| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2003;76:493-498
© 2003 The Society of Thoracic Surgeons
a Departments of Thoracic and Cardiovascular Surgery, Frankfurt/Main, Germany
b Diagnostic and Interventional Radiology, Johann Wolfgang Goethe-University Frankfurt/Main, Frankfurt/Main, Germany
* Address reprint requests to Dr Doss, Department of Thoracic and Cardiovascular Surgery, Johann Wolfgang Goethe-University Frankfurt/Main, Theodor-Stern-Kai 7, 60590 Frankfurt, Main, Germany
e-mail: mirkodoss{at}aol.com
Presented at the Forty-ninth Annual Meeting of the Southern Thoracic Surgical Association, Miami Beach, FL, Nov 7–9, 2002.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
METHODS: From January 2001 to October 2001, we deployed 11 emergent endovascular stent grafts into the thoracic aorta. We treated seven ruptured aortic aneurysms and four acutely perforated type B dissections. Aortic rupture was confirmed preoperatively by spiral computed tomography. In all cases, hemothorax was present. The average interval from onset of symptoms to treatment was 28.5 hours. We used nine Talent and two Excluder stent grafts.
RESULTS: Deployment of the stent grafts was successful in nine cases. There were two cases of access failure due to small caliber of iliac arteries, and 1 of these patients died shortly after the procedure was abandoned, At 12 months of follow-up, there were no cases of paraplegia, stent migration, or endoleaks. There was, however, one temporary renal failure, and 2 patients required mechanical ventilation for more than 48 hours.
CONCLUSIONS: Our experiences with emergency endovascular stent grafting show that the procedure is technically feasible, with less morbidity and mortality than conventional open surgery, in high-risk patients.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
Conventionally, the preferred treatment of aortic type B dissection is medical therapy, essentially consisting of antihypertensive therapy. Surgical treatment is reserved for those cases that are complicated by progression of the disease, impending rupture, formation of a pseudoaneurysm, persistent thoracic pain, drug-resistant hypertension, or end-organ ischemia [5, 6]. Medical treatment is associated with a mortality rate of 20%, whereas the mortality rate for surgical repair is 35% to 50%. Thus, both conservative treatment and surgery in patients with lesions of the descending thoracic aorta are associated with a high mortality rate [6, 7].
In patients with leaking thoracic aortic aneurysms, immediate surgical repair is required. Surgery usually consists of the replacement of the proximal part of the descending aorta. Despite recent advances in peri- and postoperative care in these patients, surgical repair remains associated with a high rate of mortality and morbidity [8, 9].
In this study, we report our experiences with emergent endovascular stent graft deployment in 11 patients who presented with perforated acute type B dissections and leaking aneurysms of the descending thoracic aorta.
|
Material and methods |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
A patient was considered urgent if he or she had a recent onset of thoracic pain and if hemothorax was present on the preoperative computed tomagrahpy (CT) scan, confirming leakage. To keep time needed for diagnostic imaging as low as possible, a volume CT scan reaching from the apex of the thorax down to the vessels of the groin was performed preoperatively in all patients and used as the sole source of data for diagnosis and stent graft planning. Principally, all patients presenting with a lesion of the descending thoracic aorta were considered potential candidates for stent graft deployment. Exclusion criteria for stent graft implantation were a diameter of the thoracic aorta proximal and distal to the lesion of greater than or equal to 44 mm, due to the lack of suitable stent graft prostheses; a lesion originating from the ascending aorta; a lesion too close to the orifice of the left common carotid artery (length 
5 mm), as we would routinely overstent the left subclavian artery in an emergency setting; and a diameter of the common iliac arteries less than or equal to 7 mm, as well as a heavily tortuous course of the abdominal or extreme kinking of the thoracic aorta. Concomitant diseases, increasing the risk for surgical repair, were arterial hypertension (84.6%), chronic obstructive pulmonary disease (30.8%), stroke (7.7%), cardiac disease with a low ejection fraction (< 30%) (11.5%), and renal insufficiency (11.5%). One patient had Marfan’s syndrome with previous surgical repair of the ascending aorta as well as the aortic arch.
Resuscitation protocol
Depending on their clinical status on admission to the hospital, most patients required some form of immediate medical attention to stabilize their hemodynamic condition to allow for the planning of an endovascular procedure. Most patients presented in a state of shock (n = 5) and needed inotropic resuscitation with dopamine or noradrenaline in moderate dosages. Furthermore, due to their blood loss, another 4 patients were stabilized with volume resuscitation, using colloids and blood products. Two patients were severely hypertensive at admission, with systolic blood pressures of up to 200 mm Hg. To control their hypertension ß-blockers, 
-blockers, and nitrates were used. A total of 7 patients had symptoms of acute back or chest pain upon presentation, and required intravenous opioids. To proceed with an endovascular procedure, patients had to respond to our resuscitation therapy, fulfilling the following criteria: systolic blood pressure of 90 to 120 mm Hg, a heart rate less than 110 per minute, central venous pressure of 7 to 14 cm H2O, and a hemoglobin of more than 10 g/dL.
Stent graft deployment
All stent graft implantations were performed in a digital subtraction angiography (DSA) suite (Multistar Plus, Siemens, Erlangen, Germany) under general anesthesia by a team of cardiovascular surgeons and interventional radiologists. A heart-lung machine and a perfusionist were in standby in all cases.
For aortic aneurysms, a healthy proximal and distal landing zone was a prerequisite. In type B dissections, only the entry tear was overstented and the estimated diameter of the healthy aorta (distal aortic arch) oversized by 5 mm. Stent grafts of various sizes were readily available off the shelf or were delivered within 12 hours to our hospital.
Two different commercially available thoracic stent graft systems were implanted in this study: Talent LPS (n = 9) (Medtronic World Medical, Sunrise, FL) and Excluder (n = 2) (W.L. Gore & Associates Inc., Sunnyvale, CA). The stent graft consisted of a self-expanding Nitinol stent covered externally with Dacron graft (Talent LPS) or internally by a PTFE graft (Excluder). All Talent stent grafts had a bare spring design at the proximal end, 15 mm in length. The mean length of stent grafts implanted was 113.8 ± 21.9 mm (range, 94 to 153 mm). Stent diameter (mean 34.3 ± 5.9 mm) was oversized by 4 to 6 mm based on the diameter of the aortic neck in order to allow an appropriate fixation of the stent graft.
The DSA unit was equipped with a double C-arm construction, allowing rotation of the entire C-arm (Roentgenogram tube and image intensifier) at speeds up to 25 revolutions per second. Additionally, the system is road map capable and offers the fading-in of DSA images during fluoroscopy. No external skin markers were used to locate the orifice of the large supraaortic branches.
The patients were placed in a supine position. In all cases, a percutaneous access through the brachial artery (left, nine cases; right, two cases) was obtained. After placement of a 5F introducer sheath (Radiofocus, Terumo Co, Tokyo, Japan), a 5F pigtail catheter (Nylex, Cordis Endovascular, Waterloo, Belgium) was positioned in the ascending thoracic aorta, thus allowing contrast medium application throughout the entire process of stent graft deployment. Arterial access for stent graft insertion was achieved through surgical cut-down in all cases (left groin, n = 3; right groin, n = 8). In all cases, we used our standard technique of elective stent graft insertion and deployment. In short, after successful surgical exposure of the femoral artery and transverse arteriotomy, a pigtail catheter was positioned in the ascending thoracic aorta over a soft, angled guidewire (Radiofocus standard guidewire M, Terumo). The guidewire was exchanged to a super-stiff guidewire (Lunderquist Extra Stiff, Cook Inc, Bloomington, IN) for stent graft insertion. After an initial, angulated DSA of the thoracic aorta with 30 mL contrast material (Visipaque 320, Nycomed Amersham, Buchler, Germany) and a flow rate of 15 mL/s, 5,000 IU Heparin was administered intravenously. The stent graft system was then advanced into the desired position and correct location verified by an additional DSA immediately before stent graft deployment. To avoid downstream migration of the device during deployment, systolic arterial blood pressure was lowered to 70 mm Hg using sodium nitroprusside (n = 7) just before device release. In 5 patients, short cardiac arrest was achieved by intravenous injection of adenosine (6 to12 mg) in order to ease device deployment and allow exact device positioning at the orifice of the left subclavian or carotid artery. In all cases, postdilation was performed using an endoballoon in order to obtain optimal shape and sealing of the implanted stent graft.
A final DSA was performed to verify appropriate stent graft localization and to demonstrate free perfusion of the supraaortic vessels as well as the stent graft. All interventional materials were removed and the arteriotomy was closed with a continuous 5.0 Prolene suture. All patients were transferred to the intensive care unit (ICU) for postoperative surveillance for at least 4 hours. All patients had a single-shot broad-spectrum antibiotic prophylaxis (third-generation cephalosporins) just before the initial incision. Follow-up examinations were performed using MD-CTA (multi-detector computed tomography angiography) scans before discharge, and at 3, 6, and 12 months after implantation.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
Overall, stent graft deployment was successful in 9 patients. Insertion was difficult in another 2 patients with a diameter of the pelvic vessels not exceeding 8 mm, and a device size of 25 F. After previous dilatation with 8- and 9-mm endoballoons and subsequent careful device advancement, stent graft delivery was possible in the appropriate position in both patients. Seven patients showed an aneurysm of the descending aorta (Figs 1, 2), two caused by trauma after a road traffic accident and five ruptured spontaneously in the course of an increase in aneurysm size and hypertensive peaks. Four patients had an acutely perforated type B dissection of the descending aorta (Fig 3). In 2 of these patients, the dissection reached down to the aortic bifurcation, however, not causing any immediate malperfusion of the extremities.
|
|
|
In 2 patients, two stent grafts had to be inserted in an overlapping manner to seal off the primary lesion. In 1 of these patients, the aneurysm was excluded angiographically after the first stent graft implantation, but revealed a reperfusion during the 1-month follow-up CT scan. After successful placement of a second stent graft, this endoleak was successfully sealed.
The distance from the proximal part of the lesion to the origin of the left subclavian artery ranged from 8 to 85 mm (mean, 30.5 ± 25.8 mm). The mean length of the lesions (for ruptured thoracic aortic aneurysms) was 129.7 ± 74.3 mm.
Upon conclusion of the procedure, no antegrade perfusion of the false aortic lumen in the thoracic segment or the aneurysm could be documented angiographically. Mean procedure time was 93 ± 15.8 minutes and mean fluoroscopy time was 14.9 ± 9.2 minutes. A mean of 216.5 ± 75.5 mL contrast medium was administered.
Postinterventional course
All patients were observed in the ICU for at least 4 hours. No neurologic deficit was observed. However, 2 patients suffered from groin infection, which normalized under intravenous antibiotics. Two patients with initial aortic perforation and lung parenchyma bleeding developed pneumonia after the operation, which necessitated prolonged intubation and mechanical ventilation. These 2 patients were transferred to a peripheral ward 6 days after successful intervention. In another 2 patients (1 with extensive pleural hematoma, 1 with multiple bone injuries after road traffic accident), prolonged intubation and ICU observation was necessary for 3 and 4 weeks, respectively, due to concomitant diseases or injuries. In all other patients, the immediate postinterventional course was uneventful.
Seven patients have completed the 12 months of follow-up. All patients were free of symptoms with regard to the lesions of the descending aorta. Also, no cases of stent migration or stent fracture were observed. In none of the patients was an endoleak observed. In 4 patients, the size of the aneurysm had already decreased.
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
Safe aortic stent graft deployment in the emergency setting constitutes a special situation. It requires careful and quick screening of the lesion’s morphology and extension, as well as of the probable access site. Most studies published so far used thoracic helical CT and calibrated intrarterial DSA to access the lesion of the thoracic descending aorta as well as the pelvic and groin vessels [10–12]. Although these techniques were used, White and associates [13] reported 27% of access complications (eg, femoral artery reconstruction or iliac artery grafts). In our patient population, we only had 1 patient (3.8%) with access failure and another 2 patients (7.7%) with slight difficulty to advance the system without any severe vascular complications. One of the reasons for the lower percentage of vascular access complications might be the more accurate depiction of pelvic artery morphology through employment of a volume CT scan reaching from the apex of the thorax down to the vessels of the groin. This technique enables not only the accurate depiction of the lesion of the thoracic aorta but also gives an exact view of the entire course of the aorta as well as the pelvic arteries. The knowledge of tortuosity of the aorta and pelvic arteries is essential for a safe stent graft delivery [10].
One of our patients died before the stent graft could be deployed. Preoperatively, she did not respond adequately to our resuscitation measures. Her systolic blood pressure remained at 75 mm Hg and her heart rate was more than 120 per minute. At the induction of anesthesia, she required maximum doses of inotropic agents and eventually died of cardiac failure. Achieving preoperative hemodynamic stability in these critically ill patients is essential, as the appropriate-sized stent graft may not be available off the shelf and the time span until delivery has to be bridged without subjecting the patient to an excessive risk. Only in four cases was a stent graft of the right size available off the shelf. In 7 other patients, the prosthesis had to be ordered on an emergency basis. We have set up an agreement with our suppliers that guarantees us the delivery of a stent graft of almost any standard size within 12 hours. If the appropriate size is not available, the patient is scheduled for surgery. A prerequisite to allow for such a time delay is the careful monitoring of the patient around the clock, on an intermediate care ward, with staff that is trained to deal with such a challenge. Dake and associates [4] reported on 19 patients with acute aortic dissections treated with stent grafts in two different centers. All patients had an intimal tear distal to the left subclavian artery; 4 had a retrograde dissection into the aortic arch and were classified as type A dissection, and the other patients had type B dissections. In 15 of 19 patients, thrombosis of the false lumen in the thoracic aorta occurred after sealing the primary tear with a stent graft. However, only in 1 patient did thrombosis of the false lumen in the abdominal aorta occur. No rupture of the false lumen was observed during a maximum follow-up of 28 months. Nienaber and associates [14] reported on 12 patients with subacute or chronic type B dissections treated with stent grafts in two different centers. Sealing of the entry tear was successful in all patients, and thrombosis of the false lumen in the thoracic aorta occurred in all patients. No complications were observed. We made similar observations in our patients. Sealing of the primary tear was successful in all 4 patients. In contrast to the series of Dake and associates [4], the primary tear was close to the left subclavian artery in 2 patients. All abdominal organs were perfused by the true lumen. In all patients, successful stent graft implantation in the proximal (n = 2) and central (n = 2) part of the thoracic aorta, covering the entire segment, also induced partial thrombosis of the false lumen within 3 months. Our results support the previously reported hypothesis [4, 14] that covering the primary aortic intimal tear in acute type B aortic dissection and thereby obliterating the false lumen prevents subsequent dilatation, and therefore appears to be a promising new treatment technique.
Because of the absence of aortic cross-clamping and reperfusion, endovascular stent graft repair of the descending thoracic aorta may lower the incidence of spinal cord ischemia [1, 11, 12]. Concurrent or previous abdominal aortic repair, however, has been associated with an increased risk of paraplegia [12]. In our series, there were no neurologic deficits, paraplegia, or paraparesis. As all of the cases were performed immediately after referral to our hospital, spinal vessel supply was not investigated preoperatively. Nevertheless, in the case of immediate or delayed onset of paraplegia, the current design of the stent grafts does not allow an endoluminal removal of the endoprosthesis once completely deployed. Successful reversal of paraplegia has been reported by both emergent conversion to open surgery [15] and cerebrospinal fluid (CSF) drainage [16]. Monitoring evoked spinal cord potentials during temporary interruption of the intercostal arteries by a specially designed retrievable occlusive device may aid in identifying the patients at risk for spinal cord ischemia, in whom conversion to traditional open repair or prophylactic CSF drainage may be performed to avoid this complication [17].
The incidence of endoleaks continues to be a cause for concern. So far, we have not observe any endoleaks in our patient population, whereas Heijmen and associates [10] reported 29% of endoleaks during follow-up, although the majority of these endoleaks sealed spontaneously within 3 months.
In our patient population, we had no conversion to surgical repair and only one case of reintervention. These results support the conclusion of other studies that this procedure is effective and safe for the treatment of urgent thoracic aortic lesions.
In conclusion, our midterm data illustrate that endovascular stent grafting for ruptured descending thoracic aortic aneurysms and perforated type B dissections is technically feasible with a low morbidity and mortality rate as well as a low rate of reintervention or conversion to surgical repair. The occurrences of endoleaks, aortic neck dilation, and stent graft migration remain a cause for extreme concern, and further evaluation of this new technique is indicated to determine its effectiveness and durability in the longer term.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
Two questions. One, in your abstract, it states that 7 of the patients were judged not to be suitable candidates for conventional surgery, and yet you treated 11 patients. How was the decision made in the additional 4 patients to use this rather than the conventional procedure?
My second question relates to the time of treatment. What was the time interval between diagnosis (not onset of symptoms) and treatment? I was surprised to see that one of your criteria was an available stent within 12 hours. That does not sound very prompt for a ruptured aorta.
Thanks very much. I enjoyed your paper.
DR DOSS: First of all, refering to the 4 patients who could have been treated surgically, they were in fact the last 4 patients in the series, and by then we tried to treat all patients that came to our practice, if the anatomic conditions allowed it, by endovascular stent grafting. We have had good experiences with it, so that is why we feel confident in managing them that way.
Dr Sundt: So that is now your standard of care?
Dr Doss: That is correct, yes. That is what we try to do now. And what was the second question?
Dr Sundt: The second question related to the time between diagnosis and treatment.
Dr Doss: Yes, diagnosis and treatment. What I said was we had a mean interval of 28.5 hours from onset of symptoms to treatment. Once the patient came to the hospital, it usually took 4 to 12 hours of getting the stent graft, and again there, we had to assess if the patient was responding to resuscitation treatment. That is what I tried to say in the beginning. If you can hemodynamically stabilize the patient for some time, then obviously you can wait for 12 hours. For that reason, we put them in the intensive care ward and we monitor their blood pressure and treat them aggressively. Also, we have a contract with our suppliers that usually can bring any stent graft size within 4 hours to our hospital. So we think that this is an arrangement that we can live with.
|
References |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. Doss, J. P. Wood, J. Balzer, S. Martens, H. Deschka, and A. Moritz Emergency endovascular interventions for acute thoracic aortic rupture: Four-year follow-up J. Thorac. Cardiovasc. Surg., March 1, 2005; 129(3): 645 - 651. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Morishita, Y. Kurimoto, N. Kawaharada, J. Fukada, Y. Hachiro, Y. Fujisawa, and T. Abe Descending Thoracic Aortic Rupture: Role of Endovascular Stent-Grafting Ann. Thorac. Surg., November 1, 2004; 78(5): 1630 - 1634. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. F. Duebener, P. Lorenzen, G. Richardt, M. Misfeld, A. Notzold, F. Hartmann, H.-H. Sievers, and V. Geist Emergency Endovascular Stent-Grafting for Life-Threatening Acute Type B Aortic Dissections Ann. Thorac. Surg., October 1, 2004; 78(4): 1261 - 1266. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. K. Hatfield, G. X. Zaleski, D. Kozlov, T. Woo, E. Gentile, and J. Sinnen Angio-Seal Device Used for Hemostasis in the Descending Aorta Am. J. Roentgenol., September 1, 2004; 183(3): 612 - 614. [Full Text] [PDF]
|
|
|
|
|
|
M. Doss, J. Balzer, S. Martens, J. P. Wood, G. Wimmer-Greinecker, H.-G. Fieguth, and A. Moritz Surgical versus endovascular treatment of acute thoracic aortic rupture: a single-center experience Ann. Thorac. Surg., November 1, 2003; 76(5): 1465 - 1470. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
