Perioperative stroke remains a major complication of cardiac surgery, and atheroemboli are believed to be a significant etiologic factor. Epiaortic (EPI) ultrasonic scanning has been advocated as a means of detecting ascending aortic atherosclerotic plaque in order to avoid inducing intraoperative embolization . The use of EPI scanning, however, lengthens the surgical procedure, introduces the risk of contaminating the surgical field, and requires both special equipment and extensive training. Transesophageal echocardiography (TEE) does not interrupt the surgery or invade the surgical field, but allows only limited visualization of a significant portion of the thoracic aorta. This study investigated the ability of TEE imaging of all visualized portions of the thoracic aorta to predict the presence of atherosclerotic plaque in the ascending aorta.
The protocol was approved by our institutional review board, and written informed consent was obtained from 81 patients scheduled for cardiac surgery requiring cardiopulmonary bypass. There were 57 males and 24 females aged 32-88 yr old (mean age 64 yr). Fifty-one patients underwent coronary artery bypass grafting, 15 patients underwent single or double valve replacement, 11 had combined valve/coronary artery bypass grafting procedures, 2 had Bentall procedures, 1 had an ascending aortic aneurysm repair, and 1 had an atrial septal defect repair.
After tracheal intubation, a multifrequency adult biplane TEE probe (Acuson, Mountain View, CA) was inserted into the esophagus and a comprehensive interrogation of the thoracic aorta was performed in the following sequence. At a depth of approximately 25-30 cm in the transverse plane, a short axis view of the ascending aorta was obtained. The probe was advanced down to the aortic valve, and in the longitudinal plane the ascending aorta was then imaged. Next, the probe was switched back to the transverse imaging plane and rotated posteriorly to image the descending aorta and advanced into the stomach until the image of the descending aorta disappeared from view. The probe was then withdrawn slowly, allowing visualization of the complete descending aorta. At the level of the aortic arch, the probe was rotated clockwise to image the transverse aortic arch. Additional views of the arch were obtained by changing to the longitudinal plane and slowly rotating the probe through the arch. After pericardiotomy, the EPI examination was performed. A 7-MHz phased array Microcase Registered Trademark probe (Acuson, Mountain View, CA) was placed in an ultrasonic gel containing probe cover and applied directly to the ascending aorta at the level of the aortic valve. Short axis and long axis views of the entire ascending aorta were then obtained. Near field enhancement of the anterior wall of the ascending aorta was achieved by either a gel standoff, fluid in the pericardium, or application of a fluid containing sterile glove placed between the probe and the aorta. The presence and severity of atherosclerosis was evaluated at the time of the echo examinations and both the TEE and EPI examinations were recorded on video tape for later review.
According to the grading system proposed by Wareing et al. , the presence and severity of atherosclerosis in the thoracic aorta was noted in both the TEE and EPI examinations. Intimal thickening <or=to3 mm was defined as mild disease and was not considered significant. Echodense focal areas of atherosclerosis or intimal thickening >3 mm and <5 mm were defined as moderate disease, and severe disease was defined as multiple areas of plaque or intimal thickening >5 mm. These were considered significant atherosclerotic disease. Thickness of the disease was measured by freezing the image and using the digital calipers of the internal calculation software. These measurements were performed when any disease was detected. Four aortic segments were graded for each patient: three TEE (ascending, arch, and descending) and one EPI (ascending).
The original on-line evaluations were entered into a database and constituted the first reading. A second reading of all of the videotapes was performed by a single observer 3-12 mo later. This observer was blinded to the pairings of the TEE and EPI examinations. Any disparities in plaque severity interpretation were resolved by an additional observer (also blinded to the pairings). Using EPI scanning as the "gold standard" and the standard formulas, the sensitivity, specificity, positive predictive value, and negative predictive value of TEE for detecting moderate to severe ascending aortic atherosclerosis were calculated. This was performed for all three segments of the thoracic aorta individually (i.e., ascending, transverse arch, and descending) and for the thoracic aorta as a whole. Ninety-five percent confidence limits were calculated for the sensitivity and negative predictive value of TEE-visualized plaque anywhere in the thoracic aorta for EPI-visualized ascending aortic plaque.
Interpretable images were obtained in all patients, and a total of 324 aortic segments (81 patients times 4 segments per patient) were graded. Disparities between the on-line reading and the second reading occurred in only six segments (98.1% concordance). Fourteen of the 81 patients (17%) had either moderate or severe atherosclerosis of the ascending aorta by EPI scanning, and 41 of the 81 patients (51%) had significant atherosclerosis of the thoracic aorta by TEE. There were no complications attributable to the study.
(Table 1) shows the following results: In the comparison of TEE scanning of the ascending aorta and EPI scanning, there were 4 true positives, 66 true negatives, 1 false positive, and 10 false negatives. This yielded a sensitivity of 29%, a specificity of 99%, a positive predictive value of 80%, and a negative predictive value of 87%.
In the comparison of TEE scanning of the aortic arch and EPI scanning, there were 10 true positives, 55 true negatives, 12 false positives, and 4 false negatives. This yielded a sensitivity of 71%, a specificity of 82%, a positive predictive value of 45%, and a negative predictive value of 93%.
In the comparison of TEE scanning of the descending aorta and EPI scanning, there were 10 true positives, 42 true negatives, 25 false positives, and 4 false negatives. This yielded a sensitivity of 71%, a specificity of 63%, a positive predictive value of 29%, and a negative predictive value of 91%.
In the comparison of TEE scanning of the entire thoracic aorta and EPI scanning, there were 14 true positives, 40 true negatives, 27 false positives, and 0 false negatives. This yielded a sensitivity of 100%, a specificity of 60%, a positive predictive value of 34%, and a negative predictive value of 100%. The onetailed 95% lower bound is 78.3% for the sensitivity and 91.4% for the negative predictive value.
Stroke after adult cardiac surgery is a major cause of morbidity and mortality, and autopsy studies have shown that atheroemboli from the ascending aorta are a major etiologic factor . Plaque in the transverse aortic arch has also been shown to be a risk factor for stroke . Intraoperative detection and avoidance of the diseased portions of the ascending aorta may reduce the incidence of perioperative stroke . Intraoperatively, aortic plaque can be detected by palpation, TEE, or placement of a sterilely wrapped echocardiographic probe directly on the aorta--EPI scanning. Previous work has indicated that palpation detects only 25% of atherosclerotic lesions and thus is of little clinical value . Because TEE does not lengthen the surgical procedure and does not have the potential of contaminating the surgical field, it would be the preferred method of aortic evaluation. However, due to the interposition of the trachea and bronchi between the esophagus and the aorta, TEE may not visualize as much as 42% of the ascending aorta and may miss significant aortic atherosclerotic lesions . Therefore, EPI scanning currently seems to be the optimal intraoperative method of detecting aortic atherosclerotic lesions.
It is desirable to limit the number of EPI examinations because many patients do not have aortic plaques, the ultrasonic probe and disposables are expensive, and surgical cooperation and training are necessary. Thus, it is desirable to determine the need for EPI scanning. In this study, no single section of biplane TEE evaluation of the thoracic aorta absolutely predicted the presence of ascending aortic plaque as detected by EPI scanning. The present study demonstrates that significant plaque observed anywhere in the thoracic aorta by TEE is a very sensitive screening examination for significant ascending aortic plaque in a wide cross-section of patients. This application of TEE is limited, however, by the moderate specificity.
The ability to screen patients for ascending aortic plaque using biplane TEE of the entire thoracic aorta does not imply that TEE is capable of visualizing or localizing all of the plaque in the ascending aorta. EPI scanning has been previously shown to be superior for this application [7,8]. The low sensitivity (29%) of the ascending aortic TEE examination (for ascending aortic atherosclerosis) in the current study confirms others' findings that TEE is not clinically reliable for detecting significant atherosclerotic disease of the ascending aorta. Others have reported using TEE of the ascending aorta alone as a decision criterion for performing endarterectomy of the ascending aorta under hypothermic circulatory arrest . This application of TEE is not supported by the present study.
Interestingly, one patient Figure 1 had TEE findings of moderate intimal thickening (4.6 mm) in the longitudinal view of the ascending aorta with only minimal intimal thickening (2.3 mm) present on the transverse aortic view by EPI scanning. Because of the potential for a tangential cross-section of the aortic wall by TEE giving a distorted enlarged impression of the disease and the superiority of short axis views in determining wall thickness and overall superior image quality of EPI scanning, it was assumed that the EPI image was correct. Unfortunately, no anatomic verification was possible. In addition to the potential for tangential cross-sections, air/fluid interfaces between the probe and the target can create other artifacts .
The limited patient sample in this study limits the power of the findings. Therefore, we estimated lower confidence bounds for the sensitivity and negative predictive value of examinations of the entire aorta. The true sensitivity could be anywhere between 78% and 100% (95% confidence) and be compatible with these data. This study looked at a cross-section of all patients coming for cardiac surgery; it is possible that the results would be different in a "high-risk" group of patients.
This study only investigated the use of biplane TEE to predict the presence of plaque in the ascending aorta. Similar results would be expected with multiplane TEE probes, but this may not be the case for monoplane probes. Monoplane TEE probes have a far more limited view of the ascending and transverse aorta. It is possible that significant plaque could have been missed by monoplane imaging.
In conclusion, the role of TEE and EPI scanning in reducing the incidence of perioperative stroke remains controversial. This study showed that biplane TEE of the entire thoracic aorta is a very sensitive, but only moderately specific, method of predicting the presence of atherosclerotic plaque in the ascending aorta.
1. Marshall WG, Barzilai B, Kouchoukous NT, Saffitz J. Intraoperative ultrasonic imaging of the ascending aorta. Ann Thorac Surg 1989;48:339-44.
2. Wareing TH, Davila-Roman VG, Barzilai B, et al. Management of the severely atherosclerotic ascending aorta during cardiac surgery. J Thorac Cardiovasc Surg 1992;103:453-62.
3. Blauth CI, Cosgrove C, Webb BW, et al. Atheroembolism from the ascending aorta. J Thorac Cardiovasc Surg 1992;103:1104-12.
4. Amarenco P, Cohen A, Tzourio C, et al. Atherosclerotic disease of the aortic arch and the risk of stroke. N Engl J Med 1994;331:1474-9.
5. Wareing TH, Davila-Roman VG, Daily BB, et al. Strategy for reduction of stroke in cardiac surgical patients. Ann Thorac Surg 1993;55:1400-8.
6. Ohteki H, Itoh T, Natsuaki M, et al. Intraoperative ultrasonic imaging of the ascending aorta in ischemic heart disease. Ann Thorac Surg 1990;50:539-42.
7. Konstadt SN, Reich DL, Quintana C, Levy M. The ascending aorta: how much does transesophageal echocardiography see? Anesth Analg 1994;78:240-4.
8. Black IW, Stewart WJ, Grimm R, et al. Direct epivascular echocardiography is essential for complete intraoperative assessment of aortic atheroma [abstract]. Circulation 1993;88:I386.
9. Katz ES, Tunick PA, Rusinek H, et al. Protruding aortic atheromas predict stroke in elderly patients undergoing cardiopulmonary bypass: experience with intraoperative transesophageal echocardiography. J Am Coll Cardiol 1992;20:70-7.
10. Appelbe AF, Walker PG, Yeoh JK, et al. Clinical significance and origin of artifacts in transesophageal echocardiography of the thoracic aorta. J Am Coll Cardiol 1993;21:754-60.