Background
The aims of this study were to determine the usefulness of hand-carried ultrasound devices in pediatric cardiology and to compare the performance of three different hand-carried ultrasound devices in a pediatric cardiology outpatient clinic and intensive care unit.
Methods
One hundred ten patients (49 male; mean age, 6.4 ± 5.2 years; range 0.1–38 years) with congenital heart defects or innocent heart murmurs were examined using Siemens Acuson P10, Siemens Acuson P50, and Philips CX 50 systems. The quality of images and the accuracy of B-mode measurements were compared with those obtained using a standard echocardiographic system (Philips iE33).
Results
Fifty-nine patients were examined with the Siemens Acuson P10, 29 with the Siemens Acuson P50, and 22 with the Philips CX 50 system. There were no significant differences in B-mode measurements. The Acuson P10 system, however, showed significantly lower image quality, with 64.54% of all studies considered of excellent quality compared with 92.83% with the Acuson P50 and 95.52% with the CX 50 ( P < .05) and a mean quality score (1 = fair, 5 = excellent) of 3.5 versus 4.57 with the Acuson P50 and 4.86 with the CX 50 ( P < .05). This was attributed to the limited capacity for accurate diagnosis in children with body weights < 10 kg and complex heart defects.
Conclusion
Hand-carried ultrasound devices represent a valuable alternative to standard echocardiographic systems in pediatric cardiology. In particular, systems including all echocardiographic modalities offer unlimited versatility in outpatient and intensive care.
Modern techniques of ultrasound transmission and signal processing allow for continuing improvements in noninvasive imaging. Especially in pediatric cardiology, such systems must provide high spatial and temporal resolution, because the anatomic structures shown are small and heart rates are high. High-end echocardiographic systems have the disadvantage of being bulky and heavy systems that are difficult to handle in the setting of neonatal and pediatric intensive care units (ICUs) with limited space. Thus, hand-carried ultrasound (HCU) devices should reduce this disadvantage and combine good images with device flexibility. Several devices are commercially available; we tested three in the setting of a tertiary health care center, examining 110 children with cardiac disease in an outpatient clinic and ICU. The aim of this investigation was the comparison of image quality among the HCU devices, with images obtained with a full-sized ultrasound machine.
Methods
Patients
In total, a group of 110 patients was recruited from the patients seen in the outpatient clinic and the ICU of the Department of Pediatric Cardiology, Ludwig-Maximilians-University (Munich, Germany). All patients were consecutively included in our study during a period of 2 weeks as they were admitted to our ICU or presented to our outpatient clinic. All patients were screened following the same standard protocol of echocardiography. Patients were excluded from the study if a full echocardiographic examination was not possible with the full-sized ultrasound machine (see below for details) for technical reasons, such as difficult conditions in the ICU or poor patient collaboration in the outpatient clinic. The patients’ characteristics are given in Table 1 . Because our institution focuses on patients with congenital and acquired heart defects, these represented the majority in our patient groups. Some patients presented for the evaluation for heart murmurs at the outpatient clinic. In the ICU, the patients were examined shortly after cardiac surgery or transplantation. The heart defects diagnosed are given in Table 2 . All patients or their relatives gave written informed consent; the study was approved by the local ethics committee and followed the Declaration of Helsinki.
| Siemens Acuson P10 | Siemens Acuson P50 | Philips CX 50 | |
|---|---|---|---|
| Variable | ( n = 59) | ( n = 29) | ( n = 22) |
| Male/female | 33/26 | 17/12 | 9/13 |
| Age (years) | 7.6 ± 6.4 (0.1–22.4) | 5.1 ± 7.8 (range 0.1–38) | 5.4 ± 6.4 (range 0.1–20.1) |
| Weight (kg) | 26.6 ± 21.6 (2–82) | 17.8 ± 18.2 (2.4–78) | 18.5 ± 18.6 (2.8–67) |
| Height (cm) | 116.3 ± 40 (45–176) | 93.9 ± 39 (42–181) | 97.9 ± 39 (52–173) |
| Siemens Acuson P10 | Siemens Acuson P50 | Philips CX 50 | Inpatients | Outpatients | Total | |
|---|---|---|---|---|---|---|
| Diagnosis | ( n = 59) | ( n = 29) | ( n = 22) | ( n = 22) | ( n = 88) | ( n = 110) |
| Orthotopic heart transplantation | 3 | 2 | 1 | 6 | 6 | |
| VSD | 7 | 6 | 1 | 14 | 14 | |
| Tetralogy of Fallot/double-outlet right ventricle | 5 | 1 | 1 | 7 | 7 | |
| Transposition of the great arteries | 1 | 1 | 2 | 2 | ||
| Univentricular heart | 20 | 8 | 8 | 15 | 21 | 36 |
| Truncus arteriosus communis | 1 | 1 | 2 | 2 | ||
| Coarctation of the aorta | 2 | 2 | 4 | 4 | ||
| Interrupted aortic arch | 2 | 1 | 3 | 3 | ||
| Innocent heart murmur | 4 | 4 | 1 | 9 | 9 | |
| Bicuspid aortic valve | 1 | 1 | 2 | 2 | ||
| Atrioventricular septal defect | 2 | 1 | 1 | 4 | 4 | |
| Patent foramen ovale | 1 | 1 | 2 | 2 | ||
| Third-degree atrioventricular block | 1 | 1 | 2 | 2 | ||
| Dilated cardiomyopathy | 1 | 1 | 2 | 2 | ||
| Idiopathic pulmonary hypertension | 1 | 1 | 2 | 2 | ||
| Pulmonary valvar stenosis | 1 | 1 | 2 | 2 | ||
| Supraventricular tachycardia | 1 | 1 | 2 | 2 | ||
| Pulmonary atresia/VSD | 1 | 1 | 2 | 2 | ||
| Atrial septal defect | 4 | 1 | 2 | 7 | 7 |
Echocardiography
All patients were examined by a single sonographer with the assistance of the respective manufacturer’s sales representative to optimize the images. For each patient, the first examination was performed with one of the following HCU devices, and the second examination was performed with the reference system. (For a summary of the technical specifications of the three HCU systems, see Table 3 .)
| Specification | Siemens Acuson P10 | Siemens Acuson P50 | Philips CX 50 |
|---|---|---|---|
| Weight (kg) | 0.725 | 5.4 | 6.1 |
| Dimensions (cm) | 5.6 × 9.7 × 14.6 | 5.1 × 24.9 × 36.4 | 7.5 × 26 × 38 |
| Display diagonal size (cm) | 9.4 | 39.12 | 39.12 |
| Display resolution (pixels) | 640 × 480 | 1,280 × 768 | 1,400 × 1,050 |
| Frame rate (frames/sec) | 28 | 150 | 90 |
| Transducer (MHz) | 2–4 | 1.67–4 | 1–5 |
Siemens Acuson P10
The Siemens Acuson P10 (Siemens Medical Systems, Erlangen, Germany) was tested, with two-dimensional mode. This device weighs 725 g (including the transducer), with a 3.7-inch liquid crystal display, display resolution of 640 × 480 pixels, and dimensions of 56 × 97 × 146 mm. The system operates a 2-Mhz to 4-MHz phased-array transducer with 64 elements and a 16.6 × 14 mm array footprint. The images were displayed on the liquid crystal display with a maximal frame rate of 28 frames/sec. A lithium ion battery allows for 1 hour of work; fully charging the battery in the docking station requires 2 hours. The software installed includes application presets for cardiac, abdominal, and obstetric imaging. Images are processed using adjustable gain and automatic time gain compensation. Storage and retrieval of patient and study data are performed using digital acquisition of still-frame and cine imaging data (2-sec clips) with up to 30,000 still frames or 500 cine loops on a 1-GB memory card. Transfer to a PC via a universal serial bus 1.1 interface can be performed to review studies using the Acuson P10 Viewer ultrasound software (Siemens Medical Systems) in offline mode. For image acquisition, the preinstalled cardiac preset was used ( Figure 1 ).
Siemens Acuson P50
The Siemens Acuson P50 (Siemens Medical Systems) was tested, with all modalities of modern echocardiography, including two-dimensional mode, color Doppler mode, M- ode, pulsed-wave and continuous-wave Doppler modes, and tissue Doppler. This system is connected to an Apple MacBook Pro (Apple Computer, Cupertino, CA) notebook with 4 GB of random-access memory, a 250-GB hard drive, a 2.53-GHz Intel Core Duo processor (Intel Corporation, Santa Clara, CA), and Windows XP (Microsoft Corporation, Redmond, WA) and Mac OS X Snow Leopard (Apple Computer) operating systems for electronic operation. The device weighs 5.4 kg, its size is 364 × 249 × 51 mm with a 15.4-inch display, and it has display resolution of 1,280 × 768 pixels (ultrasound) and of 2,048 × 1,536 pixels (Apple MacBook Pro) and a response time of 1 msec. In two-dimensional mode, the frame rate is 150 frames/sec, automatic image optimization is performed by tissue harmonic imaging and SieVision software (Siemens Medical Systems), and manual corrections are possible with time gain compensation (eight-zone slide controls) and lateral gain compensation (eight-zone slide controls). This system was operating a 4V2 phased-array transducer at 1.67 to 4 MHz, with 64 elements and a footprint of 16.3 mm. Digital storage of the still frames and cine loops was performed on the local hard disk of the notebook in Digital Imaging and Communications in Medicine format. For review and offline processing purposes, the Syngo US Workplace (Siemens Medical Systems) could be used ( Figure 2 ).
Philips CX 50
The Philips CX 50 (Philips Medical Systems, Andover, MA) system was tested, with all modalities of modern echocardiography (two-dimensional B mode, M mode, color Doppler, and pulsed-wave and continuous-wave Doppler) operating an S5 1-MHz phased-array pure-wave crystal transducer with a footprint of 16.6 mm. The device weighs 6.1 kg (7.3 kg including its battery), its size is 380 × 260 × 75 mm, and it has a 15.4-inch liquid crystal display and a resolution of 1,400 × 1,050 pixels. The maximal frame rate of the system is 90 frames/sec at a 15° sector size at 30-cm image depth and 90 frames/sec at a 90° sector size at 10-cm image depth. Image optimization is performed online using manual time gain compensation (eight sliders) and lateral gain compensation and automatically using the Philips iScan automatic image optimization module and the Philips XRES adaptive image processing technique for noise and artifact reduction. The images were stored digitally in Digital Imaging and Communications in Medicine format on the local hard disk of the device ( Figure 3 ).
The second echocardiographic examination was performed in all patients by the same sonographer using a Philips iE33 system of the latest generation (Philips Medical Systems) operating S12 8-MHz, S8 4-MHz, and S5 1-MHz phased-array transducers. This system is routinely used for echocardiography at our institution and was used as a reference system for this study, with the images and loops stored on a central server. The evaluation of the echocardiograms for final diagnoses was performed using the Philips Xcelera platform (Philips Medical Systems) ( Figure 4 ).
A subgroup of patients with univentricular heart (21 of 36 patients, including 15 patients with hypoplastic left-heart syndrome, five with tricuspid atresia, and one with double-outlet right ventricle and mitral atresia) and with ventricular septal defects (VSDs; 10 of 14 patients) was reinvestigated by the same sonographer with all three HCU devices and the full-sized ultrasound system randomly.
In all patients, the images obtained by standard echocardiographic cross-sectional views were classified by an external observer (an experienced pediatric cardiologist with special training in noninvasive imaging). Because the images were displayed best on the systems’ own screens (especially for the Acuson P10, as Acuson P10 Viewer shows only low-resolution images), the revision of the images was performed offline for each system used. The quality of the images was quantified using a scoring system that included the visibility of key structures in the four-chamber view and parasternal long-axis cine loop and still frame at end-diastole and end-systole. The key structures were defined as the leaflets and chordae tendineae of the mitral and tricuspid valves (four-chamber view) and the leaflets of the aortic and mitral valve (parasternal long-axis view). One of five scores was assigned: 1 = poor quality, leaflets not clearly visible in still frames; 2 = fair quality, leaflets clearly visible in still frames, not visible in cine loops; 3 = good quality, leaflets clearly visible in still frames and cine loops; 4 = very good quality, leaflets clearly visible, chordae tendineae visible in still frames; and 5 = excellent quality, leaflets and chordae tendineae clearly visible in still frames and cine loops. Because the analysis was performed on the screens of the systems, a blinded analysis was not possible. We therefore performed the analysis of the images obtained in unblinded paired data (first the HCU images, then the iE33 images). Last, the diameters of the four cardiac valves were compared: the diameters of the tricuspid valve ring and of the mitral valve ring were measured in the four-chamber view as the maximal distance between the hinge points of the leaflets in diastole, and the diameter of the aortic valve ring was measured in the parasternal long-axis view and that of the pulmonary valve ring in the parasternal short-axis view, both in systole. In patients with univentricular hearts, the valve diameters were measured in the existing valves (i.e., the pulmonary valve, tricuspid valve, and aortic valve in hypoplastic left-heart syndrome with mitral atresia).
Statistical Analysis
Calculations were performed using SPSS for Windows version 16.0 (SPSS, Inc., Chicago, IL). Differences between patients’ first and second echocardiographic examination results were tested using a unifactorial analysis of variance. Correlations were analyzed using Pearson’s correlation coefficient. All significance testing was fixed at P < .05 (two sided).
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
