A comparative analysis of open and endovascular repair for the ruptured descending thoracic aorta

Article Outline

• Abstract
• Methods
  • Statistical analysis
• Results
  • Early results
  • Late results
• Discussion
• Author contributions
• References
• Copyright

Background

Successful repair of the ruptured (non-traumatic) descending thoracic aorta (rTA) remains a formidable clinical challenge. Although effective for rTA, traditional open repair (DTAR) has significant associated morbidity. With expanding indications for thoracic endovascular aortic repair (TEVAR), we describe our experience with TEVAR and DTAR in this high-risk setting to elucidate their evolving roles.

Methods

Since the inception of our thoracic aortic endovascular program in 1993, 69 patients underwent DTAR (34) or TEVAR (35) for rTA. Patients underwent TEVAR if they were considered nonoperative candidates because of extensive comorbidities (n = 31; 88.6%) or had extremely favorable anatomy for endovascular repair (eg, mid-descending saccular aneurysm, n = 4). Aortic pathology causing rupture was fusiform aneurysm (18), saccular aneurysm/ulcer (22), and dissection (29). Associated aortobronchial fistulae (12) and aortoesophageal (1) fistulae were also present in 18.8%. Arch repair was needed in 46; total descending repair was needed in 33. Follow-up was 100% complete (mean 37.4 months).

Results

Mean age was 65.9 years (DTAR 60.3 year vs TEVAR 71.3 years, P = .005). In-hospital or 30-day mortality was seen in 13 patients (TEVAR n = 4; 11.4% vs DTAR n = 9; 26.5%, P = .13). Median length of stay was shorter after TEVAR (8 days vs DTAR 15 days, P = .02). Mean Kaplan-Meier survival was similar between groups (TEVAR 67.4 months vs DTAR 65.0 months, P = .7). By multivariate analysis, independent predictors of a composite outcome of early mortality, stroke, permanent spinal cord ischemia, or need for dialysis or tracheostomy included the presentation with hemodynamic instability (P < .001) and treatment with conventional open repair (P = .02).

Conclusion

An endovascular approach for the ruptured (non-traumatic) descending thoracic aorta reduces early morbidity, mortality, and duration of hospitalization, while providing equivalent late outcomes even in an older group largely considered high risk for open repair. These data support a paradigm shift, with TEVAR emerging as the preferred therapy for all patients presenting with descending aortic rupture.

The gold standard therapy for descending thoracic aneurysms has traditionally been open repair (DTAR). However, recent work by our group as well as others has suggested that high-risk patients, whether related to comorbidities or presentation, may be more appropriately treated with an endoluminal approach (TEVAR).¹, ², ³ Reported outcomes following DTAR for the non-traumatic ruptured thoracic aorta have been poor, with documented mortality rates exceeding 20%.⁴, ⁵, ⁶ Although comparative analyses of open and endovascular thoracic aortic repair have been conducted in the elective setting or for subgroups of patients with thoracic aortic pathology, no report thus far has concurrently compared outcomes in the setting of the non-traumatic ruptured descending aorta.⁷ The purpose of this study is to focus on these outcomes.

Back to Article Outline

Methods 

This study was approved by the Institutional Review Board (IRB) of the University of Michigan Hospitals (IRB #2003-0128, informed consent requirements waived).

Data from all patients (n = 69) who underwent operative therapy for ruptured descending thoracic aortic pathology (excluding blunt traumatic aortic injury) at the University of Michigan Hospitals between 1993 and February 2009 were prospectively collected and retrospectively analyzed. The anatomic extent of pathology requiring intervention in all instances was located between the left carotid artery and the celiac axis. Patients presenting with thoracoabdominal aneurysm rupture were excluded from this analysis. The diagnosis of a ruptured aorta was made by computed tomography (CT) demonstration of contrast extravasation outside the aorta (uncommon) or (more common) by aneurysmal enlargement of the aorta with high-attenuation crescents in thickened wall (crescent sign), soft-tissue infiltration of peri-aortic fat, and inability to define the outer wall of the aorta (peri-aortic hematoma) and/or pseudoaneurysm formation. For those patients undergoing open aneurysm repair, the intraoperative findings of rupture were confirmed. All patients were evaluated for open repair initially by a surgeon with specific expertise in thoracic aortic reconstruction. Patients who were prospectively deemed high-risk candidates for traditional open repair (n = 31) or had extremely favorable anatomy for endovascular repair (eg, mid-descending saccular aneurysm, n = 4) underwent TEVAR as previously described.² High-risk comorbidities identified in this group included prior history of myocardial infarction, renal failure, stroke, malignancy, or chronic obstructive pulmonary disease (COPD), or poor preoperative functional status.

All open thoracic aortic repairs (n = 34) were performed with extracorporeal perfusion support. Left heart bypass or partial cardiopulmonary bypass was utilized in eight patients. The remaining 26 patients had adjunctive use of deep hypothermic circulatory arrest (HCA) as previously described.⁸ Indications for HCA included the presence of aortic pathology precluding use of aortic cross clamp, or the need to extend the resection into either the arch aorta or the entire descending thoracic aorta.

Postoperative management for prevention of spinal cord ischemia for both open and endovascular repairs was conducted according to standardized protocols as previously described.³ Assuming hemodynamic stability allowing placement of lumbar drains, both groups of patients received spinal drainage at the discretion of the surgeon for similar indications and included those patients needing repair beyond the proximal third of the descending aorta or in the distal half of the descending aorta. Those patients with previous infrarenal aortic repair also preferentially had placement of lumbar drains. In total, 20 patients underwent placement of lumbar drains (DTAR n = 12, 35.3% vs TEVAR n = 8, 22.9%).

The primary outcomes of this study were: 1) a composite outcome of early (in-hospital or 30-day) mortality, stroke, permanent spinal cord ischemia, need for dialysis, or prolonged ventilatory support requiring tracheostomy; and 2) vital status at last follow-up. Data were collected from clinic visit notes, hospital charts, and imaging studies, and mortality was verified by interrogation of the National Death Index. Follow-up for these primary outcomes was 100% complete as of February 2009 (mean follow-up 37.4 ± 41.3 months).

Statistical analysis 

Data were analyzed using SPSS (SPSS Inc., Chicago Ill). Dichotomous variables were evaluated using chi-square analysis; continuous variables were evaluated using one-way analysis of variance. Multivariate models (binary logistic regression) were constructed using a forward conditional process to identify factors that were independently associated with each of the outcomes of interest. Models were tested for goodness of fit using the Hosmer-Lemeshow statistic. Factors utilized in multivariate analysis included using those with P ≤ .1 significance on univariate analysis. Survival was analyzed by Kaplan-Meier methods. All results with P < .05 were considered statistically significant.

Back to Article Outline

Results 

The mean age of the entire cohort was 65.7 ± 16.5 years (59.4% male). Demographics and comorbidities for the two groups are listed in Table I. The DTAR group was younger, had larger aortic diameters, had a lower frequency of COPD, and more frequently required intervention in the arch aorta. Underlying pathology causing rupture, extent of repair, and procedural details are also listed in Table I.

Table I. Demographics and comorbidities with univariate analysis of the study groups

	Open repair (n = 34)	TEVAR (n = 35)	P value
Demographics
Age (years)	60.4±14.2	71.3±16.9	0.005
Male gender	24(70.5%)	17(48.6%)	0.087
Mean aortic dimension (cm)	6.8±2.0	5.6±2.0	0.04
Comorbidities
CAD	7(20.6%)	11(31.4%)	0.41
History of CHF	2(5.6%)	0(0%)	0.24
COPD	4(11.7%)	13(37.1%)	0.02
Diabetes	4(11.7%)	4(11.4%)	1.0
Hypertension	24(70.5%)	27(32.6%)	0.59
Preoperative creatinine(mg/dL)	1.0±0.5	1.0±0.6	0.85
Prior AAA repair	4(11.7%)	5(14.3%)	1.0
PVOD	10(29.4%)	13(37.1%)	0.61
History of tobacco abuse	16(47.1%)	21(60%)	0.34
High risk from comorbidities	2(5.6%)	31(88.6%)	<0.001
Aortic pathology causing rupture
Hemodynamic instability at presentation	3(8.8%)	3(8.6%)	1.0
Fusiform aneurysm	8(23.6%)	10(28.6%)	0.79
Acute aortic dissection	18(52.9%)	11(31.4%)	0.09
Saccular aneurysm/pseudoaneurysm	8(23.6%)	14(40%)	0.2
Aortic fistulae	6(17.6%)	7(20%)	1.0
Treated aortic segments⁎
Arch aorta	27(79.4%)	19(54.3%)	0.04
Total descending aorta	17(50%)	16(45.7%)	0.8

AAA, Abdominal aortic aneurysm; CAD, coronary artery disease; CHF, congestive heart failure; COPD, chronic obstructive pulmonary disease; PVOD, peripheral vascular occlusive disease; TEVAR, thoracic endovascular aortic repair.

Technical success in TEVAR was achieved in 33 patients (94.3%). The remaining two patients presented with contained ruptures. One had a slow-filling proximal type 1 endoleak in a highly curved arch and the other had a slow-filling distal type 1 endoleak with a marginal distal neck at the celiac axis. Neither patient was deemed an open operative candidate; both were lost to imaging follow-up after discharge and survived 53 and 2.5 months, respectively. Devices utilized included Gore TAG (WL Gore and Associates, Flagstaff, Ariz) (19), Medtronic Talent (Medtronic, Inc, Minneapolis, Minn) (9), custom fabricated (3), AneuRx aortic cuff (Medtronic, Inc) (3) and Cook TX2 (Cook Inc, Bloomington, Ind) (1). Device delivery was via a transfemoral approach in 27, an iliac approach utilizing a conduit in five, and ascending aorta and carotid artery in one each. Of the 19 patients who had coverage of the left subclavian artery, 10 underwent left carotid-subclavian bypass prior to TEVAR.

Early results 

Understanding baseline differences in this cohort, there was a trend toward a reduction in early mortality (defined as either in-hospital or within 30 days) in the TEVAR group (n = 4; 11.4% vs DTAR n = 9; 26.5%, P = .13). The causes of early mortality are listed in Table II. The only univariate correlate of this outcome was the presentation with hemodynamic instability (P < .0001).

Table II. Cause of in-hospital or 30-day mortality

Patient	Thoracic aortic pathology	Cause of mortality
DTAR group
1	Saccular aneurysm	Fulminant pneumonia. death on POD #35⁎
2	Aortic dissection	Superior mesenteric artery embolus, death from sepsis POD #4
3	Saccular aneurysm	Intraoperative death from cardiogenic shock
4	Aortic dissection	Intraoperative death from cardiogenic shock
5	Aortic dissection	Retrograde type A dissection, death POD #2⁎
6	Aortobronchial fistula, saccular aneurysm	Brainstem infarct, death POD #4⁎
7	Hemodynamically unstable, fusiform aneurysm	Death from multisystem organ failure on POD #8
8	Hemodynamically unstable, acute dissection	Multisystem organ failure, death on first operative day
9	Hemodynamically unstable, acute on chronic dissection	Intraoperative death, unable to separate from cardiopulmonary bypass
TEVAR group
1	Rupture at aberrant right subclavian artery	Intraoperative death from iliac artery rupture
2	Hemodynamically unstable, penetrating ulcer/intramural hematoma	Ruptured thoracic aorta on night of TEVAR
3	Hemodynamically unstable, aortobronchial fistula from post-coarctation pseudoaneurysm	Death from hypoxemia and multisystem organ failure on POD#2
4	Aorto-esophageal fistula from saccular aneurysm	Death from development of tracheo-esophageal fistula on POD #23

DTAR, Open descending thoracic aneurysm repair; POD, post-operative day; TEVAR, thoracic endovascular aortic repair.

The incidence of stroke was 2.9% (n = 2). Both patients had undergone open resection of the distal arch aorta for contained ruptures, one with the use of hypothermic arrest. This latter patient had a protracted postoperative course requiring temporary dialysis and tracheostomy but was ultimately discharged on postoperative day 88. The other patient, who had aortic resection using partial cardiopulmonary bypass for a ruptured pseudoaneurysm with aortobronchial fistula, sustained a brainstem stroke. He had no neurological recovery, and care was withdrawn on postoperative day four. The incidence of renal failure needing dialysis was seen in 10 patients (14.5%), with no significant difference found between groups (P = .19).

Permanent lower extremity paralysis or paresis was seen in one patient (1.4%) who had undergone open repair of the distal arch and total descending aorta for an acute on chronic dissection with contained rupture (with lumbar drain). This patient developed a compartment syndrome in the left leg (site of femoral artery cannulation). He had a protracted postoperative course and returned to the operating room for muscle debridement on postoperative day 10. He had several bouts of hypotension during this procedure and awoke thereafter with lower extremity paralysis. Despite re-insertion of a lumbar drain and permissive hypertension, he never recovered his spinal cord function.

In order to generate a sufficient event rate for multivariate analysis, a composite end-point representing this early morbidity and mortality (early death, stroke, permanent paraplegia, a need for dialysis or tracheostomy) was derived. Univariate correlates with this outcome measure are listed in Table III. Independent predictors of this poor composite end-point included presentation with hemodynamic instability (P < .001) and repair with a conventional open approach (P = .02). Surprisingly, the pathology underlying the rupture did not correlate on univariate analysis with the risk for a poor composite outcome (all P > .24). Finally, the TEVAR group had a significantly shorter postoperative length of stay (median TEVAR 8 days vs DTAR 15 days, P = .02).

Table III. Univariate analysis of early and late outcomes^⁎

	P value
Univariate correlates of early poor composite outcome
Age	.04
COPD	.02
Group considered high risk for open repair secondary to comorbid conditions	.07
Hemodynamic instability at presentation	<.0001
Treatment group	.04
Univariate correlates of late mortality
Age	.12
Need for arch repair	.04
Acute dissection pathology	.09
Presence of aortic fistulae or mycotic aneurysm	.06
Hemodynamic instability at presentation	.03

COPD, Chronic obstructive pulmonary disease.

Late results 

The overall crude mortality rate for the entire cohort at last follow-up was 50.7% and did not differ between groups (DTAR n = 18 vs TEVAR n = 17, P = .81). Univariate correlates of late mortality are included in Table III. Independent predictors of late mortality included age (P = .017) and presentation with hemodynamic instability (P < .0001). By Kaplan-Meier analysis (Fig 1), actuarial survival was also similar between open and endovascular groups (P = .72).

• 
  • View Large Image
  • Download to PowerPoint

• Fig 1. 

  A Kaplan-Meier survival analysis comparing open descending thoracic aortic repair to thoracic aortic endovascular repair. This actuarial analysis demonstrates that following either open or endovascular thoracic aortic repair, there is no significant difference in Kaplan-Meier survival for patients presenting with descending aortic rupture. The 10-year survival for TEVAR is 21.3% vs that for DTAR at 30.8% (log rank P = .72). The survival curves have been truncated at seven years, where the standard error exceeds 10%.

Endoleaks were seen in nine patients (25.7%), and included type 1 (3), type 2 (2), type 3 (2), and indeterminant type (2). Two of those with type 1 endoleaks required open repair but were not treated, as they were not considered open operative candidates. One expired of aortic rupture at 39 months, and the other died of unknown cause at 13 months. The remaining patient underwent a successful proximal extension. One patient with a type 3 endoleak was successfully treated during the same hospitalization as the primary procedure. Another patient who had presented with an aortobronchial fistula had an early type 3 endoleak, refused therapy, and expired in hospice at four months. Finally, one with a type 2 endoleak underwent successful coil embolization of the left subclavian artery, while the other patient was observed, and had a stable sac size at last follow-up. There were no instances of migration identified. A Kaplan-Meier curve, shown in Fig 2, was constructed to examine the risk for aortic reintervention at four years in any aortic segment, including treated, adjacent, or remote segments. This analysis suggested a higher need for aortic reintervention in the TEVAR group.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 2. 

  A Kaplan-Meier analysis describing the need for reintervention in any aortic segment. This analysis suggests that the need for aortic reintervention at any aortic segment (treated, adjacent, or remote) is significantly higher in the TEVAR group. Freedom from reintervention at four years was 87.4% for DTAR vs 61.2% for TEVAR (P = .037). In this analysis, if patients were deemed to be nonoperative, or refused further intervention, the date at which point the need for reintervention was identified was used as the time of treatment failure.

Back to Article Outline

Discussion 

Thoracic aortic rupture remains a formidable clinical entity. A population-based study from Sweden suggested that presentation with thoracic aortic rupture (including ascending, arch and descending aorta) was associated with a mortality rate of 54% in the first six hours following onset of symptoms.⁹ Therapy in this setting has traditionally been considered open repair. Despite improvements in perioperative care, outcomes following open repair of the ruptured descending thoracic aorta remain poor and ill defined.⁴, ⁵, ⁶ Although several reports have attempted to describe these outcomes, they have been confounded with inclusion of thoracoabdominal aortic pathology. These studies have suggested that early mortality rates, variably reported as in-hospital or 30-day, range from 20% to 50% following open repair. Rates of spinal cord ischemia, myocardial infarction, and renal failure are also significantly elevated, when compared with that obtained with elective repair.

With the advent of an endovascular solution for descending thoracic aortic pathology, several investigators have reported successful outcomes with TEVAR for the ruptured aorta.¹⁰, ¹¹, ¹² Scheinert and colleagues reported an early mortality rate of 9.7% with TEVAR for rupture in 31 patients.¹⁰ Recent comparative series have included a report from Morishita et al describing mortality rates that were slightly higher in the stent grafted group (17% vs DTAR 9%) in 29 patients.¹¹ In contrast, Doss et al, in another comparative analysis in 60 patients, reported a reduced early mortality in the endovascular arm (3.1%) vs open group (17.8%).¹² They suggested that late complications including need for reintervention were higher in the TEVAR group, and that continued surveillance was necessary. However, these and other reports have included patients presenting with traumatic aortic rupture. The addition of blunt aortic trauma in an appraisal of aortic rupture may not be valid, as there have been multiple reports describing the initial successful non-operative treatment for blunt aortic trauma.¹³ This suggests that the natural history of patients with untreated non-traumatic and traumatic ruptures may be quite different, with the latter group having a more benign course.⁹, ¹³

The current analysis focuses on patients with non-traumatic rupture, and describes a 15-year comparison of open and endovascular repair in this setting. Both groups were not directly comparable, with the endovascular group presenting at an advanced age, less frequently needing arch repair and more frequently designated as high-risk open operative candidates. Despite these differences, TEVAR in this setting was independently predictive of an improved early outcome as defined by a composite endpoint of early death, stroke, paraplegia, or need for dialysis or tracheostomy. Importantly, late survival was not significantly different between groups, although the risk for aortic reintervention was higher after TEVAR.

The rates of mortality in this series do not differ from other published reports.⁴, ⁵, ⁶, ¹⁰, ¹¹, ¹² However, they are markedly different than those obtained with elective aneurysm repair, highlighting the importance of early detection. Typically, the decision to proceed with aneurysm repair is predicated on the ratio of the risk of rupture or dissection to the risk of repair, with the absolute aortic diameter as the most important determinant of rupture.¹⁴ It is interesting to note that the mean diameter at the time of rupture in this series was 6.2 cm, not significantly greater than the typical size criterion for intervention for descending aortic pathology (6.0 cm). This suggests that with the perceived reduction in morbidity and mortality with an endovascular approach, a randomized trial of intervention vs standard medical therapy for smaller thoracic aneurysms (ie, 5.0 to 6.0 cm) is now warranted.

At the University of Michigan, patients with elective indications for intervention are evaluated for TEVAR on a selective basis, with younger age (< 65 years), physiologic status of the patient, and anatomic suitability for TEVAR deemed among the most important criteria for a recommendation to proceed with open repair. In contrast, for patients presenting with rupture, we now preferentially evaluate all patients for an endovascular approach regardless of age. Specific exclusion criteria for TEVAR in this setting include the presence of rupture in a pre-existing chronic dissection or in patients with connective tissue disease, lack of access vessels for delivery, or the presence of compromised landing zones. The latter is a relative contraindication for TEVAR, however, as compromised landing zones may occasionally be used in patients not considered suitable for open operation.

Although open repair is still considered the gold standard therapy, the need for urgent or emergent intervention in the setting of rupture may preclude a full appraisal of comorbidities, particularly a cardiac evaluation. With the significant additional physiological stress of open repair (vs TEVAR), it is not surprising that two of the nine DTAR deaths were cardiac related (vs none for TEVAR). This has also been identified as a cause of mortality in up to 75% of patients in other reports of DTAR in rupture.⁴, ⁵ Understanding the exceedingly poor prognosis with a postoperative myocardial infarction, we now consider performance of a cardiac catheterization immediately prior to open repair. This delay is undertaken only in selected patients with a significant prior cardiac history (ie, previous myocardial infarction with reduced ejection fraction, or active cardiac symptoms). These patients must also present with stable hemodynamics with a contained rupture and must also be anatomically unsuitable for TEVAR. If they are found to have significant coronary artery disease, we perform open repair with simultaneous coronary artery bypass graft (CABG), consider percutaneous angioplasty without stenting of the target coronary artery (ie, no clopidogrel requirement), or deem them to be too high risk for open surgery. One patient in this report underwent a DTAR with concomitant CABG for left main disease. He succumbed to a retrograde type A dissection on the second postoperative day; the family refused further intervention and withdrew care. In contrast, for those patients who are to undergo TEVAR, cardiac evaluation is only undertaken if: 1) these patients are not considered open operative candidates; 2) if they are hemodynamically stable and have a contained rupture; and 3) if they have a significant prior cardiac history. Cardiac evaluation in this setting consists of a nuclear stress test. TEVAR is then performed urgently if no contraindication exists. If a significant cardiac ischemic burden is identified, we then decide whether to perform angioplasty without stenting of the coronary artery with TEVAR thereafter, perform CABG, and deliver the endograft in an antegrade fashion via the ascending aorta, or deem them unsuitable for endovascular repair.

Limitations of this study include its retrospective nature, small sample size, and baseline differences between groups. Indeed, the differences in these groups primarily reflect a selection bias preferentially shunting older patients with more complex comorbidities to the TEVAR arm. Despite this, we were able to identify the importance of a TEVAR strategy in reducing early morbidity even in this older and higher-risk group. However, baseline differences in pathologic extent (ie, more frequent arch repair in DTAR) may also have contributed to the increased morbidity seen in the open repair group. Another important limitation is that although the study spans the entire TEVAR era at the University of Michigan, access to the endovascular arm in this retrospective analysis varied by date of operation (Fig 3). The TEVAR “learning curve” likely affected who were considered suitable anatomic candidates (including proportion of open repair patients anatomically eligible for TEVAR), and the availability of endografts at our institution was markedly different after the FDA approval of the Gore TAG device in 2005.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 3. 

  The evolution of therapy for descending aortic rupture at the University of Michigan. This graph divides the entire TEVAR era at the University of Michigan into three time periods. The years 1993-1999 reflect the time when no commercial endografts were generally available, and the dominant procedure is open aortic repair. During the years 2000-2004, endografts were typically available as part of clinical trials, although we did selectively utilize custom-fabricated devices. Finally, 2005-2008 reflects the time period when thoracic endografts were commercially available. This graph demonstrates an increasing shift toward endovascular repair during the study period, likely reflecting the ability to now offer a therapeutic option to patients previously considered non-operative candidates (ie, previously referred to medical therapy alone). Note, however, during this period, open repair is still considered an important option during this period, representing the primary mode of therapy for patients considered suitable candidates for open repair.

In conclusion, presentation with a ruptured descending aorta portends a poor prognosis for early and late survival. This comparative analysis of open and endovascular strategies for the ruptured (non-traumatic) aorta suggests that TEVAR reduces early morbidity while yielding equivalent late survival even in a group largely considered high risk for open surgery. These results support a paradigm shift, with TEVAR emerging as the therapy of choice for all patients presenting with descending aortic rupture.

Back to Article Outline

Author contributions 

Back to Article Outline

References 

1. Greenberg RK, O'Neill S, Walker E, Haddad F, Lyden SP, Svensson LG, et al. Endovascular repair of thoracic aortic lesions with the Zenith TX1 and TX2 thoracic endografts: intermediate-term results. J Vasc Surg. 2005;41:589–596
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (283 KB)
   • CrossRef
2. 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
3. Patel HJ, Williams DM, Upchurch GR, Dasika NL, Passow MC, Prager RL, et al. A comparison of open and endovascular descending thoracic aortic repair in patients older than 75 years of age. Ann Thorac Surg. 2008;85:1597–1603
   • View In Article
   • CrossRef
4. Crawford ES, Hess KR, Cohen ES, Coselli JS, Safi HJ. Ruptured aneurysm of the descending thoracic and thoracoabdominal aorta; analysis according to size and treatment. Ann Surg. 1991;213:417–425
   • View In Article
   • MEDLINE
   • CrossRef
5. Girardi LN, Krieger KH, Altorki NK, Mack CA, Lee LY, Isom OW. Ruptured descending and thoracoabdominal aneurysms. Ann Thorac Surg. 2002;74:1066–1070
   • View In Article
   • MEDLINE
   • CrossRef
6. Barbato JE, Kim JY, Zenati M, Abu-Hamad G, Rhee RY, Makaroun MS, et al. Contemporary results of open repair of ruptured descending thoracic and thoracoabdominal aortic aneurysms. J Vasc Surg. 2007;45:667–676
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (152 KB)
   • CrossRef
7. Bavaria JE, Appoo JJ, Makaroun MS, Verter J, Yu ZF, Mitchell RS, et al. 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
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (142 KB)
   • CrossRef
8. Patel HJ, Shillingford MS, Mihalik S, Proctor MC, Deeb GM. Resection of the descending thoracic aorta: outcomes after use of hypothermic circulatory arrest. Ann Thorac Surg. 2006;82:90–96
   • View In Article
   • CrossRef
9. Johansson G, Markstrom U, Swedenborg J. Ruptured thoracic aortic aneurysms: a study of incidence and mortality rates. J Vasc Surg. 1995;21:985–988
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (354 KB)
   • CrossRef
10. Scheinert , Krakenberg H, Schmidt A, Gummert JF, Nitzsche S, Braunlich S, et al. Endoluminal stent-graft placement for acute rupture of the descending thoracic aorta. Eur Heart J. 2004;25:694–700
    • View In Article
    • MEDLINE
    • CrossRef
11. Morishita K, Kurimoto Y, Kawaharada N, Fukada J, Hachiro Y, Fujisawa Y, et al. Descending thoracic aortic rupture: role of endovascular stent-grafting. Ann Thorac Surg. 2004;78:1630–1634
    • View In Article
    • CrossRef
12. 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
13. Demetriades D, Velmahos GC, Scalea TM, Jurkovich GH, Karmy-Jones R, Teixeira PG, et al. Diagnosis and treatment of blunt thoracic aortic injuries: changing perspectives. J Trauma. 2008;64:1415–1418
    • View In Article
14. Davies RR, Goldstein LJ, Coady MA, Tittle SL, Rizzo JA, Kopf GS, et al. Yearly rupture or dissection rates for thoracic aortic aneurysms: simple prediction based on size. Ann Thorac Surg. 2002;73:17–28
    • View In Article
    • MEDLINE
    • CrossRef