The effect of left subclavian artery coverage on morbidity and mortality in patients undergoing endovascular thoracic aortic interventions: A systematic review and meta-analysis

Article Outline

• Abstract
• Methods
  • Eligibility criteria
  • Study identification
  • Data collection
  • Statistical analysis
    • Meta-analyses
    • Subgroup and sensitivity analyses
• Results
  • Study identification
  • Study quality
  • Meta-analysis
  • Subgroup and sensitivity analysis
• Discussion
  • Our findings
  • Strengths, limitations, and comparison with other systematic reviews
  • Implications for practice and research
• Conclusion
• Author contributions
• References
• Copyright

Objectives

Thoracic endografts (stent grafts) have emerged as a less invasive modality to treat various thoracic aortic lesions. The intentional coverage of the left subclavian artery (LSA) during the placement of these endografts is associated with several complications including stroke, spinal cord ischemia, and arm ischemia. In this review, we synthesize the available evidence regarding the complications associated with LSA coverage.

Methods

We searched electronic databases (MEDLINE and EMBASE) from January 1990 through February 2008 for studies that included patients who received thoracic endografts and had intentional LSA coverage. Eligible studies had a control group that either received the endograft without LSA coverage or had primary revascularization prior to coverage. Two independent reviewers determined trial eligibility and extracted descriptive, methodological and outcome data from each eligible study. Meta-analyses estimated Peto odds ratio (OR) and 95% confidence intervals (CI) to describe the strength of association between coverage and complications; the I² statistic described the proportion of inconsistency of treatment effect among studies not due to chance.

Results

We found 51 eligible observational studies. LSA coverage was associated with significant increase in the risk of arm ischemia (OR 47.7; CI, 9.9-229.3; I² = 72%, 19 studies) and vertebrobasilar ischemia (OR 10.8; CI, 3.17-36.7; I² = 0%; eight studies); and nonsignificant increase in the risk of spinal cord ischemia (OR 2.69; CI, 0.75-9.68; I² = 40%; eight studies) and anterior circulation stroke (OR 2.58; CI, 0.82-8.09; I² = 64%, 13 studies). There were no significant associations between LSA coverage and death, myocardial infarction, or transient ischemic attacks. The incidence of phrenic nerve injury as a complication of primary revascularization was 4.40% (CI, 1.60%-12.20%). Data on perioperative infection were sparse and rarely reported.

Conclusions

Very low quality evidence suggests that LSA coverage increases the risk of arm ischemia, vertebrobasilar ischemia, and possibly spinal cord ischemia and anterior circulation stroke.

Over the past decade, the use of thoracic endografts (stent grafts) has emerged as a less invasive modality to treat various thoracic aortic pathologies, including aneurysm, dissection, trauma, fistula, and penetrating ulcer. To expand the anatomic limits of this technology, intentional coverage of the left subclavian artery with the proximal aspect of the endograft without revascularization (carotid-subclavian artery bypass or transposition) has been performed.

Intentional covering of the left subclavian artery may lead to a higher incidence of extremity ischemia, spinal cord ischemia, or stroke. A few studies report that intentional coverage without revascularization is not associated with additional morbidity,¹, ², ³ whereas other studies report a higher incidence of postoperative arm ischemia and posterior circulation strokes compared with patients that did not undergo intentional coverage.⁴, ⁵ The discrepancy in the outcomes of intentional LSA coverage may be due to the primary etiology being treated, underlying patient comorbidities, or other patient and/or anatomic factors.

The Society for Vascular Surgery has formed a committee of experts in the treatment of thoracic aortic lesions to formulate clinical practice guidelines to guide patients and surgeons in making treatment decisions. This committee commissioned us to conduct a systematic review and meta-analysis to support the formulation of evidence-based recommendations. Thus, we sought to summarize the best available evidence about the effect of intentional LSA coverage on patients' morbidity and mortality.

Back to Article Outline

Methods 

The report of this protocol-driven systematic review was approved by the Committee on Thoracic Aortic Disease from the Society for Vascular Surgery and is in adherence with the standards for reporting Meta-analyses of Observational Studies in Epidemiology (MOOSE).⁶ The quality of evidence was rated using the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) methods.⁷

Eligibility criteria 

Eligible studies were controlled studies that enrolled patients receiving an endovascular device to repair descending thoracic aortic pathologies such as transection, aneurysm or dissection. Eligible studies enrolled patients in whom the LSA was intentionally covered to extend the proximal seal-zone for endovascular repair. Eligible studies included a comparison group including patients in whom the endograft placement did not cover the LSA or patients who underwent a primary revascularization procedure such as carotid-subclavian bypass or transposition.

We included studies that measured the outcomes of interest including death, arm ischemia, vertebrobasilar ischemia, anterior circulation stroke, transient ischemic attack (TIA), spinal cord ischemia, myocardial infarction, phrenic nerve paralysis, and infection. We defined spinal cord ischemia as permanent decrease or loss of lower extremity neurological function in the immediate postoperative period. Vertebrobasilar ischemia was defined as permanent loss of neurological function related to the posterior circulation. Arm ischemia was defined as any symptoms of hand ischemia that occurred at rest or with arm exertion and required revascularization. Studies were included regardless of their language, sample size, or duration of patient follow-up. Single cohort studies (ie, studies in which all patients received coverage without a concurrent comparison group) were excluded.

Study identification 

An expert reference librarian (P.J.E) designed and conducted the electronic search strategy with input from study investigators with expertise in conducting systematic reviews. To identify eligible studies, we searched electronic databases (MEDLINE and EMBASE) from January 1990 through February 2008. We chose the year 1990 as the start point for literature search because the use of thoracic endografts started in the late 1990s in Europe and in the early 2000s in the US; and because surgical techniques, device technology, and peri-operative care have likely changed sufficiently to decrease the relevancy of literature published prior to this date. We also sought references from experts, bibliographies of included trials, and the ISI (Institute for Scientific Information) Science Citation Index for publications that cited included studies.

MeSH and EMBASE subject headings were primarily used to describe the aorta, with subheadings to focus upon surgery or repair: aorta, thoracic/sugery, therapy, aortic surgery, therapy, aortic aneurysm/surgery, therapy, aortic disease, or aorta dissecting aneurysm or aortic rupture. A combination of subject headings and text words were used for the type of intervention: stent* or stentgraft, in conjunction with endograft, endovascul*, endolumin*, intralumin*, endoprosth*, blood vessel prosthesis implantation, or blood vessel prosthesis. Text words were used to describe the placement: covered, uncovered, anchor*, planned within 2 words of transpos*, subclavian*, lsa. The outcomes of concern (risk*, postoperative complications, mortality, brain ischemia, cerebrovascular accident, cerebrovascular trauma, hypoxia, intracranial embolis and thrombosis, bacterial infections, intracranial hemorrhages, or paralysis) were combined with all terms, and limited to clinical trials, comparative studies, practice guidelines, and other clinical studies. Detailed search strategy is available from study authors upon request.

Paired reviewers working independently screened all titles and abstracts for eligibility. References that were deemed potentially relevant were retrieved in full text for full text evaluation against eligibility criteria. The chance adjusted inter-reviewer agreement (kappa statistic) about study eligibility was 0.96 (95% confidence interval (CI) 0.96-1.0). Disagreements were resolved by consensus (the two reviewers discussed the study and reached a consensus), and when disagreement continued, by arbitration (a third reviewer adjudicated the study).

Data collection 

Two reviewers working independently and using a standardized form extracted data from all eligible studies. We extracted descriptive data (number of patients in each study arm, patients' age, gender, indication for endograft placement, urgency of procedure, control group description, type of endograft, comorbidities such as diabetes, coronary artery disease, hypertension, and history of a previous stroke); methodological data (study design, randomization procedures if any, bias protection measures, proportions of patients lost to follow up, funding source, the prognostic comparability of the two study groups, ascertainment of exposure and outcome, and blinding of outcome assessors); and outcome data (death, arm ischemia, vertebrobasilar ischemia, anterior circulation stroke, TIA, spinal cord ischemia, myocardial infarction, phrenic nerve paralysis, and infection). We contacted authors of included studies by e-mail for clarification and to obtain missing data.

Statistical analysis 

Anticipating that studies will have rare events and report sparse data, we used Peto odds ratio (OR)⁸ as the measure of effect for dichotomous outcomes and estimated the 95% confidence intervals (CI) for each outcome to reflect the uncertainty of point estimates of effect. When data are sparse and events are rare, Peto odds ratio method is associated with less bias, more power, and better confidence interval coverage than other methods, provided that study arms are balanced and treatment effects are not large,⁹ which we anticipated to be the case in this meta-analysis. In this review, OR of 1.0 indicates no association between LSA coverage and a particular outcome whereas OR above 1.0 indicates that coverage without revascularization increased the risk for this outcome. If the confidence interval of the OR overlaps the value of 1.0, this would indicate that the results are not statistically significant.

We used the I² statistic, which estimates the percentage of heterogeneity across studies that is due to heterogeneity rather than chance (ie, the percentage of variability in treatment effects across trials that is not due to chance or random error, but rather due to real differences in study patients, design or interventions).¹⁰ I² values of ≤ 25%, 50%, and ≥ 75% represent low, moderate, and high inconsistency, respectively. Statistical analysis was conducted using Comprehensive Meta-Analysis, Version 2 (Biostat Inc., Englewood, New Jersey, USA).

We established a priori hypotheses to explore subgroup interactions and explain inconsistency in the direction and magnitude of effect among studies. These subgroup analyses were based on the indication for aortic repair (aneurysm vs. dissection vs. transsection) and the urgency of repair (urgent vs. elective), and the control intervention (no coverage vs. coverage after primary revascularization). We tested the hypotheses of a subgroup effect using a test of interaction.¹¹ We also planned to conduct meta-regression using the effect size as the dependent outcome variable and the duration of follow up as the independent continuous variable.

Anticipating that the outcomes of interest would be rare and that many studies would report zero events in both arms, we planned to conduct sensitivity analysis using the risk difference (RD) as the measure of effect. RD value of 0 indicates that the absolute risk of a particular outcome is similar between the two study arms. RD >0 indicates that coverage without revascularization increased the risk for this outcome. If the confidence interval of the RD overlaps the value of 0, this would indicate that the results are not statistically significant.

Using RD allows the inclusion of these “no event” studies, which increases sample size, improves precision, and includes the totality of evidence in generating meta-analytic estimates. RD has the shortcomings of being associated with higher inconsistency among studies and is very dependent on the control group event rates in the different studies.¹², ¹³ We planned to explore whether study conclusions were sensitive to the choice of outcome measure used.

Back to Article Outline

Results 

Study identification 

Fig 1 depicts our search and selection procedures. Table I summarizes the characteristics of the 51 eligible studies in terms of patients' age, gender, comorbidities, type of endograft, indications for placement, and elements of study quality. These studies enrolled 3365 patients with a mean study size of 70 patients. The majority of patients (75%) were males and the mean age was 58 years.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 1. 

  Process of study selection.

Table I. Description of included studies

CAD, Coronary artery disease; DM, diabetes mellitus; FP, for profit; HTN, hypertension; M, multi-center study; NFP, not-for-profit; NR, not reported; OB, observational or nonrandomized study; P, prospective study; R, retrospective study; S, single center study; UC, unclear.

Study quality 

All eligible studies were observational, comprising five prospective cohort studies and 46 retrospective chart reviews. Researchers ascertained the outcomes, which were mostly death or major morbidities, and the exposure, by reviewing medical charts and perioperative records. Reviewers determining the quality of the eligible studies had perfect agreement on study design, exposure and outcome ascertainment; however, the prognostic comparability of study cohorts at baseline was difficult to assess because there were multiple indications for endograft placement and these indications conferred variable prognosis given in part by the original aortic pathology regardless of the procedure patients received. Loss of follow-up was in general low and only exceeded 10% in two studies. Study funding was often not reported but on two occasions was provided by endograft manufacturers.¹⁴, ¹⁵ Outcomes were poorly described in the included studies and we had to accept the definitions of the original studies; therefore, it was not always clear whether cases of spinal cord, vertebrobasilar, and hand ischemia were permanent or that they required interventions/revascularization.

Meta-analysis 

LSA coverage was associated with significant increase in the risk of arm ischemia (OR 47.69; CI, 9.92-229.34; I² = 72%, 19 studies; Fig 2) and vertebrobasilar ischemia (OR 10.78; CI, 3.17-36.69; I² = 0%; eight studies; Fig 3); and nonsignificant increase in the risk of spinal cord ischemia (OR 2.69; CI, 0.75-9.68; I² = 40%; eight studies; Fig 4) and anterior circulation stroke (OR 2.58; CI, 0.82-8.09; I² = 64%, 13 studies; Fig 5).

• 
  • View Large Image
  • Download to PowerPoint

• Fig 2. 

  Arm ischemia: random-effects meta-analysis. Squares are odds ratios of individual studies, lines are 95% confidence intervals, diamond represents the pooled effect, and the width of the diamond is the 95% confidence interval of the pooled estimate.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 3. 

  Vertebrobasilar ischemia: random-effects meta-analysis. Squares are odds ratios of individual studies, lines are 95% confidence intervals, diamond represents the pooled effect, and the width of the diamond is the 95% confidence interval of the pooled estimate.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 4. 

  Spinal cord ischemia: random-effects meta-analysis. Squares are odds ratios of individual studies, lines are 95% confidence intervals, diamond represents the pooled effect, and the width of the diamond is the 95% confidence interval of the pooled estimate.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 5. 

  Anterior circulation stroke: random-effects meta-analysis. Squares are odds ratios of individual studies, lines are 95% confidence intervals, diamond represents the pooled effect, and the width of the diamond is the 95% confidence interval of the pooled estimate.

LSA coverage did not significantly affect the risk of death (OR 1.14; CI, 0.49-2.67; I² = 28%, 17 studies), myocardial infarction (OR 0.47; CI, 0.02-9.26; I² = 11%, two studies), or TIA (OR 0.94; CI, 0.19-4.74; I² = 12%, six studies). The incidence of phrenic nerve injury as a complication of primary revascularization was 4.40% (CI, 1.60%-12.20%). The baseline risks of outcomes of interest in patients who underwent LSA coverage are presented in Table II. Data on perioperative infection were sparse and rarely reported.

Table II. Baseline risk of outcomes of interest in patients who underwent LSA coverage

Outcome	Overall risk	Aneurysm	Dissection	Trauma
Arm ischemia	38/687(6%)	2/82(2%)	7/127(6%)	5/57(9%)
Spinal cord ischemia	27/673(4%)	NR	NR	NR
Vertebrobasilar ischemia	7/442(2%)	1/82(1%)	6/127(5%)	1/47(2%)
Anterior circulation stroke	33/676(5%)	1/26(4%)	3/99(3%)	NR
Death	29/447(6%)	2/18(11%)	1/60(2%)	4/50(8%)

LSA, Left subclavian artery; NR, no events reported/incalculable.

Data on myocardial infarction, infection, transient ischemic attacks are sparse and not well-reported.

Subgroup and sensitivity analysis 

When RD was used as the measure of effect instead of Peto OR, study conclusions did not change in terms of the outcomes of death (RD −0.49%; CI, −3.94%-2.96%; I² = 0%, 32 studies), myocardial infarction (RD −0.45%; CI, −3.41%-2.52%; I² = 0%, 31 studies), anterior circulation stroke (RD 1.76%; CI, −0.45%-3.97%; I² = 0%, 40 studies), TIA (RD −0.15%; CI, −2.07%-1.76%; I² = 0%, 41 studies), spinal cord ischemia (RD 1.18%; CI, −0.79%-3.14%, I² = 0%, 47 studies) or arm ischemia (RD 2.14%; CI, 0.15%-4.13%; I² = 4%, 31 studies). However, the outcome of vertebrobasilar ischemia became less precise due to the inclusion of numerous studies with zero events in both arms (RD 0.64%; CI, −1.15%-2.43%).

We found no significant subgroup interactions across subgroups defined by whether the aortic intervention was done on an urgent or elective basis or whether the indication for the intervention was aneurysm or dissection. Data on trauma/transsection was insufficient for subgroup analysis. When we tested for subgroup interaction based on the control group in the included studies (patients received aortic endograft without coverage vs. they received primary revascularization before coverage), we found no significant interaction for all outcomes except arm ischemia (P value for interaction = 0.01). This interaction means that the magnitude of increased risk of arm ischemia due to LSA coverage was greater when coverage is compared with no coverage; than when coverage is compared with coverage preceded by primary revascularization. Several other subgroup analyses were not conducted due to insufficient data. Table III depicts all these subgroup analyses. Meta-regression showed no significant association between the length of study follow-up (expressed in days) and outcomes of interest (P > .05 for all outcomes).

Table III. Subgroup analyses

Outcome	Peto OR (95% CI)	P value for interaction
Death
Control group: revascularization	0.76(0.18-3.29)	.50
Control group: no coverage	1.43(0.46-4.42)
Elective procedure	327.20(0.28-∞)	.61
Urgent procedure	38.74(0.61-∞)
Indication: aneurysm	1.25(0.05-30.93)	.78
Indication: dissection	2.41(0.10-56.32)
Arm ischemia
Control group: revascularization	3.05(0.26-35.99)	.01
Control group: no coverage	144.43(27.86-748.08)
Elective procedure	0.87(0.24-3.19)	.31
Urgent procedure	2.36(0.55-10.09)
Indication: aneurysm	786.70(1.08-∞)	.79
Indication: dissection	3156.10(1.23-∞)
Anterior circulation stroke
Control group: revascularization	1.84(0.35-9.74)	.59
Control group: no coverage	3.48(0.72-16.67)
Elective procedure	1.44(0.10-21.43)	.65
Urgent procedure	0.28(0.00-228.00)
Indication: aneurysm	1.39(0.11-16.87)	.31
Indication: dissection	7.68(0.85-69.55)
TIA
Control group: revascularization	0.58(0.05-7.44)	.59
Control group: no coverage	1.53(0.14-17.13)
Vertebrobasilar insufficiency
Control group: revascularization	7.73(1.54-38.68)	.67
Control group: no coverage	17.02(2.58-112.14)
Indication: aneurysm	25.79(0.16-∞)	.92
Indication: dissection	19.18(1.45-254.21)
Spinal cord ischemia
Control group: revascularization	3.38(0.64-18.02)	.74
Control group: no coverage	1.57(0.14-18.27)
Indication: aneurysm	0.31(0.01-18.98)	.59
Indication: dissection	0.06(0.00-3.63)

CI, Confidence interval; OR, odds ratio; TIA, transient ischemic attack.

Back to Article Outline

Discussion 

Our findings 

We conducted a systematic review of the literature to evaluate the effect of the intentional coverage of the left subclavian artery associated with endograft placement in the descending thoracic aorta. We found very-low-quality evidence suggesting that coverage increased the risk of arm ischemia and vertebrobasilar ischemia with a strong trend for increased risk of spinal cord ischemia and anterior circulation stroke. We found no difference in other outcomes such as death, myocardial infarction, or TIA. The incidence of phrenic nerve injury associated with primary revascularization was fairly low. Data on perioperative infection were sparse and inconclusive.

The evidence in this review is of very low quality due to several reasons. First, the included studies are observational; hence, patients were assigned to the different interventions based on their morbidity and surgeon's preferences/expertise. Second, it is apparent that the patients within each study are heterogenous in terms of diagnoses, morbidity, and aortic pathology. It is possible that the underlying aortic pathology, rather than the procedure itself, had played a significant role in the overall outcome. Third, data are in general sparse with low number of events. Fourth, the outcomes were poorly described in many cases; therefore, transient or less clinically important outcomes (eg, a case of hand ischemia that is not severe or does not require revascularization) may have been included in the analysis, adding an element of indirectness to this evidence. Thus, the quality of the evidence is downgraded due to methodological limitations of included studies, heterogeneity, imprecision, and indirectness.

Strengths, limitations, and comparison with other systematic reviews 

A systematic review and meta-analysis conducted by Cooper et al¹⁶ found that LSA coverage, with or without revascularization, was associated with increased risk of stroke, whereas the risk of spinal cord ischemia is only increased when revascularization is not done. In this report, we differentiated between anterior circulation stroke, transient ischemic attacks and vertebrobasilar ischemia; and assessed several additional outcomes such as arm ischemia, death, and myocardial infarction. Our findings are also in agreement with a narrative review by Noor et al;¹⁷ our review, however, explicitly describes the search process as well as the eligibility criteria and offers a meta-analytic estimate to assist decision makers in balancing the risks and benefits of this procedure. In addition to these strengths, our team reviewed evidence in blinded pairs of reviewers with adequate inter-reviewer agreement. Also, we have shown that our findings are robust to the use of different statistical pooling methods and consistent across several subgroups of patients, interventions, and study designs. The inferences drawn are weak given the very low quality of the best available evidence. Indeed, the primary studies summarized here enrolled a heterogeneous patient population, have important methodological limitations, and report inconsistent findings. Furthermore, the overall data is sparse and the estimates imprecise even after pooling.

Implications for practice and research 

The increased risk of arm ischemia, vertebrobasilar ischemia and, possibly, spinal cord ischemia and anterior circulation stroke associated with LSA coverage need to be weighed against other factors such as urgency of aortic repair, availability of surgical expertise, feasibility of carotid-subclavian bypass or transposition before coverage, patient anatomy, and other risks and burdens associated with these procedures. The accompanying guideline document prepared by the Committee on Thoracic Aortic disease from the Society for Vascular Surgery contains the practical and clinical implications of our results.

Considering that some of the indications for these procedures, such as dissection or transsection, are rare and data is overall sparse, it is apparent that future studies are needed to confirm or dispute the weak inferences we have drawn from the available very-low-quality evidence. Multicenter research with collaboration among surgeons and researchers is paramount to accrue the large number of patients and events necessary to improve the precision of future studies and the overall evidence base. Random or protocol-driven allocation stratified by procedure urgency and aortic pathology will decrease the heterogeneity of future evidence and improve its quality.

Back to Article Outline

Conclusion 

Very low quality evidence suggests that the intentional coverage of the left subclavian artery during placement of an endovascular stent graft in the descending thoracic aorta increases the risk of arm ischemia, vertebrobasilar ischemia, and possibly of spinal cord ischemia and anterior circulation stroke. When feasible, primary revascularization procedures may reduce this risk.

Back to Article Outline

Author contributions 

Back to Article Outline

References 

1. Riesenman PJ, Farber MA, Mendes RR, Marston WA, Fulton JJ, Keagy BA. Coverage of the left subclavian artery during thoracic endovascular aortic repair. J Vasc Surg. 2007;45:90–94discussion 94-5
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (239 KB)
   • CrossRef
2. Galili O, Fajer S, Eyal A, Karmeli R. Left subclavian artery occlusion by thoracic aortic stent graft: long-term clinical and duplex follow-up. Isr Med Assoc J. 2007;9:668–670
   • View In Article
3. Woo E, Carpenter J, Jackson B, Pochettino A, Bavaria J, Szeto W, et al. Left subclavian artery coverage during thoracic endovascular aortic repair: a single-center experience. J Vasc Surg. 2008;48:555–560
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (448 KB)
   • CrossRef
4. Peterson BG, Eskandari MK, Gleason TG, Morasch MD. 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
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (265 KB)
   • CrossRef
5. Reece TB, Gazoni LM, Cherry KJ, Peeler BB, Dake M, Matsumoto AH, et al. Reevaluating the need for left subclavian artery revascularization with thoracic endovascular aortic repair. Ann Thorac Surg. 2007;84:1201–1205
   • View In Article
   • CrossRef
6. Stroup DF, Berlin JA, Morton SC, Olkin I, Williamson GD, Rennie D, et al. Meta-analysis of observational studies in epidemiology: a proposal for reporting (Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group). JAMA. 2000;283:2008–2012
   • View In Article
   • MEDLINE
   • CrossRef
7. Murad MH, Swiglo BA, Sidawy AN, Ascher E, Montori VM. Methodology for clinical practice guidelines for the management of arteriovenous access. J Vasc Surg. 2008;48(5Supp):26–30
   • View In Article
8. Yusuf S, Peto R, Lewis J, Colins R, Sleight P. Beta blockade during and after myocardial infarction (An overview of randomized trials). Prog Cardiovasc Dis. 1985;27:335–371
   • View In Article
   • MEDLINE
   • CrossRef
9. Bradburn M, Deeks J, Berlin J, Russell Localio A. Much ado about nothing: a comparison of the performance of meta-analytical methods with rare events. Stat Med. 2007;26:53–77
   • View In Article
   • MEDLINE
   • CrossRef
10. Higgins J, Thompson S, Deeks J, Altman D. Measuring inconsistency in meta-analyses. BMJ. 2003;327:557–560
    • View In Article
    • CrossRef
11. Altman DG, Bland JM. Interaction revisited: the difference between two estimates. [see comment] BMJ. 2003;326:219
    • View In Article
    • CrossRef
12. Engels EA, Schmid CH, Terrin N, Olkin I, Lau J. Heterogeneity and statistical significance in meta-analysis: an empirical study of 125 meta-analyses. Stat Med. 2000;19:1707–1728
    • View In Article
    • MEDLINE
    • CrossRef
13. Egger M, Smith GD, Phillips AN. Meta-analysis: Principles and procedures. BMJ. 1997;315:1533–1537
    • View In Article
    • MEDLINE
    • CrossRef
14. Song TK, Donayre CE, Walot I, Kopchok GE, Litwinski RA, Lippmann M, et al. Endograft exclusion of acute and chronic descending thoracic aortic dissections. J Vasc Surg. 2006;43:247–258
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (497 KB)
    • CrossRef
15. Thompson M, Ivaz S, Cheshire N, Fattori R, Rousseau H, Heijmen R, et al. Early results of endovascular treatment of the thoracic aorta using the Valiant endograft. Cardiovasc Intervent Radiol. 2007;30:1130–1138
    • View In Article
    • CrossRef
16. Cooper DG, Walsh SR, Sadat U, Noorani A, Hayes PD, Boyle JR. 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
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (877 KB)
    • CrossRef
17. Noor N, Sadat U, Hayes P, Thompson M, Boyle J. Management of the left subclavian artery during endovascular repair of the thoracic aorta. J Endovasc Ther. 2008;15:168–176
    • View In Article
    • CrossRef
18. Mitchell RS, Dake MD, Sembra CP, Fogarty TJ, Zarins CK, Liddel RP, et al. Endovascular stent-graft repair of thoracic aortic aneurysms. J Thorac Cardiovasc Surg. 1996;111:1054–1062
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (1969 KB)
    • CrossRef
19. Kato N, Dake MD, Miller DC, Semba CP, Mitchell RS, Razavi MK, et al. Traumatic thoracic aortic aneurysm: treatment with endovascular stent-grafts. [see comment] Radiology. 1997;205:657–662
    • View In Article
    • MEDLINE
20. Hausegger KA, Tiesenhausen K, Schedlbauer P, Oberwalder P, Tauss J, Rigler B. Treatment of acute aortic type B dissection with stent-grafts. Cardiovasc Intervent Radiol. 2001;24:306–312
    • View In Article
    • MEDLINE
    • CrossRef
21. Criado FJ, Clark NS, Barnatan MF. Stent graft repair in the aortic arch and descending thoracic aorta: a 4-year experience. J Vasc Surg. 2002;36:1121–1128
    • View In Article
    • Abstract
    • Full-Text PDF (225 KB)
    • CrossRef
22. Fattori R, Napoli G, Lovato L, Russo V, Pacini D, Pierangeli A, et al. Indications for, timing of, and results of catheter-based treatment of traumatic injury to the aorta. AJR American Journal of Roentgenology. 2002;179:603–609
    • View In Article
    • MEDLINE
23. Orend KH, Scharrer-Pamler R, Kapfer X, Liewald F, Gorich J, Sunder-Plassmann L. Endoluminal stent-assisted management of acute traumatic aortic rupture. Chirurg. 2002;73:595–600
    • View In Article
    • MEDLINE
    • CrossRef
24. Pamler RS, Kotsis T, Gorich J, Kapfer X, Orend KH, Sunder-Plassmann L. Complications after endovascular repair of type B aortic dissection. J Endovasc Ther. 2002;9:822–828
    • View In Article
    • MEDLINE
    • CrossRef
25. Totaro M, Mazzesi G, Marullo AG, Neri E, Fanelli F, Miraldi F. Endoluminal stent grafting of the descending thoracic aorta. Ital Heart J. 2002;3:366–369
    • View In Article
    • MEDLINE
26. Balzer JO, Doss M, Thalhammer A, Fieguth HG, Moritz A, Vogl TJ. Urgent thoracic aortal dissection and aneurysm: treatment with stent-graft implantation in an angiographic suite. Eur Radiol. 2003;13:2249–2258
    • View In Article
    • MEDLINE
    • CrossRef
27. Lambrechts D, Casselman F, Schroeyers P, De Geest R, D'Haenens P, Degrieck I. Endovascular treatment of the descending thoracic aorta. Eur J Vasc Endovasc Surg. 2003;26:437–444
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (1070 KB)
    • CrossRef
28. Matravers P, Morgan R, Belli A. The use of stent grafts for the treatment of aneurysms and dissections of the thoracic aorta: a single centre experience. Eur J Vasc Endovasc Surg. 2003;26:587–595
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (1199 KB)
    • CrossRef
29. Orford VP, Atkinson NR, Thomson K, Milne PY, Campbell WA, Roberts A, et al. Blunt traumatic aortic transection: The endovascular experience. Ann Thorac Surg. 2003;75:106–112
    • View In Article
    • MEDLINE
    • CrossRef
30. Scharrer-Pamler R, Kotsis T, Kapfer X, Gorich J, Orend KH, Sunder-Plassmann L. Complications after endovascular treatment of thoracic aortic aneurysms. J Endovasc Ther. 2003;10:711–718
    • View In Article
    • MEDLINE
    • CrossRef
31. Sunder-Plassmann L, Scharrer-Pamler R, Liewald F, Gorich XKJ, Orend KH. Endovascular exclusion of thoracic aortic aneurysms: Mid-term results of elective treatment and in contained rupture. J Card Surg. 2003;18:367–374
    • View In Article
    • MEDLINE
    • CrossRef
32. Tiesenhausen K, Hausegger KA, Oberwalder P, Mahla E, Tomka M, Allmayer T, et al. Left subclavian artery management in endovascular repair of thoracic aortic aneurysms and aortic dissections. J Card Surg. 2003;18:429–435
    • View In Article
    • MEDLINE
    • CrossRef
33. Amabile P, Collart F, Gariboldi V, Rollet G, Bartoli JM, Piquet P. Surgical versus endovascular treatment of traumatic thoracic aortic rupture. J Vasc Surg. 2004;40:873–879
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (181 KB)
    • CrossRef
34. Leurs LJ, Bell R, Degrieck Y, Thomas S, Hobo R, Lundbom J, et al. Endovascular treatment of thoracic aortic diseases: combined experience from the EUROSTAR and United Kingdom Thoracic Endograft registries. J Vasc Surg. 2004;40:670–679discussion 679-80
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (190 KB)
    • CrossRef
35. Nemes A, Dzsinich C, Huttl K, Acsady G. New chapter of vascular surgery; stent-graft placement in the treatment of aortic aneurysm. Orv Hetil. 2004;145:2075–2083
    • View In Article
    • MEDLINE
36. Tse LW, MacKenzie KS, Montreuil B, Obrand DI, Steinmetz OK. The proximal landing zone in endovascular repair of the thoracic aorta. Ann Vasc Surg. 2004;18:178–185
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (840 KB)
    • CrossRef
37. Dagenais F, Normand JP, Turcotte R, Mathieu P. Changing trends in management of thoracic aortic disease: where do we stand with thoracic endovascular stent grafts?. Can J Cardiol. 2005;21:173–178
    • View In Article
    • MEDLINE
38. Dong ZH, Fu WG, Wang YQ, Guo DQ, Xu X, Chen B, et al. Strategies for handling the insufficiency of the proximal landing zone during endovascular thoracic aortic repair. Zhonghua Wai Ke Za Zhi. 2005;43:857–860
    • View In Article
    • MEDLINE
39. Doss M, Wood JP, Balzer J, Martens S, Deschka H, Moritz A. Emergency endovascular interventions for acute thoracic aortic rupture: Four-year follow-up. J Thorac Cardiovasc Surg. 2005;129:645–651
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (147 KB)
    • CrossRef
40. Fu WG, Dong ZH, Wang YQ, Guo DQ, Chen B, Jiang JH, et al. Strategies for managing the insufficiency of the proximal landing zone during endovascular thoracic aortic repair. Chin Med J. 2005;118:1066–1071
    • View In Article
41. Guo W, Gai LY, Liu XP, Zhang GH, Liang FQ, Li R. The endovascular repair of aortic dissection: early clinical results of 178 cases. Zhonghua Wai Ke Za Zhi. 2005;43:921–925
    • View In Article
    • MEDLINE
42. Lawlor DK, Ott M, Forbes TL, Kribs S, Harris KA, DeRose G. Endovascular management of traumatic thoracic aortic injuries. [see comment] Can J Surg. 2005;48:293–297
    • View In Article
    • MEDLINE
43. Melissano G, Civilini E, Bertoglio L, Setacci F, Chiesa R. Endovascular treatment of aortic arch aneurysms. Eur J Vasc Endovasc Surg. 2005;29:131–138
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (301 KB)
    • CrossRef
44. Pogorzelski R, Jakuczun W, Szostek M, Rowinski O. The use of stent-grafts in case of descending aortic dissections. [Polish, English] Pol Przegl Chir. 2005;77:264–272
    • View In Article
45. Amabile P, Rollet G, Vidal V, Collart F, Bartoli JM, Piquet P. Emergency treatment of acute rupture of the descending thoracic aorta using endovascular stent-grafts. Ann Vasc Surg. 2006;20:723–730
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (952 KB)
    • CrossRef
46. Caronno R, Piffaretti G, Tozzi M, Lomazzi C, Lagana D, Carrafiello G, et al. Emergency endovascular stent-graft treatment for acute thoracic aortic syndromes. Surgery. 2006;140:58–65
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (334 KB)
    • CrossRef
47. Ferrari E, Tozzi P, Von Segesser L. Thoracic aorta emergencies: is the endovascular treatment the new gold standard?. Interact Cardiovasc Thorac Surg. 2006;5:730–734
    • View In Article
    • CrossRef
48. Kaya A, Heijmen RH, Overtoom TT, Vos JA, Morshuis WJ, Schepens MA. Thoracic stent grafting for acute aortic pathology. Ann Thorac Surg. 2006;82:560–565
    • View In Article
    • CrossRef
49. Marcheix B, Dambrin C, Bolduc JP, Arnaud C, Cron C, Hollington L, et al. Midterm results of endovascular treatment of atherosclerotic aneurysms of the descending thoracic aorta. Ann Thorac Cardiovasc Surg. 2006;132:1030–1036
    • View In Article
50. Patel HJ, Williams DM, Upchurch GR, Shillingford MS, Dasika NL, Proctor MC, et al. Long-term results from a 12-year experience with endovascular therapy for thoracic aortic disease. Ann Thorac Surg. 2006;82:2147–2153
    • View In Article
    • CrossRef
51. Schoder M, Grabenwoger M, Holzenbein T, Cejna M, Ehrlich MP, Rand T, et al. Endovascular repair of the thoracic aorta necessitating anchoring of the stent graft across the arch vessels. J Thorac Cardiovasc Surg. 2006;131:380–387e381
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (234 KB)
    • CrossRef
52. Verhoye JP, de Latour B, Heautot JF, Vola M, Langanay T, Corbineau H, et al. Mid-term results of endovascular treatment for descending thoracic aorta diseases in high-surgical risk patients. Ann Vasc Surg. 2006;20:714–722
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (760 KB)
    • CrossRef
53. Buth J, Harris PL, Hobo R, van Eps R, Cuypers P, Duijm L, 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;46:1103-11.e1102
    • View In Article
54. Feezor RJ, Martin TD, Hess PJ, Klodell CT, Beaver TM, Huber TS, et al. Risk factors for perioperative stroke during thoracic endovascular aortic repairs (TEVAR). J Endovasc Ther. 2007;14:568–573
    • View In Article
    • CrossRef
55. Ferreira M, Monteiro M, Lanziotti L, Abuhadba G, Capotorto L. Deliberate subclavian artery occlusion during aortic endovascular repair: is it really that safe?. Eur J Vasc Endovasc Surg. 2007;33:664–667
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (418 KB)
    • CrossRef
56. Juszkat R, Jemielity M, Pukacki F, Oszkinis G, Staniszewski R, Kulesza J, et al. Endovascular treatment of traumatic injuries of thoracic aorta. [Polish, English] Acta Angiologica. 2007;13:65–74
    • View In Article
57. McPhee JT, Asham EH, Rohrer MJ, Singh MJ, Wong G, Vorhies RW, et al. The midterm results of stent graft treatment of thoracic aortic injuries. J Surg Res. 2007;138:181–188
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (366 KB)
    • CrossRef
58. Neschis DG, Moaine S, Gutta R, Charles K, Scalea TM, Flinn WR, et al. Twenty consecutive cases of endograft repair of traumatic aortic disruption: lessons learned. J Vasc Surg. 2007;45:487–492
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (398 KB)
    • CrossRef
59. Orend KH, Zarbis N, Schelzig H, Halter G, Lang G, Sunder-Plassmann L. Endovascular treatment (evt) of acute traumatic lesions of the descending thoracic aorta - 7 years' experience. Eur J Vasc Endovasc Surg. 2007;34:666–672
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (1085 KB)
    • CrossRef
60. Rehders TC, Petzsch M, Ince H, Kische S, Korber T, Koschyk DH, 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
    • View In Article
    • MEDLINE
    • CrossRef
61. Rodriguez JA, Olsen DM, Shtutman A, Lucas LA, Wheatley G, Alpern J, et al. Application of endograft to treat thoracic aortic pathologies: a single center experience. J Vasc Surg. 2007;46:413–420
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (639 KB)
    • CrossRef
62. Saratzis NA, Saratzis AN, Melas N, Ginis G, Lioupis A, Lykopoulos D, et al. Endovascular repair of traumatic rupture of the thoracic aorta: single-center experience. Cardiovasc Intervent Radiol. 2007;30:370–375
    • View In Article
    • MEDLINE
    • CrossRef