- Email: [email protected]
3D color-Doppler echocardiography and chronic aortic regurgitation: A novel approach for severity assessment
International Journal of Cardiology 166 (2013) 640–645
Contents lists available at SciVerse ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
3D color-Doppler echocardiography and chronic aortic regurgitation: A novel approach for severity assessment ☆ Leopoldo Perez de Isla ⁎, Jose Zamorano, Covadonga Fernandez-Golfin, Sara Ciocarelli, Cecilia Corros, Tibisai Sanchez, Joaquín Ferreirós, Pedro Marcos-Alberca, Carlos Almeria, Jose Luis Rodrigo, Carlos Macaya Unidad de Imagen Cardiovascular, Hospital Clínico San Carlos, Madrid, Spain
a r t i c l e
i n f o
Article history: Received 8 April 2010 Received in revised form 20 July 2010 Accepted 26 November 2011 Available online 20 December 2011 Keywords: Doppler-echocardiography Three-dimensional Aortic regurgitation
a b s t r a c t Background: 3D echocardiography provides a complete evaluation of the aortic valve and adjacent structures and it improves the assessment of this cardiac region. Three-dimensional color-Doppler echocardiography (3DCDE) evaluation might improve the measurements of the functional regurgitant orifice in patients with Chronic Aortic Regurgitation (CAR). Objectives: Our aim was to compare the accuracy of current echo-Doppler methods and 3DCDE for the assessment of CAR severity. The reference method used in this work was the CAR severity determined by means of cardiac magnetic resonance (CMR) Methods: Thirty-two consecutive patients with an established diagnosis of CAR recruited in our institution comprised our study group. CAR severity was determined by conventional Echo-Doppler methods and by 3DCDE and their results were compared with those obtained by means of CMR. Results: Mean age was 63.0 ± 13.5 years. Twenty-two patients (68.8%) were men. Compared with the traditional echo-Doppler methods, 3DCDE evaluation had the best linear association with CMR results (3D vena contracta cross sectional area method: r = 0.88; r square = 0.77; p b 0.001. 3D vena contracta cross sectional area/left ventricular outflow tract cross sectional area method: r = 0.87; r square = 0.75; p b 0.001). The ROC analysis showed an excellent area under curve for detection of severe CAR (3D vena contracta cross sectional area method = 0.97; 3D vena contracta cross sectional area/left ventricular outflow tract cross sectional area method = 0.98). Inter- and intra-observer variability for the 3DCDE evaluation was good (ICC = 0.89 and ICC = 0.91 for inter and intra observer variability respectively). Conclusions: 3DCDE is an accurate and highly reproducible diagnostic tool for estimating CAR severity. Compared with the traditional echo-Doppler methods, 3DCDE has the best agreement with the CMR determined CAR severity. Thus, 3DCDE is a diagnostic method that may improve the therapeutic management of patients with CAR. © 2011 Elsevier Ireland Ltd. All rights reserved.
1. Introduction and objectives The therapeutic management of patients with chronic aortic regurgitation (CAR) represents a challenge for the clinician [1,2]. This entity is responsible of quality of life impairment and life expectancy reduction in a huge number of patients. To define the best therapeutic strategy, clinical data and accurate measurements of the CAR severity
Abbreviations: 2D, Two Dimensional Echocardiography; 3DCDE, Three Dimensional Color-Doppler Echocardiography; ARFr, Aortic Regurgitant Fraction; CAR, Chronic Aortic Regurgitation; CMR, Cardiac Magnetic Resonance; EDV, End Diastolic Volume; ESV, End Systolic Volume; ICC, Intra-class Correlation Coefficient; LVOT, left ventricular outflow tract. ☆ There is no conflict of interest concerning this manuscript. ⁎ Corresponding author at: Unidad de Imagen Cardiovascular, Hospital Clínico San Carlos, Plaza Cristo Rey, 28040-Madrid, Spain. Tel./fax: + 34 913303290. E-mail address: [email protected] (L. Perez de Isla). 0167-5273/$ – see front matter © 2011 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ijcard.2011.11.094
are needed. Nevertheless, currently employed methods are not free from limitations [3]. Three-dimensional echocardiography provides a unique “en-face” view of the complete aortic, sub-aortic and aortic root structures and could therefore improve the assessment of this cardiac region. Furthermore, three-dimensional color-Doppler evaluation might improve the measurements of the functional regurgitant orifice [4]. However, it has not been routinely used due to the prolonged data acquisition and off-line processing time. With the new transthoracic 3D matrix array probe (Philips, Bothell, Washington) that allows 3D color-Doppler-Echocardiography (3DCDE) rendering and the new analysis software, this technique could be included in the routine assessment of this valvular heart disease. Our aim was to compare the accuracy of traditional echo-Doppler methods and 3DCDE for the assessment of CAR severity. The reference method used in this work was the CAR severity determined by means of cardiac magnetic resonance (CMR) [5,6].
L. Perez de Isla et al. / International Journal of Cardiology 166 (2013) 640–645 2. Methods 2.1. Patient population Thirty-two consecutive patients with an established diagnosis of CAR [1], recruited in our institution, comprised our study group. Exclusion criteria were more than mild valvular heart disease different of aortic regurgitation (mild valvular lesions were not considered as having a relevant impact on the study results), congenital heart disease, the presence of cardiomyopathies and patient unwillingness to be included in the study. Every patient was given verbal and written information about the techniques they would undergo. They were also informed and gave consent for anonymous use of their data for research purposes.
2.2. Conventional echo evaluation A complete conventional Echo-Doppler study was performed in all patients using a IE-33 ultrasound machine and a S 5–1 probe (Philips, Bothell, Washington). M-mode echo, 2D echo, pulsed and continuous spectral Doppler and color-Doppler recordings were obtained from standard paraesternal and apical windows and measurements of the different structures were obtained, following the current recommendations [3]. Three cardiac cycles for patients in sinus rhythm and five for patients in atrial fibrillation were recorded and their results averaged for every patient. The presence of aortic regurgitation was defined by means of echo-Doppler and its severity was assessed by the current recommendations [3]. Left ventricular volumes and ejection fraction were evaluated by means of Simpson method. Linear dimensions were obtained by using M mode echocardiography. Mitral e and a wave peak velocities were obtained by measuring velocities with pulsed wave Doppler at the level of the tips of the mitral valve. Pulmonary artery systolic pressure was derived from the tricuspid systolic gradient and the estimated right atrial pressure. Jet width and vena contracta were obtained following the current recommendations with M-mode color-Doppler. Regurgitant volume and ejection fraction was obtained by means of the volumetric method as described elsewhere [3].
2.3. 3DCDE evaluation 3DCDE was performed immediately after the 2D study. Data were recorded using an IE33 ultrasound machine and an X 3-1 probe (Philips, Bothell, Washington). Settings were optimized for 3D resolution. The full-volume method provides a pyramidal data set of approximately 90°×90° and allows inclusion of a larger cardiac volume, although it requires ECG gating, because this technique compiles 4 to 7 pyramidal scans obtained during 4 consecutive heartbeats. The full volume method was used for left ventricular volume and systolic function analysis. 3D color-Doppler was employed for the CAR vena contracta measurement (Fig. 1). The capture was performed by placing the region of interest in the aortic valve region from an apical view. Settings were optimized for color-Doppler resolution. The protocol used for patients in atrial fibrillation was the same as for patients in sinus rhythm because the software used is able to improve the acquisition in patients with arrhythmias. The software automatically rejects the RR intervals that are very different from the mean RR interval.
641
All images were sent to the central DICOM server for off-line analysis using Q-Lab® software (Philips, Bothell, Washington). 3DCDE LVOT and aortic regurgitation vena contracta measurements were performed “en-face” at the ideal cross section, just immediately below the aortic valvular plane in mid-diastole (Figs. 2 and 3). In case of two or more regurgitant orifices, cross sectional areas were independently evaluated and their areas were added. Left ventricular volumes and ejection fraction were evaluated using the 3DQ-Advance software, integrated in the Q-lab software platform. 2.4. Cardiac magnetic resonance study protocol CMR were performed within 15 days of the echocardiographic study. No clinical event appeared within this period. Furthermore, no change in treatment was done. All CMR exams were performed at 1.5 T Signa EXCITE 11.0 (General Electrics, Milwaukee, WI) whole body scanner with a dedicated cardiac 8 channel surface coil. After real time initial localizing scans, breath-hold, electrocardiographically triggered steady state free precession cine images (echo time/repetition time 1.74/3.92, flip angle 45°) were acquired in four chamber long axis view and short axis view from base to apex (8 mm thickness, gap 2 mm). Quantitative measures of aortic flow and aortic regurgitation volume and fraction were assessed using breath-hold electrocardiographically triggered phase contrast velocity sequence with the following parameters: TR 12 ms, TE= 4.4 ms, flip angle = 17°, FOV= 320 × 240, matrix = 256 × 128 and slice thickness 6 mm. Velocity encoding (VENC) was initially set in 200 cm/s with further 30 cm/s increments in case of aliasing. The plane for aortic flow measurements was localized at the level of the aortic root parallel and just above the aortic valve (Fig. 4). CRM image analyses were performed using a commercial work station (ReportCard, Version 2.0, General Electrics, Milwaukee, WI). Endocardial LV borders were manually traced at end diastole and at end-systole. Papillary muscles were included as part of LV cavity volume. Left ventricular end-diastolic volume (EDV) and endsystolic volume (ESV) were determined using a Simpson's rule method. Stroke volume was calculated as the difference between EDV and ESV. Ejection fraction was calculated as EF = (EDV − ESV)/EDV × 100.Aortic regurgitation volume was directly calculated from the aortic flow curve by integrating diastolic reverse flow. Aortic regurgitation volume was calculated from the aortic flow curve by tracing the diastolic reverse flow. Aortic regurgitation fraction (ARFr) was calculated as the ratio of aortic regurgitant volume and aortic forward flow. Aortic regurgitation severity was graded as follows: mild ARFr b 15%, moderate ARFr 15–30% and severe ARFr > 30% (Fig. 5). 2.5. Inter- and intra-observer variability 3DCDE datasets of the CAR vena contracta of the first 15 enrolled patients were analyzed offline at different times by two independent observers. The same images were also analyzed on a different day by one of these observers. 2.6. Statistical analysis The statistical package used was SPSS version 15.0. Quantitative data were expressed as mean ± standard deviation. Qualitative data were expressed as absolute number (percentage). Inter- and intra-observer reproducibility was evaluated by means of
Fig. 1. Color-Doppler echocardiography images of a patient with chronic aortic regurgitation.
642
L. Perez de Isla et al. / International Journal of Cardiology 166 (2013) 640–645
Fig. 2. Aortic regurgitation vena contracta planes orientation in a 3D color-Doppler echocardiography dataset. the Intra-class Correlation Coefficient (ICC). Inter-methods linear association was assessed by using linear regression analysis. Comparisons were considered significant in presence of a p value b 0.05.
3. Results Thirty-two consecutive non-selected patients with CAR were prospectively enrolled. They were the analyzed group. Before ending the inclusion period, five patients were not enrolled due to the unwillingness to undergo a CMR study and one patient was not included due to the presence of an inadequate acoustic window. There were 22 men (68.8%) and mean age was 63.0 ± 13.5 years. CAR was the predominant valvular lesion in all of them but concomitant mild mitral regurgitation was present in 18 patients (56.3%). Aortic stenosis and mitral stenosis were not present in any patient. Thirty patients
(93.7%) were in normal sinus rhythm and 2 (6.3%) in atrial fibrillation. Main descriptive variables are depicted in Table 1. The value of the regurgitant fraction of the CAR obtained by using CMR divided the severity of the studied population into: 14 patients with mild CAR (43.8%), 11 patients with moderate CAR (34.4%) and 7 patients with severe CAR (21.9%). The aetiology of the aortic regurgitation was degenerative (associated to age) in every patient. The frame rate in the 3D studies was between 12 and 16 volumes per second. And in the 2D studies between 20 and 30 frames per second. 3.1. Comparison of classic Echo-Doppler methods, 3DCDE and CMR CAR severity was evaluated by using the previously described methods [3]. Mean results for each studied variable are shown in Tables 2 and 3. Linear regression analysis was used to establish the
Fig. 3. Aortic regurgitation vena contracta 3D echo-Doppler quantification in a 3D color-Doppler echocardiography dataset.
L. Perez de Isla et al. / International Journal of Cardiology 166 (2013) 640–645
643
Fig. 4. Phase-contrast velocity sequence for aortic regurgitation volume quantification at the level of the aortic root, just above the valvular plane. Diastolic magnitude and phase (left and right respectively) images with aortic regurgitation central jet clearly seen.
linear association between each method and the values obtained with CMR. Results are described in Table 3. Time required to obtain and analyze the 3DCDE images, evaluated in all the enrolled patients was 16.2 ± 2.6 minutes. Main results for the evaluation of aortic regurgitation vena contracta cross sectional area are shown in Table 3. Linear regression analysis results between 3DCDE and the values obtained with CMR are described in Table 4. It is of note that the best linear association is obtained with the 3Decho-Doppler method (see Table 4).
3.2. Diagnostic accuracy of 3DCDE to detect severe CAR In order to evaluate the diagnostic accuracy of conventional echoDoppler methods and 3DCDE to detect severe CAR, a ROC curves analysis was performed. Results for every method are depicted in Table 5 and detailed results for the 3D echo-Doppler methods evaluation are shown in Table 6. The best diagnostic accuracy was obtained for the 3DCDE, with an area under curve 0.97 when using the 3D vena contracta cross sectional area method and an area under curve 0.98 when the vena contracta cross sectional area/left ventricular outflow tract cross sectional area method were used. These results are clearly superior to those obtained using the conventional echo-Doppler techniques.
3.3. Inter- and intra-observer variability Inter-observer agreement of 3DCDE CAR vena contracta cross sectional area measurement was excellent: ICC was 0.89 for the measurement of the aortic regurgitation vena contracta area by using 3DCDE data sets. Similar results were noted for intra-observer agreement: ICC was 0.91.
4. Discussion To our knowledge, this is the first study that evaluates the usefulness of 3DCDE to assess the CAR severity by measuring its functional orifice using as the gold-standard method the CMR evaluation. Furthermore, the results of the present work show that 3DCDE has an excellent correlation with CMR severity evaluation. Looking at these results, 3DCDE assessment is superior to echo-Doppler traditional methods commonly used to evaluate CAR severity. There are no large-scale studies evaluating the natural history of patients with CAR. The data obtained in the pre-surgical era indicate that symptomatic patients have a poor outcome with medical therapy, with mortality rates as high as 20% per year [7,8]. Nevertheless, to define the best therapeutic strategy in patients with CAR, accurate measurements of the lesion severity are needed. Traditional echoDoppler methods for the evaluation of CAR severity are not free from pitfalls [3,9]. Methods based on direct measurement of valvular orifice should be more accurate. To date, direct measurements of the CAR vena contracta area can only be performed using planimetry traced on M-mode or 2D echo images. However, these methods have multiple limitations, the major one being the correct image plane orientation. 3DCDE improves the operator's ability to perform a well oriented planimetry [4]. It has been demonstrated in different cardiac conditions [10–17]. 3DCDE allows a direct evaluation of the CAR vena contracta cross sectional area. Furthermore, rotation and orientation of the aortic valve and adjacent cardiac structures to the desired plane are easy and independent of the orientation of the acoustic window where image acquisition was obtained. The estimation of vena contracta in eccentric jets that splays out along the undersurface of the valve or prosthesis is usually very difficult to assess. This could be
Table 1 Patients characteristics.
Fig. 5. Aortic flow curve derived from phase-contrast velocity sequence (same patient as in Fig. 4). Reversal diastolic flow is delimited, corresponding to aortic regurgitant volume of 26 ml.
Variable
Mean ± Standard deviation or absolute number (percentage)
n Mean age (years) Male Hypertension DM Dyslipidemia Smoking Atrial fibrillation Coronary artery disease Previous coronary artery revascularization Mild mitral regurgitation
32 63.0 ± 13.5 22 (68.8) 23 (71.9) 12 (37.5) 17 (53.1) 9 (28.1) 2 (6.3) 4 (12.5) 3 (9.4) 18 (56.3)
644
L. Perez de Isla et al. / International Journal of Cardiology 166 (2013) 640–645
Table 2 M-mode and 2D echocardiography main results.
2D EDLVD (mm) 2D ESLVD (mm) LVEF (%) Left atrium diameter (mm) Interventricular septum thickness (mm) Posterior wall thickness (mm) Mitral e wave peak velocity (cm/s) Mitral a wave peak velocity (cm/s) Pulmonary artery systolic pressure (mm Hg) Long-axis view M-mode jet width (mm) Long-axis view M-mode LVOT width (mm) Long-axis view M-mode jet width/long-axis view M-mode LVOT width Long-axis 2D jet width (mm) Long-axis 2D LVOT width (mm) Long-axis 2D jet width/long-axis 2D LVOT width Short-axis 2D vena contracta CSA (cm2) Short-axis 2D LVOT CSA (cm2) Short-axis 2D vena contracta CSA/short-axis 2D LVOT CSA Regurgitant volume 2D echo-Doppler (ml) Regurgitant fraction 2D echo-Doppler (%)
N
Mean
Std. Deviation
32 32 32 32 32 32 32 30 27 32 32 32
52.8 32.0 62.9 39.8 12.3 10.4 60.2 143.6 32.5 0.6 2.1 0.3
8.1 8.3 13.7 6.0 1.9 1.6 14.1 232.8 9.4 0.2 0.1 0.1
32 32 32 32 32 32
0.5 2.1 0.2 0.5 3.5 0.1
0.2 0.1 0.1 0.2 0.3 0.06
32 32
29.0 29.8
3.2 0.1
2D EDLVD: 2D echo end-diastolic left ventricular diameter; 2D ESLVD: 2D echo endsystolic left ventricular diameter; LVEF: left ventricular ejection fraction; LVOT: Left ventricular outflow tract;
one of the advantages of 3D echo over 2D echo and it could partially explain the results of the study It is of note the high association of 3DCDE results and CMR when compared with traditional methods. Some of them, such as regurgitant volume and regurgitant fraction obtained with 2D echoDoppler show regression coefficients as low as 0.42 and 0.40 respectively. Probably, the relatively high number of measurements that this method requires are sources of a poor reproducibility and a low accuracy. This problem could be avoided by the use of 3DCDE, based on a fast, simple and isolated measurement. This study shows, for the first time, that 3DCDE echo is the most accurate echocardiography parameter for measuring CAR severity using CMR data as the gold-standard method when it is compared with other traditional non-invasive echo-Doppler methods to evaluate the severity of CAR [18]. Furthermore, 3DCDE measurements have excellent inter and intra-observer variability. However, other 3D measurements are possible. Sugeng et al. [19] demonstrated feasibility and comparison of measurement of 3D regurgitant volumes of tricuspid regurgitation and mitral regurgitation to 2D quantitative methods; Pirat et al. [20] demonstrated direct measurement of proximal isovelocity surface area (PISA) by real time 3D color-Doppler is feasible and calculations of aortic regurgitant volumes using this
Table 3 3D color-Doppler echocardiography and cardiac magnetic resonance main results.
3D vena contracta CSA (cm2) 3D LVOT CSA (cm2) 3D vena contracta CSA / 3D LVOT CSA 3D EDLVV (ml) 3D ESLVV (ml) Regurgitant volume CMR (ml) Regurgitant fraction CMR (%) CMR EDLVV (ml) CMR ESLVV (ml) CRM LVEF (%)
N
Mean
Std. Deviation
32 32 32 32 32 32 32 32 32 32
0.4 3.2 0.12 180.9 81.6 22.1 20.0 204.3 93.8 55.8
0.2 0.2 0.06 72.5 45.6 21.5 14.3 78.4 50.9 12.3
3D EDLVV: 3D echo end-diastolic left ventricular volume; 3D ESLVV: 3D echo endsystolic left ventricular volume; CMR: Cardiac magnetic resonance; CMR EDLVD: end-diastolic left ventricular volume evaluated by CMR; CMR ESLVD: end-systolic left ventricular volume evaluated by CMR; CSA: Cross sectional area; LVEF: left ventricular ejection fraction; LVOT: Left ventricular outflow tract;
Table 4 Linear regression analysis. Dependent variable: cardiac magnetic resonance-derived regurgitant fraction. Variable
r
R square
p
Long-axis view M-mode jet width (mm) Long-axis view M-mode jet width / Long-axis view M-mode LVOT width Long-axis 2D jet width (mm) Long-axis 2D jet width/Long-axis 2D LVOT width Short-axis 2D vena contracta CSA (cm2) Short-axis 2D vena contracta CSA / Short-axis 2D LVOT CSA Regurgitant volume 2D echo-Doppler (ml) Regurgitant fraction 2D echo-Doppler (%) 3D vena contracta CSA (cm2) 3D vena contracta CSA/3D LVOT CSA
0.66 0.69
0.44 0.48
b 0.001 b 0.001
0.65 0.66 0.48 0.42
0.42 0.44 0.23 0.18
b 0.001 b 0.001 b 0.001 b 0.001
0.45 0.40 0.88 0.87
0.20 0.16 0.77 0.76
b 0.001 b 0.001 b 0.001 b 0.001
CSA: Cross sectional area; LVOT: Left ventricular outflow tract; r: regression coefficient.
method is more accurate than conventional 2D methods. Thus, other methods based on 3D imaging should be possible and are areas of research which could be explored. It is important how the different variables were obtained. A complete description of how they were obtained may be consulted elsewhere [3]. The methods used are crucial in assessing the validity of the results and understanding the limitations of each modality.
5. Study limitations CMR analysis is not free from limitations [5,6]. Ideally, the desired gold standard method should have been the direct anatomic measurement performed in the surgical specimen inspection. Another limitation is that the echocardiography delineation of the cardiac structures is always dependent on the quality of the image, in the same manner as every echocardiographic technique. Furthermore, the relatively low temporal resolution of the 3D studies could be considered as other limitation. The number of patients with severe aortic regurgitation was relatively small. Thus, this study results should be considered as preliminary data. Further studies should be carried out to definitively validate this new technique. The reproducibility was evaluated on the same obtained images. Thus, variability of 3D dataset acquisition was not assessed.
6. Clinical implications 3DCDE can improve the CAR severity assessment and consequently the management of patients suffering from this valvular heart disease. Thus, 3DCDE is becoming a useful diagnostic method and it could be able to replace other non-invasive methods.
Table 5 ROC curves analysis results for detection of severe chronic aortic regurgitation. Reference method: aortic regurgitant fraction obtained by means of cardiac magnetic resonance. Variable
AUC
p
Long-axis view M-mode jet width (mm) Long-axis view M-mode jet width/long-axis view M-mode LVOT width Long-axis 2D jet width (mm) Long-axis 2D jet width/long-axis 2D LVOT width Short-axis 2D vena contracta CSA (cm2) Short-axis 2D vena contracta CSA / Short-axis 2D LVOT CSA Regurgitant volume 2D echo-Doppler (ml) Regurgitant fraction 2D echo-Doppler (%) 3D vena contracta CSA (cm2) 3D vena contracta CSA / 3D LVOT CSA
0.81 0.81
b 0.001 b 0.001
0.82 0.80 0.86 0.87 0.73 0.76 0.97 0.98
b 0.001 b 0.001 0.003 0.003 0.009 0.01 b 0.001 b 0.001
AUC: Area under curve; CSA: Cross sectional area; LVOT: Left ventricular outflow tract;
L. Perez de Isla et al. / International Journal of Cardiology 166 (2013) 640–645 Table 6 ROC curves analysis results for detection of severe chronic aortic regurgitation by means of 3D color-Doppler echocardiography assessment. Reference method: aortic regurgitant fraction obtained by means of cardiac magnetic resonance. Variable
AUC Cut-off point
Sensitivity Specificity PPV
3D vena contracta CSA 0.97 0.50 cm2 100% 3D vena contracta CSA/ 0.98 0.19 100% 3D LVOT CSA
92.59% 77.78%
NPV
81.82% 100% 60.00% 100%
AUC: Area under curve; CSA: Cross sectional area; LVOT: Left ventricular outflow tract; NPV: Negative predictive value; PPV: Positive predictive value.
7. Conclusions 3DCDE is an accurate and highly reproducible diagnostic tool for estimating CAR severity. Compared with the traditional echoDoppler methods, 3DCDE has the best agreement with the CMR determined CR severity. Thus, 3DCDE is a diagnostic method that may improve the therapeutic management of patients with CAR. Acknowledgements The authors of this manuscript have certified that they comply with the Principles of Ethical Publishing in the International Journal of Cardiology. References [1] Vahanian A, Baumgartner H, Bax J, et al. Task Force on the Management of Valvular Hearth Disease of the European Society of Cardiology; ESC Committee for Practice Guidelines. Guidelines on the management of valvular heart disease: The Task Force on the Management of Valvular Heart Disease of the European Society of Cardiology. Eur Heart J 2007 Jan;28(2):230–68. [2] American College of Cardiology; American Heart Association Task Force on Practice Guidelines (Writing Committee to revise the 1998 guidelines for the management of patients with valvular heart disease); Society of Cardiovascular Anesthesiologists, Bonow RO, Carabello BA, Chatterjee K, de Leon AC Jr, Faxon DP, Freed MD, Gaasch WH, Lytle BW, Nishimura RA, O'Gara PT, O'Rourke RA, Otto CM, Shah PM, Shanewise JS, Smith SC Jr, Jacobs AK, Adams CD, Anderson JL, Antman EM, Fuster V, Halperin JL, Hiratzka LF, Hunt SA, Lytle BW, Nishimura R, Page RL, Riegel B. ACC/AHA 2006 guidelines for the management of patients with valvular heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (writing Committee to Revise the 1998 guidelines for the management of patients with valvular heart disease) developed in collaboration with the Society of Cardiovascular Anesthesiologists endorsed by the Society for Cardiovascular Angiography and Interventions and the Society of Thoracic Surgeons. J Am Coll Cardiol. 2007 Mar 6;49(9):1014.
645
[3] Zoghbi WA, Enriquez-Sarano M, Foster E, et al. American Society of Echocardiography. Recommendations for evaluation of the severity of native valvular regurgitation with two-dimensional and Doppler echocardiography. J Am Soc Echocardiogr Jul 2003;16(7):777–802. [4] Hung J, Lang R, Flachskampf F, et al. 3D echocardiography: a review of the current status and future directions. J Am Soc Echocardiogr 2007 Mar;20(3):19A–20A. [5] Ley S, Eichhorn J, Ley-Zaporozhan J, et al. Evaluation of aortic regurgitation in congenital heart disease: value of MR imaging in comparison to echocardiography. Pediatr Radiol 2007 May;37(5):426–36. [6] Gelfand EV, Hughes S, Hauser TH, et al. Severity of mitral and aortic regurgitation as assessed by cardiovascular magnetic resonance: optimizing correlation with Doppler echocardiography. J Cardiovasc Magn Reson 2006;8(3):503–7. [7] Hegglin R, Scheu H, Rothlin M. Aortic insufficiency. Circulation 1968;38:77–92. [8] Spagnuolo M, Kloth H, Taranta A, Doyle E, Pasternack B. Natural history of rheumatic aortic regurgitation: criteria predictive of death, congestive heart failure, and angina in young patients. Circulation 1971;44:368–80. [9] Oh J, Seward J, Tajik A. Valvular heart disease. The echo manual. 3rd ed. Lippincott; 1999. p. 190. [10] Zamorano Jose, Cordeiro Pedro, Sugeng Lissa, et al. Real-time three-dimensional echocardiography for rheumatic mitral valve stenosis evaluation. J Am Coll Cardiol Jun 2 2004;43(11):2091–6. [11] Zamorano José, Pérez de Isla Leopoldo, Sugeng Lissa, et al. Non Invasive Assessment of Mitral Valve Area during Percutaneous Balloon Mitral Valvuloplasty: Role of Real Time 3D Echocardiography. Eur Heart J Dec 2004;25(23): 2086–91. [12] Gutierrez-Chico JL, Zamorano JL, Pérez de Isla L, et al. Comparison of left ventricular volumen and ejection fraction measured by three-dimensional echocardiography and cardiac magnetic resonante in patients with various cardiomyopathies. Am J Cardiol Mar 15 2005;95(6):809–13. [13] de Isla Leopoldo Pérez, José Zamorano, Gracieta Malangatana, et al. Morphologic determinants of subaortic stenosis in hypertrophic cardiomyopathy: insights from real time 3D echocardiography. J Am Soc Echocardiogr 2005 Aug;18(8): 802–4. [14] Jacobs LD, Salgo IS, Goonewardena S, et al. Real-time three-dimensional echocardiographic quantification of left ventricular volumes using a rapid tissue tracking algorithm. Eur Heart J 2006 Feb;27(4):460–8. [15] Pérez de Isla L, Casanova C, Almerıía C, et al. Which method should be the reference method to evaluate the severity of rheumatic mitral stenosis? Gorlin's method versus 3D-echo. Eur J Echocardiogr 2007 Dec;8(6):470–3. [16] Gutiérrez-Chico JL, Zamorano JL, Prieto-Moriche E, Hernández-Antolín RA, BravoAmaro M, Pérez de Isla L, Sanmartín-Fernández M, Baz-Alonso JA, Iñiguez-Romo A. Real-time three-dimensional echocardiography in aortic stenosis: a novel, simple, and reliable method to improve accuracy in area calculation. bhttp://www.ncbi. nlm.nih.gov/pubmed/17989075> Eur Heart J May 2008;29(10):1296–306. [17] Gutiérrez-Chico JL, Zamorano Gómez JL, Rodrigo-López JL, et al. Accuracy of real-time 3-dimensional echocardiography in the assessment of mitral prolapse. Is transesophageal echocardiography still mandatory? Am Heart J Apr 2008;155(4):694–8. [18] Fang L, Hsiung MC, Miller AP, et al. Assessment of aortic regurgitation by live three dimensional transthoracic echocardiographic measurement of vena contracta area: usefulness and validation. Echocardiography 2005;22:775–81. [19] Sugeng L, Weinert L, Lang RM. Real time 3-dimensional color Doppler flow of mitral and tricuspid regurgitation: feasibility and initial quantitative comparison with 2-dimensional methods. J Am Soc Echocardiogr 2007;20:1050–7. [20] Pirat B, Little S, McCulloch M, Nose Y, Hartley C, Zoghbi W. Direct measurement of proximal isovelocity surface area by real-time three dimensional color Doppler for quantification of aortic regurgitant volume: an in vitro validation. J Am Soc Echocardiogr 2009;22:306–13.
Contents lists available at SciVerse ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
3D color-Doppler echocardiography and chronic aortic regurgitation: A novel approach for severity assessment ☆ Leopoldo Perez de Isla ⁎, Jose Zamorano, Covadonga Fernandez-Golfin, Sara Ciocarelli, Cecilia Corros, Tibisai Sanchez, Joaquín Ferreirós, Pedro Marcos-Alberca, Carlos Almeria, Jose Luis Rodrigo, Carlos Macaya Unidad de Imagen Cardiovascular, Hospital Clínico San Carlos, Madrid, Spain
a r t i c l e
i n f o
Article history: Received 8 April 2010 Received in revised form 20 July 2010 Accepted 26 November 2011 Available online 20 December 2011 Keywords: Doppler-echocardiography Three-dimensional Aortic regurgitation
a b s t r a c t Background: 3D echocardiography provides a complete evaluation of the aortic valve and adjacent structures and it improves the assessment of this cardiac region. Three-dimensional color-Doppler echocardiography (3DCDE) evaluation might improve the measurements of the functional regurgitant orifice in patients with Chronic Aortic Regurgitation (CAR). Objectives: Our aim was to compare the accuracy of current echo-Doppler methods and 3DCDE for the assessment of CAR severity. The reference method used in this work was the CAR severity determined by means of cardiac magnetic resonance (CMR) Methods: Thirty-two consecutive patients with an established diagnosis of CAR recruited in our institution comprised our study group. CAR severity was determined by conventional Echo-Doppler methods and by 3DCDE and their results were compared with those obtained by means of CMR. Results: Mean age was 63.0 ± 13.5 years. Twenty-two patients (68.8%) were men. Compared with the traditional echo-Doppler methods, 3DCDE evaluation had the best linear association with CMR results (3D vena contracta cross sectional area method: r = 0.88; r square = 0.77; p b 0.001. 3D vena contracta cross sectional area/left ventricular outflow tract cross sectional area method: r = 0.87; r square = 0.75; p b 0.001). The ROC analysis showed an excellent area under curve for detection of severe CAR (3D vena contracta cross sectional area method = 0.97; 3D vena contracta cross sectional area/left ventricular outflow tract cross sectional area method = 0.98). Inter- and intra-observer variability for the 3DCDE evaluation was good (ICC = 0.89 and ICC = 0.91 for inter and intra observer variability respectively). Conclusions: 3DCDE is an accurate and highly reproducible diagnostic tool for estimating CAR severity. Compared with the traditional echo-Doppler methods, 3DCDE has the best agreement with the CMR determined CAR severity. Thus, 3DCDE is a diagnostic method that may improve the therapeutic management of patients with CAR. © 2011 Elsevier Ireland Ltd. All rights reserved.
1. Introduction and objectives The therapeutic management of patients with chronic aortic regurgitation (CAR) represents a challenge for the clinician [1,2]. This entity is responsible of quality of life impairment and life expectancy reduction in a huge number of patients. To define the best therapeutic strategy, clinical data and accurate measurements of the CAR severity
Abbreviations: 2D, Two Dimensional Echocardiography; 3DCDE, Three Dimensional Color-Doppler Echocardiography; ARFr, Aortic Regurgitant Fraction; CAR, Chronic Aortic Regurgitation; CMR, Cardiac Magnetic Resonance; EDV, End Diastolic Volume; ESV, End Systolic Volume; ICC, Intra-class Correlation Coefficient; LVOT, left ventricular outflow tract. ☆ There is no conflict of interest concerning this manuscript. ⁎ Corresponding author at: Unidad de Imagen Cardiovascular, Hospital Clínico San Carlos, Plaza Cristo Rey, 28040-Madrid, Spain. Tel./fax: + 34 913303290. E-mail address: [email protected] (L. Perez de Isla). 0167-5273/$ – see front matter © 2011 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ijcard.2011.11.094
are needed. Nevertheless, currently employed methods are not free from limitations [3]. Three-dimensional echocardiography provides a unique “en-face” view of the complete aortic, sub-aortic and aortic root structures and could therefore improve the assessment of this cardiac region. Furthermore, three-dimensional color-Doppler evaluation might improve the measurements of the functional regurgitant orifice [4]. However, it has not been routinely used due to the prolonged data acquisition and off-line processing time. With the new transthoracic 3D matrix array probe (Philips, Bothell, Washington) that allows 3D color-Doppler-Echocardiography (3DCDE) rendering and the new analysis software, this technique could be included in the routine assessment of this valvular heart disease. Our aim was to compare the accuracy of traditional echo-Doppler methods and 3DCDE for the assessment of CAR severity. The reference method used in this work was the CAR severity determined by means of cardiac magnetic resonance (CMR) [5,6].
L. Perez de Isla et al. / International Journal of Cardiology 166 (2013) 640–645 2. Methods 2.1. Patient population Thirty-two consecutive patients with an established diagnosis of CAR [1], recruited in our institution, comprised our study group. Exclusion criteria were more than mild valvular heart disease different of aortic regurgitation (mild valvular lesions were not considered as having a relevant impact on the study results), congenital heart disease, the presence of cardiomyopathies and patient unwillingness to be included in the study. Every patient was given verbal and written information about the techniques they would undergo. They were also informed and gave consent for anonymous use of their data for research purposes.
2.2. Conventional echo evaluation A complete conventional Echo-Doppler study was performed in all patients using a IE-33 ultrasound machine and a S 5–1 probe (Philips, Bothell, Washington). M-mode echo, 2D echo, pulsed and continuous spectral Doppler and color-Doppler recordings were obtained from standard paraesternal and apical windows and measurements of the different structures were obtained, following the current recommendations [3]. Three cardiac cycles for patients in sinus rhythm and five for patients in atrial fibrillation were recorded and their results averaged for every patient. The presence of aortic regurgitation was defined by means of echo-Doppler and its severity was assessed by the current recommendations [3]. Left ventricular volumes and ejection fraction were evaluated by means of Simpson method. Linear dimensions were obtained by using M mode echocardiography. Mitral e and a wave peak velocities were obtained by measuring velocities with pulsed wave Doppler at the level of the tips of the mitral valve. Pulmonary artery systolic pressure was derived from the tricuspid systolic gradient and the estimated right atrial pressure. Jet width and vena contracta were obtained following the current recommendations with M-mode color-Doppler. Regurgitant volume and ejection fraction was obtained by means of the volumetric method as described elsewhere [3].
2.3. 3DCDE evaluation 3DCDE was performed immediately after the 2D study. Data were recorded using an IE33 ultrasound machine and an X 3-1 probe (Philips, Bothell, Washington). Settings were optimized for 3D resolution. The full-volume method provides a pyramidal data set of approximately 90°×90° and allows inclusion of a larger cardiac volume, although it requires ECG gating, because this technique compiles 4 to 7 pyramidal scans obtained during 4 consecutive heartbeats. The full volume method was used for left ventricular volume and systolic function analysis. 3D color-Doppler was employed for the CAR vena contracta measurement (Fig. 1). The capture was performed by placing the region of interest in the aortic valve region from an apical view. Settings were optimized for color-Doppler resolution. The protocol used for patients in atrial fibrillation was the same as for patients in sinus rhythm because the software used is able to improve the acquisition in patients with arrhythmias. The software automatically rejects the RR intervals that are very different from the mean RR interval.
641
All images were sent to the central DICOM server for off-line analysis using Q-Lab® software (Philips, Bothell, Washington). 3DCDE LVOT and aortic regurgitation vena contracta measurements were performed “en-face” at the ideal cross section, just immediately below the aortic valvular plane in mid-diastole (Figs. 2 and 3). In case of two or more regurgitant orifices, cross sectional areas were independently evaluated and their areas were added. Left ventricular volumes and ejection fraction were evaluated using the 3DQ-Advance software, integrated in the Q-lab software platform. 2.4. Cardiac magnetic resonance study protocol CMR were performed within 15 days of the echocardiographic study. No clinical event appeared within this period. Furthermore, no change in treatment was done. All CMR exams were performed at 1.5 T Signa EXCITE 11.0 (General Electrics, Milwaukee, WI) whole body scanner with a dedicated cardiac 8 channel surface coil. After real time initial localizing scans, breath-hold, electrocardiographically triggered steady state free precession cine images (echo time/repetition time 1.74/3.92, flip angle 45°) were acquired in four chamber long axis view and short axis view from base to apex (8 mm thickness, gap 2 mm). Quantitative measures of aortic flow and aortic regurgitation volume and fraction were assessed using breath-hold electrocardiographically triggered phase contrast velocity sequence with the following parameters: TR 12 ms, TE= 4.4 ms, flip angle = 17°, FOV= 320 × 240, matrix = 256 × 128 and slice thickness 6 mm. Velocity encoding (VENC) was initially set in 200 cm/s with further 30 cm/s increments in case of aliasing. The plane for aortic flow measurements was localized at the level of the aortic root parallel and just above the aortic valve (Fig. 4). CRM image analyses were performed using a commercial work station (ReportCard, Version 2.0, General Electrics, Milwaukee, WI). Endocardial LV borders were manually traced at end diastole and at end-systole. Papillary muscles were included as part of LV cavity volume. Left ventricular end-diastolic volume (EDV) and endsystolic volume (ESV) were determined using a Simpson's rule method. Stroke volume was calculated as the difference between EDV and ESV. Ejection fraction was calculated as EF = (EDV − ESV)/EDV × 100.Aortic regurgitation volume was directly calculated from the aortic flow curve by integrating diastolic reverse flow. Aortic regurgitation volume was calculated from the aortic flow curve by tracing the diastolic reverse flow. Aortic regurgitation fraction (ARFr) was calculated as the ratio of aortic regurgitant volume and aortic forward flow. Aortic regurgitation severity was graded as follows: mild ARFr b 15%, moderate ARFr 15–30% and severe ARFr > 30% (Fig. 5). 2.5. Inter- and intra-observer variability 3DCDE datasets of the CAR vena contracta of the first 15 enrolled patients were analyzed offline at different times by two independent observers. The same images were also analyzed on a different day by one of these observers. 2.6. Statistical analysis The statistical package used was SPSS version 15.0. Quantitative data were expressed as mean ± standard deviation. Qualitative data were expressed as absolute number (percentage). Inter- and intra-observer reproducibility was evaluated by means of
Fig. 1. Color-Doppler echocardiography images of a patient with chronic aortic regurgitation.
642
L. Perez de Isla et al. / International Journal of Cardiology 166 (2013) 640–645
Fig. 2. Aortic regurgitation vena contracta planes orientation in a 3D color-Doppler echocardiography dataset. the Intra-class Correlation Coefficient (ICC). Inter-methods linear association was assessed by using linear regression analysis. Comparisons were considered significant in presence of a p value b 0.05.
3. Results Thirty-two consecutive non-selected patients with CAR were prospectively enrolled. They were the analyzed group. Before ending the inclusion period, five patients were not enrolled due to the unwillingness to undergo a CMR study and one patient was not included due to the presence of an inadequate acoustic window. There were 22 men (68.8%) and mean age was 63.0 ± 13.5 years. CAR was the predominant valvular lesion in all of them but concomitant mild mitral regurgitation was present in 18 patients (56.3%). Aortic stenosis and mitral stenosis were not present in any patient. Thirty patients
(93.7%) were in normal sinus rhythm and 2 (6.3%) in atrial fibrillation. Main descriptive variables are depicted in Table 1. The value of the regurgitant fraction of the CAR obtained by using CMR divided the severity of the studied population into: 14 patients with mild CAR (43.8%), 11 patients with moderate CAR (34.4%) and 7 patients with severe CAR (21.9%). The aetiology of the aortic regurgitation was degenerative (associated to age) in every patient. The frame rate in the 3D studies was between 12 and 16 volumes per second. And in the 2D studies between 20 and 30 frames per second. 3.1. Comparison of classic Echo-Doppler methods, 3DCDE and CMR CAR severity was evaluated by using the previously described methods [3]. Mean results for each studied variable are shown in Tables 2 and 3. Linear regression analysis was used to establish the
Fig. 3. Aortic regurgitation vena contracta 3D echo-Doppler quantification in a 3D color-Doppler echocardiography dataset.
L. Perez de Isla et al. / International Journal of Cardiology 166 (2013) 640–645
643
Fig. 4. Phase-contrast velocity sequence for aortic regurgitation volume quantification at the level of the aortic root, just above the valvular plane. Diastolic magnitude and phase (left and right respectively) images with aortic regurgitation central jet clearly seen.
linear association between each method and the values obtained with CMR. Results are described in Table 3. Time required to obtain and analyze the 3DCDE images, evaluated in all the enrolled patients was 16.2 ± 2.6 minutes. Main results for the evaluation of aortic regurgitation vena contracta cross sectional area are shown in Table 3. Linear regression analysis results between 3DCDE and the values obtained with CMR are described in Table 4. It is of note that the best linear association is obtained with the 3Decho-Doppler method (see Table 4).
3.2. Diagnostic accuracy of 3DCDE to detect severe CAR In order to evaluate the diagnostic accuracy of conventional echoDoppler methods and 3DCDE to detect severe CAR, a ROC curves analysis was performed. Results for every method are depicted in Table 5 and detailed results for the 3D echo-Doppler methods evaluation are shown in Table 6. The best diagnostic accuracy was obtained for the 3DCDE, with an area under curve 0.97 when using the 3D vena contracta cross sectional area method and an area under curve 0.98 when the vena contracta cross sectional area/left ventricular outflow tract cross sectional area method were used. These results are clearly superior to those obtained using the conventional echo-Doppler techniques.
3.3. Inter- and intra-observer variability Inter-observer agreement of 3DCDE CAR vena contracta cross sectional area measurement was excellent: ICC was 0.89 for the measurement of the aortic regurgitation vena contracta area by using 3DCDE data sets. Similar results were noted for intra-observer agreement: ICC was 0.91.
4. Discussion To our knowledge, this is the first study that evaluates the usefulness of 3DCDE to assess the CAR severity by measuring its functional orifice using as the gold-standard method the CMR evaluation. Furthermore, the results of the present work show that 3DCDE has an excellent correlation with CMR severity evaluation. Looking at these results, 3DCDE assessment is superior to echo-Doppler traditional methods commonly used to evaluate CAR severity. There are no large-scale studies evaluating the natural history of patients with CAR. The data obtained in the pre-surgical era indicate that symptomatic patients have a poor outcome with medical therapy, with mortality rates as high as 20% per year [7,8]. Nevertheless, to define the best therapeutic strategy in patients with CAR, accurate measurements of the lesion severity are needed. Traditional echoDoppler methods for the evaluation of CAR severity are not free from pitfalls [3,9]. Methods based on direct measurement of valvular orifice should be more accurate. To date, direct measurements of the CAR vena contracta area can only be performed using planimetry traced on M-mode or 2D echo images. However, these methods have multiple limitations, the major one being the correct image plane orientation. 3DCDE improves the operator's ability to perform a well oriented planimetry [4]. It has been demonstrated in different cardiac conditions [10–17]. 3DCDE allows a direct evaluation of the CAR vena contracta cross sectional area. Furthermore, rotation and orientation of the aortic valve and adjacent cardiac structures to the desired plane are easy and independent of the orientation of the acoustic window where image acquisition was obtained. The estimation of vena contracta in eccentric jets that splays out along the undersurface of the valve or prosthesis is usually very difficult to assess. This could be
Table 1 Patients characteristics.
Fig. 5. Aortic flow curve derived from phase-contrast velocity sequence (same patient as in Fig. 4). Reversal diastolic flow is delimited, corresponding to aortic regurgitant volume of 26 ml.
Variable
Mean ± Standard deviation or absolute number (percentage)
n Mean age (years) Male Hypertension DM Dyslipidemia Smoking Atrial fibrillation Coronary artery disease Previous coronary artery revascularization Mild mitral regurgitation
32 63.0 ± 13.5 22 (68.8) 23 (71.9) 12 (37.5) 17 (53.1) 9 (28.1) 2 (6.3) 4 (12.5) 3 (9.4) 18 (56.3)
644
L. Perez de Isla et al. / International Journal of Cardiology 166 (2013) 640–645
Table 2 M-mode and 2D echocardiography main results.
2D EDLVD (mm) 2D ESLVD (mm) LVEF (%) Left atrium diameter (mm) Interventricular septum thickness (mm) Posterior wall thickness (mm) Mitral e wave peak velocity (cm/s) Mitral a wave peak velocity (cm/s) Pulmonary artery systolic pressure (mm Hg) Long-axis view M-mode jet width (mm) Long-axis view M-mode LVOT width (mm) Long-axis view M-mode jet width/long-axis view M-mode LVOT width Long-axis 2D jet width (mm) Long-axis 2D LVOT width (mm) Long-axis 2D jet width/long-axis 2D LVOT width Short-axis 2D vena contracta CSA (cm2) Short-axis 2D LVOT CSA (cm2) Short-axis 2D vena contracta CSA/short-axis 2D LVOT CSA Regurgitant volume 2D echo-Doppler (ml) Regurgitant fraction 2D echo-Doppler (%)
N
Mean
Std. Deviation
32 32 32 32 32 32 32 30 27 32 32 32
52.8 32.0 62.9 39.8 12.3 10.4 60.2 143.6 32.5 0.6 2.1 0.3
8.1 8.3 13.7 6.0 1.9 1.6 14.1 232.8 9.4 0.2 0.1 0.1
32 32 32 32 32 32
0.5 2.1 0.2 0.5 3.5 0.1
0.2 0.1 0.1 0.2 0.3 0.06
32 32
29.0 29.8
3.2 0.1
2D EDLVD: 2D echo end-diastolic left ventricular diameter; 2D ESLVD: 2D echo endsystolic left ventricular diameter; LVEF: left ventricular ejection fraction; LVOT: Left ventricular outflow tract;
one of the advantages of 3D echo over 2D echo and it could partially explain the results of the study It is of note the high association of 3DCDE results and CMR when compared with traditional methods. Some of them, such as regurgitant volume and regurgitant fraction obtained with 2D echoDoppler show regression coefficients as low as 0.42 and 0.40 respectively. Probably, the relatively high number of measurements that this method requires are sources of a poor reproducibility and a low accuracy. This problem could be avoided by the use of 3DCDE, based on a fast, simple and isolated measurement. This study shows, for the first time, that 3DCDE echo is the most accurate echocardiography parameter for measuring CAR severity using CMR data as the gold-standard method when it is compared with other traditional non-invasive echo-Doppler methods to evaluate the severity of CAR [18]. Furthermore, 3DCDE measurements have excellent inter and intra-observer variability. However, other 3D measurements are possible. Sugeng et al. [19] demonstrated feasibility and comparison of measurement of 3D regurgitant volumes of tricuspid regurgitation and mitral regurgitation to 2D quantitative methods; Pirat et al. [20] demonstrated direct measurement of proximal isovelocity surface area (PISA) by real time 3D color-Doppler is feasible and calculations of aortic regurgitant volumes using this
Table 3 3D color-Doppler echocardiography and cardiac magnetic resonance main results.
3D vena contracta CSA (cm2) 3D LVOT CSA (cm2) 3D vena contracta CSA / 3D LVOT CSA 3D EDLVV (ml) 3D ESLVV (ml) Regurgitant volume CMR (ml) Regurgitant fraction CMR (%) CMR EDLVV (ml) CMR ESLVV (ml) CRM LVEF (%)
N
Mean
Std. Deviation
32 32 32 32 32 32 32 32 32 32
0.4 3.2 0.12 180.9 81.6 22.1 20.0 204.3 93.8 55.8
0.2 0.2 0.06 72.5 45.6 21.5 14.3 78.4 50.9 12.3
3D EDLVV: 3D echo end-diastolic left ventricular volume; 3D ESLVV: 3D echo endsystolic left ventricular volume; CMR: Cardiac magnetic resonance; CMR EDLVD: end-diastolic left ventricular volume evaluated by CMR; CMR ESLVD: end-systolic left ventricular volume evaluated by CMR; CSA: Cross sectional area; LVEF: left ventricular ejection fraction; LVOT: Left ventricular outflow tract;
Table 4 Linear regression analysis. Dependent variable: cardiac magnetic resonance-derived regurgitant fraction. Variable
r
R square
p
Long-axis view M-mode jet width (mm) Long-axis view M-mode jet width / Long-axis view M-mode LVOT width Long-axis 2D jet width (mm) Long-axis 2D jet width/Long-axis 2D LVOT width Short-axis 2D vena contracta CSA (cm2) Short-axis 2D vena contracta CSA / Short-axis 2D LVOT CSA Regurgitant volume 2D echo-Doppler (ml) Regurgitant fraction 2D echo-Doppler (%) 3D vena contracta CSA (cm2) 3D vena contracta CSA/3D LVOT CSA
0.66 0.69
0.44 0.48
b 0.001 b 0.001
0.65 0.66 0.48 0.42
0.42 0.44 0.23 0.18
b 0.001 b 0.001 b 0.001 b 0.001
0.45 0.40 0.88 0.87
0.20 0.16 0.77 0.76
b 0.001 b 0.001 b 0.001 b 0.001
CSA: Cross sectional area; LVOT: Left ventricular outflow tract; r: regression coefficient.
method is more accurate than conventional 2D methods. Thus, other methods based on 3D imaging should be possible and are areas of research which could be explored. It is important how the different variables were obtained. A complete description of how they were obtained may be consulted elsewhere [3]. The methods used are crucial in assessing the validity of the results and understanding the limitations of each modality.
5. Study limitations CMR analysis is not free from limitations [5,6]. Ideally, the desired gold standard method should have been the direct anatomic measurement performed in the surgical specimen inspection. Another limitation is that the echocardiography delineation of the cardiac structures is always dependent on the quality of the image, in the same manner as every echocardiographic technique. Furthermore, the relatively low temporal resolution of the 3D studies could be considered as other limitation. The number of patients with severe aortic regurgitation was relatively small. Thus, this study results should be considered as preliminary data. Further studies should be carried out to definitively validate this new technique. The reproducibility was evaluated on the same obtained images. Thus, variability of 3D dataset acquisition was not assessed.
6. Clinical implications 3DCDE can improve the CAR severity assessment and consequently the management of patients suffering from this valvular heart disease. Thus, 3DCDE is becoming a useful diagnostic method and it could be able to replace other non-invasive methods.
Table 5 ROC curves analysis results for detection of severe chronic aortic regurgitation. Reference method: aortic regurgitant fraction obtained by means of cardiac magnetic resonance. Variable
AUC
p
Long-axis view M-mode jet width (mm) Long-axis view M-mode jet width/long-axis view M-mode LVOT width Long-axis 2D jet width (mm) Long-axis 2D jet width/long-axis 2D LVOT width Short-axis 2D vena contracta CSA (cm2) Short-axis 2D vena contracta CSA / Short-axis 2D LVOT CSA Regurgitant volume 2D echo-Doppler (ml) Regurgitant fraction 2D echo-Doppler (%) 3D vena contracta CSA (cm2) 3D vena contracta CSA / 3D LVOT CSA
0.81 0.81
b 0.001 b 0.001
0.82 0.80 0.86 0.87 0.73 0.76 0.97 0.98
b 0.001 b 0.001 0.003 0.003 0.009 0.01 b 0.001 b 0.001
AUC: Area under curve; CSA: Cross sectional area; LVOT: Left ventricular outflow tract;
L. Perez de Isla et al. / International Journal of Cardiology 166 (2013) 640–645 Table 6 ROC curves analysis results for detection of severe chronic aortic regurgitation by means of 3D color-Doppler echocardiography assessment. Reference method: aortic regurgitant fraction obtained by means of cardiac magnetic resonance. Variable
AUC Cut-off point
Sensitivity Specificity PPV
3D vena contracta CSA 0.97 0.50 cm2 100% 3D vena contracta CSA/ 0.98 0.19 100% 3D LVOT CSA
92.59% 77.78%
NPV
81.82% 100% 60.00% 100%
AUC: Area under curve; CSA: Cross sectional area; LVOT: Left ventricular outflow tract; NPV: Negative predictive value; PPV: Positive predictive value.
7. Conclusions 3DCDE is an accurate and highly reproducible diagnostic tool for estimating CAR severity. Compared with the traditional echoDoppler methods, 3DCDE has the best agreement with the CMR determined CR severity. Thus, 3DCDE is a diagnostic method that may improve the therapeutic management of patients with CAR. Acknowledgements The authors of this manuscript have certified that they comply with the Principles of Ethical Publishing in the International Journal of Cardiology. References [1] Vahanian A, Baumgartner H, Bax J, et al. Task Force on the Management of Valvular Hearth Disease of the European Society of Cardiology; ESC Committee for Practice Guidelines. Guidelines on the management of valvular heart disease: The Task Force on the Management of Valvular Heart Disease of the European Society of Cardiology. Eur Heart J 2007 Jan;28(2):230–68. [2] American College of Cardiology; American Heart Association Task Force on Practice Guidelines (Writing Committee to revise the 1998 guidelines for the management of patients with valvular heart disease); Society of Cardiovascular Anesthesiologists, Bonow RO, Carabello BA, Chatterjee K, de Leon AC Jr, Faxon DP, Freed MD, Gaasch WH, Lytle BW, Nishimura RA, O'Gara PT, O'Rourke RA, Otto CM, Shah PM, Shanewise JS, Smith SC Jr, Jacobs AK, Adams CD, Anderson JL, Antman EM, Fuster V, Halperin JL, Hiratzka LF, Hunt SA, Lytle BW, Nishimura R, Page RL, Riegel B. ACC/AHA 2006 guidelines for the management of patients with valvular heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (writing Committee to Revise the 1998 guidelines for the management of patients with valvular heart disease) developed in collaboration with the Society of Cardiovascular Anesthesiologists endorsed by the Society for Cardiovascular Angiography and Interventions and the Society of Thoracic Surgeons. J Am Coll Cardiol. 2007 Mar 6;49(9):1014.
645
[3] Zoghbi WA, Enriquez-Sarano M, Foster E, et al. American Society of Echocardiography. Recommendations for evaluation of the severity of native valvular regurgitation with two-dimensional and Doppler echocardiography. J Am Soc Echocardiogr Jul 2003;16(7):777–802. [4] Hung J, Lang R, Flachskampf F, et al. 3D echocardiography: a review of the current status and future directions. J Am Soc Echocardiogr 2007 Mar;20(3):19A–20A. [5] Ley S, Eichhorn J, Ley-Zaporozhan J, et al. Evaluation of aortic regurgitation in congenital heart disease: value of MR imaging in comparison to echocardiography. Pediatr Radiol 2007 May;37(5):426–36. [6] Gelfand EV, Hughes S, Hauser TH, et al. Severity of mitral and aortic regurgitation as assessed by cardiovascular magnetic resonance: optimizing correlation with Doppler echocardiography. J Cardiovasc Magn Reson 2006;8(3):503–7. [7] Hegglin R, Scheu H, Rothlin M. Aortic insufficiency. Circulation 1968;38:77–92. [8] Spagnuolo M, Kloth H, Taranta A, Doyle E, Pasternack B. Natural history of rheumatic aortic regurgitation: criteria predictive of death, congestive heart failure, and angina in young patients. Circulation 1971;44:368–80. [9] Oh J, Seward J, Tajik A. Valvular heart disease. The echo manual. 3rd ed. Lippincott; 1999. p. 190. [10] Zamorano Jose, Cordeiro Pedro, Sugeng Lissa, et al. Real-time three-dimensional echocardiography for rheumatic mitral valve stenosis evaluation. J Am Coll Cardiol Jun 2 2004;43(11):2091–6. [11] Zamorano José, Pérez de Isla Leopoldo, Sugeng Lissa, et al. Non Invasive Assessment of Mitral Valve Area during Percutaneous Balloon Mitral Valvuloplasty: Role of Real Time 3D Echocardiography. Eur Heart J Dec 2004;25(23): 2086–91. [12] Gutierrez-Chico JL, Zamorano JL, Pérez de Isla L, et al. Comparison of left ventricular volumen and ejection fraction measured by three-dimensional echocardiography and cardiac magnetic resonante in patients with various cardiomyopathies. Am J Cardiol Mar 15 2005;95(6):809–13. [13] de Isla Leopoldo Pérez, José Zamorano, Gracieta Malangatana, et al. Morphologic determinants of subaortic stenosis in hypertrophic cardiomyopathy: insights from real time 3D echocardiography. J Am Soc Echocardiogr 2005 Aug;18(8): 802–4. [14] Jacobs LD, Salgo IS, Goonewardena S, et al. Real-time three-dimensional echocardiographic quantification of left ventricular volumes using a rapid tissue tracking algorithm. Eur Heart J 2006 Feb;27(4):460–8. [15] Pérez de Isla L, Casanova C, Almerıía C, et al. Which method should be the reference method to evaluate the severity of rheumatic mitral stenosis? Gorlin's method versus 3D-echo. Eur J Echocardiogr 2007 Dec;8(6):470–3. [16] Gutiérrez-Chico JL, Zamorano JL, Prieto-Moriche E, Hernández-Antolín RA, BravoAmaro M, Pérez de Isla L, Sanmartín-Fernández M, Baz-Alonso JA, Iñiguez-Romo A. Real-time three-dimensional echocardiography in aortic stenosis: a novel, simple, and reliable method to improve accuracy in area calculation. bhttp://www.ncbi. nlm.nih.gov/pubmed/17989075> Eur Heart J May 2008;29(10):1296–306. [17] Gutiérrez-Chico JL, Zamorano Gómez JL, Rodrigo-López JL, et al. Accuracy of real-time 3-dimensional echocardiography in the assessment of mitral prolapse. Is transesophageal echocardiography still mandatory? Am Heart J Apr 2008;155(4):694–8. [18] Fang L, Hsiung MC, Miller AP, et al. Assessment of aortic regurgitation by live three dimensional transthoracic echocardiographic measurement of vena contracta area: usefulness and validation. Echocardiography 2005;22:775–81. [19] Sugeng L, Weinert L, Lang RM. Real time 3-dimensional color Doppler flow of mitral and tricuspid regurgitation: feasibility and initial quantitative comparison with 2-dimensional methods. J Am Soc Echocardiogr 2007;20:1050–7. [20] Pirat B, Little S, McCulloch M, Nose Y, Hartley C, Zoghbi W. Direct measurement of proximal isovelocity surface area by real-time three dimensional color Doppler for quantification of aortic regurgitant volume: an in vitro validation. J Am Soc Echocardiogr 2009;22:306–13.