Methods

This was a single-center prospective research trial. The institutional review board approved the study protocol. Informed written consent was obtained from all patients. Any adolescent or young adult followed in our center (1) with a cardiac diagnosis of d-TGA repaired with an atrial switch and (2) >13 years of age was eligible to be included. Specific exclusion criteria consisted of (1) known arrhythmias, (2) presence of implanted pacemaker leads or cardioverter–defibrillator, or (3) lack of consent to participate in the protocol. Patients were recruited for 2DE-KBR and CMR imaging on the same day. All studies were performed exclusively for research purposes. Patients were not screened for difficult acoustic windows or 2D image quality before inclusion.

CMR imaging was performed immediately before 2DE-KBR. CMR imaging used a 1.5-T scanner (Achieva 3.2.1.0, Philips, Best, the Netherlands) with a 32-channel cardiac coil. Cardiac synchronization was performed with vector electrocardiography. Ventricular volumes and function were assessed using steady-state free-precession cine sequences (repetition time 500 ms, echo time 60 ms, flip angle 60°, field of view 180 × 180 mm, voxel size 1.71) during brief periods of breath-holding in the following planes: ventricular 2-chamber, 4-chamber, left and right ventricular outflow tracts, short-axis stack (12 to 14 equidistant slices with slice thickness 7 mm, no interslice gap), axial stack (slice thickness 8 mm, interslice gap 1), and multiple oblique coronal planes parallel to the superior and inferior cavae.

Echocardiographic examinations were performed using a Vivid 7 system (GE, Milwaukee, Wisconsin) connected to a Windows XP-based computer (Ventripoint, Seattle, Washington). A single sonographer (A.P.) with 5 years of experience acquired images in all patients. This sonographer had received 5 hours of hands-on training with 2DE-KBR acquisition in 3 healthy volunteers before scanning the first patient enrolled. The 2DE-KBR system employs a commercially available electromagnetic field generator and a field disturbance sensor firmly attached to the transducer to enable high-precision positional information for each scan plane in 3-dimensional space. Ultrasound and positional data are then stored for processing into 3-dimensional representations and surface rendering of the RV.

Acoustic captures (3 seconds) with the best possible coverage of the systemic RV and RV-specific rotated (foreshortened apical) and inflow–outflow (parasternal long-axis and foreshortened apical) views were obtained. The patient was asked to hold his/her breath for up to 5 seconds during the acquisition of each image and was not allowed to change position during the entire acquisition. The times taken by the sonographer from start to finish for 2 separate 2DE-KBR acquisitions in each patient were noted.

KBR postprocessing was performed independently by 2 blinded observers (A.P. and P.G.). These observers had 1 year of cardiac imaging research experience and received 6 hours of training in 2DE-KBR analysis. All acquired views in each study were used for analysis. Analysis involved 3 steps as shown in Figure 1 : (1) selection of end-diastolic and end-systolic frames, (2) point placement at designated anatomic RV landmarks, and (3) 3-dimensional reconstruction. Using chamber size and valve motion, the accuracy of the default selection of end-diastolic and end-systolic frames in all cine loops was confirmed or modified by the user as necessary.

Three-dimensional reconstruction of the systemic right ventricle in a 26-year old patient with atrial switch for d-transposition of great arteries. (Top panel) Determination of end-diastole and end-systole on each required view and point placements in these views. (Middle panel) Right ventricular endocardial and septal borders are shown on each image. (Bottom panel) Two-dimensional echocardiography with knowledge-based reconstructions in end-diastolic and end-systolic frames.

Points were placed on anatomic structures for the RV as mandated by protocol with the aid of a 3-dimensional image viewer. This ensured that the points were well distributed around the structure ( Figure 2 ). The structures were the pulmonary annulus (2 points), tricuspid annulus (3 points), RV apex (1 point), subpulmonary region (1 point), RV septum (4 to 5 points), and RV endocardium (4 to 5 points) as described previously. Endocardial points were placed at the base of the RV trabeculations. A minimum of 15 points was placed for each study. Point placements by the 2 observers in all studies were checked for anatomic accuracy and corrections were made if necessary. The reconstructed RV volume overlaid on each 2D image frame was reviewed in end-diastole and end-systole and optimized as needed by addition or deletion of points and revised reconstruction. The system communicated by a secured internet connection to a Web-based server where surface reconstruction of the RV was created using the points entered on the 2D images and the 3-dimensional coordinates acquired from the tracking system ( Figure 2 ).

Right ventricular knowledge-based reconstruction model with different views enables evaluation of the 3-dimensional shape of the reconstruction. The reconstruction can be repeated (if necessary) by placing additional points on the 2-dimensional image. (A) Three-dimensional right ventricular reconstruction model with scan plane. (B) Right ventricular reconstruction mesh view with spatially localized points. (C) Right ventricular reconstruction mesh view with intersections. (D) Combined end-diastolic and end-systolic image of the reconstruction.

For CMR analysis measurements of left ventricular and RV end-diastolic and end-systolic volumes were obtained from the short-axis cine stack by manual tracing of endocardial contours. Using commercially available software package (QMass MR 7.2, Medis, Leiden, the Netherlands) all measurements were made by a third observer (L.L.) blinded to the results of 2DE-KBR analyses. Papillary muscles and trabeculae were included in the volume to provide better reproducibility for RV measurements. To ensure accuracy, CMR tracings were performed with the assistance of the long-axis images to visually confirm the basal and apical slices. Volumes measured by the 2 methods were indexed to body surface area. Times taken for postprocessing 2DE-KBR and CMR images in each study were recorded. 2DE-KBR–derived RV volumes and EF were compared to the CMR measurements. 2DE-KBR measurements in all patients were compared between the 2 observers to assess interobserver variability. The first observer performed repeated 2DE-KBR measurements 4 weeks apart in all patients to assess intraobserver variability.

RV volumes and EFs by 2DE-KBR and CMR are reported as mean ± SD. For comparisons between methods, the average of the 2 observers was used as the 2DE-KBR measurement and paired-samples t test was applied. The Bland–Altman method was used to determine inter- and intraobserver agreements for repeated KBR measurements. Statistical significance was defined as a p value <0.05. Statistical analysis was performed with Minitab 16.1 (Minitab, Inc., State College, Pennsylvania).

Results

From January 2011 through July 2011, 15 patients were studied. Examinations were done 23 ± 6 years after atrial switch. Median patient age was 24 ± 7 years (13 to 46); median weight was 67 ± 12 kg (50 to 95). The qualitative 2D echocardiographic grade of tricuspid regurgitation was mild in all patients. All patients completed the 2DE-KBR protocol and image analysis was feasible in all studies. One patient developed claustrophobia during the CMR component of the study. Although the acquisition protocol was incomplete in this patient, the cine images required for measuring RV volumes and function had been obtained before termination of the study and were therefore suitable for analysis. Demographics and acquisition and postprocessing times are listed in Table 1 .

Measurements of RV end-diastolic volume, end-systolic volume, and EF ( Table 2 ) correlated well between methods ( Figure 3 ). Mean indexed RV end-diastolic volumes were 122 ± 4 ml/m ² with 2DE-KBR and 128 ± 4 ml/m ² with CMR (r = 0.80); mean indexed RV end-systolic volumes were 71 ± 5 ml/m ² with 2DE-KBR and 73 ± 4 ml/m ² with CMR (r = 0.81); and mean EFs were 42 ± 10% with 2DE-KBR and 43 ± 7% with CMR (r = 0.86). Paired t testing showed good agreement of 2DE-KBR with CMR for end-diastolic volume, end-systolic volume, and EF. KBR tended to underestimate end-diastolic volume and end-systolic volume compared to CMR; mean underestimations were 4.3% for end-diastolic volume and 2.3% for end-systolic volume.

Table 2

Measurements from two-dimensional echocardiography with knowledge-based reconstruction and cardiovascular magnetic resonance

Patient No.	2DE-KBR					CMR
	EDV (ml)	ESV (ml)	EDVI (ml/m 2 )	ESVI (ml/m 2 )	EF (%)	EDV (ml)	ESV (ml)	EDVI (ml/m 2 )	ESVI (ml/m 2 )	EF (%)
1	211	100	113	54	53	223	119	119	64	47
2	195	120	117	78	39	182	101	109	61	44
3	215	130	130	78	40	203	111	123	67	46
4	162	116	105	75	29	184	126	119	82	31
5	228	136	122	73	40	214	131	114	70	39
6	283	182	150	96	36	307	190	162	101	38
7	139	60	87	37	57	215	111	135	70	49
8	212	120	112	63	43	234	126	124	67	46
9	230	138	112	67	40	223	119	108	58	47
10	233	150	130	84	36	258	160	144	90	38
11	236	153	121	79	35	259	158	133	81	39
12	189	62	112	37	67	194	76	115	45	61
13	260	146	151	85	44	252	131	147	76	48
14	264	154	138	81	42	272	174	142	91	36
15	257	174	133	89	33	246	151	127	78	39

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Meta-Analysis of Ten Trials on the Effectiveness of the Radial Versus the Femoral Approach in Primary Percutaneous Coronary Intervention
2. Relation of Left Atrial Spontaneous Echocardiographic Contrast in Patients With Mitral Stenosis to Inflammatory Markers
3. Incidence of Cardiac Arrhythmias in Asymptomatic Hereditary Hemochromatosis Subjects With C282Y Homozygosity
4. Goldberger’s Electrocardiographic Triad in Patients With Echocardiographic Severe Left Ventricular Dysfunction

Stay updated, free articles. Join our Telegram channel

Join