- Email: [email protected]
The intraoperative assessment of ascending aortic atheroma: Epiaortic imaging is superior to both transesophageal echocardiography and direct palpation
The Intraoperative Assessment of Ascending Aortic Atheroma: Epiaortic Imaging Is Superior to Both Transesophageal Echocardiography and Direct Palpation Stephen Sylivris, FRACP, Paul Calafiore, FRACP, George Matalanis, FRACS, Alexander Rosalion, FRACS, Hok Pan Yuen, MSc, Brian F. Buxton, MS, and Andrew M. Tonkin, MD Objectives: To determine the optimal method for detecting ascending aortic atheroma intraoperatively by comparing manual palpation by the operating surgeon, intraoperative transesophageal echocardiography, and epiaortic ultrasound (linear and phased-array imaging); and to assess risk factors for severe aortic atheroma. Design: A longitudinal prospective study. Assessment of the atheroma by manual palpation was blinded to the results of the ultrasound images. Setting: The study was performed in a single university tertiary referral hospital. Participants: One hundred consecutive patients undergoing coronary bypass or valve surgery were studied after their written, informed consent. Interventions: Potential risk factors were evaluated by both a patient questionnaire and examination of prior hospital records. The ascending aorta was assessed by the following methods: manual palpation by the operating surgeon, intraoperative transesophageal echocardiography, and epiaortic ultrasound (linear and phased-array imaging) performed by an echocardiologist. For analysis, the ascending aorta was divided into three equal segments: proximal, mid, and distal, corresponding to regions of different operative manipulations.
Measurements and Main Results: Age older than 70 years and hypertension were significant risk factors for severe ascending aortic atheroma with adjusted odds ratios of 3.3 (95% CI, 1.2 to 9.3) and 3,9 (95% CI, 1.3 to 12.0), respectively. There was no significant difference in atheroma detection between the two ultrasonic epiaortic probes in any segment; however, epiaortic probes were superior to manual palpation in all segments and also superior to transesophageal echocardiography in the mid and distal segments of the ascending aorta. Conclusions: Age older than 70 years and hypertension are significant risk factors for severe ascending aortic atheroma. Intraoperative detection of ascending aortic atheroma is best achieved by epiaortic ultrasound with either a linear or phased array transducer. Transesophageal echocardiography is an insensitive technique for evaluation of mid and distal ascending aortic atheroma and, therefore, of little value in guiding surgical manipulations such as cross-clamping. Copyright © 1997 by W.B. Saunders Company
TROKES are a major cause of perioperative morbidity and mortality in cardiac surgical patients.1 Recently, numerous studies have shown that ascending aortic and aortic arch atheroma is an independent risk factor for stroke. 2,3 Currently, surgeons rely on manual palpation to determine the location and severity of ascending aortic atheroma and, consequently, the safest sites of cannulation, cross-clamping, graft insertion, and suturing. Because of the advent of newer ultrasound technology, however, transesophageal and direct epiaortic ultrasound techniques have made it possible to noninvasively image this area. Some recent studies have shown that epiaortic ultrasound is far more sensitive in atheroma detection than manual palpation, 4-6 and one study suggested that it may be more sensitive than transesophageal scanning. 7 However, no studies have compared all three modalities. One purpose of this study was to compare the ability of manual palpation, biplane transesophageal echocardiography, and epiaortic ultrasound in the detection of ascending aortic atheroma. Furthermore, a prototype epiaortic 7-MHz phasedarray transducer and a current conventional 7-MHz linear transducer were also compared. The phased array transducer, because of its miniaturized size, theoretically has greater maneuverability in the confines of a limited surgical field and
may detect atheroma when optimal images cannot be obtained by the bulkier linear transducer. The other major aim of this study was to identify the risk factors associated with severe ascending aortic atheroma. This study is unique in that no other project has compared all of these noninvasive techniques in this clinical context or used the findings to assess potential risk factors for severe ascending aortic atheroma.
S
From the Departments of Cardiology and Cardiothoracic Surgery, Austin and Repatriation Medical Centre, Heidelberg, Australia, and the Statistical Consulting Centre, University of Melbourne, Australia. Address reprint requests to Stephen Sylivris, FRACP, c/o Austin and Repatriation Medical Centre, Studley Rd, Heidelberg 3084, Victoria, Australia, Copyright © 1997 by W.B. Saunders Company 1053-0770/97/1106-000453.00/0 704
KEY WORDS: transesophageal echocardiography, epiaortic ultrasound, ascending aortic atheroma
METHODS Consecutive patients undergoing cardiac surgery under cardiopulmonary bypass at the Austin and Repatriation Medical Centre Hospital were considered for recruitment into this study. Patients were excluded if they had a contraindication to transesophageal echocardiography (esophageal varices or esophageal obstruction), if they did not provide informed consent, or if they were having redo coronary bypass grafting. All potential risk.factors were evaluated by both a patient questionnaire and from examination of prior hospital records. Six variables were prospectively identified: age older than 70 years, diabetes mellitus, any prior smoking history, hypertension (two or more prior blood pressure readings >140/90 or currently on drug treatment for hypertension), serum cholesterol >5.5 mmoi/L, and extracardiac vascular disease on clinical examination, as potential risk factors. After anesthesia and endotracheal intubation, all enrolled subjects underwent assessment of their aorta. First, a transesophageal echocardiograph was performed during graft harvesting. A multifrequency biplane transducer (Acuson Computed Sonography, Mountain View, CA) using varying frequencies (to obtain optimal short- and long-axis views) was used. Both the ascending and descending aorta were imaged. The surgeons were blinded to the results of the ultrasound examination. After exposure of the ascending aorta, the surgeons then graded any ascending aortic atheroma by manual palpation. Direct epiaortic ultrasound imaging was then performed by an echocardiologist. Both the standard 7-MHz transducer and the prototype 7-MHz phased-array transducer (with a gel standoff to enhance the near field) were used, and
Journal of Cardiothoracic and VascularAnesthesia, Vol 11, No 6 (October), 1997: pp 704-707
INTRAOPERATIVE ASSESSMENT OF AORTIC ATHEROMA
705
short- and long-axis views were again obtained. The prototype phasedarray probe is significantly smaller than the standard linear probe (5.0 × 1.5 × 2.0 cm compared with 5.5 × 2.5 × 10.0 cm) and was thought to have greater maneuverability in the confines of a limited surgical field. Consequently, it may detect atheroma when optimal images cannot be obtained by the bulkier linear transducer. All ultrasonic images were recorded on standard VHS videotape for off-line measurements and review by two independent cardiologists. The ascending aorta was divided into three equal segments between the aortic root and the innominate artery, as described by DavillaRoman et al.4 These three segments correspond to sites of differing surgical manipulations and may consequently have varying degrees of importance in embolus generation. The proximal and distal segments are the usual sites of cardioplegia and arterial cannulation, respectively, and the mid segment the site of vein graft anastamosis. Quantification of the severity of ascending aortic atheroma was performed by dividing it into three grades: Grade 0, localized thickening less than 1.0 mm thick (absent or minimal on palpation), Grade 1, atheroma 1.0 to 3.9 mm thick (moderate on palpation), and Grade 2, 4.0 or more mm thick or mobile atheroma (severe on palpation). This ultrasound grading system is the same as that used by Amarenco et al,2 who showed that patients with severe ascending atheroma have a markedly increased odds ratio of stroke. Stepwise logistic regression analysis was performed using the statistical computing package GLIM 3.77 (Royal Statistical Society, London, 1985). Sign test analyses were performed to compare the ability of the different modalities in atheroma detection. Ap value of <0.05 was considered significant. RESULTS
A total of 100 patients were enrolled, and ultrasonic scans were obtained in all. Their mean age was 69 _+ 8 years, and they included 75 men and 25 women. Eighty-three patients underwent coronary bypass surgery only, whereas 10 had aortic valve replacement and seven had combined bypass surgery and aortic valve replacement. Overall, 90% of subjects had atheroma detected by ultrasonic techniques, whereas only 16% had atheroma detected by the standard method of manual palpation. (p = 0.02, Table 1.) Severe (2+) atheroma was detected in 26 of 100 patients and involved 38 different segments of the ascending aorta. Both forms of epiaortic ultrasound (linear and phased-array imaging) were able to detect the severe grade of atheroma to an equal extent. However, manual palpation and transesophageal echocardiography were unable to detect this grade of atheroma in most cases (severe atheroma was detected in only two [p < 0.0001] and three [p < 0.0001] segments, respectively). These data are presented in Table 2. Risk factor analysis for severe (2+) ascending aortic atheroma is displayed in Table 3. Age older than 70 years and hypertension were shown to be significant risk factors. The
Table 1. Operative Procedure and Atheroma Detection Surgical Procedure
No, of Patients
CABG AVR CABG & A V R Total
83 10 7 100
Atheroma by Ultrasound n (%) 75 8 7 90
(90) (80) (100) (90)
Atheroma by Palpationn (%) 14 2 0 16
(17) (20) (0) (16)
Abbreviations: CABG, coronary artery bypass grafting; AVR, aortic valve replacement. p = 0.02.
Table 2. Detection of Severe (2+) Atheroma in 28 Subjects (38 Segments) Technique
No. of Segments
p (v Linear Array)
Palpation Transesophageal echo Phased array (epiaertic) Linear array (epiaortic)
2 3 35 36
<0,0001 <0,0001 NS
Abbreviation: NS, not significant; p > 0.05.
adjusted odds ratio for age was 3.3 (95% CI, 1.2 to 9.3); whereas for hypertension, the adjusted odds ratio was 3.9 (95% CI, 1.3 to 12.0). There was also a significant association between the presence of severe descending atheroma and severe ascending atheroma. Risk factors that were not significant included diabetes, extracardiac vascular disease, smoking history, and cholesterol >5.5 mmol/L. When analyzing the three ascending aortic segments, the epiaortic ultrasound was used with the linear probe as the baseline with which the other techniques were compared (Table 4). In the proximal segment, there was no significant difference in atheroma grading between the transesophageal echocardiograph and either form of epiaortic imaging. However, manual palpation significantly underestimated the atheroma grade in 69 of ]00 cases (p < 0.0001). In the mid and distal ascending aortic segments, there was again no significant difference between atheroma grading by either form of epiaortic ultrasound. However, in these segments, not only did manual palpation underestimate atheroma grade (in 53 of 100 cases, p < 0.0001), but transesophageal echocardiography also significantly underrated the atheroma grade (in 51 of 100 cases, p < 0.0001). Intrarater and interrater agreement were assessed by reviewing 54 and 45 segments, respectively (Table 5). There was an excellent agreement on atheroma grading, with Kappa values all greater than 0.9 with each of the ultrasound methods. DISCUSSION
These studies clearly show that patients undergoing cardiac surgery have a high incidence of coexisting ascending aortic atheroma. Moreover, the best modality to assess this atheroma, especially in the mid and distal segments of the ascending aorta, is direct epiaortic ultrasound. Table 3. Risk Factor Analysis
Risk Factor
Grade 2+ Disease (n - 26)
GradeO/+ Disease (n = 74)
Age >70 yrs 19 (73%) 34 (46%) Smoking 17 (65%) 49 (66%) Diabetes 8 (31%) 17 (23%) Hypertension 22 (85%) 34 (46%) Cholesterol >5.5 mmol/L 15/24 (63%) 32/47 (68%) Extracardiac vascular disease 9 (35%) 13 (18%) Descending aortic atheroma I+ 11 (42%) 44 (59%) 2+ 13 (50%) 18 (24%)
Odds Ratio Crude Adjusted
95% Cl
2.4 1.0 1.3 4.3
3.3 1.0 1.2 3.9
1.2-9.3 0.4-2.8 0,4-3.6 1.3-12.0
0.8
1.7
0.5-5.7
1.9
2.2
0.7-6.5
1.4 3.0
0.9 3.4
0.4-2.2 1.3-8.9
706
SYLIVRIS ET AL Table 4, Technique Comparison (With Linear Epiaortic) No. of Patients With UndergradedAtheroma
Technique
Proximal Ascending Segment(p)
Mid Ascending Segment (p)
Distal Ascending Segment (p)
69/100 (<0.0001) 53/100 (<0.0001) 51/100 (<0.0001) Palpation Transesoph9/100 (NS) 50/100 (<0.0001) 50/100 (<0.0001) ageal Epiaortic (phased7/100 (NS) 5/100 (NS) 4/100 (NS) array) Abbreviation: NS, not significant; p > 0.05.
Recent studies 8,9 using epiaortic ultrasound have shown that between 60% and 90% of patients undergoing cardiac surgery have some degree of ascending aortic atheroma. In this study, 90% of cases had some degree of atheroma detected by ultrasound examination, and 28% of patients had severe disease. This was significantlygreater than the detection rate using the current method of manual palpation. The trial population was older than the subjects studied by Ohteki et al, 8 89% of whom had ascending aortic atheroma, Previous studies that have shown a lower prevalence of atheroma also included subjects with a slightly lower mean age, and this may partially explain the higher atheroma rate. Confounding of the echocardiographic interpretations by artefact is an unlikely explanation, because images were obtained in two perpendicular planes. Moreover, although pathological examinations of the aorta were not routinely performed, aortic inspection during the aortic valve replacement cases and the one postmortem case (after death secondary to a mechanical valve malfunction) confirmed the echocardiographic appearances. Blinded intrarater and interrater agreement analyses were also performed, and showed an excellent agreement between the echocardiographic interpretations. Consequently, it is believed that 90% is a true reflection of the incidence of ascending aortic atheroma in this population. Identification of patients at higher risk of severe ascending aortic atheroma should enable surgeons to pay extreme care in surgical manipulations such as cannulation and cross-clamping. It may even alert them to the need for intraoperative ultrasound in cases in which severe atheroma is suspected based on the presence of multiple risk factors. Previously, it has been shown that advanced age and diabetes are significant risk factors for aortic atheroma. 4 This study confirms that age is a significant risk factor for severe ascending aortic atheroma; however, hypertension was a significant and independentpredictor. These results support the widely held opinion that, because the elderly have more advanced atherosclerotic disease, they are at higher risk of perioperative stroke. In fact, Loop et all0 showed that the Table 5. Statistical Agreement Method
[ntrarater Agreement (Kappa)
Interrater Agreement(Kappa)
Transesophageal Linear (epiaortic) Phased-array (epiaortic)
0,91 0.97 0.90
0.94 0.95 0.92
stroke risk in cardiac surgical patients older than 75 years of age is twice that of patients younger than 65 years. When comparing the different modalities for assessment of the ascending aorta, these studies show that overall epiaortic ultrasound with either probe is the most sensitive technique. Although, subjectively, the miniaturized phased-array probe allowed greater maneuverability,there was no significant difference in atheroma grading. Analysis of different segments of the aorta showed that manual palpation was a very poor predictor of both presence and grade of atheroma throughout its entire length. Transesophageal echocardiography, which was as sensitive as epiaortic ultrasound in assessing the proximal ascending aorta, was poor at detecting and grading mid- and distalsegment disease. In fact, there were no cases in which either transesphageal echocardiology or manual palpation detected severe atheroma when the epiaortic probes failed to do this. This study is similar to that of Konstadt et a111 in that it displays that overall, transesophageal echocardiography has a low sensitivity (29% in their study) for detecting ascending aortic atheroma. However, the atheroma grading system used by their group differed from this one, and the mean age of their group was also much younger. Consequently, a comparison between these two studies must be limited. Moreover, this study specifically examines the different segments of the ascending aorta, which the Konstadt group did not do. They considered the whole ascending aorta as one section and thus were unable to comment on the fact that (as shown in this study) transesophageal echocardiography and manual palpation were equally as poor at evaluating the mid and distal ascending aortic segments when compared with epiaortic scanning. The importance of atheroma detection is in direction of surgical techniques. Currently, many surgeons usually rely on palpation to determine the safest areas to cannulate and cross-clamp. Transcranial Doppler assessments of the middle cerebral artery have also shown that, during surgery, a preponderance of microembolic events occur in association with cross-clamp manipulations.12 This would suggest that embolic events are arising from the dislodgment of atheromatous debris from the distal segment. Consequently, it seems likely that accurate detection and grading of the distal segments are very important in prevention of emboli, and this study suggests that in these distal segments manual palpation and transesophageal echocardiography are too insensitive to be used as intraoperafive guides. One limitationof this study was that only biplane transesophageal echocardiography was used, because the multiplane probe was not available. The images obtained may be slightly improved with a multiplane probe. However, because of the anatomic relationship between the distal ascending aorta and the esophagus (with the interference of the airway passages), it is very unlikely that difference in sensitivity between the transesophageal and the epiaortic approach could be breached by use of this probe. In summary, age older than 70 years and hypertension are independent and significant risk factors for severe ascending aortic atheroma. There is also a significant association between severe ascending and descending aortic atheroma. Epiaortic ultrasound, which is safe and reliable, is the best technique for
INTRAOPERATIVE ASSESSMENT OF AORTIC ATHEROMA
707
assessing this atheroma. Transesophageal echocardiography severely underestimates mid and distal ascending aortic atheroma and therefore has little value in guiding surgical manipulations such as cross-clamping. Further prospective studies are
needed, using epiaortic ultrasound (especially in patients with significant risk factors for severe atheroma) to definitively assess whether this technique actually reduces perioperative strokes.
REFERENCES
1. Mills SA, Prough DS: Cerebral complications following cardiac surgery, in Cardiac Surgery: State of the Art Reviews. Philadelphia, PA, Hanley and Belfus, 1991, p 455 2. Amarenco P, Cohen A, Tzourio C, et al: Atherosclerotic disease of the aortic arch and the risk of ischaemic stroke. N Engl J Med 331:1474-1479, 1994 3. Jones E, Kalman J, Calafiore P, et al: Proximal aortic atheroma: An independent risk factor for cerebral ischaemia. Stroke 26:218-224, 1995 4. Davila-Roman VE, Barzilai B, Wavering TH, et al: Intraoperative ultrasonographic evaluation of the ascending aorta in 100 consecutive patients undergoing cardiac surgery. Circulation 84:47-53, 1991 (suppl 5) 5. Wareing TH, Davila-Roman VE, Daily BB, et al: Strategy for the reduction of stroke incidence in cardiac surgical patients. Ann Thorac Surg 55:1400-1408, 1993 6. Wareing TH, Davila-Roman VE, Barzilai B, et al: Management of the severely atherosclerotic ascending aorta during cardiac operations: A strategy for detection and treatment. J Thorac Cardiovasc Surg 103:453-462, 1992
7. Konstadt SN, Reich DL, Quintana C, Levy M: The ascending aorta: How much do transesophageal echocardiographs see? Anaesth Analg 78:240-244, 1994 8. Ohteki H, Itoh Y, Natsuaki M, et al: Intraoperative ultrasonic imaging of the ascending aorta in ischaemic heart disease. Ann Thorac Surg 50:539-542, 1990 9. Marshall W, Barzilai B, Kouchoukos N, Saffitz J: Intraoperative ultrasonic imaging of the ascending aorta. Ann Thorac Surg 48:339344, 1989 10. Loop F, Lytle B, Cosgrove D, et al: Coronary artery bypass graft surgery in the elderly. Cleve Clin J Med 55:23-34; 1988 11. Konstadt S, Reich D, Kahn R, et al: Transesophageal echocardiography can l~e used to screen for ascending aortic atherosclerosis. Anesth Analg 81:225-228, 1995 12. Barbut D, Hint0n RB, Szatrowski TP, et al: Cerebral emboli detected during bypass surgery are associated with clamp removal. Stroke 25:2398-2402, 1994
Measurements and Main Results: Age older than 70 years and hypertension were significant risk factors for severe ascending aortic atheroma with adjusted odds ratios of 3.3 (95% CI, 1.2 to 9.3) and 3,9 (95% CI, 1.3 to 12.0), respectively. There was no significant difference in atheroma detection between the two ultrasonic epiaortic probes in any segment; however, epiaortic probes were superior to manual palpation in all segments and also superior to transesophageal echocardiography in the mid and distal segments of the ascending aorta. Conclusions: Age older than 70 years and hypertension are significant risk factors for severe ascending aortic atheroma. Intraoperative detection of ascending aortic atheroma is best achieved by epiaortic ultrasound with either a linear or phased array transducer. Transesophageal echocardiography is an insensitive technique for evaluation of mid and distal ascending aortic atheroma and, therefore, of little value in guiding surgical manipulations such as cross-clamping. Copyright © 1997 by W.B. Saunders Company
TROKES are a major cause of perioperative morbidity and mortality in cardiac surgical patients.1 Recently, numerous studies have shown that ascending aortic and aortic arch atheroma is an independent risk factor for stroke. 2,3 Currently, surgeons rely on manual palpation to determine the location and severity of ascending aortic atheroma and, consequently, the safest sites of cannulation, cross-clamping, graft insertion, and suturing. Because of the advent of newer ultrasound technology, however, transesophageal and direct epiaortic ultrasound techniques have made it possible to noninvasively image this area. Some recent studies have shown that epiaortic ultrasound is far more sensitive in atheroma detection than manual palpation, 4-6 and one study suggested that it may be more sensitive than transesophageal scanning. 7 However, no studies have compared all three modalities. One purpose of this study was to compare the ability of manual palpation, biplane transesophageal echocardiography, and epiaortic ultrasound in the detection of ascending aortic atheroma. Furthermore, a prototype epiaortic 7-MHz phasedarray transducer and a current conventional 7-MHz linear transducer were also compared. The phased array transducer, because of its miniaturized size, theoretically has greater maneuverability in the confines of a limited surgical field and
may detect atheroma when optimal images cannot be obtained by the bulkier linear transducer. The other major aim of this study was to identify the risk factors associated with severe ascending aortic atheroma. This study is unique in that no other project has compared all of these noninvasive techniques in this clinical context or used the findings to assess potential risk factors for severe ascending aortic atheroma.
S
From the Departments of Cardiology and Cardiothoracic Surgery, Austin and Repatriation Medical Centre, Heidelberg, Australia, and the Statistical Consulting Centre, University of Melbourne, Australia. Address reprint requests to Stephen Sylivris, FRACP, c/o Austin and Repatriation Medical Centre, Studley Rd, Heidelberg 3084, Victoria, Australia, Copyright © 1997 by W.B. Saunders Company 1053-0770/97/1106-000453.00/0 704
KEY WORDS: transesophageal echocardiography, epiaortic ultrasound, ascending aortic atheroma
METHODS Consecutive patients undergoing cardiac surgery under cardiopulmonary bypass at the Austin and Repatriation Medical Centre Hospital were considered for recruitment into this study. Patients were excluded if they had a contraindication to transesophageal echocardiography (esophageal varices or esophageal obstruction), if they did not provide informed consent, or if they were having redo coronary bypass grafting. All potential risk.factors were evaluated by both a patient questionnaire and from examination of prior hospital records. Six variables were prospectively identified: age older than 70 years, diabetes mellitus, any prior smoking history, hypertension (two or more prior blood pressure readings >140/90 or currently on drug treatment for hypertension), serum cholesterol >5.5 mmoi/L, and extracardiac vascular disease on clinical examination, as potential risk factors. After anesthesia and endotracheal intubation, all enrolled subjects underwent assessment of their aorta. First, a transesophageal echocardiograph was performed during graft harvesting. A multifrequency biplane transducer (Acuson Computed Sonography, Mountain View, CA) using varying frequencies (to obtain optimal short- and long-axis views) was used. Both the ascending and descending aorta were imaged. The surgeons were blinded to the results of the ultrasound examination. After exposure of the ascending aorta, the surgeons then graded any ascending aortic atheroma by manual palpation. Direct epiaortic ultrasound imaging was then performed by an echocardiologist. Both the standard 7-MHz transducer and the prototype 7-MHz phased-array transducer (with a gel standoff to enhance the near field) were used, and
Journal of Cardiothoracic and VascularAnesthesia, Vol 11, No 6 (October), 1997: pp 704-707
INTRAOPERATIVE ASSESSMENT OF AORTIC ATHEROMA
705
short- and long-axis views were again obtained. The prototype phasedarray probe is significantly smaller than the standard linear probe (5.0 × 1.5 × 2.0 cm compared with 5.5 × 2.5 × 10.0 cm) and was thought to have greater maneuverability in the confines of a limited surgical field. Consequently, it may detect atheroma when optimal images cannot be obtained by the bulkier linear transducer. All ultrasonic images were recorded on standard VHS videotape for off-line measurements and review by two independent cardiologists. The ascending aorta was divided into three equal segments between the aortic root and the innominate artery, as described by DavillaRoman et al.4 These three segments correspond to sites of differing surgical manipulations and may consequently have varying degrees of importance in embolus generation. The proximal and distal segments are the usual sites of cardioplegia and arterial cannulation, respectively, and the mid segment the site of vein graft anastamosis. Quantification of the severity of ascending aortic atheroma was performed by dividing it into three grades: Grade 0, localized thickening less than 1.0 mm thick (absent or minimal on palpation), Grade 1, atheroma 1.0 to 3.9 mm thick (moderate on palpation), and Grade 2, 4.0 or more mm thick or mobile atheroma (severe on palpation). This ultrasound grading system is the same as that used by Amarenco et al,2 who showed that patients with severe ascending atheroma have a markedly increased odds ratio of stroke. Stepwise logistic regression analysis was performed using the statistical computing package GLIM 3.77 (Royal Statistical Society, London, 1985). Sign test analyses were performed to compare the ability of the different modalities in atheroma detection. Ap value of <0.05 was considered significant. RESULTS
A total of 100 patients were enrolled, and ultrasonic scans were obtained in all. Their mean age was 69 _+ 8 years, and they included 75 men and 25 women. Eighty-three patients underwent coronary bypass surgery only, whereas 10 had aortic valve replacement and seven had combined bypass surgery and aortic valve replacement. Overall, 90% of subjects had atheroma detected by ultrasonic techniques, whereas only 16% had atheroma detected by the standard method of manual palpation. (p = 0.02, Table 1.) Severe (2+) atheroma was detected in 26 of 100 patients and involved 38 different segments of the ascending aorta. Both forms of epiaortic ultrasound (linear and phased-array imaging) were able to detect the severe grade of atheroma to an equal extent. However, manual palpation and transesophageal echocardiography were unable to detect this grade of atheroma in most cases (severe atheroma was detected in only two [p < 0.0001] and three [p < 0.0001] segments, respectively). These data are presented in Table 2. Risk factor analysis for severe (2+) ascending aortic atheroma is displayed in Table 3. Age older than 70 years and hypertension were shown to be significant risk factors. The
Table 1. Operative Procedure and Atheroma Detection Surgical Procedure
No, of Patients
CABG AVR CABG & A V R Total
83 10 7 100
Atheroma by Ultrasound n (%) 75 8 7 90
(90) (80) (100) (90)
Atheroma by Palpationn (%) 14 2 0 16
(17) (20) (0) (16)
Abbreviations: CABG, coronary artery bypass grafting; AVR, aortic valve replacement. p = 0.02.
Table 2. Detection of Severe (2+) Atheroma in 28 Subjects (38 Segments) Technique
No. of Segments
p (v Linear Array)
Palpation Transesophageal echo Phased array (epiaertic) Linear array (epiaortic)
2 3 35 36
<0,0001 <0,0001 NS
Abbreviation: NS, not significant; p > 0.05.
adjusted odds ratio for age was 3.3 (95% CI, 1.2 to 9.3); whereas for hypertension, the adjusted odds ratio was 3.9 (95% CI, 1.3 to 12.0). There was also a significant association between the presence of severe descending atheroma and severe ascending atheroma. Risk factors that were not significant included diabetes, extracardiac vascular disease, smoking history, and cholesterol >5.5 mmol/L. When analyzing the three ascending aortic segments, the epiaortic ultrasound was used with the linear probe as the baseline with which the other techniques were compared (Table 4). In the proximal segment, there was no significant difference in atheroma grading between the transesophageal echocardiograph and either form of epiaortic imaging. However, manual palpation significantly underestimated the atheroma grade in 69 of ]00 cases (p < 0.0001). In the mid and distal ascending aortic segments, there was again no significant difference between atheroma grading by either form of epiaortic ultrasound. However, in these segments, not only did manual palpation underestimate atheroma grade (in 53 of 100 cases, p < 0.0001), but transesophageal echocardiography also significantly underrated the atheroma grade (in 51 of 100 cases, p < 0.0001). Intrarater and interrater agreement were assessed by reviewing 54 and 45 segments, respectively (Table 5). There was an excellent agreement on atheroma grading, with Kappa values all greater than 0.9 with each of the ultrasound methods. DISCUSSION
These studies clearly show that patients undergoing cardiac surgery have a high incidence of coexisting ascending aortic atheroma. Moreover, the best modality to assess this atheroma, especially in the mid and distal segments of the ascending aorta, is direct epiaortic ultrasound. Table 3. Risk Factor Analysis
Risk Factor
Grade 2+ Disease (n - 26)
GradeO/+ Disease (n = 74)
Age >70 yrs 19 (73%) 34 (46%) Smoking 17 (65%) 49 (66%) Diabetes 8 (31%) 17 (23%) Hypertension 22 (85%) 34 (46%) Cholesterol >5.5 mmol/L 15/24 (63%) 32/47 (68%) Extracardiac vascular disease 9 (35%) 13 (18%) Descending aortic atheroma I+ 11 (42%) 44 (59%) 2+ 13 (50%) 18 (24%)
Odds Ratio Crude Adjusted
95% Cl
2.4 1.0 1.3 4.3
3.3 1.0 1.2 3.9
1.2-9.3 0.4-2.8 0,4-3.6 1.3-12.0
0.8
1.7
0.5-5.7
1.9
2.2
0.7-6.5
1.4 3.0
0.9 3.4
0.4-2.2 1.3-8.9
706
SYLIVRIS ET AL Table 4, Technique Comparison (With Linear Epiaortic) No. of Patients With UndergradedAtheroma
Technique
Proximal Ascending Segment(p)
Mid Ascending Segment (p)
Distal Ascending Segment (p)
69/100 (<0.0001) 53/100 (<0.0001) 51/100 (<0.0001) Palpation Transesoph9/100 (NS) 50/100 (<0.0001) 50/100 (<0.0001) ageal Epiaortic (phased7/100 (NS) 5/100 (NS) 4/100 (NS) array) Abbreviation: NS, not significant; p > 0.05.
Recent studies 8,9 using epiaortic ultrasound have shown that between 60% and 90% of patients undergoing cardiac surgery have some degree of ascending aortic atheroma. In this study, 90% of cases had some degree of atheroma detected by ultrasound examination, and 28% of patients had severe disease. This was significantlygreater than the detection rate using the current method of manual palpation. The trial population was older than the subjects studied by Ohteki et al, 8 89% of whom had ascending aortic atheroma, Previous studies that have shown a lower prevalence of atheroma also included subjects with a slightly lower mean age, and this may partially explain the higher atheroma rate. Confounding of the echocardiographic interpretations by artefact is an unlikely explanation, because images were obtained in two perpendicular planes. Moreover, although pathological examinations of the aorta were not routinely performed, aortic inspection during the aortic valve replacement cases and the one postmortem case (after death secondary to a mechanical valve malfunction) confirmed the echocardiographic appearances. Blinded intrarater and interrater agreement analyses were also performed, and showed an excellent agreement between the echocardiographic interpretations. Consequently, it is believed that 90% is a true reflection of the incidence of ascending aortic atheroma in this population. Identification of patients at higher risk of severe ascending aortic atheroma should enable surgeons to pay extreme care in surgical manipulations such as cannulation and cross-clamping. It may even alert them to the need for intraoperative ultrasound in cases in which severe atheroma is suspected based on the presence of multiple risk factors. Previously, it has been shown that advanced age and diabetes are significant risk factors for aortic atheroma. 4 This study confirms that age is a significant risk factor for severe ascending aortic atheroma; however, hypertension was a significant and independentpredictor. These results support the widely held opinion that, because the elderly have more advanced atherosclerotic disease, they are at higher risk of perioperative stroke. In fact, Loop et all0 showed that the Table 5. Statistical Agreement Method
[ntrarater Agreement (Kappa)
Interrater Agreement(Kappa)
Transesophageal Linear (epiaortic) Phased-array (epiaortic)
0,91 0.97 0.90
0.94 0.95 0.92
stroke risk in cardiac surgical patients older than 75 years of age is twice that of patients younger than 65 years. When comparing the different modalities for assessment of the ascending aorta, these studies show that overall epiaortic ultrasound with either probe is the most sensitive technique. Although, subjectively, the miniaturized phased-array probe allowed greater maneuverability,there was no significant difference in atheroma grading. Analysis of different segments of the aorta showed that manual palpation was a very poor predictor of both presence and grade of atheroma throughout its entire length. Transesophageal echocardiography, which was as sensitive as epiaortic ultrasound in assessing the proximal ascending aorta, was poor at detecting and grading mid- and distalsegment disease. In fact, there were no cases in which either transesphageal echocardiology or manual palpation detected severe atheroma when the epiaortic probes failed to do this. This study is similar to that of Konstadt et a111 in that it displays that overall, transesophageal echocardiography has a low sensitivity (29% in their study) for detecting ascending aortic atheroma. However, the atheroma grading system used by their group differed from this one, and the mean age of their group was also much younger. Consequently, a comparison between these two studies must be limited. Moreover, this study specifically examines the different segments of the ascending aorta, which the Konstadt group did not do. They considered the whole ascending aorta as one section and thus were unable to comment on the fact that (as shown in this study) transesophageal echocardiography and manual palpation were equally as poor at evaluating the mid and distal ascending aortic segments when compared with epiaortic scanning. The importance of atheroma detection is in direction of surgical techniques. Currently, many surgeons usually rely on palpation to determine the safest areas to cannulate and cross-clamp. Transcranial Doppler assessments of the middle cerebral artery have also shown that, during surgery, a preponderance of microembolic events occur in association with cross-clamp manipulations.12 This would suggest that embolic events are arising from the dislodgment of atheromatous debris from the distal segment. Consequently, it seems likely that accurate detection and grading of the distal segments are very important in prevention of emboli, and this study suggests that in these distal segments manual palpation and transesophageal echocardiography are too insensitive to be used as intraoperafive guides. One limitationof this study was that only biplane transesophageal echocardiography was used, because the multiplane probe was not available. The images obtained may be slightly improved with a multiplane probe. However, because of the anatomic relationship between the distal ascending aorta and the esophagus (with the interference of the airway passages), it is very unlikely that difference in sensitivity between the transesophageal and the epiaortic approach could be breached by use of this probe. In summary, age older than 70 years and hypertension are independent and significant risk factors for severe ascending aortic atheroma. There is also a significant association between severe ascending and descending aortic atheroma. Epiaortic ultrasound, which is safe and reliable, is the best technique for
INTRAOPERATIVE ASSESSMENT OF AORTIC ATHEROMA
707
assessing this atheroma. Transesophageal echocardiography severely underestimates mid and distal ascending aortic atheroma and therefore has little value in guiding surgical manipulations such as cross-clamping. Further prospective studies are
needed, using epiaortic ultrasound (especially in patients with significant risk factors for severe atheroma) to definitively assess whether this technique actually reduces perioperative strokes.
REFERENCES
1. Mills SA, Prough DS: Cerebral complications following cardiac surgery, in Cardiac Surgery: State of the Art Reviews. Philadelphia, PA, Hanley and Belfus, 1991, p 455 2. Amarenco P, Cohen A, Tzourio C, et al: Atherosclerotic disease of the aortic arch and the risk of ischaemic stroke. N Engl J Med 331:1474-1479, 1994 3. Jones E, Kalman J, Calafiore P, et al: Proximal aortic atheroma: An independent risk factor for cerebral ischaemia. Stroke 26:218-224, 1995 4. Davila-Roman VE, Barzilai B, Wavering TH, et al: Intraoperative ultrasonographic evaluation of the ascending aorta in 100 consecutive patients undergoing cardiac surgery. Circulation 84:47-53, 1991 (suppl 5) 5. Wareing TH, Davila-Roman VE, Daily BB, et al: Strategy for the reduction of stroke incidence in cardiac surgical patients. Ann Thorac Surg 55:1400-1408, 1993 6. Wareing TH, Davila-Roman VE, Barzilai B, et al: Management of the severely atherosclerotic ascending aorta during cardiac operations: A strategy for detection and treatment. J Thorac Cardiovasc Surg 103:453-462, 1992
7. Konstadt SN, Reich DL, Quintana C, Levy M: The ascending aorta: How much do transesophageal echocardiographs see? Anaesth Analg 78:240-244, 1994 8. Ohteki H, Itoh Y, Natsuaki M, et al: Intraoperative ultrasonic imaging of the ascending aorta in ischaemic heart disease. Ann Thorac Surg 50:539-542, 1990 9. Marshall W, Barzilai B, Kouchoukos N, Saffitz J: Intraoperative ultrasonic imaging of the ascending aorta. Ann Thorac Surg 48:339344, 1989 10. Loop F, Lytle B, Cosgrove D, et al: Coronary artery bypass graft surgery in the elderly. Cleve Clin J Med 55:23-34; 1988 11. Konstadt S, Reich D, Kahn R, et al: Transesophageal echocardiography can l~e used to screen for ascending aortic atherosclerosis. Anesth Analg 81:225-228, 1995 12. Barbut D, Hint0n RB, Szatrowski TP, et al: Cerebral emboli detected during bypass surgery are associated with clamp removal. Stroke 25:2398-2402, 1994