Eur J Cardiothorac Surg 2003;24:119-124
© 2003 Elsevier Science NL
Cardiac function predicts mortality following thoracoabdominal and descending thoracic aortic aneurysm repair
Shinichi Suzuki,
Cornelius A. Davis, III,
Charles C. Miller, III,
Tam T.T. Huynh,
Anthony L. Estrera,
Eyal E. Porat,
Anders Vinnerkvist,
Hazim J. Safi*
Department of Cardiothoracic and Vascular Surgery, The University of Texas Health Science Center at Houston Medical School, Memorial Hermann Hospital, Houston, TX 77030, USA
Received 26 September 2002;
received in revised form 13 March 2003;
accepted 17 March 2003.
* Corresponding author. Tel.: +1-713-500-5304; fax: +1-713-500-0647
e-mail: hazim.j.safi{at}uth.tmc.edu
 |
Abstract
|
|---|
Objective: Previous studies have identified age, renal failure and aneurysm extent as predictors of mortality following thoracoabdominal and descending thoracic aortic aneurysm (TAA) repair. We studied the impact of coronary artery disease (CAD) and cardiac function on 30-day mortality following TAA repair. Methods: Between February 1991 and May 2001, we performed 854 TAA repairs. Two hundred ninety-one patients (34%) had a history of coronary artery disease. One hundred forty-one/291 (49%) had undergone coronary artery bypass surgery (CAB) prior to TAA repair. We conducted multivariable analyses of known risk factors along with the left ventricular ejection fraction (EF) and prior CAB to determine the adjusted effect of CAD on outcome. Results: Mortality in patients with CAD was 54/291 (18%) compared to 75/563 (13%) without CAD (P<0.05). In patients who had prior CAB, mortality was 31/141 (22%) compared to 98/713 (14%) patients without prior CAB, (P<0.02). In multivariable analysis, the effects of CAD and CAB on mortality were eliminated by consideration of a low EF (defined as less than 50%). Conclusion: Impaired left ventricular function appears to be the strongest cardiac predictor of mortality for TAA repair, independent of the presence of coronary artery disease or coronary artery bypass revascularization.
Key Words: Cardiac • Thoracoabdominal aortic aneurysm • Mortality
|
1. Introduction
|
|---|
Surgical repair of thoracoabdominal and descending thoracic aortic aneurysms remains a formidable task despite significant reductions in morbidity and mortality rates associated with the recent advances in anesthesia and surgical adjuncts. Patient age, renal failure, aneurysm extent, and emergency presentation have been identified previously as risk factors for mortality following thoracic aortic aneurysm repair [1,2]. Coronary artery disease is prevalent in patients with aortic aneurysm and associated with increased mortality following major aortic surgery [3–6]. Preoperative cardiac evaluation of patients requiring abdominal aortic or peripheral arterial reconstructions has been well described but remains controversial [4–10]. The optimal management of coronary artery disease in patients with thoracoabdominal and descending thoracic aortic aneurysms is still unclear. The aim of this study was to evaluate the impact of coronary artery disease and cardiac function in patients undergoing thoracoabdominal and descending thoracic aortic aneurysm surgery.
|
2. Methods and materials
|
|---|
Between February 1991 and May 2001, we performed 854 operations for descending thoracic or thoracoabdominal aortic aneurysm. Two hundred ninety-nine patients were female and 555 were male. The average age was 68 years (range, 8–90 years). Coronary artery disease was defined as historical evidence of recent or remote angina, myocardial infarction, ischemic electrocardiographic changes, and previous coronary artery percutaneous angioplasty or bypass revascularization. Two hundred ninety-one of 854 patients (34.1%) had coronary artery disease (Table 1). One hundred forty-one of 291 (49%) underwent coronary artery bypass at some time prior to descending thoracic or thoracoabdominal aortic aneurysm surgery.
A team of experienced cardiologists evaluated all patients. Preoperative cardiac assessment included 12-lead electrocardiography, two-dimensional echocardiography, chemical nuclear or echocardiographic stress test, and cardiac catheterization. The best estimate of the left ventricular ejection fraction was recorded preoperatively either by two-dimensional echocardiography or contrast ventriculography. Coronary artery revascularization was undertaken either by percutaneous intervention or surgical bypass when indicated. An interval of 4–6 weeks was usually allowed between coronary artery revascularization and subsequent thoracoabdominal or descending thoracic aortic aneurysm surgery, unless there were symptoms related to the aneurysm in which case urgent repair was performed.
2.1. Surgical technique
The autologous reversed saphenous vein was the preferred conduit used for coronary artery bypass prior to thoracoabdominal or descending thoracic aortic aneurysm repair. Use of the left internal mammary artery prohibits placement of the aortic crossclamp proximal to the left subclavian artery at the time of thoracoabdominal or descending thoracic aortic aneurysm repair (Fig. 1)
.
View larger version (46K):
[in this window]
[in a new window]
|
Fig. 1. Aortic crossclamp proximal to the left subclavian artery (upper right), unlike distal crossclamp (lower right), can cause cardiac ischemia following coronary artery bypass using left internal mammary artery (left).
|
|
The details of our thoracoabdominal or descending thoracic aortic aneurysm operative techniques have been described previously [2,11–13] but are reviewed here. The patient is anesthetized and intubated, using a double-lumen endotracheal tube. An upper extremity arterial catheter is placed and a pulmonary artery catheter is floated to allow continuous pressure monitoring. An epidural drain is placed percutaneously between the third and fourth lumbar space for monitoring of cerebrospinal fluid (CSF) pressure and drainage (Fig. 2)
. CSF pressure is maintained at less than 10 mm Hg throughout the procedure. The patient is positioned in a right lateral decubitus position. The incision is tailored to complement the extent of the aneurysm (Fig. 3)
. The diaphragm is preserved, exposing the aortic hiatus and incising only the muscular portion of the diaphragm around the aorta. The patient is anticoagulated with heparin at a dose of 1 mg/kg of body weight. The pericardium is opened posterior to the phrenic nerve, and the left inferior pulmonary vein or the left atrial appendage is cannulated for distal aortic perfusion. A BioMedicus pump (BioMedicus, Minneapolis, MN) with an in-line heat exchanger is attached to this cannula and the arterial inflow is established through the left common femoral artery, or the descending thoracic aorta when the femoral artery is not accessible (Fig. 4)
.
View larger version (40K):
[in this window]
[in a new window]
|
Fig. 3. Artist's illustration of normal aorta and modified Crawford classification of aneurysm extent, extent I to extent V.
|
|
View larger version (47K):
[in this window]
[in a new window]
|
Fig. 4. Distal aortic perfusion from the left inferior pulmonary vein to the left common femoral artery.
|
|
Distal aortic perfusion is begun as the crossclamp is applied. For the proximal anastomosis in repair of the descending thoracic aorta, extent I, or II thoracoabdominal aorta, the proximal aortic neck is completely transected and separated from the esophagus. For extensive aneurysms, sequential clamping is used after completion of the proximal anastomosis; the mid-descending aortic clamp is moved distally to the infrarenal aorta to accommodate intercostal and visceral reattachment. Patent lower intercostal arteries (T8–T12) are reattached, except in cases of acute aortic dissection, heavily calcified or diseased aorta, or when technically not feasible. With completion of intercostal artery reattachment, the abdominal aorta is clamped and the aorta is opened, and the visceral arteries are identified. The celiac, superior mesenteric and renal arteries are perfused with no. 9 Pruitt catheters. The cold perfusate flow (blood at 4°C) delivered to the viscera is dependent on the proximal aortic pressure and is maintained between 300 and 600 ml/min (Fig. 5) . Renal temperature is directly monitored and maintained between 15 and 20°C. Because cold visceral perfusion may cause hypothermia, core body temperature is kept between 32 and 33°C by warming the lower body. If a distal clamp cannot be placed due to technical reasons and warming of the lower body cannot be performed, cold visceral perfusion is avoided to prevent core body cooling that may result in cardiac dysrhythmias. Before completion of the distal anastomosis, the graft is flushed proximally and the aorta distally. The patient is weaned from bypass once the nasopharyngeal temperature has reached 36–37°C. The atrial and femoral cannulae are removed and protamine is administered (Figs. 6A,B)
.
View larger version (61K):
[in this window]
[in a new window]
|
Fig. 6. (A) Descending thoracic aortic aneurysm, illustrated from left to right by: (1) computerized tomography; (2) artist's rendition; and (3) saggital section computerized tomography. (B) Completed graft replacement, type II thoracoabdominal aortic aneurysm illustrated from left to right by: (1) computed tomography; (2) artist's rendition; and (3) three dimensional saggital section computerized tomography.
|
|
Postoperatively, the mean arterial pressure is maintained between 80 and 100 mmHg. CSF is drained intermittently by gravity to maintain a CSF pressure of less than 10 mmHg for 3 days. If a delayed neurological deficit appears after removal of the drain, the CSF drain is immediately reinserted.
2.2. Statistical methods
Data were collected prospectively by a trained nurse abstractor and entered into a custom-designed Microsoft Access database. Univariate analyses were conducted using contingency table methods with continuous variables divided into quartiles. Odds ratios with test-based confidence intervals were computed for 2x2 tables. Univariate odds ratios for continuous variables were computed by logistic regression. P values for contingency tables were chi squares or Fisher's exact tests as appropriate; for continuous variables they were Wald chi square tests based on logistic maximum-likelihood estimates. Multivariable analyses were computed by logistic regression with stepwise screening as a first pass. Variables significant by stepwise selection were stepped into the model manually, and those remaining significant at P<0.05 were retained in the final models. All computations were performed using SAS software version 8.02 running under Microsoft Windows 2000.
|
3. Results
|
|---|
Using univariate analysis, coronary artery disease, coronary artery bypass, a left ventricular ejection fraction of less than 50%, renal insufficiency, and emergency presentation all emerged as significant risk factors for 30-day mortality following thoracoabdominal and descending thoracic aortic aneurysm repair (Table 1). Preoperative renal insufficiency and emergency presentation, as in previous studies, proved to be the strongest predictors, with P-values <0.0001. The mortality rate for patients with coronary artery disease was 54 of 291 patients (18.5%) compared to 75 of 563 (13.3%) in patients without coronary artery disease (P<0.05). For coronary artery bypass patients the mortality rate was 21%, compared to 13.7% in non-bypass patients (P<0.02). Twenty-four of 94 (25.5%) patients with an ejection fraction less than 50% died within 30 days after thoracoabdominal or descending thoracic aortic aneurysm surgery, compared to 105/760 patients (13.8%) with normal ejection fraction (P<0.005). Without the left ventricular ejection fraction in the analysis, prior coronary artery bypass, renal insufficiency, and emergency presentation remained significant in multivariable analysis (Table 2A). When the left ventricular ejection fraction was included in the analysis, prior coronary artery bypass lost predictive force and ejection fraction of less than 50% emerged as a significant predictor of 30-day mortality (odds ratio 1.85, P<0.03), (Table 2B). Renal insufficiency and emergency presentation retained significance.
|
4. Discussion
|
|---|
In the last decade major advances in anesthesia, surgical techniques, and the use of distal aortic perfusion and cerebrospinal fluid drainage have reduced mortality and complications in patients undergoing thoracoabdominal and descending thoracic aortic aneurysm surgery [2,11–15]. In prior reports we found that age, aneurysm extent and emergency presentation predict increased mortality. Coronary artery disease as a specific risk factor has not been examined previously, although it is recognized as one of the leading causes of death in patients with thoracoabdominal aortic aneurysms [1,3,16,17]. Unlike age and aneurysm extent, which are fixed and unalterable risk factors, coronary artery disease can potentially be modified with appropriate treatment prior to thoracoabdominal or descending thoracic aortic aneurysm surgery.
There are some case reports of concomitant coronary artery bypass and thoracoabdominal aortic aneurysm repair [18–20], but the management of patients with coronary artery disease and thoracoabdominal or descending thoracic aortic aneurysm is underreported and the optimal treatment is debatable. Our approach has been to evaluate the left ventricular function, assess for the presence of correctible coronary artery lesions, and perform revascularization and aneurysm surgery in stages. If there are correctible coronary artery lesions, then coronary artery revascularization is undertaken prior to aneurysm surgery. In general, percutaneous coronary intervention with intraluminal stenting is performed for single vessel disease, and bypass surgery is reserved for multi-vessel disease. An interval of 4–6 weeks is observed after coronary artery revascularization before proceeding with thoracoabdominal or descending thoracic aortic aneurysm surgery. During this period, the patient is maintained on antiplatelet therapy, particularly after percutaneous intraluminal coronary artery stenting, or allowed to recover from coronary artery bypass surgery. However, if the patient has symptoms related to the thoracoabdominal or descending thoracic aortic aneurysm, the interval period may be shortened. Future studies will attempt to determine the optimal interval between coronary artery bypass and thoracoabdominal repair. When a patient requires coronary artery bypass surgery, the preferred conduit is the autologous reversed saphenous vein. Use of the left internal mammary artery is discouraged because the subsequent thoracoabdominal or descending thoracic aortic surgery may require crossclamping of the aorta proximal to the left subclavian artery, a maneuver that can lead to fatal cardiac ischemia if there is a patent left internal mammary artery graft to the left anterior descending coronary artery. Furthermore, the internal mammary artery may be an important collateral blood supply to the spinal cord.
The incidence of coronary artery disease in our patient population was 34% with a mortality rate of 18.5% within this group, compared to 13.3% for patients without coronary artery disease. Although in the univariate analysis coronary artery disease and coronary artery bypass were found as contributing to mortality, these effects were displaced by low ejection fraction in multivariable analysis. The left ventricular ejection fraction turned out to be the most important element of cardiac function to predict mortality following thoracoabdominal and descending thoracic aortic aneurysm repair. It is conceivable that the group of patients who had coronary artery bypass included patients with more severe and multi-vessel disease, a confounder for increased mortality in the multivariable analysis without inclusion of low ejection fraction. Similarly, the increased mortality risk in patients with coronary artery disease without consideration of the left ventricular ejection fraction in the analysis may reflect the heterogeneity of this group of patients that included severe non-correctible lesions.
In conclusion, a thorough evaluation of cardiac function, along with appropriate treatment, in patients with thoracoabdominal or descending thoracic aortic aneurysms is essential. Although the presence of coronary artery disease or coronary artery bypass may be associated with increased mortality following thoracoabdominal or descending thoracic aortic aneurysm repair, it is the underlying cardiac function – as measured by the left ventricular ejection fraction – that really determines the risk. We use a staged coronary disease treatment approach to restore as much cardiac function as possible before proceeding with aneurysm repair.
|
Footnotes
|
|---|
Presented at the 16th Annual Meeting of the European Association for Cardio-thoracic Surgery, Monte Carlo, Monaco, September 22–25, 2002.
|
Appendix A. Conference discussion
|
|---|
Dr T. Wahlers (Jena, Germany): As a consequence of your paper, do you routinely perform cardiac caths in these patients irrespective of the age?
Dr Safi: We ask our cardiologists to evaluate our patients. I don't do the cardiac cath. And cardiologists do cardiac caths in about 80% of cases. So it's up to our experienced cardiologists who have been with me for the last 20 years dealing with these cases.
Mr S. Westaby (Oxford, United Kingdom): If you identify poor cardiac function, whether it's impaired left ventricular function or severe coronary disease, would you be inclined to use cardiopulmonary bypass and perhaps circulatory arrest to support those patients as you do the thoracic aneurysm?
Dr Safi: In our experience using profound hypothermia and circulatory arrest, patients bleed through the left lung, and our mortality rate is higher than if we used distal aortic perfusion and CSF drainage. If you have an experienced cardiovascular anesthesiologist, I think he can get you through the operation with minimal risk.
Dr N. Mitety (Baghdad, Iraq): Have you made any bowel ischemia following this large number of series, and how do you try to prevent this type of ischemia in your patients?
Dr Safi: In our thoracoabdominal aortic aneurysms, the incidence of atherosclerosis affecting the visceral vessels is about 5%. These patients need endarterectomy of the celiac and superior mesenteric or renal arteries.
We find that endarterectomy of the celiac and superior mesenteric arteries is well tolerated, and we have less than 1% incidence of gut ischemia. But our problem is endarterectomy of the renal arteries. The incidence of renal failure, (defined by either dialysis or an increase in creatinine of 1 mg/dl over 2 days), is about 20–25%. It's a vexing problem that we continue to struggle with.
|
References
|
|---|
- Svensson L.G., Crawford E.S., Hess K.R., Coselli J.S., Safi H.J. Experience with 1509 patients undergoing thoracoabdominal aortic operations. J Vasc Surg 1993;17(2):357-368.[CrossRef][Medline]
- Safi H.J., Campbell M.P., Miller C.C., 3rd, Iliopoulos D.C., Khoynezhad A., Letsou G.V., Asimacopoulos P.J. Cerebral spinal fluid drainage and distal aortic perfusion decrease the incidence of neurological deficit: the results of 343 descending and thoracoabdominal aortic aneurysm repairs. Eur J Vasc Endovasc Surg 1997;14(2):118-124.[CrossRef][Medline]
- Hollier L.H., Symmonds J.B., Pairolero P.C., Cherry K.J., Hallett J.W., Gloviczki P. Thoracoabdominal aortic aneurysm repair. Analysis of postoperative morbidity. Arch Surg 1988;123(7):871-875.[Abstract]
- Beven E.G. Routine coronary angiography in patients undergoing surgery for abdominal aortic aneurysm and lower extremity occlusive disease. J Vasc Surg 1986;3(4):682-684.[CrossRef][Medline]
- Jones S.E., Raymond R.E., Simpfendorfer C.C., Whitlow P.L. Cardiac outcome of major non-cardiac surgery in patients undergoing preoperative coronary angioplasty. J Invasive Cardiol 1993;5(6):212-218.[Medline]
- Farkouh M.E., Rihal C.S., Gersh B.J., Rooke T.W., Hallett J.W., Jr, O'Fallon W.M., Ballard D.J. Influence of coronary heart disease on morbidity and mortality after lower extremity revascularization surgery: a population-based study in Olmsted County, Minnesota (1970–1987). J Am Coll Cardiol 1994;24(5):1290-1296.[Abstract]
- Wolf Y.G., Landersberg G., Mosseri M., Schechter D., Anner H., Weissman C., Berlatzky Y. Preoperative dipyridamole-thallium scanning, selective coronary revascularization and long-term survival in patients with critical lower limb ischemia. J Cardiovasc Surg (Torino) 2001;42(1):89-95.[Medline]
- Kresowik T.F., Hoballah J.J., Sharp W.J., Corson J.D. Cardiac screening tests prior to lower extremity revascularization: routine versus selective application. Semin Vasc Surg 1997;10(1):55-60.[Medline]
- Van Damme H., Pierard L., Gillain D., Benoit T., Rigo P., Limet R. Cardiac risk assessment before vascular surgery: a prospective study comparing clinical evaluation, dobutamine stress echocardiography, and dobutamine Tc-99m sestamibi tomoscintigraphy. Cardiovasc Surg 1997;5(1):54-64.[CrossRef][Medline]
- Kruspki W.C., Nehler M.R. How to avoid cardiac ischemic events associated with aortic surgery. Semin Vasc Surg 2001;14:235-244.[CrossRef][Medline]
- Estrera A.L., Miller C.C., 3rd, Huynh T.T., Porat E., Safi H.J. Neurologic outcome after thoracic and thoracoabdominal aortic aneurysm repair. Ann Thorac Surg 2001;72(4):1225-1230.[Abstract/Free Full Text]
- Safi H.J., Bartoli S., Hess K.R., Shenaq S.S., Viets J.R., Butt G.R., Sheinbaum R., Doerr H.K., Maulsby R., Rivera V.M. Neurologic deficit in patients at high risk with thoracoabdominal aortic aneurysms: the role of cerebral spinal fluid drainage and distal aortic perfusion. J Vasc Surg 1994;20(3):434-444.[Medline]
- Safi H.J., Hess K.R., Randel M., Iliopoulos D.C., Baldwin J.C., Mootha R.K., Shenaq S.S., Sheinbaum R., Greene T. Cerebrospinal fluid drainage and distal aortic perfusion: reducing neurologic complications in repair of thoracoabdominal aortic aneurysm types I and II. J Vasc Surg 1996;23(2):223-228.[CrossRef][Medline]
- Huynh T.T., Miller C.C., 3rd, Estrera A.L., Sheinbaum R., Allen S.J., Safi H.J. Determinants of hospital length of stay after thoracoabdominal aortic aneurysm repair. J Vasc Surg 2002;35(4):648-653.[CrossRef][Medline]
- Safi H.J., Campbell M.P., Ferreira M.L., Azizzadeh A., Miller C.C. Spinal cord protection in descending thoracic and thoracoabdominal aortic aneurysm repair. Semin Thorac Cardiovasc Surg 1998;10(1):41-44.[Medline]
- Cambria R.P., Davison J.K., Carter C., Brewster D.C., Chang Y., Clark K.A., Atamian S. Epidural cooling for spinal cord protection during thoracoabdominal aneurysm repair: a 5-year experience. J Vasc Surg 2000;31(6):1093-1102.[Medline]
- Kouchoukos N.T., Masetti P., Rokkas C.K., Murphy S.F., Blackstone E.H. Safety and efficacy of hypothermic cardiopulmonary bypass and circulatory arrest for operations on the descending thoracic and thoracoabdominal aorta. Ann Thorac Surg 2001;72(3):699-707.[Abstract/Free Full Text]
- Ener S., Yagdi T., Posacoglu H., Buket S. Combined thoracoabdominal aneurysm repair and off-pump coronary artery surgery. Thorac Cardiovasc Surg 2001;49(5):313-315.[CrossRef][Medline]
- Nakamura Y., Suzuki Y., Kamihira S., Taniguchi I., Yamaga T. Simultaneous minimally invasive direct coronary artery bypass and repair of a penetrating atherosclerotic ulcer of the descending thoracic aorta: report of a case. Surg Today 2001;31(11):999-1001.[CrossRef][Medline]
- Mihaljevic T., Tonz M., von Segesser L.K., Turina M.I. Combined coronary artery bypass grafting and repair of aneurysm of the descending aorta. Ann Thorac Surg 1999;67(5):1497-1499.[Abstract/Free Full Text]
Top
Abstract
1. Introduction
