| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Thorac Cardiovasc Surg 1994;108:567-575
© 1994 Mosby, Inc.
SURGERY FOR ACQUIRED HEART DISEASE |
Boston, Mass.
From the Department of Surgery, Harvard Medical School, and the Division of Cardiac Surgery, Brigham and Women's Hospital, Boston Mass.
Address for reprints: Robert J. Rizzo, MD, Division of Cardiac Surgery, Brigham and Women's Hospital, 75 Francis St., Boston, MA 02115.
Abstract
Angiography has been considered the gold standard for the diagnosis of acute dissection of the ascending aorta, but it may increase mortality by imposing an unnecessary delay before surgical repair. In addition, coronary angiography has often been considered essential as well. From 1988 to 1993, 37 patients (median age 61 years, 30 men and 7 women) had acute dissection of the ascending aorta. All of the initial 15 patients (group I) had angiography, even though the diagnosis of aortic dissection had already been made noninvasively in 14; six (40%) of 15 died, three of aortic rupture and none of complications of coronary artery disease. Among the next 22 patients (group II), 21 had a noninvasive diagnosis of acute dissection of the ascending aorta (eight by echocardiography; 13 by computed tomography), and 19 (86%) were operated on without angiography; two died (9%, p = 0.03 versus group I) and neither death was due to aortic rupture or coronary artery disease. Overall, either root or selective coronary angiography was attempted in 18 of 37 patients, but it documented coronary artery disease in only two patients (11%). Coronary artery disease was found in four other patients at autopsy; three of them, including two that died of aortic rupture, had angiography that failed to reveal the coronary artery disease. Noninvasive diagnosis of acute dissection of the ascending aorta is reliable and avoids the risks and delays inherent in invasive angiography. Rapid noninvasive diagnosis of aortic dissection and avoidance of routine angiography appear to improve survival by expediting surgical intervention and thus decreasing the risk of aortic rupture. (J THORAC CARDIOVASC SURG 1994;108:567-75)
Acute dissection of the ascending aorta creates an extremely tenuous condition in which the risk of mortality is increased dramatically in patients not receiving timely surgical therapy. 
1-6 In a collective review of 963 patients with aortic dissection who did not receive surgical treatment, the mortality was 38% within 1 day and 70% within 1 week; 86% of deaths were due to aortic rupture. 7 Acute obstruction of an aortic branch vessel causing organ ischemia and acute aortic valve insufficiency causing congestive heart failure are the other major contributors to morbidity and mortality in patients with acute dissection of the ascending aorta. Surgical intervention prevents aortic rupture and reverses organ ischemia and aortic valve insufficiency, but must be performed promptly before these complications occur or their effects become irreversible. Patients with suspected aortic dissection are often evaluated for coronary artery disease (CAD) in addition to aortic dissection because of their age and other cardiovascular risk factors.
Angiography had been considered the gold standard for the diagnosis of acute dissection of the ascending aorta and can also be used to evaluate CAD. However, the noninvasive techniques of echocardiography, computed tomography, and magnetic resonance imaging have recently been found to be reliable in establishing the diagnosis of aortic dissection and, particularly in the case of echocardiography, usually require less time to perform. 8-15 In addition, all of the noninvasive techniques can identify pericardial effusions, and echocardiography can detect aortic regurgitation and left ventricular dysfunction. If angiography is not needed to establish or confirm the diagnosis of acute dissection of the ascending aorta, then the benefit of angiography to evaluate CAD may not be worth the considerable risk associated with performance of the test.
The present study reviews our recent diagnostic and surgical experience with acute dissection of the ascending aorta and compares two sequential groups of patients in whom angiography was performed routinely versus selectively.
PATIENTS AND METHODS
Patient population.
Case histories of 37 consecutive patients (28 to 86 years of age, mean 58 years) with the diagnosis of acute dissection of the ascending aorta who were referred to the Cardiac Surgery Division at Brigham and Women's Hospital between January 1988 and March 1993 were reviewed and divided into two groups. Patients in group I were admitted between January 1988 and January 1991. All 15 patients had aortography: 14 had surgical repair and one had an aortic rupture in the catheterization laboratory and died before an operation could be done. Patients in group II were admitted between February 1991 and March 1993. All 22 patients had an operation but only three had angiography. Angiography was used selectively for clarification in one patient who also had a chronic aneurysmal dissection of the descending aorta; a second was thought to have a chronic dissection of the ascending aorta; and in the third the diagnosis was made by angiography at an outside hospital. The diagnosis of acute dissection of the ascending aorta was based on detection of a flap within the ascending aorta.
Data examined included clinical characteristics; diagnostic evaluations, including time from symptom onset, diagnosis, and arrival at the hospital to arrival in the operating room; prevalence of significant CAD (
50% diameter stenosis); operative techniques; and postoperative morbidity and mortality. CAD was termed acute when it was caused by the aortic dissection and chronic when it was not. Follow-up was complete for all patients.
Surgical technique.
Operation was performed via a median sternotomy. The right atrium and the femoral artery with the best pulsation were cannulated, cardiopulmonary bypass was established with the use of a membrane oxygenator, the patient was systemically cooled to 28° C or less, and the left ventricle was vented via the right superior pulmonary vein. The ascending aorta was crossclamped and opened, and cardiac arrest was induced and maintained with multidose cold blood cardioplegic solution administered via the coronary ostia or retrogradely via the coronary sinus, along with topical hypothermia. The intimal tear was resected and the dissected layers of aorta were reapproximated with circumferential sutures at the sites of the proximal and distal anastomoses, usually buttressed with a strip of pericardium or felt. If the intimal tear extended distal to the crossclamp, the patient was cooled and the tear was resected by means of an open technique. The dissected aortic valve commissure was resuspended if the valve could be repaired, or the valve was replaced in cases of chronic aortic valve disease. A combined mechanical valve–graft conduit was used if the aortic root needed replacement, and the coronary origins were usually reimplanted by the button technique. An interposition low-porosity woven Dacron aortic graft was sutured end to end proximally and distally to replace the segment of aorta containing the intimal tear and any areas of aneurysmal dilatation. More common recently was avoidance of crossclamping of the dissected aorta, use of retrograde cardioplegia, performance of the distal anastomosis first by means of the open technique 16 under circulatory arrest at nasopharyngeal temperatures of 16° C or less, use of the exclusion technique, and rewarming in the antegrade direction.
In patients who did not have angiography, the coronary arteries were evaluated at operation by epicardial palpation and by inspecting and in some cases by probing the coronary orifices from within the opened ascending aorta.
Statistics.
The group I (January 1998 through January 1991) and group II (February 1991 through March 1993) patients and the surgical and hospital survivors and nonsurvivors were compared by means of Fisher's exact test, the 
2 test, and the Wilcoxon rank-sum test. Multivariate analysis was by logistic regression. 17
RESULTS
Clinical characteristics.
No significant differences in clinical background were found between groups I and II (Table I) A history of symptomatic cardiac or peripheral vascular occlusive disease was found in only two patients (6%). Regarding clinical presentation, pericardial effusions were found in more than half of the patients in both groups, but cardiac tamponade and aortic rupture were more common in group I. All three patients with preoperative cerebral ischemia were in group II, had an emergency operation without angiography, and did very well with minimal or no residual neurologic deficit. Only one patient had acute myocardial ischemia.
|
Table II). No studies were falsely positive. In those cases in which a definitive flap could not be identified in the ascending aorta by echocardiography, one or several findings consistent with an acute dissection of the ascending aorta were often identified, such as a dilated or thickened aortic root, aortic insufficiency, or a pericardial effusion. The absence of any left ventricular dysfunction or regional wall motion abnormalities on echocardiography was also supportive of the diagnosis of aortic dissection, because these findings helped exclude the diagnosis of acute myocardial infarction. In group II, angiography was performed significantly less often and the time from symptom onset and arrival at the Brigham and Women's Hospital to arrival in the operating room was markedly decreased.
|
Table III). The only patients who required reoperative sternotomy were in group II, and one of these patients died. Arch replacements (one total arch, six hemiarch) and use of deep hypothermic circulatory arrest were also more common in group II. Coronary artery bypass grafting (CABG) was required for chronic CAD in only three patients (8%) and only once in group II.
|
Table IV). The hospital mortality resulting from aortic rupture also decreased, having accounted for half of the hospital deaths in group I but none in group II. In group I, fatal aortic rupture occurred in one patient in the cardiac catheterization laboratory and in two other patients after unsuccessful coronary angiography, either on transfer to the operating room or during anesthetic induction. The latter two patients were intubated, treated with closed chest massage, and had emergency surgical repair, but one could not be weaned from cardiopulmonary bypass and the other died of anoxic encephalopathy. In group II, the only patient in whom aortic rupture developed underwent sternotomy with relief of tamponade immediately on arrival in the operating room; this patient survived aortic repair. Noninvasive evaluation had already established the diagnosis of acute dissection of the ascending aorta and had also detected pericardial effusions in all four patients in whom aortic rupture eventually developed. These four patients represented 20% of the patients with acute dissection of the ascending aorta who had noninvasive evidence of pericardial effusions. The presence of a pericardial effusion on noninvasive evaluation was associated with some increased risk of aortic rupture (p = 0.08). None of the deaths resulted from complications of CAD (Table V).
|
|
0.05) in older patients and in those who had routine angiography or had pericardial effusions, aortic rupture, or abdominal visceral ischemia (Table VI). A suggestion (p 0.1) of increased risk was also apparent in patients with a history of hypertension or cardiac tamponade, those in whom the time from symptom onset to arrival in the operating room was 1 day or more, and in those who had any angiography.
|
Table VII). Routine angiography continued to be an independent predictor of hospital mortality even when we controlled for these confounders (all p < 0.15), which suggests that routine angiography is an additional risk for patients already at high risk of hospital death.
|
CAD.
Chronic CAD was found in two (11%) of 18 patients by angiography, in one (5%) of 19 patients evaluated at operation, and in four (80%) of five patients at autopsy, three of whom had angiography that failed to identify the CAD (Table VIII). Among the nine patients in whom selective coronary angiography was attempted, more died of preoperative aortic rupture (n = 3) than were identified by angiography to have CAD (n = 2). Thus the risk of using coronary angiography to look for CAD exceeded the probability of finding CAD.
|
Perioperative myocardial infarction was identified in four (11%) of the patients but was not fatal. The only patient who was admitted with acute coronary ischemia had acute dissection of the ascending aorta involving the right coronary artery and survived dissection repair with right CABG but did have an inferior myocardial infarction. In three patients both chronic CAD and acute subendocardial myocardial infarction were found at autopsy: two had angiography that failed to identify the CAD and had clinically insignificant myocardial infarctions, and one died during the operation after hemorrhage from the hemiarch anastomosis led to hypotension and global subendocardial ischemia. At follow-up, none of the survivors reported any angina or other manifestations of CAD.
DISCUSSION
Noninvasive diagnosis of acute dissection of the ascending aorta.
This study suggests that improved survival for patients with acute dissection of the ascending aorta can be achieved by the use of noninvasive diagnostic techniques to establish a diagnosis and expedite surgical therapy. It appears that routine angiography is not necessary and may impose greater risk than benefit. Most of the improvement in survival between groups I and II appears related to the decreased mortality from aortic rupture, which is the major cause of mortality that should be minimized by efforts to expedite surgical intervention. The diagnostic evaluation of patients in group II was more efficient, as exemplified by the fewer number of preoperative tests in general and angiography in particular and by the reduced time from symptom onset and arrival at Brigham and Women's Hospital to arrival in the operating room. Although the time from diagnosis to arrival in the operating room was similar overall for patients in groups I and II, the reason may be that the diagnosis was established at outside hospitals in more patients in group II, which incurred an obligatory transportation time. For those patients in whom the diagnosis was made at Brigham and Women's Hospital, the time from diagnosis to arrival in the operating room was indeed shorter in patients in group II.
Several recent reports in the literature include the number of patients who died in the hospital of acute dissection of the ascending aorta before an operation could be performed. Glower and associates 18 noted that 10 (11%) of 91 patients with DeBakey type I or II aortic dissections (acute and chronic combined) died before an operation could be done. In a recent multicenter Italian study in which routine angiography was performed, 23 (10%) of 242 patients considered reasonable candidates for operation died before the operation. 19 In the present study, three (20%) of 15 patients having routine angiography died of preoperative aortic rupture, whereas none of 22 patients having selective angiography died of this cause.
Although it is difficult to predict when an acute dissection of the ascending aorta will rupture, all aortic ruptures in the present study occurred in patients with known pericardial effusions. The presence of a pericardial effusion in patients with acute dissection of the ascending aorta should serve as a warning sign that hemodynamic compromise from tamponade or aortic rupture is imminent, and any delay for further diagnostic evaluation before aortic repair should be avoided especially in these patients. Rupture of an ascending aortic dissection has also been reported to occur during transesophageal echocardiography, which like angiography may be traumatic to perform in addition to causing delay. 20 Thus, although both angiography and transesophageal echocardiography have been used as a routine to diagnose or confirm acute dissection of the ascending aorta, 4,6,11 we believe that once the diagnosis of aortic dissection has been established by any method, the patient should be transferred immediately to the operating room. If further evaluation by transesophageal echocardiography is needed to assess aortic insufficiency or left ventricular function or to confirm the presence of aortic dissection, it can be performed most safely in the operating room, where anesthesia and surgical intervention are readily available.
When patients with acute dissection of the ascending aorta have profound hypotension because of tamponade, they may not survive transfer to the operating room even with ongoing closed chest cardiopulmonary resuscitation; this procedure may not be effective because the heart cannot fill adequately. In our experience, rapid surgical relief of the tamponade has been an effective way to restore cardiac function and thus preserve neurologic function in these patients. Subxiphoid pericardiotomy can be performed in the emergency room and extended into a median sternotomy in the operating room. Although frank aortic rupture with exsanguination may occur, more commonly the relief of tamponade allows these patients to undergo successful resuscitation and repair.
Our recent experience shows that a definitive noninvasive diagnosis of acute dissection of the ascending aorta can be relied on to be correct; however, if a diagnosis of aortic dissection cannot be established by one method, then further investigations should be rapidly performed, because missing a diagnosis of aortic dissection or another aortic problem could be fatal. Angiography should still play a vital role in the diagnosis of acute dissection of the ascending aorta whenever a definitive diagnosis cannot be made quickly by noninvasive techniques or in places where angiography is more available or reliable than the other modalities.
CAD.
One of the rationales for angiography in patients with suspected acute dissection of the ascending aorta has been to detect CAD in elderly patients about to undergo a major cardiovascular operation. The prevalence of CAD in patients with ascending aortic dissection may vary depending on the sensitivity of the methods used, the demographics of the population studied, and whether acute and chronic types of CAD are present. In an autopsy study by Larson and Edwards, 21 chronic CAD was found in 27 (22%) of 121 patients with DeBakey type I or II aortic dissections. On the basis of cardiac catheterization, DeBakey and colleagues 5 reported chronic CAD in 16 (8%) of 195 patients with type I or II aortic dissection. Rosenbloom and associates 22 recently reported finding CAD in 15 (41%) of 37 patients who had coronary angiography, or 15 (33%) of their total of 45 patients with acute dissection of the ascending aorta. In the present study of acute dissection of the ascending aorta, angiography revealed chronic CAD in 11% of patients who had angiography. Concomitant CABG has ranged in application from 0% to 38% in reports in the literature and was used in 14% of patients in the present study, but for chronic CAD in only 8%.
When the need for coronary angiography to assess CAD in patients with acute dissection of the ascending aorta is being evaluated, it may be helpful to separate acute from chronic CAD. Acute CAD is easy to evaluate in the operating room by direct visual inspection through the opened ascending aorta, and angiography is not necessary and merely prolongs the time to reperfusion of the ischemic myocardium. Chronic CAD is more difficult to evaluate at operation, but the opened aorta does provide access to the coronary ostia, through which the coronary arteries can be probed. Epicardial palpation can also help localize chronic CAD. Using these techniques, we have been successful in ruling out significant chronic CAD in all but one case. Angioscopy, which has been used recently for evaluation of CAD in donor hearts before heart transplantation, could play a role in the intraoperative evaluation of CAD in patients with acute dissection of the ascending aorta. 23
Although coronary angiography in general has become a safe and reliable technique, in our experience it appears to be much more difficult to complete successfully in patients with acute dissection of the ascending aorta. In the present study, the patients in whom CAD was found at autopsy did not die of complications of CAD, but several died of aortic rupture, which appeared related to delays imposed by difficult and eventually unsuccessful angiography. A subendocardial infarction was found at autopsy in two of these three cases of autopsy-proved CAD, but neither was of clinical significance. Given the relatively low prevalence of chronic CAD, the risk of potentiating fatal aortic rupture by subjecting patients to angiography, and the feasibility of intraoperative evaluation of CAD, it seems reasonable to recommend avoidance of routine angiography in patients with acute dissection of the ascending aorta. This approach has also been advocated by other groups. 24,25 Coronary angiography may be considered in the few patients with a history of angina or possibly any vascular occlusive disease, provided they are in hemodynamically stable condition and do not have pericardial effusions.
Mortality.
The overall operative mortality of 19% in the present series falls within the range of 5% to 39% reported by other groups in the surgical management of patients with acute dissection of the ascending aorta. 1-6,11,16,24,26-29 Similarly, the reduction in operative mortality in the more recent portion of this series (19% to 9%) parallels the improvements over time reported by Miller's group 28 (25% to 7%) and Svensson's group 29 (25% to 5%), which have been attributed to earlier referral for operation and advances in surgical technique. In the present series, the improvement may also be related to more expeditious diagnosis and surgical therapy, resulting in less mortality from aortic rupture.
Most of the factors found in the present series to be significantly associated with hospital mortality, such as visceral ischemia, aortic rupture, age 60 years or older, and pericardial effusion, have been found in other studies to be associated with an increased risk of mortality. Miller and associates, 28 in a multivariate analysis of 76 patients with acute dissection of the ascending aorta, identified renal dysfunction, cardiac tamponade, renal/visceral ischemia, and earlier operative period as factors associated independently with increased risk. Svensson and coworkers, 29 in a multivariate analysis of 82 patients, identified operation more than 1 day after onset of symptoms, extension of the operation into the aortic arch, cardiac complications, and earlier operative period as independent predictors of operative mortality. The present study may be the first to suggest that routine angiography in patients with acute dissection of the ascending aorta is also associated with an increased risk of hospital mortality and, in particular, mortality caused by aortic rupture. Rapid noninvasive diagnosis of acute dissection of the ascending aorta and avoidance of routine angiography appear to improve survival by expediting surgical therapy, which decreases the risk of aortic rupture.
Appendix: DISCUSSION
Dr. D. Craig Miller (Stanford, Calif.).
I agree wholeheartedly with the points Dr. Rizzo's group has raised, especially the main message that no unnecessary delay can be tolerated in getting patients with acute type A dissections to the operating room. This obviously means using coronary angiography only on a highly selective basis. This approach certainly violates the Cleveland Clinic gospel, but I think it is the way to go.
I also agree with the changes in operative methods that this group has made over time, namely, avoiding clamping the ascending aorta at all before the onset of circulatory arrest during which a hemiarch anastomosis or open anastomosis (as popularized originally by Livesay, Ott, and Cooley), is performed and, importantly, thereafter reperfusing and rewarming the patient using only antegrade perfusion by inserting a second cannula into the arch graft. I believe that is exceedingly important, and it has been used more and more often by our group.
Another helpful trick that we have used over the past 2 years, but apparently you have not, is to briefly retroperfuse the brain through the superior vena cava cannula to get rid of any occult debris or air that might be in the great vessels. I should caution, however, that it is exceedingly important to keep the pressure in the subdural sinuses under 40 mm Hg, including the hydrostatic pressure when the head is down (steep Trendelenburg position).
Do you have any comments regarding this terminal flushing of the brain?
Dr. Rizzo.
Yes. I have used it within the past year and I think it has been effective. I cannot comment on statistical significance yet, but I think it may have some value in both deairing and flushing out atherosclerotic debris.
Dr. Miller.
Sometimes it is alarming to see what comes out of the great vessels, especially in cases of atherosclerotic arch aneurysms.
Second, it is interesting to note that the number of your cases per year almost more than doubled from five per year to 11 per year from group I and group II while simultaneously the number of patients whose diagnosis was made at the Brigham decreased. This trend in more outside referrals can reflect various factors, ideally including earlier diagnosis resulting in more patients surviving to receive a correct diagnosis and then being transported to a tertiary, specialized referral center.
Do you have any other information in the Boston area that might explain this increase in outside referrals?
Dr. Rizzo.
I don't. All I can say is that I think proper management of acute aortic dissection involves a team effort including the primary physicians, the emergency room physicians, and the transferring and local cardiologists. It does seem that our volume has increased. It may be that more people are alive at diagnosis now that noninvasive diagnostic techniques are better and are better understood. One of these patients came in with a diagnosis of abdominal aneurysm on a computed tomographic scan of the abdomen. We then had to realize that the aorta appeared to be dissected. We then detected a pleural effusion on the lower cut of the scan, and that gave us a clue that there was proximal involvement. We all have to be aware of the different ways acute dissection can present. I think we are all getting better at identifying acute dissection.
Dr. Miller.
However, even in group II, the 1991 to 1993 period, only 18% of the diagnoses made noninvasively were made by means of transesophageal echocardiography. Because transesophageal echocardiography is the preferred screening and/or definitive diagnostic test for suspected aortic dissection, why wasn't this fraction closer to 100%? Did some of the outlying hospitals not have expertise with transesophageal echocardiography?
Dr. Rizzo.
Many of these patients came from community hospitals that may not have had rapid access to a transesophageal echocardiography technique. Computed tomographic scan is much more readily available now in outside community hospitals. But regardless of the method of making a diagnosis, the primary objective is for the patient to be alive on transfer.
Dr. Miller.
I agree with you wholeheartedly. Do whatever test in your local hospital that is quickest and most reliable.
You state that all patients should be transferred immediately to the operating room after the presumptive diagnosis of dissection obtained by any noninvasive diagnostic method, and I would agree. However, we should not forget that suboptimal quality of some computed tomographic scans or transesophageal exams performed at outside hospitals might lead to false positive diagnoses. What do you do in that case? Do you do a transesophageal echocardiogram in the operating room, as your manuscript suggests?
Dr. Rizzo.
Yes. If someone is admitted with a strong history, then I presume the diagnosis is acute dissection until proved otherwise. If there is a strong enough suspicion, I transfer the patient directly to the operating room and continue the workup there, ready to be able to either intubate or cannulate if necessary.
Dr. Miller.
I think that is an excellent plan.
Finally, my only criticism of this nice, retrospective investigation is that the analytical results of your multivariate logistic regression analysis are not presented explicitly. You do state that many variables confounded this analysis, looking at operative death, including older age, rupture, tamponade, time interval between symptoms and arrival in the operating room, and visceral ischemia. You further suggest, though, without proving to us how you can substantiate it, that routine angiography is an additional risk factor for patients already at high risk of operative death. I have difficulty comprehending this because this problem is exactly why multivariate models were developed— to eliminate variables carrying redundant information. I would be interested in the actual results of your multivariate analysis. Not withstanding, I submit that angiography may not actually be the real culprit, but only one that reflects inordinate delay due to whatever reason in getting the patient to the operating room.
Dr. Rizzo.
Thanks for your comments. Regarding the multivariate analysis, we did perform one. The numbers are small, including only 37 patients. I think to have an adequate multivariate analysis to either prove or disprove associations, the numbers need to be larger. We were able to show a suggestive effect related purely to angiography. We had to separate out the dependent variable, the mortality, from the independent variables. There was a fair bit of overlap among the different variables, but even when we examined two independent variables at a time, always including angiography as one of the independent variables, it always seemed that angiography had a tendency to be associated with mortality even independent of those other factors.
Dr. Miller.
That observation may reflect that coronary angiography is not only impossible in many cases, but also frequently dangerous. Those of us surgeons who look at these acutely dissected ascending aortas dread to think what the catheter is doing when the cardiologist or radiologist is trying to cannulate the coronary ostia. Furthermore, as you have proved, angiography can be fatal or also totally misleading given your three cases of coronary disease not diagnosed properly and detected only at autopsy.
I think this work contains some valuable lessons. If I might poll the audience, how many of the surgeons in the audience feel comfortable going to the operating room on the basis of transesophageal echocardiography or computed tomography alone? That looks like a large majority.
Dr. Thomas Pfeffer. (Los Angeles, Calif.).
Could you comment on about what percentage of patients arrive at your hospital with definitive diagnostic studies having already been completed? Second, what patients other than those with a known history of coronary disease do you select for angiography?
Dr. Rizzo.
The percentages of patients who were admitted with a definitive diagnosis had changed from 40% in group I to roughly 70% in group II. Definitive diagnosis was defined as a flap visualized in the ascending aorta. Others were admitted with suggestive diagnoses based on their noninvasive tests, for example, a pericardial effusion, dilated root, or thickened root.
Our method of selecting patients for coronary angiography is still evolving. We are becoming more comfortable with the fact that these patients die of aortic rupture and not of CAD. To waste time looking for CAD with an angiogram is not without risk, even if you suspect that CAD is present on the basis of the patient's age or a history of angina or peripheral vascular occlusive disease. Thus, particularly if the patient has a pericardial effusion, we would not obtain an angiogram. We would operate immediately and assess the patient's condition at operation. It may be that we have been fortunate, but we have not had to do otherwise since we began using this philosophy. I am not sure that we will ever get an angiogram again to look at the coronary arteries.
Another technique, angioscopy, might be applied in the operating room to look rapidly for significant CAD. We've had some experience in vascular surgery using angioscopy, but I have not applied it yet on the coronary arteries.
Acknowledgments
We thank Nancy Hery for assistance with preparation of the manuscript.
Footnotes
Read at Nineteenth Annual Meeting of The Western Thoracic Surgical Association, Carlsbad, Calif., June 23-26, 1993. 
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
