (3.8) | 264 (3.2) | 715 (18) | 576 (2.6) | 123 (2.3) | 746 (12) | ||
| 3 | 33,494 (74) | 5827 (71) | 2557 (65) | 15,271 (70) | 3331 (63) | 3854 (63) | ||
| 4 | 9511 (21) | 2036 (25) | 554 (14) | 5954 (27) | 1828 (34) | 1381 (23) | ||
| 5 | 14 (<0.1) | 2 (<0.1) | 3 (<0.1) | 9 (<0.1) | 2 (<0.1) | 33 (0.5) | ||
| Anesthesia type | < .001 | < .001 | ||||||
| Local/regional | 3376 (7.5) | 1331 (16) | 2997 (77) | 1314 (6.0) | 872 (16) | 4523 (74) | ||
| General | 41,601 (92) | 6820 (84) | 912 (23) | 20,579 (94) | 4436 (84) | 1582 (26) | ||
| Procedure time, minutes | 107 (52) | 67 (34) | 60 (38) | < .001 | 113 (55) | 67 (33) | 63 (44) | < .001 |
ASA, American Society of Anesthesiologists; CEA, carotid endarterectomy; TCAR, transcarotid artery revascularization; TFCAS, transfemoral carotid artery stenting.
Data are presented as number (%) or median (interquartile range).
Boldface P values indicate statistical significance.
Includes prior neck surgery, neck stoma, contralateral laryngeal nerve palsy, common carotid artery lesions below the clavicle, or a combination thereof.
PERIOPERATIVE OUTCOMES
Thirty-day mortality.
In asymptomatic patients, unadjusted analysis revealed differences in 30-day mortality (CEA, 0.5%; TCAR, 0.3%; TFCAS, 0.5%; P = .01) (Table V). After adjustment, TCAR was associated with reduced mortality compared with CEA (adjusted odds ratio [aOR], 0.4; 95% confidence interval [CI], 0.3–0.7) and TFCAS (aOR, 0.5; 95% CI, 0.2–0.95) (Table VI).
Table V.
Univariate analysis of outcomes
| Asymptomatic (n = 57,037) | Symptomatic (n = 33,306) | |||||||
|---|---|---|---|---|---|---|---|---|
| CEA (n = 44,977) |
TCAR (n = 8151) |
TFCAS (n = 3909) |
P value | CEA (n = 21,893) |
TCAR (n = 5308) |
TFCAS (n = 6105) |
P value | |
| 30-day mortality | 240 (0.5) | 23 (0.3) | 18 (0.5) | .01 | 211 (1.0) | 26 (0.5) | 75 (1.2) | < .001 |
| Stroke | 423 (0.9) | 79 (1.0) | 54 (1.4) | .03 | 434 (2.0) | 101 (1.9) | 158 (2.6) | .01 |
| 30-day mortality/stroke | 613 (1.4) | 100 (1.2) | 69 (1.8) | .06 | 593 (2.7) | 123 (2.3) | 227 (3.7) | < .001 |
| TIA | 182 (0.4) | 36 (0.4) | 22 (0.6) | .3 | 154 (0.7) | 55 (1.0) | 61 (1.0) | .01 |
| MIa | 303 (0.7) | 46 (0.6) | 15 (0.4) | .06 | 178 (0.8) | 37 (0.7) | 34 (0.6) | .11 |
| Other cardiac complicationsb | 841 (1.9) | 118 (1.4) | 55 (1.4) | .01 | 448 (2.0) | 120 (2.3) | 134 (2.2) | .54 |
| Length of stay (>2 days) | 7505 (17) | 1051 (13) | 454 (12) | < .001 | 9490 (43) | 2057 (39) | 2816 (46) | < .001 |
| Access site complications | – | 237 (2.9) | 103 (2.6) | .43 | – | 165 (3.1) | 210 (3.4) | .35 |
| Reperfusion injuryc | 49 (0.1) | 14 (0.2) | 8 (0.2) | .11 | 51 (0.2) | 4 (<0.1) | 54 (0.9) | < .001 |
| Cranial nerve injuryd | 1031 (2.3) | 19 (0.2) | – | < .001 | 667 (3.0) | 22 (0.4) | – | < .001 |
| Return to operating roome | 801 (1.8) | 73 (0.9) | 15 (0.4) | < .001 | 545 (2.5) | 56 (1.1) | 24 (0.4) | < .001 |
| IV blood pressure medication (indications) | <0.001 | <0.001 | ||||||
| Hypotension | 4254 (9.5) | 1369 (17) | 586 (15) | 2307 (11) | 839 (16) | 946 (15) | ||
| Hypertension | 8479 (19) | 1117 (14) | 351 (9.0) | 4765 (22) | 803 (15) | 813 (13) | ||
| Both | 641 (1.4) | 107 (1.3) | 33 (0.8) | 327 (1.5) | 104 (2.0) | 82 (1.3) | ||
CEA, Carotid endarterectomy; MI, myocardial infarction; TCAR, transcarotid artery revascularization; TFCAS, transfemoral carotid artery stenting; TIA, transient ischemic attack.
Data are presented as number (%).
Boldface P values indicate statistical significance.
Diagnosed with troponin only or with electrocardiogram/clinical diagnosis.
New postoperative dysrhythmia and/or postoperative congestive heart failure.
Includes retroocular/ipsilateral headache, seizure or hemorrhage, or both.
Indications: hematoma/bleeding, stenosis, pseudoaneurysm, arteriovenous fistula, or infection-related.
TFCAS excluded (0%). Includes transient cranial nerve injury and presence of deficit at discharge.
Table VI.
Multivariable outcomes by operative approach for patients with chronic kidney disease (CKD)
| Asymptomatic | Symptomatic | Asymptomatic | Symptomatic | |
|---|---|---|---|---|
| OR (95% CI) Ref: CEA |
OR (95% CI) Ref: TFCAS |
|||
| 30-day mortality | ||||
| CEA | – | – | 1.1 (0.6–2.0) | 0.6 (0.4–0.8) |
| TCAR | 0.4 (0.3–0.7) | 0.5 (0.3–0.7) | 0.5 (0.2–0.95) | 0.3 (0.2–0.4) |
| TFCAS | 0.9 (0.5–1.6) | 1.8 (1.2–2.5) | – | – |
| 30-day mortality/stroke | ||||
| CEA | – | – | 0.79 (0.58–1.08) | 0.7 (0.5–0.8) |
| TCAR | 0.87 (0.69–1.08) | 0.8 (0.7–1.00) | 0.7 (0.5–0.97) | 0.6 (0.4–0.7) |
| TFCAS | 1.3 (0.93–1.7) | 1.5 (1.2–1.9) | – | – |
| Stroke | ||||
| CEA | – | – | 0.7 (0.5–1.04) | 0.8 (0.6–0.98) |
| TCAR | 1.04 (0.8–1.3) | 0.95 (0.8–1.2) | 0.8 (0.5–1.1) | 0.7 (0.5–0.97) |
| TFCAS | 1.4 (1.00–1.9) | 1.3 (1.02–1.7) | – | – |
| TIA | ||||
| CEA | – | – | 0.7 (0.4–1.3) | 0.7 (0.4–0.98) |
| TCAR | 1.06 (0.7–1.5) | 1.4 (0.99–1.89) | 0.8 (0.4–1.4) | 0.9 (0.6–1.4) |
| TFCAS | 1.4 (0.8–2.3) | 1.5 (1.02–2.3) | – | – |
| MIa | ||||
| CEA | – | – | 1.3 (0.7–2.4) | 1.3 (0.8–2.2) |
| TCAR | 0.8 (0.6–1.1) | 0.8 (0.6–1.1) | 1.00 (0.5–2.0) | 1.07 (0.6–1.9) |
| TFCAS | 0.8 (0.4–1.4) | 0.8 (0.5–1.2) | – | – |
| Other cardiac complicationsb | ||||
| CEA | – | – | 1.1 (0.8–1.5) | 0.8 (0.6–1.00) |
| TCAR | 0.7 (0.6–0.9) | 1.03 (0.8–1.3) | 0.8 (0.6–1.2) | 0.8 (0.6–1.1) |
| TFCAS | 0.9 (0.7–1.3) | 1.3 (1.00–1.7) | – | – |
| Length of stay >2 days | ||||
| CEA | – | – | 1.4 (1.2–1.5) | 0.7 (0.6–0.8) |
| TCAR | 0.7 (0.7–0.8) | 0.8 (0.8–0.9) | 0.98 (0.9–1.1) | 0.6 (0.5–0.6) |
| TFCAS | 0.7 (0.7–0.8) | 1.4 (1.3–1.5) | – | – |
| Reperfusion injuryc | ||||
| CEA | – | – | 0.4 (0.2–1.1) | 0.2 (0.1–0.2) |
| TCAR | 1.5 (0.7–2.7) | 0.4 (0.1–0.9) | 0.6 (0.2–1.8) | 0.1 (0.02–0.2) |
| TFCAS | 2.4 (0.9–5.7) | 6.7 (4.1–10.8) | – | – |
| Return to operating roomd | ||||
| CEA | – | – | 4.4 (2.7–8.0) | 6.0 (3.9–9.7) |
| TCAR | 0.5 (0.4–0.6) | 0.4 (0.3–0.6) | 2.2 (1.3–4.1) | 2.6 (1.6–4.4) |
| TFCAS | 0.2 (0.1–0.4) | 0.2 (0.1–0.3) | – | – |
| Cranial nerve injurye | ||||
| CEA | – | – | ||
| TCAR | 0.1 (0.1–0.2) | 0.1 (0.1–0.2) | ||
CEA, Carotid endarterectomy; CI, confidence interval; MI, myocardial infarction; OR, odds ratio; TCAR, transcarotid artery revascularization; TFCAS, transfemoral carotid artery stenting; TIA, transient ischemic attack.
Boldface values indicate statistical significance.
Adjusted for CKD stage 2–5, dialysis, age, gender, prior congestive heart failure, diabetes, chronic obstructive pulmonary disease, and anesthesia type (general vs non-general).
Diagnosed with troponin only or with electrocardiogram/clinical diagnosis.
New postoperative dysrhythmia and/or postoperative congestive heart failure.
Includes retroocular/ipsilateral headache, seizure or hemorrhage, or both.
Indications: hematoma/bleeding, stenosis, pseudoaneurysm, arteriovenous fistula, or infection-related.
TFCAS excluded (0%). Includes transient cranial nerve injury and presence of deficit at discharge.
Among symptomatic patients, univariate analysis also showed differences in mortality between approaches (CEA, 1.0%; TCAR, 0.5%; TFCAS, 1.2%; P < .001). In multivariate analysis TCAR was associated with a lower mortality risk compared with both CEA (aOR, 0.5; 95% CI, 0.3–0.7) and TFCAS (aOR, 0.3; 95% CI, 0.2–0.4). TFCAS was associated with a greater mortality risk than CEA (aOR, 1.8; 95% CI, 1.2–2.5).
Stroke.
Univariate analysis in the asymptomatic cohort showed significant differences (CEA, 0.9%; TCAR, 1.0%; TFCAS, 1.4%; P = .03); however, this was not seen after adjustment of potential confounders. In symptomatic patients, univariate analysis also showed differences (CEA, 2.0%; TCAR, 1.9%; TFCAS, 2.6%; P = .01). This persisted in multivariable analysis with CEA and TCAR both associated with a lower stroke risk when compared with TFCAS (aOR, 0.8; 95% CI, 0.6–0.98 and aOR, 0.7; 95% CI, 0.5–0.97, respectively). TCAR and CEA did not differ in postoperative stroke risk in asymptomatic or symptomatic patients.
Composite 30-day mortality/stroke.
Among asymptomatic patients, unadjusted analysis demonstrated no significant differences among approach types in the composite outcome (CEA, 1.4%; TCAR, 1.2%; TFCAS, 1.8%). After adjustment, TCAR was found to have a lower risk of composite 30-day mortality/stroke compared with TFCAS (aOR, 0.7; 95% CI, 0.5–0.97). CEA and TCAR did not differ in risk of the composite outcome.
For symptomatic patients, univariate analysis identified differences (CEA, 2.7%; TCAR, 2.3%; TFCAS, 3.7%; P < .001). This persisted in multivariable analysis. Both TCAR (aOR, 0.6; 95% CI, 0.4–0.7) and CEA (aOR, 0.7; 95% CI, 0.5–0.8) were associated with lower rates of stroke/death compared with TFCAS. There were no differences in risk of the composite outcome between TCAR and CEA (aOR, 0.8; 95% CI, 0.7–1.00) in the symptomatic cohort.
Cardiac complications.
Postoperative MI did not differ by symptom status or by approach type in either symptom cohort in univariate or multivariable analysis (asymptomatic: CEA, 0.7%; TCAR, 0.6%; TFCAS, 0.4%; P = .06; symptomatic: CEA, 0.8%; TCAR, 0.7%; TFCAS, 0.6%; P = .11).
In terms of other cardiac complications, univariate analysis demonstrated significant differences in the asymptomatic cohort (CEA, 1.9%; TCAR, 1.4%; TFCAS, 1.4%; P = .01). Multivariable analysis showed that asymptomatic patients treated with TCAR had a lower risk of cardiac complications compared with CEA (aOR, 0.7; 95% CI, 0.6–0.9). No other differences were identified.
In the symptomatic group, no significant differences between approach type were noted (CEA, 2.0%; TCAR, 2.3%; TFCAS, 2.2%; P = .54). However, after adjustment, TFCAS was associated with a higher risk of cardiac complications than CEA (aOR, 1.3; 95% CI, 1.00–1.7). Cardiac complication risk in symptomatic patients did not differ between TCAR and CEA after adjustment.
Survival analysis.
On assessment of long-term survival, significant differences in survival across approach types were observed for both symptomatic (P < .0001) and asymptomatic (P = .001) patients (Fig). In adjusted analysis, there were no differences in survival between TCAR or TFCAS compared with CEA (Table VII) among asymptomatic patients. However, TCAR was associated with improved survival compared with TFCAS (hazard ratio [HR], 0.8; 95% CI, 0.6–0.9). When symptomatic patients were assessed, mortality was similar between patients treated with CEA and TCAR. However, both CEA (HR, 0.6; 95% CI, 0.5–0.7) and TCAR (HR, 0.6; 95% CI, 0.5–0.8) were associated with improved mortality compared with TFCAS.
Fig.
Survival analyses (log-rank score) by symptom status. CEA, Carotid endarterectomy; TFCAS, transfemoral carotid artery stenting; TCAR, transcarotid artery revascularization.
Table VII.
Long-term survival by operative approach among patients with chronic kidney disease (CKD)
| Asymptomatic HR (95% CI)a | Symptomatic HR (95% CI)b | |||
|---|---|---|---|---|
| CEA | – | 0.9 (0.7–1.01) | – | 0.6 (0.5–0.7) |
| TCAR | 0.9 (0.8–1.03) | 0.8 (0.6–0.9) | 1.01 (0.9–1.2) | 0.6 (0.5–0.8) |
| TFCAS | 1.2 (0.99–1.4) | – | 1.6 (1.4–1.9) | – |
CEA, Carotid endarterectomy; CI, confidence interval; HR, hazard ratio; TCAR, transcarotid artery revascularization; TFCAS, transfemoral carotid artery stenting.
Boldface values indicate statistical significance.
Adjusted for CKD stage 2–5, dialysis, age, gender, prior congestive heart failure, diabetes, chronic obstructive pulmonary disease, and anesthesia type (general vs non-general).
C-index: 0.67.
C-index: 0.68.
DISCUSSION
This study demonstrated the safety and efficacy of TCAR and CEA in patients with CKD with rates of stroke/death less than 1% for asymptomatic patients and less than 3% for symptomatic patients. In this patient population, TCAR was associated with improved 30-day mortality in both asymptomatic and symptomatic patients and was comparable to CEA with regard to postoperative stroke, composite stroke/mortality, and MI. Importantly, TFCAS was associated with worse 30-day mortality and stroke with no improvement in MI compared with alternative revascularization approaches.
CKD is historically linked to increased morbidity and reduced life expectancy, with current research consistently categorizing patients with CKD as high-risk with varying degrees of risk and clinical outcomes.11–14 However, recent studies have also reported that TCAR achieves acceptable outcomes in treating carotid stenosis in other high-risk groups, such as patients with CHF, severe CAD, COPD requiring oxygen therapy, and those aged over 80 years.14,15 This trend in outcomes likely stems from providers favoring TCAR in those with significant medicalrisks. This is reflected in our results, as the CKD population is likely to be at higher risk due to concomitant medical comorbidities.
The guidelines for patients with renal insufficiency are notably limited, but this study found that carotid revascularization can be achieved with acceptable perioperative risk. Although TCAR and CEA were comparable in most outcomes, TCAR was associated with improved 30-day mortality among patients with CKD compared with CEA. Importantly, TFCAS had inferior outcomes compared to both alternative approaches. This finding fills a crucial gap in existing guidelines but also suggests that TCAR is a viable and safe option for this high-risk patient group and may provide some benefit compared with CEA. As cited in current guidelines, a single-center, retrospective study focusing on patients with CKD reported a 6-month risk of composite death, stroke, and MI looking at TFCAS in CKD HRs exceeding 2.5.16 Worse composite outcomes with TFCAS were also demonstrated in our study, further emphasizing the likely need to stop performing TFCAS in this population if anatomically feasible to perform TCAR or CEA. Current guidelines recommend medical management over carotid revascularization for asymptomatic high-risk patients, unless the perioperative morbidity and mortality rate is below 3%.8 Our study found that asymptomatic patients with CKD can safely undergo TCAR and CEA, with rates of stroke/death below 2%. Notably, cardiovascular risk is elevated in patients with CKD compared with the general population, and historically, these individuals have faced greater risks of cardiac complications from carotid interventions.2,10,17–19 Our findings showed asymptomatic patients undergoing TCAR had similar risk of MI, and lower risk of other cardiac complications compared with CEA. Consequently, these patients should not be excluded from carotid revascularization based on renal function.
Prior to the recent Centers for Medicare & Medicaid Services expansion of TCAR for standard surgical risk patients, TCAR has been limited for reimbursement to high surgical risk patients as reflected in this study’s analysis of comorbidities. Despite this, TCAR has shown superior or equivalent outcomes to CEA in this population, despite having a larger proportion of patients with higher surgical risk. Our findings suggest that revascularization with CEA and TCAR can be performed safely and aligns with established guidelines. Moreover, given its reduced 30-day mortality and other associated benefits, such as a lower risk of prolonged length of stay, cranial nerve injury, and returning to the operating room, TCAR may provide advantage over CEA among patients with CKD (Table VI).
Limitations.
There are several limitations to this study that should be noted. First, while this study reflects a large proportion of TCARs performed due to Centers for Medicare & Medicaid Services’ reporting standard, patients treated with CEA and TFCAS do not have mandated reporting. This selective enrollment could introduce some selection bias, as these centers might have different patient demographics, surgeon expertise, or resource availability compared with those that do not participate in such monitoring. Furthermore, there are variables that are unaccounted for in our analysis that likely have influenced the surgeon’s decision to preferentially select one procedure over another (ie, high-risk anatomy, history of neck radiation, etc).
Another limitation of the current study is the constraint imposed by the variables available in our dataset. These unmeasured variables might include patient genetic predispositions or even specifics of operative techniques and postoperative care, which could significantly influence the results. Additionally, the absence of data on medically managed patients restricts our ability to comment on the selection criteria for these individuals, potentially limiting the generalizability of our findings.
Perhaps the most critical limitation is the difficulty in reliably assessing long-term follow-up and long-term stroke-free survival. Although the immediate postoperative outcomes provide valuable insights between these approaches, long-term data are needed to comprehensively understand the efficacy and safety of these interventions over time. The lack of such data inhibits our ability to make definitive conclusions about the long-term benefits, risks, and overall survival rates following different carotid revascularization techniques. This gap highlights the need for improved extended follow-up periods and data collection.
CONCLUSIONS
Among patients with renal insufficiency, TCAR and CEA are associated with acceptable outcomes for the treatment of carotid artery stenosis. Despite similar rates of stroke and MI, TCAR was associated with a 30-day mortality benefit. TFCAS, however, is associated with increased morbidity and mortality, including 30-day mortality and stroke, and similar rates of MI. Given the increased risk of major morbidity and mortality, TFCAS should be avoided in patients with CKD who are other-wise anatomic candidates for TCAR or CEA.
ARTICLE HIGHLIGHTS.
Type of Research: Retrospective study of prospectively collected Vascular Quality Initiative data
Key Findings: In a study of 90,343 patients with chronic kidney disease (CKD), transcarotid artery revascularization (TCAR) showed lower 30-day mortality and stroke/death rates than carotid endarterectomy (CEA) and transfemoral carotid artery stenting (TFCAS), with similar risks for stroke, myocardial infarction, and cardiac complications, demonstrating its comparative safety and efficacy in both asymptomatic and symptomatic cohorts.
Take Home Message: Among patients with CKD, TCAR is associated with improved 30-day mortality and cardiac complications with no differences in stroke rates compared with CEA, whereas TFCAS is associated with worse mortality and stroke rates compared with both TCAR and CEA. Given the increased risk of major morbidity and mortality, TFCAS should not be performed in patients with CKD who are otherwise anatomic candidates for TCAR or CEA.
Footnotes
DISCLOSURES
None.
Presented at the Thirty-seventh Annual Meeting of the Western Vascular Society, Victoria, British Columbia, Canada, September 17–20, 2022
The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a conflict of interest.
REFERENCES
- 1.Deery SE, Holscher CM, Nejim B, et al. In-hospital and one-year outcomes are similar for women and men following transcarotid artery revascularization in symptomatic and asymptomatic patients. J Vasc Surg. 2022;75:572e580.e3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Zanoli L, Mikhailidis DP. Narrative Review of carotid disease and the kidney. Ann Transl Med. 2021;9:1210. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Schermerhorn ML, Liang P, Dakour-Aridi H, et al. In-hospital out-comes of transcarotid artery revascularization and carotid endar-terectomy in the Society for Vascular Surgery Vascular Quality Initiative. J Vasc Surg. 2020;71:87e95. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Li Y, Yang JJ, Zhu SH, Xu B, Lian W. Long-term efficacy and safety of carotid artery stenting versus endarterectomy: a meta-analysis of randomized controlled trials. PLoS One. 2017;12:e0180804. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Kashyap VS, So K, Schneider P, Rathore R, Pham T, Raghu M. One-year outcomes after transcarotid artery revascularization (TCAR) in the ROADSTER 2 trial. J Vasc Surg. 2022;76:466e473.e1. [DOI] [PubMed] [Google Scholar]
- 6.Kwolek CJ, Jaff MR, Leal JI, et al. Results of the ROADSTER multicenter trial of transcarotid stenting with dynamic flow reversal. J Vasc Surg. 2015;62:1227e1234. [DOI] [PubMed] [Google Scholar]
- 7.Elsayed N, Vasudevan RS, Zarrintan S, Barleben A, Kashyap VS, Malas MB. TransCarotid artery revascularization can Be safely performed in patients undergoing dialysis. Ann Vasc Surg. 2023;92: 57e64. [DOI] [PubMed] [Google Scholar]
- 8.AbuRahma AF, Avgerinos ED, Chang RW, et al. The Society for Vascular Surgery implementation document for management of extracranial cerebrovascular disease. J Vasc Surg. 2022;75:26Se98S. [DOI] [PubMed] [Google Scholar]
- 9.Bensley RP, Beck AW. Using the Vascular Quality Initiative to improve quality of care and patient outcomes for vascular surgery patients. Semin Vasc Surg. 2015;28:97e102. [DOI] [PubMed] [Google Scholar]
- 10.Inker LA, Eneanya ND, Coresh J, et al. New creatinine- and cystatin C-based equations to estimate GFR without race. N Engl J Med. 2021;385:1737e1749. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Ascher E, Marks NA, Schutzer RW, Hingorani AP. Carotid endarterectomy in patients with chronic renal insufficiency: a recent series of 184 cases. J Vasc Surg. 2005;41:24e29. [DOI] [PubMed] [Google Scholar]
- 12.Latz CA, Boitano LT, Wang LJ, et al. Perioperative outcomes for carotid revascularization on asymptomatic dialysis-dependent patients meet Society for Vascular Society guidelines. J Vasc Surg. 2021;74:195e202. [DOI] [PubMed] [Google Scholar]
- 13.Blecha M, DeJong M, Carlson K. Risk factors for mortality within 5 years of carotid endarterectomy for asymptomatic stenosis. J Vasc Surg. 2022;75:1945e1957. [DOI] [PubMed] [Google Scholar]
- 14.AbuRahma AF, AbuRahma ZT, Santini A, et al. A single-center experience of 30-day perioperative and one year clinical outcomes of transcarotid artery revascularization in 100 consecutive patients. Vascular. 2023;31:1161e1172. [DOI] [PubMed] [Google Scholar]
- 15.Kibrik P, Stonko DP, Alsheekh A, et al. Association of carotid revascularization approach with perioperative outcomes based on symptom status and degree of stenosis among octogenarians. J Vasc Surg. 2022;76:769e777.e2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Saw J, Gurm HS, Fathi RB, et al. Effect of chronic kidney disease on outcomes after carotid artery stenting. Am J Cardiol. 2004;94: 1093e1096. [DOI] [PubMed] [Google Scholar]
- 17.Yuo TH, Sidaoui J, Marone LK, Makaroun MS, Chaer RA. Revascularization of asymptomatic carotid stenosis is not appropriate in patients on dialysis. J Vasc Surg. 2015;61:670e674. [DOI] [PubMed] [Google Scholar]
- 18.Reil T, Shekherdimian S, Golchet P, Moore W. The safety of carotid endarterectomy in patients with preoperative renal dysfunction. Ann Vasc Surg. 2002;16:176e180. [DOI] [PubMed] [Google Scholar]
- 19.Rigdon EE, Monajjem N, Rhodes RS. Is carotid endarterectomy justified in patients with severe chronic renal insufficiency? Ann Vasc Surg. 1997;11:115e119. [DOI] [PubMed] [Google Scholar]