HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Thomas Walther Volkmar Falk Friedrich W. Mohr Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Fabricius, A. M. Articles by Mohr, F. W. Search for Related Content PubMed PubMed Citation Articles by Fabricius, A. M. Articles by Mohr, F. W. Related Collections Valve disease

Ann Thorac Surg 2004;78:575-578
© 2004 The Society of Thoracic Surgeons

---

Original article: cardiovascular

Three-dimensional echocardiography for planning of mitral valve surgery: Current applicability?

^a Division of Cardiovascular Surgery, Herzzentrum University of Leipzig, Leipzig, Germany

Accepted for publication October 3, 2003.

^* Address reprint requests to Dr Fabricius, Division of Cardiovascular Surgery, Herzzentrum Leipzig, Strümpelstr 39, 04289 Leipzig, Germany.
e-mail: faba{at}medizin.uni-leipzig.de

	Abstract

Top Abstract Introduction Material and methods Results Comment References

 
BACKGROUND: Two-dimensional transesophageal echocardiographic (2D TEE) assessment of the mitral valve requires mental integration of a limited number of 2D imaging planes. Structural display in three dimensions from any perspective may be of advantage to the surgeon for better judgment and planning.

METHODS: Feasibility, accuracy, and limitations of preoperative three-dimensional transesophageal echocardiography (3D TEE) was assessed in 51 patients with mitral valve disease. The width of the anterior mitral valve was measured with either method and compared with the operative finding. Three-dimensional dynamic sequences of the reconstructed mitral valve were shown preoperatively to the surgeon and later compared with the intraoperative finding.

RESULTS: The quality of the 3D reconstruction was graded as good in 25 patients (49.0%), fair in 16 patients (31.4%), and poor in 10 patients (19.6%) where atrial fibrillation did not allow ECG gating. Thirty-nine patients had successful mitral valve repair and twelve patients required valve replacement. Based on intraoperative findings, sensitivity for the diagnosis of mitral valve prolapse using 2D TEE and 3D TEE was 97.7% and 92.9% (p = ns) respectively and specificity was 100% by both methods. Sensitivity for the diagnosis of rupture of chordae tendineae using 2D TEE and 3D TEE was 92.3% and 30.8% respectively (p < 0.05) and specificity was 100% by both methods.

CONCLUSIONS: Dynamic 3D echocardiography is feasible and can provide good insight into valvular motion and allows adequate preoperative planning when reconstruction is being considered. However dynamic 3D reconstruction is currently limited by the quality of the original 2D echo cross sectional images which can be adversely affected by minimal patient movements, breathing, or cardiac arrhythmia, thus limiting accuracy of the 3D TEE significantly compared with 2D TEE.

	Introduction

Top Abstract Introduction Material and methods Results Comment References

 
Conventional two-dimensional (2D) echocardiography has proven to give excellent insight into complex cardiac anatomy, in particular the mitral valve [1, 2]. However a great deal of experience is required in order to mentally integrate the number of 2D planes assessed into a three-dimensional (3D) structure, followed by a good description to the surgeon thereafter. Due to the complex nature of some delicate cardiac structures, accuracy and variability of conventional 2D echocardiography are often limited to geometric assumptions rather than on anatomical reality. Moreover the complex interactions of all cardiac structures will likely never be fully grasped by 2D echocardiography.

A static or dynamic 3D presentation of cardiac structures to the surgeon from any viewpoint desired could be of great advantage in planning the operative procedure, especially for valve surgery [3–5]. However the efficiency and applicability of a new diagnostic tool such as 3D reconstruction depends on its feasibility and accuracy. The purpose of this study, therefore, was to assess feasibility and accuracy of 3D echo by comparing preoperative results to intraoperative findings.

	Material and methods

Top Abstract Introduction Material and methods Results Comment References

 
The preoperative characteristics for 51 patients requiring mitral valve procedures are depicted in Table 1. After obtaining informed consent all patients were examined preoperatively by transthoracic and transesophageal echocardiography using a conventional Hewlett-Packard SONOS 2500 (Hewlett Packard Co., Medical Products Division, Andover, MA) echocardiography system for diagnosis of mitral stenosis and mitral regurgitation.

Mitral valve regurgitation was diagnosed if an improper closure of the leaflets resulted in systolic regurgitation. Severity of mitral regurgitation was measured by pulsed, continuous wave, color flow Doppler and color M-Mode. Three-dimensional color flow Doppler was not used to quantify the regurgitation. Anatomical description of the mitral valve was done by segmental analysis of six different sections [6, 7] (A1–A3 for the anterior and P1–P3 for the posterior leaflet). The width of the anterior valve leaflet was measured three times (measurement from annulus to leaflet edge in middle scallop A2) on a 2D TEE systolic frame and on a systolic frame of a 3D reconstruction.

After complete examination of the heart a four chamber view (transducer angle of 0°) and a two chamber view (transducer angle of 65°–100°) were used to acquire the set of data required. Before commencing an acquisition a single complete rotation of the transducer plane from 0°7ndash;180° was performed to ensure visualization of the entire mitral valve apparatus. Multiple 2D echocardiograms were then registered by rotation of the ultrasonic crystal at the tip of a conventional multiplane transesophageal probe at fixed increments of 3°–180° across the cardiac cycle with electrocardiogram (ECG) and respiratory gating. Acquisition of all images was done around a central axis oriented perpendicular to the scanface thus sweeping out a conical volume encompassing the entire mitral valve. All images were then continuously displayed as a cineloop with the first acquisition at 0° and ideally being the mirror image of the last acquisition at 180°. In the case of major transducer motion during acquisition (eg, due to coughing or choking of the patient) the registered images would appear misaligned and the scan was repeated. After data collection, the 3D reconstruction process was initiated by transferring the set of digitized 2D images into a 3D image (MedCom/Fraunhofer Institute, Darmstadt, Germany). Briefly the reconstruction algorithm consists of mapping the source 2D images acquired in cylindrical coordinates into a regular Cartesian grid of voxels by interpolation of the missing information in between the single scans [8, 9]. The reconstructed 3D image is mapped onto a polyhedron on the screen and can be viewed interactively by rotating it in any direction using the screen mouse. The faces of the polyhedron act as cutting planes that can be used to slice the 3D image in any orientation thus enabling the examiner to assess structures statically or dynamically from any viewpoint desired—something impossible in conventional 2D echocardiography.

Data acquisition
Two-dimensional TEE assessment and 3D TEE reconstruction of the mitral valve was performed by one person (A.M.F.) 1 day before surgery and the time needed for reconstruction was noted. Quality of 3D reconstruction was graded on a scale from 1–3 (1 for good, 2 for fair, and 3 for poor quality).

Dynamic views of the reconstructed mitral valve were presented to the surgeon before the operation (Figs 1, 2) and later compared with the intraoperative findings. The topography of the prolapse was noted intraoperatively using the same standard classification as for 2D and 3D echocardiography and the presence or absence of ruptured chordae tendineae was noted. Before replacement or reconstruction of the mitral valve, the anterior valve was measured using sizers with known dimensions.

View larger version (127K): [in this window] [in a new window]   Fig 1. Regression analysis comparing anatomic and three-dimensional transesophageal echocardiographic measurements of anterior widths of the mitral valve (r = 0.89; 95% confidence limits; –9 and 11 mm, p < 0.05). Prolapse of A2 and P2 segment (AL = anterior leaflet; PL = posterior leaflet).

View larger version (171K): [in this window] [in a new window]   Fig 2. Left atrial view of mitral valve with prolapse P2 (AL = anterior leaflet; PL = posterior leaflet).

View larger version (25K): [in this window] [in a new window]   Fig 3. Regression analysis comparing anatomic and three-dimensional transesophageal echocardiographic (3D TEE) measurements of anterior widths of the mitral valve (r = 0.89; 95% confidence limits; –9 and 11 mm, p < 0.05).

	Results

Top Abstract Introduction Material and methods Results Comment References

 
Raw data acquisition by conventional 2D TEE for later 3D reconstruction was performed in all 51 patients. Mean time needed for reconstruction was 85 ± 25 minutes. In 19 patients (37.3%) data acquisition had to be repeated due to coughing or movement of the patient; in 10 patients (19.6%) atrial fibrillation precluded ECG gating. The quality of the 3D reconstruction was graded as good in 25 patients (49.0%), fair in 16 patients (31.4%), and poor in those 10 patients (19.6%) where atrial fibrillation had precluded ECG gating. Mitral valve stenosis was diagnosed by 2D and 3D TEE in 8 patients (15.7%), all confirmed by intraoperative visual inspection. Forty-three patients (84.3%) presented with mitral valve prolapse from either the posterior leaflet (n = 28), the anterior leaflet (n = 10), or both leaflets (n = 5). Intraoperative diagnosis of prolapse matched the preoperative 2D TEE diagnosis in 42 patients (97.6%) compared with 39 patients (90.6%) using 3D TEE (p = ns). Sensitivity for the diagnosis of mitral valve prolapse using 2D and 3D TEE was 97.7% and 92.9% respectively (p = ns) and specificity was 100% by both methods.

Rupture of one or more chordae tendinae was diagnosed by intraoperative inspection in 13 patients, of which the diagnosis had been made by preoperative 2D TEE in 12 patients (92.3%). In comparison, the preoperative diagnosis of ruptured chordae was made in only 4 patients (30.8%) using 3D reconstruction, thus significantly underestimating the number of ruptures compared with 2D TEE (p < 0.03). Specificity for the diagnosis of rupture of chordae tendinae was 100% by both methods.

The width of the anterior mitral valve was 31.2 ± 1.8 mm when measured intraoperatively compared with 31.4 ± 1.8 mm by 2D TEE and 32.0 ± 1.9 mm by 3D TEE. The 95% confidence limits of agreement for echocardiographic and intraoperative measurement were –0.48–0.42 (r = 0.94, p < 0.0001) for 2D TEE and –0.66–0.86 (r = 0.89, p < 0.0001, Fig 3) for 3D TEE (Fig 3).

	Comment

Top Abstract Introduction Material and methods Results Comment References

 
Improved anatomic representation as compared with 2D helps to eliminate variability due to operator effects which limits reproducibility of results with 2D echo. In the present study we have been able to demonstrate that 3D TEE is feasible for morphologic evaluation of the mitral valve especially of the prolapsing segment and measurement of the mitral annulus which can help plan the surgical intervention.

However there were discrepancies between the evaluation based on the 3D reconstruction and the operative finding in two patients (5%). In both cases a large but eccentric prolapse of the middle scallop of the posterior mitral valve paired with a small posteromedial segment gave the impression of a large posteromedial prolapse which was described before [4, 10]. In addition smaller anatomic structures like ruptured free-floating chordae tendinae were frequently missed by 3D TEE. We believe that accuracy was reduced when measurements were performed on the 3D reconstructed anterior mitral valve compared with measurements on the 2D image due to the interpolation of missing information in between single scans.

Reconstructions of all patients presenting with atrial arrhythmias (19%) were graded as poor. In fact a good reconstruction was nearly impossible due to major misalignment of the previously acquired single images. The misalignment could not be corrected by currently available reconstruction software thus precluding 3D TEE application in patients with atrial arrhythmias. In fact we have shown that the greater the heart rate variability the poorer the quality of the reconstruction. Three-dimensional TEE reconstruction depends critically on the quality of the original 2D sectional images which, in turn, were affected by important procedure-related limitations. Minor patient movements during coughing or choking, irregular movements of the heart during arrhythmias, or a variable respiratory pattern distort the images and result in artifacts. The influence of respiratory variability can be eliminated if acquisition is done in the operating room under general anesthesia: the patient can be preoxygenated and briefly disconnected from the respirator whereas images are acquired. We have performed this technique with promising results, however, the time required for image acquisition and high-quality 3D reconstruction is substantial.

Certain technical limitations should be kept in mind when acquiring the basic image frames. The selection of an optimal threshold level for best boundary detection affects detection of the mitral orifice and valve structure. Dropouts and shadowing due to attenuation of the 2D ultrasound beam are also present after 3D reconstruction resulting in a poorly defined endocardial boundary, particularly in patients with a calcified mitral valve annulus or spontaneous echo contrast in the left atrium. Tachycardia also decreased 3D reconstruction quality. With increasing velocity of highly mobile structures such as valve leaflets over a set time interval, the information to be recorded and processed increases resulting in lesser accuracy of the 3D reconstruction.

The time required for reconstruction is an important limiting factor when evaluating the clinical applicability of this technique. In our hands reconstructions took between 60–110 minutes to complete limiting its intraoperative application. These findings are comparable to what has been published in the literature with times ranging from 30–90 minutes [10–12].

A systematic limitation of the current study was that the measurements of the mitral annulus were performed on a static 2D image. These measurements do not completely correspond with the 3D reality of the mitral valve which is shaped like a saddle rather than like a flat oval [13]. Additional limitations in comparing these two techniques include the fact that all evaluations were performed by a single unblinded echocardiographer.

In conclusion dynamic 3D echo was helpful to describe the lesions and to quantify the mitral annulus to the surgeon. However it cannot replace current 2D TEE evaluations as the gold standard and therefore has limited applicability for mitral valve surgery. Despite these current limitations and the time needed for data acquisition and analysis we believe that 3D echocardiography has great potential to facilitate preoperative planning of mitral valve repair in case of valve prolapse. We were able to show that 3D echo is very accurate for localizing the involved portion of the prolapsing mitral valve. Perhaps with expected advances in computer technology, real-time 3D echo may become the procedure of choice for examinations of the mitral valve and better reconstruction quality will lead to more accurate measurements and descriptions. Future developments may also lead to digital "on-screen" mitral valve repair thereby enhancing preoperative planning for cardiac surgeons.

	References

Top Abstract Introduction Material and methods Results Comment References

 

1. Stewart W.J., Currie P.J., Salcedo E.E., Klein A.L., Marwick T., Agler D.A., Homer D., Cosgrove D.M. Evaluation of the mitral leaflet motion by echocardiography and jet direction by Doppler color flow mapping to determine the mechanism of mitral regurgitation. J Am Coll Cardiol 1992;20:1353-1361.[Abstract]
2. Foster G.P., Isselbacher E.T.I., Rose G.A. Multiplane transesophageal echocardiographic evaluation of mitral valve disease. Am J Card Imag 1995;9:121-128.[Medline]
3. Salustri A., Becker A.E., van Herwerden L., Vletter W.B., Ten Cate F.J., Roelandt J.R.T.C. Three-dimensional echocardiography of normal and pathologic mitral valve: a comparison with two-dimensional transesophageal echocardiography. J Am Coll Cardiol 1996;27:1502-1510.[Abstract]
4. Chauvel C., Bogino E., Clerc P., Fernandez G., Vernhet J.C., Becat A., Dehant P. Usefulness of three-dimensional echocardiography for the evaluation of mitral valve prolapse: An intraoperative study. J Heart Valve Dis 2000;9:341-349.[Medline]
5. Sutaria N., Northridge D., Masani N., Pandian N. Three dimensional echocardiography for the assessment of mitral valve disease. Heart 2000;84(suppl II):7-10.[Free Full Text]
6. Kumar N., Kumar M., Duran C.M. A revised terminology for recording surgical findings of the mitral valve. J Heart Valve Dis 1995;4:70-75.[Medline]
7. Foster G.P., Isselbacher E.M., Rose G.A., Torchiana D.F., Akins C.W., Picard M.H. Accurate localization of mitral regurgitation defects using multiplane transesophageal echocardiography. Ann Thorac Surg 1998;65:1025-1031.[Abstract/Free Full Text]
8. Fenster A, Dunne S, Chan TK, Downey D. Methods and system for constructing and displaying three dimensional ultrasound images. United States Patent 5454371 1995
9. Fenster A., Downey D. 3D Ultrasound imaging 1996: a review. IEEE Eng Med Biol Mag 1996;15:41-51.
10. Hozumi T., Yoshikawa J., Yoshida k, Akasaka T., Takagi T., Yamamuro A. Assessment of flail mitral leaflets by dynamic three-dimensional echocardiographic imaging. Am J Cardiol 1997;79:223-225.[Medline]
11. Sugeng L., Spencer K.T., Mor-Avi V., DeCara J.M., Bednarz J.E., Weinert L., Korcarz C.E., Lammertin G., Balasia B., Jayakar D., Jeevanandam V., Lang R.M. Dynamic three-dimensional color flow Doppler: an improved technique for the assessment of mitral regurgitation. Echocardiography 2003;20:265-273.[Medline]
12. Aikawa K., Sheehan F.H., Otto C.M., Coady K., Bashein G., Bolson E.L. The severity of functional mitral regurgitation depends on the shape of the mitral apparatus: a three-dimensional echo analysis. J Heart Valve Dis 2002;11:627-636.[Medline]
13. Levine R.A., Handschumaker M.A., Sanfilippo A.J., Hagege A.A., Harrigan P., Marshall J.E., Weymann A.E. Three dimensional reconstruction of the mitral valve, with implications for the diagnosis of the mitral valve prolapse. Circulation 1989;80:589-598.[Abstract/Free Full Text]

This article has been cited by other articles:

				B. Jungwirth and G. B. Mackensen Real-Time 3-Dimensional Echocardiography in the Operating Room Seminars in Cardiothoracic and Vascular Anesthesia, December 1, 2008; 12(4): 248 - 264. [Abstract] [PDF]

				T. Bharucha, K. S. Roman, R. H. Anderson, and J. J. Vettukattil Impact of Multiplanar Review of Three-Dimensional Echocardiographic Data on Management of Congenital Heart Disease Ann. Thorac. Surg., September 1, 2008; 86(3): 875 - 881. [Abstract] [Full Text] [PDF]

				T. Bharucha, S. Y. Ho, and J. J. Vettukattil Multiplanar review analysis of three-dimensional echocardiographic datasets gives new insights into the morphology of subaortic stenosis Eur J Echocardiogr, September 1, 2008; 9(5): 614 - 620. [Abstract] [Full Text] [PDF]

				S. P. Hoole, T. V. Liew, J. Boyd, F. C. Wells, and R. A. Rusk Transthoracic real-time three-dimensional echocardiography offers additional value in the assessment of mitral valve morphology and area following mitral valve repair Eur J Echocardiogr, September 1, 2008; 9(5): 625 - 630. [Abstract] [Full Text] [PDF]

				A. Kondur, S. Pitta, and L. Afonso Incremental utility of real-time three-dimensional echocardiography in the diagnosis and preoperative assessment of cleft mitral valve in adults Eur J Echocardiogr, July 1, 2008; 9(4): 586 - 588. [Abstract] [Full Text] [PDF]

				G.J. Morgan, F. Casey, B. Craig, and A. Sands Assessing ASDs prior to device closure using 3D echocardiography. Just pretty pictures or a useful clinical tool? Eur J Echocardiogr, July 1, 2008; 9(4): 478 - 482. [Abstract] [Full Text] [PDF]

				J. I. Fann, N. B. Ingels Jr., and D. C. Miller Pathophysiology of Mitral Valve Disease Card. Surg. Adult, January 1, 2008; 3(2008): 973 - 1012. [Full Text]

				L. P. Ryan, B. M. Jackson, Y. Enomoto, L. Parish, T. J. Plappert, M. G. St. John-Sutton, R. C. Gorman, and J. H. Gorman III Description of regional mitral annular nonplanarity in healthy human subjects: A novel methodology J. Thorac. Cardiovasc. Surg., September 1, 2007; 134(3): 644 - 648. [Abstract] [Full Text] [PDF]

				D. Maselli, R. De Paulis, L. Weltert, A. Salica, R. Scaffa, A. Bellisario, and A. Mingiano A new method for artificial chordae length "tuning" in mitral valve repair: Preliminary experience J. Thorac. Cardiovasc. Surg., August 1, 2007; 134(2): 454 - 459. [Abstract] [Full Text] [PDF]

				M. Pepi, G. Tamborini, A. Maltagliati, C. A. Galli, E. Sisillo, L. Salvi, M. Naliato, M. Porqueddu, A. Parolari, M. Zanobini, et al. Head-to-Head Comparison of Two- and Three-Dimensional Transthoracic and Transesophageal Echocardiography in the Localization of Mitral Valve Prolapse J. Am. Coll. Cardiol., December 19, 2006; 48(12): 2524 - 2530. [Abstract] [Full Text] [PDF]

				S. De Castro, V. Salandin, E. Cavarretta, L. Salvador, C. Valfre, S. Caselli, S. Di Michele, F. Faletra, and N. G. Pandian Epicardial Real-Time Three-Dimensional Echocardiography in Cardiac Surgery: A Preliminary Experience Ann. Thorac. Surg., December 1, 2006; 82(6): 2254 - 2259. [Abstract] [Full Text] [PDF]

				R. M. Lang, V. Mor-Avi, L. Sugeng, P. S. Nieman, and D. J. Sahn Three-Dimensional Echocardiography: The Benefits of the Additional Dimension J. Am. Coll. Cardiol., November 21, 2006; 48(10): 2053 - 2069. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Thomas Walther Volkmar Falk Friedrich W. Mohr Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Fabricius, A. M. Articles by Mohr, F. W. Search for Related Content PubMed PubMed Citation Articles by Fabricius, A. M. Articles by Mohr, F. W. Related Collections Valve disease