Association of left subclavian artery coverage without revascularization and spinal cord ischemia in patients undergoing thoracic endovascular aortic repair: A Vascular Quality Initiative® analysis

Abstract

Objectives

Investigate the impact of left subclavian artery coverage without revascularization on spinal cord ischemia development in patients undergoing thoracic endovascular aortic repair.

Methods

The Vascular Quality Initiative thoracic endovascular aortic repair module (April 2011–July 2014) was analyzed. Patients undergoing left subclavian artery coverage were divided into two groups according to revascularization status. The association between left subclavian artery revascularization with the primary outcome of spinal cord ischemia and the secondary outcome of stroke was assessed with multivariable analysis adjusting for between-group baseline differences.

Results

The left subclavian artery was covered in 508 (24.6%) of the 2063 thoracic endovascular aortic repairs performed. Among patients with left subclavian artery coverage, 58.9% underwent revascularization. Spinal cord ischemia incidence was 12.1% in the group without revascularization compared to 8.5% in the group undergoing left subclavian artery revascularization (odds ratio (95%CI): 1.48(0.82–2.68), P = 0.189). Multivariable analysis adjustment identified an independent association between left subclavian artery coverage without revascularization and the incidence of spinal cord ischemia (adjusted odds ratio (95%CI): 2.29(1.03–5.14), P = 0.043). Although the incidence of stroke was also higher for the group with a covered and nonrevascularized left subclavian artery (12.1% versus 8.5%), this difference was not statistically significant after multivariable analysis (adjusted odds ratio (95%CI): 1.55(0.74–3.26), P = 0.244).

Conclusion

For patients undergoing left subclavian artery coverage during thoracic endovascular aortic repair, the addition of a revascularization procedure was associated with a significantly lower incidence of spinal cord ischemia.

Keywords

Left subclavian artery thoracic endovascular aortic repair endovascular repair thoracic aorta spinal cord ischemia stroke

Introduction

Thoracic endovascular aortic repair (TEVAR) has become the treatment modality of choice for most pathologies of the descending thoracic aorta. The widespread use of this less invasive treatment option has contributed to a decrease in spinal cord ischemia (SCI) rates associated with the aortic repair but has not eliminated the risk of this catastrophic complication.^1,2

Because in a significant number of patients the aortic pathology involves the area adjacent to the origin of the left subclavian artery (LSCA), intentional LSCA coverage is frequently necessary to achieve adequate proximal landing zone for the endograft.^3–5

Intentional coverage of the LSCA without revascularization has been demonstrated to be an independent predictor of SCI.⁶ Others claim that although an association between coverage of the LSCA and SCI exists, it remains unclear if LSCA revascularization results in any protection,⁷ and because complications attributable to the LSCA revascularization are not negligible^4,8 a policy of selective LSCA revascularization is warranted.^5,9,10

A systematic review and meta-analysis commissioned by the Society for Vascular Surgery (SVS) demonstrating a higher but not statistically significant rate of SCI associated with LSCA coverage without revascularization¹¹ has been published in conjunction with a set of Clinical Practice Guidelines summarizing the SVS recommendations for the management of LSCA based on the best available evidence. According to those guidelines, routine LSCA revascularization is suggested for patients undergoing TEVAR who require LSCA coverage and recommended for patients who have an anatomy that would result in significantly compromised perfusion to critical organs.¹² The strength of these recommendations, however, was limited by the low level of the existing evidence.

The objective of the present study was to summarize contemporary strategies used to manage the LSCA during TEVAR among centers participating in the SVS Vascular Quality Initiative (VQI^®) and to investigate the association between LSCA coverage and SCI. We hypothesized that coverage of the LSCA without revascularization in patients undergoing TEVAR is associated with a significantly increased risk of SCI.

Methods

The Society for Vascular Surgery’s Patient Safety Organization Research Advisory Subcommittee (PSO-RAC) approved this study. The TEVAR module was abstracted from the VQI^® database for the period from April 2011 to July 2014. Institutional Review Board approval or patient consent was not necessary as all data collected under the auspices of the SVS PSO are in compliance with the Patient Safety Act, allowing patient data to be compiled for quality improvement purposes. Under this umbrella, all centers participating in the VQI^® are allowed to use nonidentifiable data both for quality improvement projects and outcomes research.¹³

The VQI^® of the SVS is designed to collect preoperative risk factors, procedural variables, outcomes, and follow-up data to be used to assess quality of care and determine best practices in vascular care through regional quality improvement projects and through collaborative research and is approved by the Agency for Healthcare Research & Quality. The extensive list of clinically relevant variables available allows risk-adjusted analysis including baseline demographic characteristics, comorbid conditions, and procedure-related factors. This initiative has been endorsed by most of the Vascular Societies in North America and participation by their members is being encouraged. Although participation in the VQI^® is voluntary, institutions electing to participate are required to comply with 100% data entry into the database, limiting the possibility of selection bias.

For the purpose of this study, all available data points from the TEVAR module were used. Patients with an aortic rupture upon presentation were not included. All patients who underwent coverage of the LSCA were identified and divided into two study groups according to how the covered subclavian artery was managed: coverage without revascularization or coverage with revascularization. Although several revascularization techniques both open and endovascular were utilized in this population no attempt was made to differentiate them during the analysis and they were all considered as a single group. Using the information available in the VQI it was not possible to determine the exact timing of the LSCA revascularization; however, all the revascularizations were performed acutely, during the index hospitalization in which the TEVAR was performed. Delayed LSCA revascularizations were not included in this study.

The two study groups were then compared to identify differences in preoperative baseline characteristics. Chi-square test or Fisher’s exact test was used for comparison of proportions and Student’s t-test or Mann–Whitney U test for comparison of means.

The primary outcome measure was the development of SCI, defined as the presence or of paraparesis or paraplegia upon hospital discharge. Stroke and in-hospital mortality were considered secondary outcomes. In order to identify independent predictors of SCI in the study population, all factors that were associated with SCI on univariate analysis at a P < .20 level were entered into a stepwise logistic regression model. The independent predictors, with adjusted odds ratio (AOR) and 95% confidence interval (CI) derived from the equation were reported. Area under the ROC curve for the regression was calculated.

To further investigate the association of LSCA management in the development of SCI, multivariable logistic regression analysis was performed to adjust for all baseline characteristics that were significantly different between the two study groups as well as all factors independently associated with SCI previously identified on stepwise logistic regression. For this analysis all independent predictors of SCI identified on stepwise regression were entered into a logistic regression model with SCI as the dependent variable and forcing the entry of the variable used to identify LSCA management in addition to all baseline factors that were significantly different between the study groups on univariate analysis.

For the secondary outcomes of stroke and in-hospital mortality, analogous analyses were performed using multivariable logistic regression analysis to derive odds ratio and 95% CI adjusting for factors significantly different between the two study groups.

Statistical significance was considered at P < .05. All statistical analysis was performed using IBM© SPSS^© for Mac, version 20.0.

Results

The TEVAR module database of the VQI^® for the study period contained 2063 patients, ruptures excluded. Coverage of the LSCA was necessary in 508 (24.6%) of these patients, which were selected as the study population (Figure 1). Among patients with LSCA coverage, 58.9% underwent revascularization and the majority of these patients (76.6%, 229/299) were revascularized using an open debranching procedure. The overall incidence of SCI and stroke for the TEVAR module are described in Figure 2. Detailed baseline characteristics for the study population and dichotomized by study group are reported in Table 1. Notable differences were noted both of statistical and clinical significance when the two study groups were compared. Patients who had the LSCA covered and revascularized were older, more likely to smoke, had higher rates of comorbidities such as hypertension, CHF, and COPD; were more likely to have had prior aortic surgery; and were more likely to have been considered not fit for open surgery and had significantly longer segments of aorta covered by the endograft.

Figure 1.

Study outline. LSCA: left subclavian artery; TEVAR: thoracic endovascular aortic repair.

Figure 2.

Crude incidence of neurologic complications after TEVAR according to LSCA management, including patients with LSCA not covered. LSCA: left subclavian artery; TEVAR: thoracic endovascular aortic repair. Spinal Ischemia rate, multiple comparisons: LSCA covered not revascularized versus LSCA covered and revascularized: OR(95% CI): 1.49 (0.83–2.70), P = .191; LSCA covered not revascularized versus LSCA not covered: OR(95% CI): 1.63 (1.03–2.60), P = .038; LSCA covered and revascularized versus LSCA not covered: OR(95% CI): 1.10 (0.70–1.73). P = .670. Stroke rate, multiple comparisons: LSCA covered not revascularized versus LSCA covered and revascularized: OR(95% CI): 1.47 (0.82–2.63), P = .198; LSCA covered not revascularized versus LSCA not covered: OR(95% CI): 5.39 (3.19–9.13), p < .001; LSCA covered and revascularized versus LSCA not covered: OR (95% CI): 3.68 (2.20–6.15). p < .001.

Table 1.

Study population baseline characteristics with dichotomized comparison.

	Total(508)	LSCA coveredand not revascularized(209, 41%)	LSCA and revascularized (299, 59%)	P value
Sex, female	40.0 (203)	40.7 (85)	39.5 (118)	.785
Race, White	71.9 (365)	73.2 (153)	70.9 (212	.570
Age > 65	50.2 (255)	44.0 (92)	54.5 (163)	.020
Transferred from another hospital or rehab facility	29.9 (149)	37.6 (76)	24.6 (73)	.002
Smoking history	66.5 (335)	51.5 (126)	69.9 (209)	.049
Currently smoking	33.9 (171)	35.6 (73)	32.8 (98)	.509
Hypertension	83.6 (424)	76.4 (159)	88.6 (265)	<.001
Diabetes	15.4 (78)	17.3 (36)	14.1 (42)	.325
Diabetes on insulin	3.6 (18)	4.8 (10)	2.7 (8)	.205
CAD	16.0 (81)	13.5 (28)	17.7 (53)	.197
CABG or PCI	14.7 (74)	12.6 (26)	16.1 (48)	.268
CHF	7.5 (38)	4.3 (9)	9.7 (29)	.022
COPD	18.8 (95)	14.4 (30)	21.8 (65)	.036
Home O₂	1.8 (9)	1.4 (3)	2.0 (6)	.743
BMI>=35	13.3 (67)	14.6 (30)	12.4 (37)	.476
Dialysis	3.5 (18)	2.9 (6)	4.0 (12)	.493
Elevated creatinine (>1.78 mg/dl)	6.6 (32)	6.1 (12)	7.0 (20)	.678
Abnormal stress test	3.2 (16)	2.0 (4)	4.0 (12)	.196
ASA Class IV or higher	55.2 (277)	50.0 (102)	58.7 (175)	.054
Hgb<10 mg/dl	19.8 (99)	18.1 (37)	20.9 (62)	.450
Prior aortic surgery	29.8 (150)	22.2 (46)	35.1 (104)	.002
Prior bypass	10.5 (53)	6.8 (14)	13.1 (39)	.023
Prior CEA/CAS	2.4 (12)	1.9 (4)	2.7 (8)	.769
Prior aneurysm repair	22.6 (114)	16.4 (34)	26.8 (80)	.006
Prior PVI	4.4 (22)	4.8 (10)	4.0 (12)	.669
Prior major amputation	0.4 (2)	0.5 (1)	0.3 (1)	1.000
Preoperative ASA	53.9 (272)	49.5 (102)	56.9 (170)	.104
P2Y12 antagonist	7.1 (36)	6.3 (13)	7.7 (23)	.546
Preoperative antiplatelet agent	55.2 (279)	50.5 (104)	58.5 (175)	.074
Statin	51.2 (258)	46.6 (96)	54.4 (162)	.087
Beta-blocker	71.2 (359)	62.1 (128)	77.5 (231)	<.001
Ace inhibitor or ARB	33.4 (125)	32.2 (49)	34.2 (76)	.688
Genetic history of aneurysm	4.0 (20)	2.0 (4)	5.4 (16)	.056
Cardiac ejection fraction <30%	0.8 (4)	1.0 (2)	0.7 (2)	1.000
TAA	39.5 (200)	34.3 (71)	43.1 (129)	.045
TAAA	8.1 (41)	7.2 (15)	8.7 (26)	.557
Trauma	9.3 (47)	20.3 (42)	1.7 (5)	<.001
Dissection	30.2 (153)	30.0 (62)	30.4 (91)	.907
Urgent/Emergent	29.7 (151)	39.7 (83)	22.7 (68)	<.001
Unfit to open surgery	34.3 (171)	26.2 (53)	39.9 (118)	.002
Length covered >20 cm	40.4 (201)	31.0 (62)	46.8 (139)	<.001
Spinal drain (prophylactic)	32.2 (160)	25.5 (51)	36.7 (109)	.009
Spinal drain (postoperative)	8.2 (41)	10.1 (30)	5.5 (11)	.068
Endoleak at completion	9.7 (49)	9.7 (20)	9.7 (29)	.989
IVUS	51.3 (260)	50.0 (104)	52.2 (156)	.630
Maximal aneurysm diameter (mm)	60.7±15.9	59.3±17.2	61.4±15.0	.697
Minor axis aneurysm diameter (mm)	49.4±16.3	47.7±15.9	50.1±16.5	.734
Contrast (ml)	139.1±81.4	140.6±78.3	138.1±83.5	.684
Fluoroscopy time (min)	24.4±27.1	22.9±25.0	25.3±28.3	.758
Procedure time (min)	203.1±120.3	173.9±103.2	223.0±127.2	<.001
Crystalloids (ml)	2167.0±1375.4	2086.5±1144.3	2223.0±1514.8	.979
PRBC in OR (units)	0.84±2.07	0.69±1.76	0.95±2.27	.148
Total PRBC (units)	1.91±4.35	1.63±3.00	2.09±5.05	.957
Transfusion >= 3 units PRBC	22.8 (112)	21.3 (197)	23.7 (70)	.532

ASA: American Society of Anesthesiology; BMI: body mass index; CABG: coronary artery bypass graft; CAD: coronary artery disease; CAS: carotid artery stent; CEA: carotid endarterectomy; CHF: congestive heart failure; COPD: chronic obstructive pulmonary disease; PCI: percutaneous coronary intervention; PRBC: packed red blood cells; PTA: percutaneous transluminal angioplasty; PVI: peripheral vascular intervention; SD: standard deviation; TAA: thoracic aortic aneurysm; TAAA: thoracoabdominal aortic aneurysm.

Continuous data are presented as mean ± standard deviation and categorical data as number (%).

The overall incidence of SCI was 9.6% (49/508) in the study population. Postprocedural stroke occurred in 10% (51/508) of the patients and the overall mortality was 6.3% (32/508). Crude comparison of outcomes did not reveal any statistically significant difference in neurologic complications, mortality, ICU days, or hospital days, as described in detail in Table 2.

Table 2.

Crude outcomes comparison according to left subclavian artery management group.

	LSCA covered and not revascularized	LSCA and revascularized	OR (95% CI)	P value
In-hospital mortality	5.8 (12)	6.7 (20)	0.85 (0.41–1.79)	.678
Spinal cord ischemia	12.1 (24)	8.5 (25)	1.48 (0.82–2.68)	.189
Paraplegia	5.6 (11)	5.1 (15)	1.09 (0.49–2.43)	.825
Stroke	12.3 (25)	8.8 (26)	1.46 (0.82–2.61)	.196
			Mean difference (95% CI)	P value
Hospital LOS, mean ± SD	12.78±11.68	11.23±10.3	0.99 (−3.50, 0.41)	.188
Surgery to discharge days, mean ± SD	10.14±10.88	8.63±8.49	0.89 (−3.28, 0.25)	.150
ICU LOS, mean ± SD	5.82±7.94	5.63±7.03	0.67 (−1.52, 1.13)	.788

CI: confidence interval; ICU: intensive care unit; LOS: length of stay; LSCA: left subclavian artery; OR: odds ratio; SD: standard deviation.

Continuous data are presented as mean ± standard deviation and categorical data as number (%).

Univariate analysis identified multiple factors potentially associated with SCI in the study population (Table 3). After stepwise logistic regression, six of these factors were noted to be independently associated with SCI, including the coverage of LSCA without revascularization (Table 4). After risk-adjustment analysis including all statistically significant between-group differences as well as the factors associated with SCI identified on stepwise logistic regression, an independent association between LSCA coverage without revascularization and the development of SCI was confirmed (AOR (95% CI): 2.29 (1.03–5.14), P = .043).

Table 3.

Factors potentially associated with spinal cord ischemia on univariate analysis.

Factor	Spinal cord ischemia	OR (95% CI)	P value
LSCA coveredLSCA revascularized	12.1 (24/198)8.5 (25/294)	1.48 (0.82–2.68)	.189*
Female genderMale gender	12.4 (24/194)8.4 (25/298)	1.54 (0.85–2.78)	.149*
WhiteNon-white	9.1 (32/352) 12.1 (17/140)	0.72 (0.38–1.35)	.308
BMI >= 35BMI < 35	10.9 (7/64) 9.6 (41/426)	1.15 (0.49–2.69)	.742
Age > 65Age <= 65	12.2 (30/246) 7.7 (19/246)	1.65 (0.90–3.03)	.098*
Transfer from other facilityNo transfer	13.7 (20/146) 8.2 (28/341)	1.77 (0.96–3.26)	.063*
Smoking historyNever smoked	9.2 (30/325) 11.0 (18/164)	0.82 (0.44–1.52)	.540
Current smokerNot smoking currently	11.4 (19/166) 9.0 (29/323)	1.31 (0.71–2.41)	.385
HypertensionNo hypertension	10.6 (44/414) 6.4 (5/78)	1.73 (0.66–4.52)	.254
DiabetesNo diabetes	9.1 (7/77) 10.1 (42/414)	0.88 (0.38/2.05)	.777
Diabetes on insulinNo diabetes on insulin	5.6 (1/18) 10.1 (48/473)	0.52 (0.06–4.00)	1.000
CADNo CAD	11.4 (9/79) 9.7 (40/413)	1.19 (0.55–2.58)	.642
Any CABG or PCINo CABG or PCI	12.2 (9/74) 9.6 (40/416)	1.30 (0.60–2.81)	.501
CHFNo CHF	13.2 (5/38) 9.7 (44/453)	1.40 (0.52–3.79)	.496
COPDNo COPD	11.0 (10/91) 9.8 (39/400)	1.14 (0.54–2.38)	.722
Home O₂No Home O₂	33.3 (3/9) 9.5 (46/482)	4.73 (1.14–19.58)	.050*
HemodialysisNo hemodialysis	0 (0/18) 10.3 (49/474)	–	.239
Creatinine > 1.78 mg/dlCreatinine <= 1.78 mg/dl	22.6 (7/31) 9.4 (41/437)	2.81 (1.14–6.93)	.029*
ASA >= 4ASA < 4	12.3 (33/268) 7.2 (16/223)	1.81 (0.97–3.39)	.059*
Prior aortic surgeryNo prior aortic surgery	8.9 (13/146) 10.5 (36/342)	0.83 (0.42–1.61)	.585
Prior peripheral bypassNo prior peripheral bypass	7.7 (4/52) 10.3 (45/438)	0.72 (0.25–2.11)	.557
Prior CEA or CASNo prior CEA or CAS	16.7 (2/12) 9.8 (47/478)	1.83 (0.39–8.62)	.341
Prior aneurysm repairNo prior aneurysm repair	9.0 (10/111) 10.3 (39/379)	0.86 (0.41–1.79)	.692
Prior PVNo prior PVI	4.5 (1/22) 10.3 (48/467)	0.41 (0.05–3.15)	.713
Prior major amputationNo prior major amputation	0 (0/2) 10.0 (49/489)	–	1.000
Pre-op ASANo pre-op ASA	11.3 (30/266) 8.4 (19/225)	1.37 (0.75–2.52)	.297
Pre-op P2Y12 antagonistNo pre-op P2Y12 antagonist	8.3 (3/36) 10.1 (46/454)	0.80 (0.23–2.73)	1.000
Any pre-op antiplatelet agentNo pre-op antiplatelet agent	11.0 (30/273) 8.7 (19/218)	1.29 (0.70–2.36)	.404
Pre-op statinNo pre-op statin	10.3 (26/253) 9.7 (23/237)	1.06 (0.59–1.92)	.833
Pre-op beta-blockerNo pre-op beta-blocker	10.8 (38/351) 7.9 (11/139)	1.41 (0.70–2.85)	.333
Pre-op ACE inhibitor or ARBNo pre-op ACE inhibitor or ARB	9.0 (11/122) 10.3 (25/242)	0.86 (0.40–1.81)	.692
Genetic history of aneurysmNo genetic history of aneurysm	10.5 (2/19) 10.0 (47/468)	1.05 (0.23–4.70)	1.000
TAANo TAA	10.3 (20/195) 9.5 (28/295)	1.09 (0.59–1.99)	.780
TraumaNo trauma	14.0 (6/43) 9.4 (42/447)	1.56 (0.62–3.92)	.294
TAAANo TAAA	9.8 (4.41) 9.8 (44/449)	0.99 (0.34–2.92)	1.000
DissectionNo dissection	9.5 (14/148) 9.9 (34/342)	0.94 (0.49–1.82)	.869
Urgent/EmergentElective	15.9 (23/145) 7.5 (26/347)	2.32 (1.27–4.23)	.005*
Considered unfit to open surgeryNot considered unfit to open surgery	15.2 (25/165) 7.5 (24/318)	2.18 (1.20–3.96)	.009*
Covered aorta > 20 cmCovered aorta < 20 cm	15.2 (30/198) 6.5 (19/293)	2.57 (1.40–4.72)	.002*
Endoleak at completionNo endoleak at completion	12.5 (6/48) 9.7 (43/442)	1.32 (0.53–3.29)	.543
IVUS usedNo IVUS used	10.4 (26/251) 9.6 (23/240)	1.09 (0.60–1.96)	.775
Prophylactic spinal drainProphylactic spinal drain	13.4 (21/157) 8.4 (28/334)	1.68 (0.93–3.07)	.085*
Transfusion of >= 3 units of PRBCsTransfusion of < 3 units of PRBCs	19.3 (21/109) 6.9 (26/375)	3.20 (1.72–5.95)	< .001*

ASA: American Society of Anesthesiology; BMI: body mass index; CABG: coronary artery bypass graft; CAD: coronary artery disease; CAS: carotid artery stent; CEA: carotid endarterectomy; CHF: congestive heart failure; CI: confidence interval; COPD: chronic obstructive pulmonary disease; LSCA: left subclavian artery; OR: odds ratio; PCI: percutaneous coronary intervention; PRBC: packed red blood cells; PVI: peripheral vascular intervention; TAA: thoracic aortic aneurysm; TAAA: thoracoabdominal aortic aneurysm;

Table 4.

Independent predictors of spinal cord ischemia.

Factor	Adjusted OR (95% CI)	Adjusted P value
LSCA coverage without revascularization	2.31 (1.13–4.73)	.021
Covered aorta > 20 cm	4.49 (2.16–9.34)	< .001
Urgent/Emergent	2.41 (1.19–4.88)	.014
Transfusion of >= 3 units of PRBCs	3.35 (1.67–6.72)	.001
Unfit to open surgery	2.81 (1.41–5.61)	.003
Home O₂	6.23 (1.10–35.17)	.038

CI: confidence interval; LSCA: left subclavian artery; OR: odds ratio; PRBC: packed red blood cells.

Variables included in the analysis: all factors associated with spinal cord ischemia at P < .2 on univariate analysis (marked with * in Table 3).

Area under the ROC curve 0.80.

R² for the logistic regression: 0.23.

Although the incidence of stroke was also higher for the group with a covered and nonrevascularized LSCA (12.1% versus 8.5%), this difference was not statistically significant on crude comparison (OR (95% CI): 1.46 (0.81–2.61), P = .196) or after multivariable analysis adjustment (AOR (95%CI): 1.55 (0.74–3.26), adjusted P = .244).

Mortality was not affected by the LSCA management strategy. Adjusted in-hospital mortality was not significantly different between the study groups with 5.8% in the no revascularization group compared to 6.7% in the revascularization group, AOR (95% CI): 0.88 (0.34–2.31), adjusted P =.797.

Although coverage of the LSCA without revascularization may result in left upper extremity arterial insufficiency, no upper extremity amputation was documented in the VQI^® database.

Discussion

On this analysis of VQI^® patients undergoing TEVAR, coverage of the LSCA without revascularization was found to be an independent predictor of SCI, defined as the presence or of paraparesis or paraplegia upon hospital discharge. Patients who did not have the covered LSCA revascularized had a significantly higher risk-adjusted incidence of SCI compared to their counterparts who underwent LSCA revascularization.

These findings are in agreement with recent analysis of the prospective database of The European Collaborators on Stent/Graft Techniques for Aortic Aneurysm Repair (EUROSTAR), which included 606 patients undergoing endovascular aortic repair. In that study, SCI was found to be independently associated with coverage of the LSCA without revascularization. Additional risk factors for SCI identified by the EUROSTAR include renal failure, concomitant open abdominal aortic surgery, and the use of three or more stent grafts.⁶ The association between length of aortic coverage and SCI after TEVAR has also been demonstrated elsewhere^14,15 and was confirmed by the findings from the present study.

Paraplegia is one of the most dreaded complications after aortic repair and remains a significant problem in the endovascular era, with a 2.9% incidence of SCI reported on the W. L. Gore TAG pivotal trial.¹ In a recent meta-analysis, Wong et al.¹⁶ reported SCI incidence of 3.9% in a pool of 4936 patients reported from 46 studies, however with significant heterogeneity in the pooled data. Since Dake et al.¹⁷ published their preliminary report demonstrating the safe and effective treatment of descending thoracic aortic aneurysms with endovascular stent grafts, the use of endovascular techniques to repair thoracic aortic pathologies has increased exponentially. Proximal landing zone extension into zone 2 of the aortic arch has contributed to further expand the number of patients who are candidates for TEVAR.^18–20 As a consequence of this approach, rates as high as 51% of LSCA during TEVAR have been reported.¹⁰ For patients requiring LSCA coverage, initial concerns were directed to the potential complication of left upper extremity arterial inflow compromise. Coverage of the origin of the LSCA, however, was found to be generally well tolerated, rarely resulting in arterial insufficiency requiring revascularization.^21–27 For the rare instances where arterial insufficiency of the left upper extremity ensues, delayed revascularization provides adequate treatment. For that reason a policy of liberal LSCA coverage without revascularization was common in the early TEVAR series. Reports of neurologic complications potentially associated with that strategy, however, prompted reassessment of this strategy.⁴ Maintenance of LSCA patency seems to be important as it contributes to collateral pathways potentially relevant to spinal cord perfusion, particularly when a number of intercostal arteries are covered. Reduced perfusion through the anterior spinal and costocervical circulation is thought to be the underlying anatomopathological basis of SCI after TEVAR with LSCA coverage.²⁸ Although multifactorial in nature, the development of SCI after TEVAR was found to be associated with coverage of LSCA.⁶

Because a selective revascularization approach cannot identify all patients at increased risk for SCI or stroke resulting from coverage of the LSCA, for patients with planned coverage of the LSCA during TEVAR, routine preoperative LSCA revascularization is suggested by the SVS clinical practice guidelines (Grade 2, Level C). For patient in which absence of collateral pathway would likely result in significantly compromised perfusion to critical organs after coverage of the LSCA, revascularization is strongly recommended (Grade 1, Level C).¹² Interestingly, despite these liberal recommendations by the SVS for LSCA revascularization, 41% of the patients requiring LSCA coverage at VQI^® participating centers did not have a revascularization procedure.

The issue of length of aortic coverage is of ultimate importance when considering the complication of SCI. Despite being one of the most important and well-described risk factors for the development of SCI, the majority of the published series in which an attempt was made to investigate the impact of LSCA revascularization on the development of SCI after TEVAR did not have any information regarding the extent of aorta being covered. The most recent meta-analysis on this topic and the only one that exclusively included patients who had the LSCA covered failed to demonstrate any benefit of LSCA revascularization in reducing SCI rates.²⁹ It is important to highlight that only one¹⁰ of the six studies included in that review reported the extent of aortic coverage. That one study contributed with only 12% (145/1161) of the pooled data. In order to properly evaluate the association between LSCA revascularization and the development of spinal ischemia, adjustment for length of aortic coverage is paramount because as demonstrated in this study patients with long segments of aorta treated are significantly more likely to undergo LSCA revascularization. In fact, long thoracic aortic segment coverage (>20 cm) has been considered an absolute indication for LSCA revascularization.^22,30 Failure to perform the aforementioned adjustment can result in underestimation of the beneficial impact of LSCA revascularization.

Additional risk factors for the development of SCI after TEVAR identified in this study were Urgent/Emergent procedure, transfusion of >= 3 units of PRBCs, patient unfit to open surgery, and use of home O₂. These are not unexpected findings. Periods of hemodynamic instability are likely to have occurred in Urgent/Emergent procedures and in those procedures requiring transfusion of several units of blood products even if this information is not available in the database. On the other hand, both patients who were considered unfit for open surgery and those on home O₂ are likely to present with compromised oxygen delivery capacity at baseline, which further increase the risk of SCI during TEVAR.

Regarding the association between coverage of the LSCA and perioperative stroke significant controversy exists. Cooper et al.⁷ demonstrated in a meta-analysis that the risk of stroke during TEVAR increases in patients undergoing LSCA coverage, but LSCA revascularization offers no significant protection against this complication. This has been corroborated by Chung et al., in a review of the ACS-NSQIP database demonstrating that coverage of the LSCA during TEVAR was associated with a significantly increased 30-day stroke risk; however, this study was limited in its ability to demonstrate stroke prevention benefit from revascularization. In the present study, although patients with LSCA coverage had significantly higher stroke rates, no benefit was achieved with subclavian revascularization, therefore LSCA revascularization cannot be considered a stroke reducing intervention. One of the limitations our study shares with most of the studies investigating this issue is that no differentiation regarding anterior and posterior circulation stroke was possible. Feezor et al.³¹ demonstrated that a strategy of revascularization guided by an aggressive preoperative evaluation of the cerebral circulation resulted in a decrease in perioperative posterior strokes from 5.5% to 1.2%. Patterson et al. demonstrated that LSCA coverage resulted in overall higher risk of stroke both in the anterior and in the posterior circulation compared to no LSCA coverage. While revascularization did not significantly affect the rate of anterior stroke among patients with LSCA coverage (2.2% versus 2.8%, P = .105), it significantly reduced the risk of posterior stroke (0.7% versus 3.8%, adjusted P = .002).³²

Significant limitations can be identified in the present database study, intrinsic to its design and type of data available in the dataset. Selection bias is a major potential problem related to the retrospective nonrandomized design. The VQI^® however has a robust and granular list of clinically relevant variables specifically chosen for each of the modules, and we made an effort to utilize statistical tools to adjust for clinically and statistically relevant confounders and prevent them from clouding our findings. Nonetheless, unmeasured variables associated with the decision of performing revascularization of the LSCA may exist that were not controlled for. Whether LSCA coverage is simply a surrogate for extent and complexity of aortic repair and not necessarily a direct cause of SCI could not be completely elucidated by the data.

Additionally, although the use of endovascular strategies to maintain LSCA patency may affect the incidence of endoleaks the present study was not designed or powered to answer that question.

It is also conceivable that patients who underwent a LSCA revascularization who developed a stroke or other major complication prior to the planned TEVAR might have never undergone stenting of their aorta. These patients would not have been included in the study, underestimating the incidence of neurological complications directly associated with the revascularization.

Conclusion

For patients from the VQI^® database undergoing LSCA coverage during TEVAR, the addition of a revascularization procedure was associated with a significantly lower incidence of SCI. In the absence of prospective randomized data, this study adds to the literature supporting that LSCA revascularization may be associated with a decrease in the risk of SCI in patients undergoing TEVAR with proximal landing zone extending into Zone 2.

Footnotes

Author contributions

Conception and design: PGRT, KW, FAW

Analysis and interpretation: PGRT, KW, AWB, STS, FAW

Data collection: PGRT

Writing the article: PGRT, FAW

Critical revision of the article: PGRT, KW, AWB, STS, FAW

Final approval of the article: PGRT, KW, AWB, STS, FAW

Statistical analysis: PGRT

Overall responsibility: PGRT

Acknowledgment

Oral presentation at the 2015 Vascular Annual Meeting, Society for Vascular Surgery, 18–20 June 2015, Chicago, IL.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Makaroun

Dillavou

Kee

et al . Endovascular treatment of thoracic aortic aneurysms: results of the phase II multicenter trial of the GORE TAG thoracic endoprosthesis. J Vasc Surg 2005; 41: 1–9.

Bavaria

Appoo

Makaroun

et al . Gore TAG investigators. Endovascular stent grafting versus open surgical repair of descending thoracic aortic aneurysms in low-risk patients: a multicenter comparative trial. J Thorac Cardiovasc Surg 2007; 133: 369–377.

Feezor

and Lee

WA.

Management of the left subclavian artery during TEVAR. Semin Vasc Surg 2009; 22: 159–164.

Peterson

Eskandari

Gleason

et al . Utility of left subclavian artery revascularization in association with endoluminal repair of acute and chronic thoracic aortic pathology. J Vasc Surg 2006; 43: 433–439.

Kotelis

Geisbüsch

Hinz

et al . Short and midterm results after left subclavian artery coverage during endovascular repair of the thoracic aorta. J Vasc Surg 2009; 50: 1285–1292.

Buth

Harris

Hobo

et al . Neurologic complications associated with endovascular repair of thoracic aortic pathology: incidence and risk factors. A study from the European Collaborators on Stent/Graft Techniques for Aortic Aneurysm Repair (EUROSTAR) registry. J Vasc Surg 2007; 46: 1103–1111.e2.

Cooper

Walsh

Sadat

et al . Neurological complications after left subclavian artery coverage during thoracic endovascular aortic repair: a systematic review and meta-analysis. J Vasc Surg 2009; 49: 1594–1601.

Woo

Carpenter

Jackson

et al . Left subclavian artery coverage during thoracic endovascular aortic repair: a single-center experience. J Vasc Surg 2008; 48: 555–560.

Reece

Gazoni

Cherry

et al . Reevaluating the need for left subclavian artery revascularization with thoracic endovascular aortic repair. Ann Thorac Surg 2007; 84: 1201–1205; discussion 1205.

10.

Lee

Andersen

Williams

et al . Results with a selective revascularization strategy for left subclavian artery coverage during thoracic endovascular aortic repair. Ann Thorac Surg 2011; 92: 97–102; discussion 102–103.

11.

Rizvi

Murad

Fairman

et al . The effect of left subclavian artery coverage on morbidity and mortality in patients undergoing endovascular thoracic aortic interventions: a systematic review and meta-analysis. J Vasc Surg 2009; 50: 1159–1169.

12.

Matsumura

Lee

Mitchell

et al . The society for vascular surgery practice guidelines: management of the left subclavian artery with thoracic endovascular aortic repair. J Vasc Surg 2009; 50: 1155–1158.

13.

Cronenwett

Kraiss

and Cambria

RP.

The society for vascular surgery vascular quality initiative. J Vasc Surg 2012; 55: 1529–1537.

14.

Gravereaux

Faries

Burks

et al . Risk of spinal cord ischemia after endograft repair of thoracic aortic aneurysms. J Vasc Surg 2001; 34: 997–1003.

15.

Feezor

Martin

Hess

et al . Extent of aortic coverage and incidence of spinal cord ischemia after thoracic endovascular aneurysm repair. Ann Thorac Surg 2008; 86: 1809–1814; discussion 1814.

16.

Wong

Healy

Canning

et al . A systematic review of spinal cord injury and cerebrospinal fluid drainage after thoracic aortic endografting. J Vasc Surg 2012; 56: 1438–1447.

17.

Dake

Miller

Semba

et al . Transluminal placement of endovascular stent-grafts for the treatment of descending thoracic aortic aneurysms. N Engl J Med 1994; 29331: 1729–1734.

18.

Czerny

Zimpfer

Fleck

et al . Initial results after combined repair of aortic arch aneurysms by sequential transposition of the supra-aortic branches and consecutive endovascular stent-graft placement. Ann Thorac Surg 2004; 78: 1256–1260.

19.

McWilliams

Murphy

Hartley

et al . In situ stent-graft fenestration to preserve the left subclavian artery. J Endovasc Ther 2004; 11: 170–174.

20.

Saito

Kimura

Odashiro

et al . Feasibility of the Inoue single-branched stent-graft implantation for thoracic aortic aneurysm or dissection involving the left subclavian artery: short- to medium-term results in 17 patients. J Vasc Surg 2005; 41: 206–212; discussion 212.

21.

Hausegger

Oberwalder

Tiesenhausen

et al . Intentional left subclavian artery occlusion by thoracic aortic stent-grafts without surgical transposition. J Endovasc Ther 2001; 8: 472–476.

22.

Tiesenhausen

Hausegger

Oberwalder

et al . Left subclavian artery management in endovascular repair of thoracic aortic aneurysms and aortic dissections. J Card Surg 2003; 18: 429–435.

23.

Riesenman

Farber

Mendes

et al . Coverage of the left subclavian artery during thoracic endovascular aortic repair. J Vasc Surg 2007; 45: 90–94; discussion 94–95.

24.

Peterson

Wheatley

Kpodonu

et al . Treatment of type II endoleaks associated with left subclavian artery coverage during thoracic aortic stent grafting. J Thorac Cardiovasc Surg 2008; 136: 1193–1199.

25.

Lambrechts

Casselman

Schroeyers

et al . Endovascular treatment of the descending thoracic aorta. Eur J Vasc Endovasc Surg 2003; 26: 437–444.

26.

Görich

Asquan

Seifarth

et al . Initial experience with intentional stent-graft coverage of the subclavian artery during endovascular thoracic aortic repairs. J Endovasc Ther 2002; 9: II39–II43.

27.

Rehders

Petzsch

Ince

et al . Intentional occlusion of the left subclavian artery during stent-graft implantation in the thoracic aorta: risk and relevance. J Endovasc Ther 2004; 11: 659–666.

28.

Noor

Sadat

Hayes

et al . Management of the left subclavian artery during endovascular repair of the thoracic aorta. J Endovasc Ther 2008; 15: 168–176.

29.

Hajibandeh

Antoniou

et al . Meta-analysis of left subclavian artery coverage with and without revascularization in thoracic endovascular aortic repair. J Endovasc Ther 2016; 23: 634–641.

30.

Holt

Johnson

Hinchliffe

et al . Outcomes of the endovascular management of aortic arch aneurysm: implications for management of the left subclavian artery. J Vasc Surg 2010; 51: 1329–1338.

31.

Feezor

Martin

Hess

et al . Risk factors for perioperative stroke during thoracic endovascular aortic repairs (TEVAR). J Endovasc Ther 2007; 14: 568–573.

32.

Patterson

Holt

Nienaber

et al . Management of the left subclavian artery and neurologic complications after thoracic endovascular aortic repair. J Vasc Surg 2014; 60: 1491–1497.