Background
With recent developments in echocardiographic technology, a new system using real-time three-dimensional echocardiography (RT3DE) that allows single-beat acquisition of the entire volume of the left ventricle and incorporates algorithms for automated border detection has been introduced. Provided that these techniques are acceptably reliable, three-dimensional echocardiography may be much more useful for clinical practice. The aim of this study was to evaluate the feasibility and accuracy of left ventricular (LV) volume measurements by RT3DE using the single-beat full-volume capture technique.
Methods
One hundred nine consecutive patients scheduled for cardiac magnetic resonance imaging and RT3DE using the single-beat full-volume capture technique on the same day were recruited. LV end-systolic volume, end-diastolic volume, and ejection fraction were measured using an auto-contouring algorithm from data acquired on RT3DE. The data were compared with the same measurements obtained using cardiac magnetic resonance imaging.
Results
Volume measurements on RT3DE with single-beat full-volume capture were feasible in 84% of patients. Both interobserver and intraobserver variability of three-dimensional measurements of end-systolic and end-diastolic volumes showed excellent agreement. Pearson’s correlation analysis showed a close correlation of end-systolic and end-diastolic volumes between RT3DE and cardiac magnetic resonance imaging ( r = 0.94 and r = 0.91, respectively, P < .0001 for both). Bland-Altman analysis showed reasonable limits of agreement. After application of the auto-contouring algorithm, the rate of successful auto-contouring (cases requiring minimal manual corrections) was <50%.
Conclusions
RT3DE using single-beat full-volume capture is an easy and reliable technique to assess LV volume and systolic function in clinical practice. However, the image quality and low frame rate still limit its application for dilated left ventricles, and the automated volume analysis program needs more development to make it clinically efficacious.
Left ventricular (LV) volume and ejection fraction (EF) are important parameters for assessing LV function and remodeling and the efficacy of treatments in clinical trials. Because serial measurements and comparisons of the above parameters are required for these purposes, a reliable and reproducible measurement of LV volume is essential. Cardiac magnetic resonance imaging (CMR) is considered the most accurate in vivo method to assess LV volume and is currently the reference method used to validate other measurement modalities. However, because CMR is not readily available in many areas, requires a substantial amount of a patient’s time and cooperation, and incurs a high cost, echocardiography remains the most widely used modality to measure LV volume in clinical practice.
Echocardiographic measurement of LV volume has been conventionally performed using two-dimensional (2D) technology. However, measurement by 2D echocardiography requires various geometric assumptions that can be sources of error in many cases. Because three-dimensional (3D) echocardiography does not require similar assumptions, it has been thought to be the ideal echocardiographic method for the measurement of LV volume. Recently, real-time 3D echocardiography (RT3DE) has become a reality, with improved image acquisition processes and rendering techniques for LV volume measurements. However, certain limitations in the 3D analysis of LV volumes have prevented this technology from moving from the bench to the bedside. One of the major limitations of 3D echocardiographic volume analysis is that most current 3D echocardiographic techniques measure chamber volumes by stitching together multiple subvolume segments acquired from a series of electrocardiographically gated beats, because the image quality and acquisition width of a single-beat capture have not been sufficient for full-volume LV evaluation. However, multiple-beat image captures cannot acquire actual real-time images and are prone to errors in measurement.
With improvements in computer technology and the development of faster memory processing techniques, a new real-time 3D echocardiographic technology has been introduced that makes it possible to acquire a full LV volume from a single heartbeat. This technique also incorporates automated volumetric cardiac ultrasound analysis with pattern recognition technology. We sought to validate the LV volume measurement technique using single-beat full-volume RT3DE by comparing it with LV volume measured by CMR and to assess the feasibility of this technique in clinical practice.
Methods
Study Population
We enrolled consecutive patients in sinus rhythm who were scheduled for both CMR and echocardiography for various indications on the same day. Subjects with poor echocardiographic windows, defined as cases in which an experienced sonographer could not sufficiently discriminate the endocardial border on 2D echocardiography, were excluded. The protocol was approved by the Samsung Medical Center institutional review board.
Real-Time 3D Echocardiographic Image Acquisition
After CMR acquisition, echocardiography was performed using the Acuson SC2000 system (Siemens Medical Solutions USA, Inc., Mountain View, CA). Real-time 3D echocardiographic images were obtained from the apical window with subjects in end-expiration, with a few seconds of breath-hold in the left lateral decubitus position. LV full-volume real-time 3D echocardiographic images were acquired in a single beat by a single experienced sonographer, using a special transducer (4Z1c) that has a matrix array with a maximum volume angle of 90° × 90°. The volume angle was 70° in most of the cases and was optimized for each patient (especially in patients with dilated hearts) for full visualization of the endocardial border. The depth was adjusted to 13 to 15 cm to cover the entire left ventricle, and the frame rate was adjusted between 12 and 15 frames/min. Analysis was performed using an embedded program (Volume Cardiac Analysis Package–Volume Left Ventricular Analysis) run on the Acuson SC2000 system.
Real-Time 3D Echocardiographic Analysis
Two experienced observers (E.-Y.K. and S.-H.H.) who were blinded to the CMR data of the patients independently performed the analyses of the 3D echocardiographic data sets. Analyses were performed online using the 3D analysis software provided on the echocardiographic system. Analysis of the 3D images was performed with the onboard LV analysis program of the system using an auto-contouring algorithm and, if needed, manually corrected after a review of the traced border ( Figure 1 ). The auto-contouring algorithm was considered successful only when LV endocardial border delineation made by the algorithm was corrected no more than five times before the results were considered acceptable for analysis.
CMR
CMR was performed using a 1.5-T scanner (Avanto; Siemens Medical Systems, Erlangen, Germany). Cine magnetic resonance imaging was performed using a steady-state free precession imaging sequence after scout and localizer image acquisition. Slice thickness was set at 6 mm and slice gap at 4 mm, and the short-axis images of the left ventricle were acquired from the apex to the base to include the entire LV volume. Repeated breath-holds were required to create adequate images. Temporal resolution was 25 to 30 frames per RR interval. LV volume analyses from magnetic resonance images were performed using a commercialized software (Argus version 4.02; Siemens Medical Systems) by a single experienced observer who was blinded to the echocardiographic data, as previously described. End-diastolic and end-systolic frames were defined as the frames in which the cavity sizes were largest and smallest by retrospective image review. The endocardial border was manually traced in the selected image frames. The papillary muscles and LV trabeculae were excluded from the endocardium and included in the LV cavity volume. At the base of the heart, slices were considered to be within the left ventricle if the blood volume was surrounded by ≥50% of ventricular myocardium.
Statistical Analysis
LV end-systolic volume (ESV), end-diastolic volume (EDV), and EF are represented as median (range). Correlations were performed between 3D echocardiography and CMR. Agreement was expressed using Bland-Altman plots with mean differences and 95% limits of agreements. Interobserver and intraobserver variability of real-time 3D echocardiographic measurement were obtained from ESV and EDV measurements from 10 randomly selected patients. Linear regression, interclass correlation, Bland-Altman analysis, and measurement of coefficients of variation were used for comparisons. P values < .05 were considered statistically significant.
Results
Feasibility of RT3DE Using the Single-Beat Full-Volume Capture System
One hundred nine patients were enrolled in the study ( Table 1 ). Forty-two patients had apparently normal hearts, which were defined according to 2D echocardiographic results when chamber size and systolic function were within normal limits, without significant valve disease, regional wall motion abnormality, or myocardial hypertrophy. The remaining 67 patients had significant heart disease according to 2D echocardiography. Thirty-two patients had ischemic cardiomyopathy, 21 had hypertrophic cardiomyopathy (HCM), eight had presumptive nonischemic cardiomyopathy, and six had other cardiac diseases.
| Variable | Value |
|---|---|
| Age (y) | 53.4 ± 13.9 |
| Men | 78 (71.6%) |
| Presumptive diagnosis | |
| No significant heart disease | 42 (38.5%) |
| Ischemic cardiomyopathy | 32 (29.4%) |
| HCM | 21 (19.3%) |
| Nonischemic cardiomyopathy | 8 (7.3%) |
| Other | 6 (5.5%) |
| LV ESV (mL) by CMR | 39.2 (11.7–282.5) |
| LV EDV (mL) by CMR | 137.0 (70.6–378.9) |
| LV EF (%) by CMR | 70.0 (19.4–87.8) |
| LV stroke volume (mL) by CMR | 87.9 (49.6–143.0) |
Although we recruited only patients with optimal 2D image quality, eight patients were excluded from final volume validation because of poor 3D echocardiographic image quality (two patients with normal left ventricles and six patients with HCM). In addition, 10 patients with large hearts could not be analyzed, because the entire LV volume could not be included in the pyramidal acquisition window (all were patients with ischemic or nonischemic cardiomyopathy). The maximum EDV and ESV by CMR that could be assessed on RT3DE were 272.4 and 201.4 mL, respectively. The time required for real-time 3D echocardiographic full-volume acquisition was 2.55 ± 0.73 min.
We analyzed 3D echocardiographic data from a total of 91 patients. In this study population, the feasibility of single-beat full-volume LV capture with RT3DE was 83.5%. The mean frame rate was 13.1 ± 1.32 frames/sec.
Validation of LV Volume and EF with CMR
ESV and EDV calculated from RT3DE showed excellent correlations with the same measurements from CMR data ( r = 0.94 and r = 0.91, respectively; Figures 2 A and 2 C). A Bland-Altman plot showed good limits of agreement ( Figure 2 B and 2 D); biases and limits of agreements were −7.91 and ±33.06 mL for ESV and −41.38 and ±36.51 mL for EDV, respectively (test for the significance of bias, P < .001 for both ESV and EDV). Bias was negative for both ESV and EDV and larger for EDV compared with ESV. LV volumes measured by RT3DE tended to be smaller than those measured by CMR, and the slopes of the lines ( Figure 2 A and 2 C) were <1 (0.66 and 0.70). LV EF values from RT3DE showed good correlation with EF measurements from CMR data ( r = 0.91), but the slope of the line was <1 ( Figure 2 E), and Bland-Altman analysis showed negative bias ( Figure 2 F).
Interobserver and Intraobserver Variability
Both interobserver and intraobserver variability showed excellent agreement. The interclass correlation was >0.99 ( P < .001). The Bland-Altman plot also showed reasonable limits of agreement for both interobserver and intraobserver variability ( Figure 3 ).
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
