DOI: 10.1111/j.1540-8175.2012.01715.xC© 2012, Wiley Periodicals, Inc.

RESEARCH FROM THE UNIVERSITY OF ALABAMA AT BIRMINGHAM

Incremental Value of Live/Real TimeThree-Dimensional TransesophagealEchocardiography over the Two-DimensionalTechnique in the Assessment of Aortic Aneurysmand Dissection

Deepak Joshi, M.D.,∗ Elif Ijlal Bicer, M.D.,∗ Cevdet Donmez, M.D.,∗ Ming C. Hsiung, M.D.,†Navin C. Nanda, M.D.,∗ Kamel Sadat, M.D.,∗ Selvin Sudhakar, M.D.,∗ Hisham Ibrahim, M.D.,∗Abhilasha Pandey, M.B.B.S.,∗ Nidhi Karia, M.B.B.S.,∗ Kunal Bhagatwala, M.B.B.S.,∗Wei-Hsian Yin, M.D., Ph.D.,†‡ Jeng-Wei, M.D., D.Sc.,† Chung-Yi-Chang, M.D.,† Yi-Cheng Chung, M.D.,†Shen-Kou Tsai, M.D., Ph.D.,† Bhavin Dumaswala, M.B.B.S.,∗ and Komal Dumaswala, M.B.B.S.∗

∗Division of Cardiovascular Disease, University of Alabama at Birmingham, Birmingham, Alabama; †Divisionof Cardiovascular Disease, Heart Center, Cheng-Hsin Medical Center, Taipei, Taiwan, Republic of China; and‡Faculty of Medicine, School of Medicine, Nation Yang-Ming University, Taipei, Taiwan, Republic of China

We compared findings from intraoperative live/real time three-dimensional transesophageal echocar-diography (3DTEE) and two-dimensional transesophageal echocardiography (2DTEE) with surgery in 67patients having aortic aneurysm and/or aortic dissection. Of these, 20 patients had aortic aneurysm with-out dissection, 21 aortic aneurysm and dissection, and 26 aortic dissection without aneurysm. 3DTEEdiagnosed the type and location of aneurysm correctly in all patients unlike 2DTEE, which missed ananeurysm in one case. There were four cases of aortic aneurysm rupture. Three of them were diagnosedby 3DTEE but only one by 2DTEE, and one missed by both techniques. The mouth of saccular aneurysm,site of aortic aneurysm rupture, and communication sites between perfusing and nonperfusing lumensof aortic dissection could be viewed en face only with 3DTEE, enabling comprehensive measurementsof their area and dimensions as well as increasing the confidence level of their diagnosis. In all patientswith aortic dissection, 3DTEE enabled a more confident diagnosis of dissection because the dissectionflap when viewed en face presented as a sheet of tissue rather than a linear echo seen on 2DTEE whichcan be confused with an artifact. 2DTEE missed dissection in one patient. In six cases the dissection flapinvolved the right coronary artery orifice by 3DTEE and surgery. These were missed by 2DTEE. Aortic re-gurgitation severity was more comprehensively assessed by 3DTEE than 2DTEE. Aneurysm size by 3DTEEcorrelated well with 2DTEE and surgery/computed tomography scan. In conclusion, 3DTEE providesincremental information over 2DTEE in patients with aortic aneurysm and dissection. (Echocardiography2012;29:620-630)

Key words: live/real time three-dimensional echocardiography, live/real time three-dimensional trans-esophageal echocardiography, two-dimensional echocardiography, transesophageal echocardiogra-phy, thoracic aorta, aortic aneurysm, aortic dissection

The role of echocardiography in the evaluationof aortic aneurysm and aortic dissection is wellestablished.1,2 When compared with computedtomography (CT) and magnetic resonance imag-ing, both two-dimensional transthoracic echocar-

Deepak Joshi, Elif Ijlal Bicer, Cevdet Donmez, and Ming C.Hsiung each contributed equally to this work.

Address for correspondence and reprint requests: Navin C.Nanda, M.D., University of Alabama at Birmingham, Echo LabSW/S102, 619 19th Street South, Birmingham, AL 35249. Fax:205-934-6747; E-mail: nanda@uab.edu

diography (2DTTE) and two-dimensional trans-esophageal echocardiography (2DTEE) have beenfound to be highly accurate with a high degreeof sensitivity and specificity in the assessment ofaortic aneurysm. The accuracy of 2DTTE for aorticdissection is modest and 2DTEE is the echocardio-graphic modality of choice for its diagnosis.3–6

Echocardiography has the advantage of beingquick, relatively inexpensive, readily available,nonionizing, and portable. But 2DTTE does notvisualize the whole course of the descending tho-racic aorta and is of limited value in patients with a

620

3D TEE in Aortic Aneurysm and Dissection

poor acoustic window. 2DTEE, in which the probeis positioned in the esophagus, obviates some ofthese limitations and the higher frequency probeused provides superior quality images.7 However,a major problem with both these techniques isthe generation of thin slice-like views of cardio-vascular structures precluding comprehensive as-sessment of aortic aneurysm and aortic dissection.The relatively recent development of live/realtime three-dimensional transthoracic echocardio-graphy (3DTTE) and transesophageal echocardio-graphy (3DTEE) has resulted in the acquisition oflarge pyramidal shaped data sets, which can con-tain large segments of the aorta.8–16 These 3Ddata sets can then be cropped in a systematicmanner using any desired angulation and are ex-pected to provide a more comprehensive and ac-curate assessment of these lesions. In the presentstudy, we attempted to determine whether intra-operative 3DTEE provided any incremental valueover intraoperative 2DTEE in the assessment ofpatients with aortic aneurysm and aortic dissec-tion who underwent surgery.

Methods:Patients:Sixty-seven adult patients with thoracic aorticaneurysm and/or aortic dissection by surgery,who underwent intraoperative 2DTEE and 3DTEE,were included in the study. Patient characteristicsare outlined in Table I. Of these, 20 had aor-tic aneurysm without associated dissection, 21aortic aneurysm with concomitant aortic dissec-tion, and 26 had aortic dissection with no aorticaneurysm. Out of 41 cases of aortic aneurysms,36 were fusiform, 4 saccular, and 1 case had bothfusiform and saccular aneurysm. The aneurysms

were located in the ascending aorta in 24 cases,aortic root in 7, aortic arch in 4, descending tho-racic aorta in 3, both aortic arch and descendingaorta in 2, and both aortic root and aortic arch inthe remaining patient. Four of the patients withaortic aneurysm were found to have aneurysmrupture at surgery. Three of these (two involvingthe aortic arch and one ascending aorta) had rup-tured into the mediastinum and one (right sinusof Valsalva) into the left ventricular outflow tract(LVOT). Among 47 patients with aortic dissection,21 were DeBakey type 3, 14 DeBakey type 2, and12 were DeBakey type 1 (using Stanford classifi-cation 26 Type A and 21 Type B). At surgery, 6of 47 patients with aortic dissection were foundto have involvement of the right coronary artery(RCA) orifice by the dissection flap.

Equipment and Method of Examination:Intraoperative 2DTEE and 3DTEE were done ina standard manner.8,17 2DTEE and 3DTEE im-ages were obtained using a 5.5 MHz and a X7-2t transducer, respectively, and a Philips iE33ultrasound system (Andover, MA, USA). 3DTEEimages were acquired in full volume mode fo-cusing on the aorta and the data sets were an-alyzed online and offline using a Philips QLABsoftware version 6.0. Using a crop box, data setswere systematically dissected in different planesand angulations to get different views from mul-tiple directions, including, en face views as wellas to obtain the maximum inner size of the aor-tic aneurysm. For 2DTEE images, aortic long- andshort-axis views were used to measure the maxi-mum inner size of the aortic aneurysm. In patientswith saccular aneurysm, the area and the maxi-mum dimensions of the mouth of the aneurysm

TABLE I

Patient Characteristics

Aortic Aneurysm, Aortic Aneurysm with Aortic Dissection,Patient Category No Dissection Associated Dissection No Aneurysm

No. of patients (67) 20 21 26Age in years (range/mean) 23–77/59.2 30–88/58 31–82/55.1Sex (males/females) 14/6 17/4 17/9Aneurysm type Fusiform (17) saccular (3) Fusiform (19), saccular (1), both

saccular and fusiform (1),DeBakey type 1 (7), DeBakeytype 2 (8), DeBakey type 3 (6)

DeBakey type 1 (5),DeBakey type 2 (6),DeBakey type 3 (15)

Aneurysm location Aortic root (5), ascendingaorta (10), aortic arch (3),aortic arch and descendingthoracic aorta (2)

Aortic root (2), ascending aorta(14), aortic arch (1), aortic archand aortic root (1), descendingthoracic aorta (3)

Aneurysm size by surgery(11)/computedtomography scan (8)

Fusiform: 5 cm–10.5 cm (10)Saccular: 2 cm–4 cm (3)

5 cm–7.5 cm (6)

() = no. of patients.

621

Joshi, et al.

were also measured by 3DTEE using en faceviews.

Aortic aneurysm was diagnosed when the aor-tic size was 50% or more of the maximum widthfor that segment or if the increment of aortic sizewas 50% or more of the adjacent aortic diame-ter, as described previously.1 Saccular aneurysmwas defined as a localized dilatation of only onewall of the aorta and the fusiform type was vi-sualized as a localized dilatation of all walls ofthe aorta.1 By 2DTEE, aortic dissection was di-agnosed by visualizing a flap of consistent, ir-regular and mobile linear echo within the aorticlumen. By 3DTEE, dissection was diagnosed byvisualizing the flap en face as a sheet of tissue.9,10

In patients with aortic dissection, a communica-tion (C) between perfusing (true) and nonperfus-ing (false) lumens was noted by both 2DTEE and3DTEE, as a consistent discontinuity in the dissec-tion flap and/or detecting color Doppler flow sig-nals moving from one lumen to the other. Using3DTEE, this communication site was viewed enface and its dimensions and area measured. Aorticaneurysm rupture was noted, by both 2DTEE and3DTEE, as a consistent discontinuity in the wall of

aneurysm. Using 3DTEE, the rupture was vieweden face and its dimensions and area measured.Aortic regurgitation (AR) was diagnosed, by colorDoppler, by noting flow signals moving from theaortic valve into LVOT in diastole. AR severity as-sessment was done by measuring the proximaljet width (vena contracta [VC]) by 2DTEE colorDoppler18,19 and by 3DTEE, by viewing the venacontracta en face and measuring its area, as de-scribed previously.20 In all patients an attemptwas made by both 2DTEE and 3DTEE to assessthe coronary arteries especially the ostia and prox-imal portions of the left main and right coronaryarteries for stenosis and involvement by the dis-section process. Aortic valve morphology assess-ment and valve area measurement were done byboth 2DTEE and 3DTEE as described before.8

Findings from 2DTEE and 3DTEE examinationswere compared with surgery in all patients. In19 of 41 patients with aortic aneurysm, aorticaneurysm size by 2DTEE and 3DTEE could becompared with surgery (11 patients) and pre-operative CT scan measurements (8 patients).Surgery or CT scan measurements were not re-ported in the remaining patients.

TABLE II

Additional Live/Real Time Three-Dimensional Transesophageal Echocardiographic Findings

1. Identification of right sinus of Valsalva aneurysm (1 case). 2DTEE showed a localized rounded area in right atrium but nocommunication with sinus.

2. Detection of right sinus of Valsalva aneurysm rupture into left ventricular outflow tract (1 case) and aortic arch rupture intomediastinum (1 case). Both not detected by 2DTEE.

3. Aneurysm rupture site visualized en face resulting in comprehensive assessment of size (0.88 cm × 1.07 cm, area =0.63 cm2; 2.34 cm × 1.75 cm, area = 3.01 cm2) and more confident diagnosis of rupture (2 cases). Rupture site notvisualized en face by 2DTEE.

4. Diagnosis of ascending aortic dissection (1 case). Dissection flap not seen by 2DTEE. In retrospect, flap was immobile andin contact with posterior aortic wall.

5. En face visualization of dissection flap as a sheet of tissue increasing the diagnostic certainty of aortic dissection (all 47cases). Flap visualized only as a linear echo by 2DTEE.

6. En face visualization of communication (C) sites between perfusing and nonperfusing lumens enabling comprehensiveassessment of size (area of C ranged from 0.01 cm2 to 6.78 cm2, with a mean of 1.23 cm2) and increased confidencelevel of diagnosis of aortic dissection (31 cases). C visualized by 2DTEE in only 19 cases, also en face viewing was notpossible.

7. Because of the ability to crop in multiple planes, clear demonstration of dissection flap involving RCA orifice in 6 cases,adjacent to RCA orifice but no involvement in 3 cases, adjacent to LMCA orifice but no involvement in 2 cases and nearthe origin of an aortic arch branch without obstruction in 1 case. In one of the cases atherosclerotic plaques were seenin LMCA, with no significant stenosis. 2DTEE could not detect any of the above findings.

8. Comprehensive assessment of aortic regurgitation (AR) severity because of the ability to view AR vena contracta (VC) enface (20 cases). Of 20 cases, 13 were grade IV AR (VCA > 0.6 cm2), 3 grade III AR (VCA 0.4–0.6 cm2) and 4 grade I AR(VCA < 0.2 cm2).20 AR assessment not reliable by 2DTEE.

9. Identification of 2 sites of graft dehiscence with en face visualization of both VCs (combined areas of both VCs measured0.58 cm2) in a patient with previous Bentall surgery (1 case). 2DTEE showed turbulent flow but exact site and severity ofleak could not be assessed.

10. Right ventricular outflow tract narrowing due to extrinsic compression by thrombus in nonperfusing lumen (1 case). Notseen by 2DTEE.

11. Identification of bicuspid aortic valve (1 case). Probable tricuspid morphology by 2DTEE.12. Detection of moderate aortic valve stenosis by en face visualization and planimetry of flow limiting aortic valve orifice area

(1 case). No aortic stenosis found by 2DTEE.

2DTEE = real time two-dimensional transesophageal echocardiography; LA = area; LMCA = left main coronary artery; RCA =right coronary artery.

622

3D TEE in Aortic Aneurysm and Dissection

Figure 1.

Figure 2.

Figure 4.

Figure 3.

Figure 5.

623

Joshi, et al.

←−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−Figure 1. Real time two- and three-dimensional transesophageal echocardiographic assessment of saccular right sinus of Valsalvaaneurysm. A. Two-dimensional study. Arrowhead points to a localized bounded space adjacent to the aortic root, but its etiologyis not clear. B. Three-dimensional study. Arrowhead points to the continuity of this space with the right sinus of Valsalva resultingin a definitive diagnosis. Movie clips 1A and 1B. AO = aorta; LA = left atrium; LV = left ventricle; MV = mitral valve; PA =pulmonary artery; RA = right atrium; RV = right ventricle; TV = tricuspid valve.

←−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−Figure 2. Live/real time three-dimensional transesophageal echocardiographic assessment of aortic arch aneurysm with rupture.The top arrowhead points to the aneurysm which contains thrombus (T) and the bottom arrowhead denotes the site of ruptureof this aneurysm into the mediastinum. Movie clips 2A and 2B. In movie clip 2B, arrowheads in the right upper panel and in theleft lower panel point to en face views of the aneurysm rupture site and mouth of the aneurysm, respectively. Note the presenceof spontaneous echo contrast suggestive of low blood flow state in the aneurysm and aortic arch (ACH).

←−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−Figure 3. Live/real time three-dimensional transesophageal echocardiographic assessment of right sinus of Valsalva aneurysmrupture. Arrowhead points to the site of aneurysm (AN) rupture into the left ventricular outflow tract. D represents the dissectionflap noted in the sinus. Movie clips 3A, 3B, and 3C. In movie clip 3B, arrowhead points to the site of rupture identified bycolor Doppler flow imaging. In movie clip 3C 1, 2, and 3 represent the right, left, and noncoronary cusps of the aortic valve,respectively. Abbreviations as in previous figures.

←−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−Figure 4. Live/real time three-dimensional transesophageal echocardiographic assessment of ascending aortic aneurysm ruptureinto the mediastinum. Arrowhead points to a large rupture visualized en face measuring 1.75 cm × 2.34 cm, area 3.01 cm2.It was not possible to visualize the rupture site en face by two-dimensional echocardiography. Movie clip 4. Abbreviation as inprevious figures.

←−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−Figure 5. Real time two-and three-dimensional transesophageal echocardiographic assessment of ascending aortic dissection.A. Two-dimensional study. Arrowhead points to a linear echo in contact with the posterior aortic wall suggestive of nonspecificthickening. There is no obvious evidence for dissection. B, C. Three-dimensional study. Long and short-axis views of ascendingaorta clearly show the dissection (D) flap. Movie clips 5A and 5B, C. Abbreviations as in previous figures.

Statistical Analysis:Pearson’s correlation coefficient (r) and Bland–Altman plot21 were used to compare 2DTEE,and 3DTEE measurements of the aortic aneurysmsize with those by surgery/CT scan in 19 pa-tients. 3DTEE measurements of aortic aneurysmsize were also compared with 2DTEE measure-ments, in all 41 patients with aortic aneurysm, byusing the same statistical tests. To determine theinterobserver and intraobserver variability, mea-surements were done by a second observer after2 weeks, blinded to the first observer’s values, andagain blindly by the first observer after 6 weeksof first measurements in randomly selected 26 of41 patients with aortic aneurysm. Correlation co-efficient (r) and linear weighted kappa tests wereused to assess interrater correlation and agree-ment. Statistical analyses were performed usinga MedCalc Software version 12.0 for Windows(Mariakerke, Belgium). A P-value <0.05 was con-sidered statistically significant.

Results (Table II):Aortic Aneurysm with or without AorticDissection:All 41 aneurysms were correctly diagnosed by3DTEE but one case of saccular aneurysm of the

right sinus of Valsalva was missed by 2DTEE. Inthis patient, 2DTEE showed a localized roundedarea in the right atrium but its communicationwith the adjacent sinus could not be delineated.On the other hand, careful cropping of the 3Ddata set clearly showed its continuity with theright sinus of Valsalva resulting in a definitive di-agnosis (Fig. 1). Additionally, 3DTEE enabled enface viewing of the mouth of the aneurysm, in allthe five cases of saccular aneurysms, facilitatingaccurate measurements of aneurysm dimensionsand area.

Three of the four cases of aortic aneurysmrupture, confirmed by surgery, were detected by3DTEE as compared to only one by 2DTEE. 2DTEEmissed aortic arch aneurysm rupture into the me-diastinum and right sinus of Valsalva aneurysmrupture into LVOT. In both cases, 3DTEE clearlyshowed the rupture by both B-mode and colorDoppler flow mapping (Figs. 2 and 3). Both3DTEE and 2DTEE missed one case of aortic archaneurysm rupture. En face views obtained by3DTEE enabled measurement of the area of therupture site in all three cases which was not possi-ble by the single slice technique of 2DTEE (Fig. 4).Since en face viewing by 3DTEE visualized theentire margin of the rupture site, this finding in-creased the confidence level of the diagnosis of

624

3D TEE in Aortic Aneurysm and Dissection

Figure 5.

Figure 6.

625

Joshi, et al.

Figure 7. Live/real time three-dimensional transesophagealechocardiographic assessment of ascending aortic dissection.Arrowhead shows a mobile dissection flap (D) protruding intothe right coronary artery (RCA) lumen. Movie clips 7A and7B. In movie clip 7B arrowhead shows the chaotic motion ofthe dissection flap in the proximal RCA lumen viewed in shortaxis. Abbreviations as in previous figures.

aortic aneurysm rupture by differentiating it froman artifactual dropout.

There were three patients diagnosed with bi-cuspid aortic valve morphology at surgery. Enface viewing by 3DTEE accurately identified thismorphology in all three cases while 2DTEE sug-gested a tricuspid morphology in one case with asurgically proven bicuspid aortic valve. Concomi-tant aortic valve stenosis (AS) was detected, atsurgery, in three patients. By 3DTEE, assessmentof the severity of AS, by planimetry of the flow lim-iting orifice area at the valve tip showed severe ASin two patients and moderate AS in the remain-ing patient. 2DTEE planimetry showed a normalaortic valve area in the patient with moderate aor-tic stenosis by 3DTEE. 2DTEE findings correlatedwell with 3DTEE in the remaining two patientswith AS. In one patient with a previous Bentalloperation both 2DTEE and 3DTEE detected leak-age from the proximal anastomosis. However,2DTEE color Doppler could not detect its exactsite or severity. In contrast, cropping of the colorDoppler 3D data set clearly showed two adjacentareas of graft dehiscence. En face views demon-strated the combined areas of the two vena con-tractas as 0.58 cm2 consistent with very signifi-cant leakage.

Figure 8. Live/real time three-dimensional transesophagealechocardiographic assessment of ascending aortic dissection.Arrowheads point to two atherosclerotic plaques in the leftmain coronary artery. The dissection flap is adjacent to itsorifice. Movie clip 8. Abbreviations as in previous figures.

Aortic Dissection with or without AorticAneurysm:Out of 47 patients with aortic dissection, 3DTEEdetected additional findings in 40 patients whencompared with 2DTEE. 3DTEE correctly diag-nosed dissection in all 47 patients whereas 2DTEEmissed aortic dissection in one patient. In thiscase of ascending aortic dissection, by 2DTEE,the posterior wall of the aorta was noted to bethickened but no obvious dissection flap wasvisualized. Cropped 3DTEE images clearly showeda dissection flap within the lumen resulting in adefinitive diagnosis (Fig. 5). 3DTEE provided amore definitive diagnosis of aortic dissection ascompared to 2DTEE, since the dissection flapcould be seen as a sheet of tissue rather than asa linear echo on 2DTEE, which could be mistakenfor an artifact and vice versa.9,10 Furthermore, enface viewing with 3DTEE visualized the entire mar-gin of the communication sites (C) between theperfusing and the nonperfusing lumens (Fig. 6).This further increased the confidence level of thediagnosis of aortic dissection in 31 of 47 patientsin whom C could be detected by 3DTEE usingboth B-mode and color Doppler flow imaging.By 2DTEE C was detected in only 19 patients.Also in all 31 cases, en face views obtained with

←−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−Figure 6. Live/real time three-dimensional transesophageal echocardiographic assessment of aortic arch and descending thoracicaortic dissection. A–D. Arrowhead in A points to a large communication (C) between the nonperfusing (NPL) and perfusing lumens(PL) at the level of the isthmus. Unlike the two-dimensional study, the communication could be viewed en face from both thenonperfusing (B) and perfusing (C) aspects using the three-dimensional approach. The C is roughly circular, measuring 1.99 cm× 2.14 cm, area 3.15 cm2. In D, the dissection (D) flap is seen at the origin of one of the aortic arch branches. This finding couldnot be detected with the two-dimensional study. Movie clips 6A, 6B, 6C, and 6D. BR = aortic arch branch.

626

3D TEE in Aortic Aneurysm and Dissection

Figure 9.

Figure 10.

627

Joshi, et al.

←−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−Figure 9. Bland–Altman plots demonstrating agreement between A. two-dimensional transesophageal echocardiography(2DTEE) and surgery/computed tomography (CT) scan, B. live/real time three-dimensional transesophageal echocardiogra-phy(3DTEE) and surgery/CT scan, and C. 2DTEE and 3DTEE. Solid line represents mean of difference between the two techniquesand dashed lines 95% limits of agreement.

←−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−Figure 10. Correlation between A. two-dimensional transesophageal echocardiography (2DTEE) and surgery/computed tomog-raphy (CT) scan, B. live/real time three-dimensional transesophageal echocardiography (3DTEE) and surgery/CT scan, and C.2DTEE and 3DTEE.

3DTEE facilitated accurate measurements of thedimensions and the area of C, which could notbe achieved with 2DTEE.

Systematic cropping and en face viewing by3DTEE showed the dissection flap in close proxim-ity or involving the orifices of RCA, left main coro-nary artery (LMCA) or an aortic arch branch in 12of 47 patients with aortic dissection. These werenot detected by 2DTEE. In 6 cases the dissectionflap was noted to involve the RCA orifice by 3DTEEand this was confirmed at surgery (Fig. 7). In theremaining patients, the RCA was not involved bythe dissection process at surgery or by 3DTEE. Inthree cases, the dissection flap was seen by 3DTEEin close proximity to the RCA orifice but there wasno involvement of the orifice. Also, with 3DTEEthe dissection flap was seen adjacent to but notinvolving the LMCA orifice in two other cases. Inone of these two cases, atherosclerotic plaquesproducing no significant obstruction were seenin the proximal LMCA lumen with 3DTEE but notby 2DTEE (Fig. 8). In none of the patients studiedby us the dissection involved the LMCA by 3DTEEor surgery. In one case the dissection flap wasnoted near the origin of an aortic branch withoutinvolvement of the vessel lumen by 3DTEE andconfirmed by surgery (Fig. 6D). Right ventricularoutflow tract narrowing due to compression by athrombus in the nonperfusing lumen was clearlydelineated by 3DTEE in one patient. This findingwas not evident with the 2DTEE study.

Aortic Regurgitation:AR was noted by both 2DTEE and 3DTEE in 20out of 67 cases (10 of 20 patients with aorticaneurysm and no dissection, 5 of 21 patients withaortic aneurysm and associated dissection, and 5of 26 patients with aortic dissection without con-comitant aortic aneurysm). 3DTEE provided su-perior assessment of the AR VC in all of the casesof AR, since it could be accurately visualized enface and its area planimetered. Of 20 cases, 13were grade IV AR (VCA >0.6 cm2), 3 grade III AR(VCA 0.4–0.6 cm2), and 4 grade I AR (VCA<0.2cm2).20 On the other hand, 2DTEE provided onlyone or two dimensions of the VC and in somecases where the short-axis views were obtained,it was difficult to determine whether the imag-

ing plane was exactly at the level of the VC andparallel to it.

Aortic Aneurysm Size:Comparison of aneurysm size by both 2DTEEand 3DTEE showed excellent correlation withsurgery/CT scan measurements (Figs. 9 and 10).Also, there was no significant difference between2DTEE and 3DTEE measurements of aneurysmsize and thus 3DTEE provided no incrementalvalue in this regard. There was no significant inter-and intraobserver variability in the measurementof aneurysm size by both 2DTEE (r = 0.99 and0.99, Kappa values = 0.89 and 0.9, respectively)and 3DTEE (r = 0.99 and 0.99, Kappa values =0.81 and 0.84, respectively).

Discussion:Our study clearly demonstrates the adjunctivevalue provided by 3DTEE over 2DTEE in theassessment of aortic aneurysm and dissection(Table II). The biggest advantage of 3DTEE over2DTEE was the ability to generate views from mul-tiple planes and angulations including en faceviews. 3DTEE was able to correctly diagnose thetype and site of all aneurysms in contrast to 2DTEEwhich missed an aneurysm of the right sinus ofValsalva. It is possible that with more meticulousexamination, 2DTEE might have detected it but itwas missed in the present study and only seen by3DTEE. We also found 3DTEE superior to 2DTEEin the diagnosis of aortic aneurysm rupture inthree cases because the site of rupture could beviewed in multiple planes including en face ori-entation. This enabled 3D assessment of the sizeof the rupture as well as contributed to increasedconfidence level in the diagnosis. However, both2DTEE and 3DTEE missed one ruptured aneurysmmost likely because the rupture site was coveredby thrombus. 2DTEE, on the other hand couldnot detect rupture in two cases. Another advan-tage of 3DTEE over 2DTEE was the more accu-rate assessment of aortic valve morphology andorifice area because the 3DTEE plane could besystematically maneuvered to evaluate the aorticvalve at different levels and also could be posi-tioned exactly parallel to the flow limiting tip of

628

3D TEE in Aortic Aneurysm and Dissection

the stenosed aortic valve. In cases of aortic dis-section, 3DTEE increased the diagnostic certaintyof dissection because the dissection flap could beviewed as a sheet of tissue and not merely as alinear echo seen on 2DTEE which may representan artifact and not dissection. This is of great clin-ical importance since occasionally patients haveundergone surgery on the basis of finding a lin-ear echo in the aorta and no dissection was foundby the surgeon. In one patient 2DTEE failed to di-agnose dissection which was easily detected by3DTEE. In retrospect, this probably resulted froma nonmobile dissection flap located very close tothe aortic wall giving an appearance of a thick-ened wall. In addition, 3DTEE was superior in de-termining the relationship of the dissection flapto the coronary artery orifices and their involve-ment in the dissection process. All six patientswith coronary involvement at surgery were cor-rectly identified by 3DTEE and none by 2DTEE.Involvement of the coronary arteries by the dis-section process is not uncommon and has impor-tant clinical implications. Mechanical obstructionof flow to the coronary arteries by the dissectionflap or extension of dissection along the wall ofthe coronary arteries producing lumen narrow-ing can lead to myocardial ischemia and eveninfarction.9 This may result in the diagnosis ofan acute coronary syndrome and the dissectionwhich led to this overlooked or missed with po-tentially disastrous consequences. Thus it is mostimportant to correctly diagnose coronary arteryinvolvement by the dissection process and for this3DTEE can provide invaluable incremental benefitover 2DTEE.

In our study, the size of aortic aneurysm by2DTEE and 3DTEE correlated very well with sur-gical/CT scan findings and hence, 3DTEE did notprovide any incremental value over 2DTEE in thisaspect. Conceptually, 3DTEE would be expectedto be superior to 2DTEE in the measurement ofaortic aneurysm size because the cropping planecan be adjusted exactly parallel to the aortic short-axis to obtain the maximum dimensions of theaneurysm accurately. On the other hand, each2DTEE view represents only a thin slice throughthe aorta and it is often difficult to determine if theslice was exactly parallel or oblique to the shortaxis of the aneurysm. This potential advantage of3DTEE over 2DTEE may not have been obvious inour study because of the relatively small numberof patients in whom measurements were availableat surgery or by CT scan.

Upon reviewing the literature, we found sev-eral case reports22–26 but few detailed studies9–11

demonstrating the role of 3DTTE or 3DTEE in theevaluation of aortic dissection or aortic aneurysm.These studies and case reports point to the use-fulness of 3D technology in accurately diagnosing

aortic dissection, in improving assessment of rela-tionship of the dissection flap with the surround-ing structures, and in more accurate quantifica-tion of the communication sites between perfus-ing and nonperfusing lumens. We could not findany study similar to ours in the literature whichcompared 3DTEE with 2DTEE in the assessmentof aortic aneurysm.

In conclusion, in our preliminary study3DTEE provided important incremental valueover 2DTEE in the assessment of aortic aneurysmand dissection. However, it did not provide anyadded benefit in the measurement of aneurysmsize. A much larger number of patients need to bestudied to define the exact role of 3DTEE in theevaluation of patients with aortic aneurysm anddissection. Also, the role of 3DTTE, which is com-pletely noninvasive, needs to be comprehensivelystudied.10

References1. Mathew T, Nanda NC: Two-dimensional and Doppler

echocardiographic evaluation of aortic aneurysm and dis-section. Am J Cardiol 1984;54:379–385.

2. Goldstein SA, Lindsay J: Thoracic aortic aneurysms: Role ofechocardiography. Echocardiography 1996;13:213–232.

3. Pemberton J, Sahn DJ: Imaging of the aorta. Int J Cardiol2004;97:(Suppl 1):53–60.

4. Moore AG, Eagle KA, Bruckman D, et al: Choice of com-puted tomography, transesophageal echocardiography,magnetic resonance imaging, and aortography in acuteaortic dissection: International Registry of Acute AorticDissection (IRAD). Am J Cardiol 2002;15:1235–1238.

5. Kabirdas D, Scridon C, Brenes JC, et al: Accuracy oftransthoracic echocardiography for the measurement ofthe ascending aorta: Comparison with transesophagealechocardiography. Clin Cardiol 2010;33(8):502–507.

6. Erbel R, Alfonso F, Boileau C, et al: Task Force on Aor-tic Dissection, European Society of Cardiology. Diagno-sis and management of aortic dissection. Eur Heart J2001;22(18):1642–1681.

7. Ballal RS, Nanda NC, Gatewood R, et al: Usefulness oftransesophageal echocardiography in assessment of aor-tic dissection. Circulation 1991;84:1903–1914.

8. Pothineni KR, Inamdar V, Miller AP, et al: Ini-tial experience with live/real time three-dimensionaltransesophageal echocardiography. Echocardiography2007;24:1099–1104.

9. Nanda NC, Khatri GK, Samal AK, et al: Three-dimensional echocardiographic assessment of aortic dis-section. Echocardiography 1998;15:745–754.

10. Htay T, Nanda NC, Agrawal G, et al: Live three-dimensional transthoracic echocardiographic assessmentof aortic dissection. Echocardiography 2003;20:573–577.

11. Evangelista A, Aguilar R, Cuellar H, et al: Usefulness of real-time three-dimensional transoesophageal echocardiogra-phy in the assessment of chronic aortic dissection. Eur JEchocardiogr 2011;12:272–277.

12. Hansalia S, Biswas M, Dutta R, et al: The value of live/realtime three-dimensional transesophageal echocardiogra-phy in the assessment of valvular vegetations. Echocardio-graphy 2009; 26:1264–1273.

13. Singh P, Manda J, Hsiung MC, et al: Live/real time three-dimensional transesophageal echocardiographic evalua-tion of mitral and aortic valve prosthetic paravalvular re-gurgitation. Echocardiography 2009; 26: 980–987.

629

Joshi, et al.

14. Mohammed Abd El-Rahman S, Khatri G, Nanda N,et al: Transesophageal three-dimensional echocardio-graphic assessment of normal and stenosed coronary ar-teries. Echocardiography 1996;13:503–510.

15. Nanda NC, Ansingkar K, Espinal M, et al: Transesophagealthree-dimensional echo assessment of sinus venosus atrialseptal defect. Echocardiography 1999;16:835–837.

16. Raslan S, Nanda NC, Lloyd L, et al: Incrementalvalue of live/real time three-dimensional transesophagealechocardiography over the two-dimensional techniquein the assessment of sinus of Valsalva aneurysm ruptureechocardiography. 2011;28:918–920.

17. Nanda NC, Domanski MJ. Atlas of TransesophagealEchocardiography, 2nd Ed. Philadelphia, PA: Lippincott,Williams and Wilkins, 2007.

18. Perry GJ, Helmcke F, Nanda NC, et al: Evaluation of aorticinsufficiency by Doppler color flow mapping. J Am CollCardiol 1987;9:952–959.

19. Willett DL, Hall SA, Jessen ME, et al: Assessment of aorticregurgitation by transesophageal color Doppler imagingof the vena contracta: Validation against an intraoperativeaortic flow probe. J Am Coll Cardiol 2001;37:1450–1455.

20. Fang L, Hsiung MC, Miller AP, et al: Assessment ofaortic regurgitation by live three-dimensional transtho-racic echocardiographic measurements of vena con-tracta area: Usefulness and validation. Echocardiography2005;22:775–781.

21. Bland JM, Altman DG: Statistical methods for assessingagreement between two methods of clinical measure-ment. Lancet 1986;1:307–310.

22. Scohy TV, Geniets B, McGhie J, et al: Feasibility of real-time three-dimensional transesophageal echocardiogra-phy in type A aortic dissection. Interact Cardiovasc ThoracSurg 2010;11(1):112–113.

23. Krapf L, Vahanian A, Messika-Zeitoun D: Three-dimensional transoesophageal echocardiography in aor-tic dissection. Heart 2009;95(1):55.

24. Nemes A, McGhie JS, ten Cate FJ: Real-time 3-dimensionalechocardiographic evaluation of aortic dissection. J AmSoc Echocardiogr 2006;19(1):108.e1–108.e3.

25. Hansalia S, Nanda NC, Bandarupalli N, et al: Live/realtime three-dimensional transthoracic echocardiographicassessment of aortic dissection rupture into right ventric-ular outflow tract: A case report and review of the litera-ture. Echocardiography 2009;26(1):100–106.

26. Yelamanchili P, Nanda NC, Patel V, et al: Definitive di-agnosis of descending thoracic aortic dissection by realtime/live three-dimensional transthoracic echocardiogra-phy. Echocardiography 2006; 23(2):158–161.

Supporting InformationAdditional Supporting Information may be foundin the online version of this article:

Movie clips 1A, 1B, 2A, 2B, 3A, 3B, 3C, 4,5A, 5B,C, 6A, 6B, 6C, 6D, 7A, 7B, 8.

Please note: Wiley-Blackwell is not responsiblefor the content or functionality of any support-ing materials supplied by the authors. Any queries(other than missing material) should be directedto the corresponding author for the article.

630