Ann Thorac Surg 2002;74:2040-2046
© 2002 The Society of Thoracic Surgeons
Original article: cardiovascular
Multimodal protocol influence on stroke and neurocognitive deficit prevention after ascending/arch aortic operations
Lars G. Svensson, MD, PhDa*,
Edward M. Nadolny, CCPb,
Wendy A. Kimmel, CCPb
a Center for Aortic Surgery, Marfan Syndrome and Connective Tissue Disorders Clinic, The Cleveland Clinic Foundation, Cleveland, Ohio, USA
b Lahey Clinic, Burlington, Massachusetts, USA
* Address reprint requests to Dr Svensson, Center for Aortic Surgery, Marfan Syndrome and Connective Tissue Disorders Clinic, The Cleveland Clinic Foundation, 9500 Euclid Avenue, F25, Cleveland OH 44195, USA.
e-mail: svenssl{at}ccf.org
Presented at the Poster Session of the Thirty-eighth Annual Meeting of The Society of Thoracic Surgeons, Fort Lauderdale, FL, Jan 28–30, 2002.
 |
Abstract
|
|---|
BACKGROUND: Various techniques are used for brain protection during aortic surgery. Rather than evaluate each factor separately, we evaluated the early outcome of a multimodal protocol (mannitol, thiopental, MgSO4, lidocaine, CO2 field flooding, Leukoguard filter, head ice packing, electroencephalographic arrest at 20°C, 
-stat, increasing right subclavian artery cannulation, and antegrade/retrograde brain perfusion) for brain protection.
METHODS: Prospectively collected data were analyzed on 403 ascending or arch aortic operations including 199 (49%) arch replacements conducted between July 25, 1991, and September 25, 2001. The mean age was 61.6 years (range 22 to 91 years); 48 (12%) had Marfan syndrome; 141 (35%) had dissection; 134 (33%) had composite grafts inserted; and 138 (34%) had concurrent coronary bypasses performed.
RESULTS: Stroke occurred in 2.0% (8/403) (3 permanent, 5 transient), clinical neurocognitive deficits in 2.5% (10/403) either by testing or patient complaint 2 to 3 weeks after surgery, and 98% (395/403) were 30-day survivors. Univariate predictors of stroke, neurocognitive decline, or death were the following: for stroke, aorta symptom severity grade (1 to 4) (p = 0.001), pump time (p = 0.001), arrest time (p = 0.001), macroscopic atheroma (p = 0.041), concurrent descending/thoracoabdominal aneurysm (p = 0.036), and highest blood rewarming temperature (p = 0.043); for neurocognitive decline, degree of cooling (p = 0.046), pump time (p = 0.001), cooling time (p = 0.001), day extubated (p = 0.042), and antegrade brain perfusion (p = 0.004); for death, pump time (p = 0.001) and clamp time (p = 0.011). The multivariable independent predictors of stroke, neurocognitive decline, or death were the following: for stroke, aorta symptoms grade (p = 0.025), peripheral vascular disease (p = 0.043), and pump time (p = 0.015); neurocognitive decline, preoperative New York Heart Association dyspnea class (p = 0.022), pump time (p = 0.05), arrest time (p = 0.06), day extubated (p = 0.042), and antegrade perfusion (p = 0.023); and for death, pump time (p = 0.018).
CONCLUSIONS: Pump time continues to be the most important predictor of adverse events. The benefit of antegrade or retrograde perfusion remains unproven, partly because of the low event rate (< 2.5%) but may be beneficial for prolonged circulatory arrest. Embolic material either from macroscopic atheroma, descending or thoracoabdominal aneurysms, or associated with peripheral vascular disease, increases the risk of stroke. Preoperative symptoms influence outcome.
|
Introduction
|
|---|
Stroke and neurocognitive deficits after ascending/arch operations, particularly after deep hypothermic circulatory arrest, remains a concern [1–27]. Although the results have continued to improve and in our recent prospective randomized study no deaths or strokes occurred [1, 19], we chose to assess the incidence of stroke and neurocognitive deficit in a larger series of patients to better identify those factors that may result in either strokes, neurocognitive deficits, or death. In addition, we used a multimodal approach to try to reduce the occurrence of these complications.
|
Material and methods
|
|---|
From July 25, 1991, through September 25, 2001, 403 patients (267 were men, 136 women) underwent ascending and or aortic arch operations. The mean age was 61.6 years (range 22 to 91 years). Marfan syndrome was present in 48 patients (12%), and 141 (35%) had aortic dissection including 64 acute aortic dissection.
During this study we used a protocol for the prevention of stroke and neurocognitive deficits described previously [1] and shown in Table 1.
Patients with and without circulatory arrest were included in this study for a number of reasons: First, opinions differ as to the indications for circulatory arrest, from some programs using arrest routinely for all ascending repairs [6–8] to those having avoided it even for proximal hemi-arch repairs [20]. We used circulatory arrest selectively when the distal ascending or arch was dilated, for reoperations, or for acute dissections, and avoided arrest in elderly patients, based on our previous study showing that with greater age, the risk of stroke is considerably increased with circulatory arrest [5]. Second, parts of the protocol we believed were valuable for noncirculatory arrest patients. This approach may have increased the risk of stroke in patients with preoperatively unknown atheroma in the arch; however, this risk was not found to be a problem. Lastly, a previous report examined use of this protocol in a smaller, prospective randomized study comparing circulatory arrest alone, antegrade brain perfusion, and retrograde brain perfusion [1, 19]. This study evaluated the results of this protocol in a larger series of patients.
Table 2
shows the preoperative presenting symptoms. Symptoms related to the aorta were graded as previously described [5], namely grade 1 had no symptoms (N = 131); grade 2, occasional pain or mild symptoms (N = 121); grade 3 had constant pain or dysphagia or aorta compression related tracheal dyspnea (N = 58); and grade 4 acute dissection, shock, or rupture (N = 90). Three patients were not graded. Chronic pulmonary disease was present in 37% (148/403), peripheral vascular disease in 21% (86/403), and 16% (64/40) had history of a previous stroke. The preoperative New York Heart Association functional class for degree of dyspnea were I, n = 68 (17.0%); II, n = 124 (31.1%); III, n = 119 (29.8%); and IV, n = 88 (22.1%). The angina classes were 0, n = 110 (27.3%); I, n = 160 (39.7%); II, n = 31 (7.7%); III, n = 46 (11.4%); and IV, n = 56 (13.9%).
Apart from ascending or arch aneurysms, associated descending aneurysms were present in 68 (16.9%) and thoracoabdominal aneurysms in 30 (7.4%). When aortic dissection was present (N = 141), the site of the tear was noted in 122, as shown in Table 3.
Tear types were classified as previously described [24]: class I, classic in 114 (81%); class II, intramural hematoma in 4 (2.7%) (operative aortic wall hematoma but no tear found. This prevalence differs from the preoperative radiologic intramural hematoma prevalence since at surgery a tear is usually found); class III, localized tear in 11 (7.8%); class IV, penetrating ulcer in 3 (2.1%); and class V, iatrogenic in 9 (6.4%). Of the patients with aortic dissection, coronary dissection was present in 39 (28%) and the aortic valve was dissected in 83 (59%). Preoperative aortic valve regurgitation was present in 109 (77%). Blood was present in the pericardium in 35 (25%), and for acute dissections in 31 of 64 patients (48%). Rupture into surrounding tissues or pericardium was present in 27 (19%) patients with dissection.
Medial degenerative disease was clinically present at surgery in 382 (95%), and macroscopic atheroma or calcification was present in 203 (50%), and rupture in 44 (11%). In 87 patients, the aortic histology revealed vasculitis or the aneurysms were inflammatory. Fifteen percent (61/403) of patients were undergoing a reoperation.
A midline incision was used in 313 (77.6%); a posterolateral left thoracotomy, with or without a midline incision in 31 (7.7%); a minimally invasive approach, mostly J/j incision in 68 (17%), and a combined thoracoabdominal incision was used in 7 (1.7%).
The extent of repair and procedures performed are shown in Table 4. Forty-seven (12%) of the entire 403 patient series and 7 (3.5%) of the arch repairs had concurrent aortic endarterectomies.
The method of cardiopulmonary bypass was aortocaval in 35.0%, femoral arterial infusion and vena caval venous drainage in 45.2%, femorofemoral bypass in 5.0%, and other method in 12.9%. Statistically, no difference was noted in the incidence of stroke based on method used when care was taken to select the appropriate method for each patient. In 7.0% the right subclavian artery was used. In the 165 circulatory arrest patients, cooling was continued with electroencephalographic electrical silence at 2 µV sensitivity unless the patient underwent emergency surgery [3]. Sometimes circulatory arrest was performed before complete cooling for rupture. Measurement of somatosensory evoked potentials was not available for establishing neurologic silence. The average temperature for circulatory arrest on rectal temperature measurement was 23.8°C (standard deviation [SD] 4.1, range 12.7° to 36.0°C), tympanic 22.2°C (SD 6.4, range 9.0° to 37.0°C), esophageal 21.2°C (SD 6.5, range 9.0° to 36.0°C), bladder 22.7°C (SD 4.9, range 8.0° to 36.0°C), and blood temperature measured by Swan-Ganz catheter 20.2°C (SD 7.4, range 7.0° to 36.0°C). The average cooling time was 47 minutes (SD 18, range 10 to 122 minutes) and warming 75 minutes (SD 27, range 24 to 215 minutes). The highest bladder temperature warmed to was an average of 36.2°C (SD 0.9, range 30° to 37.6°C) and blood temperature measured by the Swan-Ganz catheter was 37.5°C (SD 1.2, range 34.0° to 39.5°C).
In the patients who had circulatory arrest, the average circulatory arrest time was 22 minutes (SD 15), the average aortic cross clamp time was 103 minutes (SD 44), and the average pump time was 163 minutes (SD 52). Without circulatory arrest (N = 238) the cross-clamp and pump times were 54 minutes (SD 37) and 80 minutes (SD 29), respectively.
The data were prospectively collected and entered on a database and updated as patients were discharged from hospital. In addition, the postoperative complications and postoperative deaths were crosschecked against our master database and also with morbidity and mortality conference noted complications. Stroke was defined as any neurologic deficit while in the hospital, either permanent or transient, and also included coma or nonlateralizing deficits. All patients with any deficit, including amaurosis fugax, were seen by a neurologist and underwent computed tomography or magnetic resonance imaging scanning of the brain. Neurocognitive deficits were defined as decline in neurocognitive function by neurocognitive testing (35 patients of 165 circulatory arrest patients) or patient or family noting neurologic decline by direct and simple testing at 2 to 3 weeks after surgery.
Continuous variables were analyzed by the Student’s t test and categorical data by 
2 testing. Ranges and standard deviation are reported. Regression analysis was done with SPSS (SPSS Inc, Chicago, IL).
|
Results
|
|---|
Stroke occurred in 2.0% (8/403) (3 permanent, 5 transient; Table 5), clinical neurocognitive deficits in 2.5% (10/403), and 30-day survival was recorded in 98.0% (395/403). No hemorrhagic strokes occurred. Two of the patients who died did so after discharge from the hospital, both within 30 days of operation from documented ventricular tachycardia. Autopsies in both showed open coronary bypasses but extensive myocardial fibrosis. Both had preoperative ventricular ectopy on Holter examinations. Postoperative myocardial infarction occurred in 3 patients. Dialysis was required in 6 patients (0.001%–5%). Reoperation for bleeding occurred in 13 (3.2%). One patient required sternal rewiring, two had groin infections, but none had a sternal infection (Table 6)
.
On univariate analysis, the predictors of stroke were aorta symptoms severity grade (graded 1 to 4 as previously described) (p = 0.001), pump time (p = 0.001), circulatory arrest time (p = 0.001), evidence of macroscopic atheroma or calcification (p = 0.041), concurrent descending or thoracoabdominal aneurysm repair (p = 0.036), the cardiopulmonary bypass cooling time before circulatory arrest (p = 0.001), and the highest warming temperature measured in the blood (p = 0.043). The predictors of a clinical neurocognitive deficit were degree of cooling (p = 0.046), pump time (p = 0.001), cooling time (p = 0.001), day of extubation (p = 0.042), and the use of antegrade brain perfusion during circulatory arrest (p = 0.004). The univariate predictors of death were the pump time (p = 0.001) and the clamp time (p = 0.011).
On multivariable analysis the independent predictors of stroke for all 403 patients were pump time (p = 0.015), aorta symptom grade (p = 0.025), and the presence of peripheral vascular disease (p = 0.043). For the patients who had circulatory arrest, the only independent predictor was pump time (p = 0.009).
The predictors of clinical neurocognitive deficit were preoperative New York Heart Association dyspnea class (p = 0.022), pump time (p = 0.05), circulatory arrest time (p = 0.006), day of extubation after surgery (p = 0.042), and the use of antegrade brain perfusion (p = 0.023). For the circulatory arrest patients the only independent predictor was length of intensive care unit (ICU) stay (p = 0.003).
The only predictor of death for all 403 patients was pump time (p = 0.018), and for circulatory arrest patients, also pump time (p = 0.06).
Twenty-seven patients required readmission within 30 days of discharge from hospital. The multivariable predictors for readmission were homologous operative blood product transfusion (p = 0.03), NYHA dyspnea class (p = 0.05), aortic valve regurgitation grade (1 to 4; p = 0.001), and units of homologous packed red blood cells transfused (p = 0.05).
|
Comment
|
|---|
In a previous report of 656 deep hypothermia and circulatory arrest patients, the operative mortality was 10% and the stroke rate was 7% [5]. Furthermore, in a larger series of 717 ascending or arch repair patients with and without circulatory arrest examining risk factors for stroke and death, the mortality rate was 9% and stroke occurred in 3% of patients, excluding patients who had previously had a stroke before surgery [22]. Subsequent to these studies, techniques have been modified and cardiopulmonary bypass pump technology has improved considerably. Based on the above studies, other reports, and animal studies, a multimodal protocol was developed for brain protection during ascending or arch operations [3–14, 20, 22, 23]. Although each protocol factor could have been evaluated, the time and number of patients required to have sufficient statistical power to evaluate each aspect separately would have been prohibitive. We thus chose to institute the multimodal protocol based on available knowledge and then evaluate the outcome. Furthermore, because results have improved over time, we concluded that evaluating more subtle complications would be useful (Table 4), particularly neurocognitive deficits.
Thirty-day survival rate was 98% (395/403), including 2 deaths after discharge within 30 days of operation. Stroke occurred in 8 patients (2.0%) and a gross neurocognitive deficit was present in 10 patients (2.5%) at the 2- to 3-week interval of follow-up after surgery. Because of the low occurrence of death (8 patients) only one variable was predictive for operative death, namely the pump time. This factor has been noted in previous studies and reflects not only the consequences of a prolonged cardiopulmonary bypass period, but also the complexity of the care [5,22]. Thus, for example, patients who had combined mediastinal approaches for insertion of composite valve grafts and arch repairs and also concurrent thoracoabdominal repairs, generally had the longest pump times. In addition, the low incidences of multiple organ failure and systemic inflammation in this study reflect improved cardiopulmonary bypass technology and the benefit of leukocyte filtration with the multimodal protocol [1].
The declining prevalence of stroke for these operations has been documented by others and us [1,4, 6–13]. To what extent increasing use of retrograde and antegrade brain perfusion or improved cardiopulmonary bypass technology has contributed to this is difficult to ascertain. Certainly, retrospective studies using antegrade or retrograde brain perfusion have documented improved results when compared with historical controls. We were unable to show that retrograde or antegrade perfusion when used for circulatory arrest was superior to deep hypothermia and circulatory arrest alone for preventing stoke. An important caveat, however, is that because the prevalence was low (2%, 8 patients had a stroke), the statistical power was insufficient to compare groups. Furthermore, with longer circulatory arrest periods, there may have been differences.
The association with the longer pump time is probably also a marker for greater risk of embolization and hence stroke to occur. Similarly, other sources, such as macroscopic atheroma in the operative field, concurrent descending or thoracoabdominal aneurysms (with likely loose atheroma or clot), and peripheral vascular disease (possible ileo-femoral atheroma), were important risk factors for stroke. Of note, however, 47 patients had concurrent aortic endarterectomies; despite this fact none suffered a stroke, although 1 patient died after surgery from heart failure. We had hoped that by increasing use of the right subclavian artery for arterial inflow in 41 patients, including for dissection, we would reduce the risk of stroke, but in fact 2 of these patients (4.9%) had a stroke. Both patients had the aortic dissection process extending into the right subclavian artery; presumably this was a factor in the occurrence of the stroke and would negate the use of the subclavian artery when dissection is present in it.
Preoperative aorta symptom grades, as reported previously in the study of 717 patients [22], influenced outcome, particularly stroke. This is not surprising because the aorta symptom grade classified asymptomatic patients as grade 1 and the highest risk group as grade 4 (N = 90), including patients with acute dissection (N = 64, 1 stroke [1.6%], no deaths) or rupture (N = 44, 4 stroke [9%], 1 death [2.2%]). The acuity of dissection (class I to IV of tears) described and reported previously by us [24] had no influence on the incidence of stroke or death.
The occurrence of a gross neurocognitive deficit with this multimodal protocol was low (n = 10, 2.5%). The exact prevalence after deep hypothermia and circulatory arrest, or even after coronary artery bypass surgery, is difficult to evaluate because of the problems in defining when a neurocognitive deficit has occurred. This problem was discussed previously in our prospective randomized study [1,19]. Formal neurocognitive testing was not used for all the patients (35 had studies), although patients and their families were asked about any neurocognitive deficits they had noticed after surgery and simple testing was performed. Clearly such a casual evaluation would not detect subtle degrees of neurocognitive changes that may have been determined by more formal neurocognitive testing; nevertheless, the patients did not appear to have any substantial problems. Based on the multivariable analysis, an increased risk of neurocognitive deficits may be associated with antegrade brain perfusion. Antegrade brain perfusion may be needed for prolonged periods (longer than 30 minutes) of circulatory arrest; however, for short periods, as we have previously shown [1,19], the method of brain perfusion appears not to be important.
Of interest, increasing preoperative New York Heart Association class for dyspnea, and postoperative length of time of postoperative ventilation or ICU stay, was associated with increased risks of neurocognitive deficit, possibly because of hypoxia or administration of sedation in the ICU, often for prolonged periods.
Although 134 patients in this study had composite valve grafts inserted, the techniques did not seem to influence the incidence of stroke or death. In a previous study of 348 composite valve grafts, the Bentall and button techniques may have been associated with a greater risk of stroke, possibly because of intimal or atheromatous plaque disruption from mobilizing and reattaching the buttons, based on the multivariable analysis [27,28]. Similarly, of 51 patients who had valve preserving operations, either remodeling or the David type of reimplantaion of the valve [28], only 1 had a stroke (2%) and was the only death (2%). This patient had an ascending and arch repair combined with a thoracoabdominal repair, and awoke from surgery, but was found to have a large left atrial clot on a postoperative transesophageal echocardiogram performed in the ICU. During deep hypothermia and circulatory arrest the patient had been on aprotinin, which was the likely cause of the clot forming in the left atrium, which had otherwise not been opened. The risk of thrombosis associated with aprotinin and circulatory arrest has been documented previously [29]. Clearly, postoperative thrombolysis or reexploration was not an option. Subsequently, the clot embolized resulting in the patient developing a stroke, renal failure, and liver failure. No patient has required a reoperation for valve failure.
Perfusion aspects of the protocol that we believe are important include [1,19] carbon dioxide field flooding, leukocyte filtration, -stat pH control, and careful attention to perfusion temperatures. In addition to pump time, the cooling time and the highest blood temperature warmed to also appeared to influence outcome negatively. A higher temperature warmed to after deep hypothermia and circulatory arrest has previously been documented to be associated with a greater risk of stroke [20].
|
Acknowledgments
|
|---|
We gratefully acknowledge the assistance of Christopher Pierce, PhD, and Eugene Blackstone, MD, The Cleveland Clinic Foundation, with data checking and statistical review.
|
References
|
|---|
- Svensson L.G., Nadolny E.M., Penney D.L., et al. Prospective randomized neurocognitive and S-100 study of hypothermic circulatory arrest, retrograde brain perfusion, and antegrade brain perfusion for aortic arch operations. Ann Thorac Surg 2001;71:1905-1912.[Abstract/Free Full Text]
- Crawford E.S., Saleh S.A. Transverse aortic arch aneurysm: improved results of treatment employing new modifications of aortic reconstruction and hypothermic cerebral circulatory arrest. Ann Surg 1981;194:180-188.[Medline]
- Coselli J.S., Crawford E.S., Beall A.C., Jr, Mizrahi E.M., Hess K.R., Patel V.M. Determination of brain temperature for safe circulatory arrest during cardiovascular operation. Ann Thorac Surg 1988;45:638-642.[Abstract]
- Lytle B.W., McCarthy P.M., Meaney K.M., Stewart R.W., Cosgrove D.M., III Systemic hypothermia and circulatory arrest combined with arterial perfusion of the superior vena cava. Effective intraoperative cerebral protection. J Thorac Cardiovasc Surg 1995;109:738-743.[Abstract/Free Full Text]
- Svensson L.G., Crawford E.S., Hess K.R., et al. Deep hypothermia with circulatory arrest: determinants of stroke and early mortality in 656 patients. J Thorac Cardiovasc Surg 1993;106:19-31.[Abstract]
- Ergin M.A., Galla J.D., Lansman S.L., Quintana C., Bodian C., Griepp R.B. Hypothermic circulatory arrest in operations on the thoracic aorta. J Thorac Cardiovasc Surg 1994;107:788-799.[Abstract/Free Full Text]
- Griepp R.B., Ergin A., McCullough J.N. Use of hypothermic circulatory arrest for cerebral protection during aortic surgery. J Card Surgery 1997;12:312-321.
- Ergin M.A., Uysal S., Reich D.L., et al. Temporary neurological dysfunction after deep hypothermic circulatory arrest: a clinical marker of long-term functional deficit. Ann Thorac Surg 1999;67:1887-1890.[Abstract/Free Full Text]
- Juvonen T., Weisz D.J., Wolfe D., et al. Can retrograde perfusion mitigate cerebral injury after particulate embolization? A study in a chronic porcine model. J Thorac Cardiovasc Surgery 1998;115:1142-1159.[Abstract/Free Full Text]
- Ueda Y., Miki S., Kusuhara K., Okita Y., Tahata T., Yamanaka K. Deep hypothermic systemic circulatory arrest and continuous retrograde cerebral perfusion for surgery of the aortic arch aneurysm. Eur J Cardiothorac Surg 1992;6:36-41.[Abstract]
- Coselli J.S. Retrograde cerebral perfusion is an effective means of neural support during deep hypothermic circulatory arrest. Ann Thorac Surg 1997;64:908-912.
- Griepp R.B., Juvonen T., Griepp E.B., McCullough J.N., Ergin M.A. Is retrograde cerebral perfusion an effective means of neural support during deep hypothermic circulatory arrest?. Ann Thorac Surg 1997;64:913-916.
- Bachet J., Guilmet D., Goudot B., et al. Antegrade cerebral perfusion with cold blood: a 13-year experience. Ann Thorac Surg 1999;67:1874-1878.[Abstract/Free Full Text]
- Swain J.A., McDonald T.J., Griffith P.K., et al. Low-flow hypothermic cardiopulmonary bypass protects the brain. J Thorac Cardiovasc Surg 1991;102:76-84.[Abstract]
- Ueda T., Shimizu H., Ito T., et al. Cerebral complications associated with selective perfusion of the arch vessels in patients undergoing surgery of the thoracic aorta. Ann Thorac Surg 2000;70:1472-1477.[Abstract/Free Full Text]
- Mahanna E.P., Blumenthal J.A., White W.D., et al. Defining neuropsychological dysfunction after coronary artery bypass grafting. Ann Thorac Surg 1996;61:1324-1327.
- Murkin J.M. Anesthesia, the brain, and cardiopulmonary bypass. Ann Thorac Surg 1993;56:1461-1463.[Abstract]
- Engelman R.M., Pleet A.B., Rousou J.A., et al. Does cardiopulmonary bypass temperature correlate with postoperative central nervous system dysfunction?. J Card Surg 1995;10:493-497.[Medline]
- Svensson L.G., Hussain A., Penney D.L., et al. A prospective randomized study of neurocognitive function and S-100 protein after antegrade or retrograde brain perfusion with hypothermic arrest for aortic surgery. J Thorac Cardiovasc Surg 2000;119:163-166.[Free Full Text]
- Davis E.A., Gillinov A.M., Cameron D.E., Reitz B.A. Hypothermic circulatory arrest as a surgical adjunct: a 5-year experience with 60 adult patients. Ann Thorac Surg 1992;53:402-406.[Abstract]
- Safi H.J., Miller C.C., III, Estrera A.L., et al. Staged repair of extensive aortic aneurysms. Morbidity and mortality in the elephant trunk technique. Circulation 2001;104:2938-2942.[Abstract/Free Full Text]
- Crawford E.S., Svensson L.G., Coselli J.S., et al. Surgical treatment of aneurysm and/or dissection of the ascending aorta, transverse aortic arch, and ascending aorta and transverse aortic arch: factors influencing survival in 717 patients. J Thorac Cardiovasc Surg 1989;98:659-674.[Abstract]
- Safi H.J., Brien H.W., Winter J.N., et al. Brain protection via cerebral retrograde perfusion during aortic arch aneurysm repair. Ann Thorac Surg 1993;56:270-276.[Abstract]
- Svensson L.G., Labib S.B., Eisenhauer A.C., Butterly J.R. Intimal tear without hematoma: An important variant of aortic dissection that can elude current imaging techniques. Circulation 1999;99:1331-1336.[Abstract/Free Full Text]
- Heinemann M.K., Buehner B., Jurmann M.J., et al. Use of the "elephant trunk technique" in aortic surgery. Ann Thorac Surg 1995;60:2-7.[Abstract/Free Full Text]
- Svensson L.G. Rationale and technique for replacement of the ascending aorta, arch, and distal aorta using a modified elephant trunk procedure. J Card Surg 1992;7:301-312.[Medline]
- Svensson L.G., Crawford E.S., Hess K.R., Coselli J.S., Safi H.J. Composite valve graft replacement of the proximal aorta: comparison of techniques in 348 patients. Ann Thorac Surg 1992;54:427-439.[Abstract]
- Svensson L.G., Longoria J., Kimmel W.A., Nadolny E. Management of aortic valve disease during aortic surgery. Ann Thorac Surg 2000;69:778-784.[Abstract/Free Full Text]
- Sundt T.M., Kouchoukos N.T., Saffitz J.E., Murphy S.F., Wareing T.H., Stahl D.J. Renal dysfunction and intravascular coagulation with aprotinin and hypothermic circulatory arrest. Ann Thorac Surg 1993;55:1418-1424.[Abstract]
Top
Abstract
Introduction
