Application of Prospective ECG-Gated High-Pitch 128-Slice Dual-Source CT Angiography in the Diagnosis of Congenital Extracardiac Vascular Anomalies in Infants and Children

Purpose To investigate the value of prospective ECG-gated high-pitch 128-slice dual-source CT (DSCT) angiography in the diagnosis of congenital extracardiac vascular anomalies in infants and children in comparison with transthoracic echocardiography (TTE). Methods Eighty consecutive infants or children clinically diagnosed of congenital heart disease and suspected with extracardiac vascular anomaly were enrolled, and 75 patients were finally included in this prospective study. All patients underwent prospective ECG-gated high-pitch DSCT angiography after TTE with an interval of 1–7 days. The diagnostic accuracy and sensitivity of high-pitch DSCT angiography and TTE were compared according to the surgical/CCA findings. The image quality of DSCT was assessed using a five-point scale. The effective radiation dose (ED) was calculated. Results A total of 17 congenital heart diseases and 162 separate extracardiac vascular anomalies were confirmed by surgical/CCA findings in 75 patients. The diagnostic accuracy of high-pitch DSCT angiography and TTE was 99.67% and 97.89%, respectively. The sensitivity of high-pitch DSCT angiography and TTE was 97.53% and 79.62%, respectively. There was significant difference regarding to the diagnostic accuracy and the sensitivity between high-pitch DSCT angiography and TTE (χ2 = 23.561 and 28.013, P<0.05). The agreement on the image quality scoring of DSCT between the two observers was excellent (κ = 0.81), and the mean score of image quality was 4.1±0.7. The mean ED of DSCT was 0.29±0.08 mSv. Conclusions Prospective ECG-gated high-pitch 128-slice DSCT angiography with low radiation dose and high diagnostic accuracy has higher sensitivity compared to TTE in the detection of congenital extracardiac vascular anomalies in infants and children.


Introduction
Precise and comprehensive evaluation of the intracardiac and extracardiac deformities in congenital heart diseases (CHD) is critical for surgical plan. Transthoracic echocardiography (TTE) is the first-line option for children with CHD because of its availability, safety and capacity to provide hemodynamic parameters using Doppler flow studies. TTE combined with the Doppler flow imaging has the advantage in the diagnosis of intracardiac deformities, especially the small atrial septal defect and ventricular septal defect. However, TTE is not robust for evaluating extracardiac structures because it is limited by its acoustic window and relatively lower spatial resolution [1][2][3][4].
The advanced multi-detector CT (MSCT) technology with improved spatial and temporal resolution has made it an effective tool in the assessment of CHD in infants and children [5][6][7]. The value of MSCT in the evaluation of the thoracic vascular anomalies has been widely confirmed. As this imaging modality can not only provide cardiovascular information, but also offer additional information about the airway and lung parenchyma, resulting in a greater anatomic coverage. A recent introduced dual-source CT (DSCT) system (Definition Flash, Siemens Healthcare, Forchheim, Germany) provides the high-pitch scan technique. In this high-pitch mode, data acquisition is prospectively triggered with the ECG. With a very high pitch of 3.4 and a table speed of 460 mm/s, the entire heart can be covered within one single cardiac cycle. Due to the fast, non-overlapping spiral data acquisition, the radiation dose was consequently reduced [8,9].
The prospective ECG-gated sequential technique has been proven one of the most effective strategies to lower the radiation dose. The sequential mode allows high diagnostic accuracy in the assessment of intracardiac, extracadiac and coronary anatomies in children with CHD, and the image quality is seldom influenced by the variability of the heart rates [10][11][12][13]. Several studies have found that the new high-pitch mode is feasible in adults with stable and low heart rates below 65 or 70 beats per minute (bpm) [14][15][16], and its initial experience in children has been reported. With the relatively high and irregular heart rates in children with CHD, the starting position of the data acquisition in high-pitch mode usually could not correspond to the preselected starting phase, and sometimes the R wave might be covered within the acquisition window, resulting in the poor image quality of the intracardiac and coronary structures [17]. In our previous studies on high-pitch mode, we found that the image quality of extracardiac great vessels had not been influenced by the variability of heart rate, and the high-pitch mode significantly lowered the radiation dose as compared to the sequential mode. Therefore, the high-pitch mode has advantages over sequential mode in demonstrating extracardiac vascular anomalies.
The purpose of this study was to investigate the value of prospective ECG-gated high-pitch 128-slice DSCT angiography in the diagnosis of congenital extracardiac vascular anomalies in infants and children in comparison with TTE.

Patients
This study received approval from the institutional review board of Shandong Medical Imaging Research Institute and written informed consent was obtained from the parents of all patients. We have obtained written consent from the parents of the patients for publication of medical images used in the Figures.

DSCT protocol
All examinations were performed on a 128-slice DSCT scanner (Somatom Definition Flash, Siemens Healthcare, Forchheim, Germany). Short-term sedation was achieved with oral administration of chloral hydrate. All patients were freebreathing. The scans were performed in cranio-caudal direction from the thoracic outlet to the bottom of the heart. Iodinated contrast medium (Schering Ultravist, Iopromide, 350 mg I/mL, Berlin, Germany) was injected via peripheral veins at a volume of 1.5 mL/kg body weight. The injection time of contrast medium was 15 seconds. Injection rate was calculated at total injected volume of contrast medium divided by 15 seconds. In order to reduce the artifact of contrast medium in superior vena cava, a saline chaser of 1.0 mL/kg body weight was injected at the same rate of contrast medium. A ''manual'' bolus-tracking technique was used, the monitoring section was set at the four chambers of the heart and the ROI was placed at the air outside the thorax. Monitoring started 18 seconds after injection. The acquisition was manually triggered at the moment that the contrast medium opacification within both the right heart and left heart was achieved and the contrast medium artifact in the right atrium began to disappear.
The high-pitch mode was used. Data acquisition was prospectively ECGtriggered starting at 10% of the R-R interval using a pitch of 3.4. CT parameters were as follows: 0.28 s gantry rotation time, 266460.6 mm detector collimation, a slice collimation 2612860.6 mm by z-flying focal spot technique, 80 kV tube voltage and weight adapted setting for tube current (60 mAs/rotation for patients ,5 kg body weight, 60-79 mAs/rotation for patients 5-10 kg body weight, 80-120 mAs/rotation for patients.10 kg body weight).

DSCT data post-processing and analysis
Images were reconstructed with a slice thickness of 0.75 mm and increment of 0.5 mm using a medium smooth-tissue convolution kernel (B26f). All images were transferred to an external workstation (Multiple Modality Workplace, Siemens Healthcare, Forchheim, Germany) for post-processing. Multiplanar reformation (MPR), maximum intensity projection (MIP) and volume rendering (VR) were used for image interpretation.
Blinded to the results of TTE, surgical and/or CCA findings, two cardiac radiologists with more than 5 years' experience interpreted the image quality using a 5-grade scoring system (5, excellent; 4, good; 3, fair; 2, insufficient for complete evaluation; 1, not interpretable) [18]. Consensus agreement was achieved between the two observers if disagreement existed. Grades 3, 4 and 5 were considered sufficient for complete diagnosis.

Radiation dose estimations
The volume CT dose index (CTDIvol) and dose-length product (DLP) were obtained from the CT system and the effective radiation dose (mSv) was calculated from the DLP (mGy?cm) multiplied by 2.3 to adapt it to the 16-cm phantom (the DLP for the body surface area was given for a 32-cm phantom on the scanner protocol and the conversion factor of 2.3 is scanner specific for paediatric examinations at 80 kV as provided by the manufacturer). The corrected DLP value was then multiplied by the specific conversion coefficients given for a 16-cm phantom: 0.039 mSv/[mGy?cm] for children up to 4 months, 0.026 mSv/ [mGy?cm] between 4 months and 1 year of age, and 0.018 mSv/[mGy?cm] between 1 year and 6 years of age [12,13].

Transthoracic echocardiography
All patients underwent two-dimensional TTE and Doppler Flow from the parasternal, apical, subxiphoid and suprasternal approaches using a SONOS 7500 ultrasound system (Philips Medical System, Bothell, WA). The examinations were performed by an experienced echocardiographic technician, and the data were evaluated by a trained pediatrician.

Statistics
Statistical analysis was performed by SPSS 17.0 software (SPSS, Chicago, IL, USA). Results were expressed as means ¡ standard deviations for quantitative variables and as frequencies or percentages for categorical variables. With surgical and/or CCA results as the standard, the diagnostic accuracy and sensitivity of DSCT and TTE for the separate cardiovascular abnormalities was respectively calculated. Comparative analysis of the diagnostic accuracy and sensitivity between DSCT angiography and TTE was obtained using the non-parametric chi-square test. Interobserver agreement on grades of image quality was assessed by kappa statistics (k.0.81, excellent agreement; k50.61-0.80, good agreement). P,0.05 was considered statistically significant.

Diagnostic performance
Taking surgical and/or CCA findings as the reference standard, 2 BAVs and 1 single coronary artery were not identified by high-pitch DSCT angiography. Highpitch DSCT angiography misdiagnosed a normal arterial ligament as PDA and a severe pulmonary artery stenosis as pulmonary artery atresia. TTE failed to identify 2 PAPVRs, 1 aortic dysplasia, 2 CoAs, 5 pulmonary artery stenoses, 1 dilated pulmonary artery, 4 pulmonary slings, 1 PDA, 8 MAPCAs, 1 PLSVC and 3 coronary artery anomalies. TTE misdiagnosed 1 case of pulmonary artery atresia as truncus arteriosus, 2 cases of absence of one pulmonary artery as pulmonary sling, 1 CoA as IAA (type A) and 1 MAPCA as PDA. Although 3 TAPVRs were accurately diagnosed by TTE, 1 mixed type was misdiagnosed as cardiac type. Table 1 demonstrates the details on separate cardiovascular abnormalities. Four cases of congenital extracardiac vascular anomalies are shown in Figs. 1-4.

Image quality assessment
Diagnostic CT images (images graded 3 or more) were obtained in all patients. The mean image quality score was 4.1¡0.7, and distributed as score 3 (n515, 20%), score 4 (n535, 47%) and score 5 (n525, 33%). The agreement on the overall image quality scoring between the two observers was excellent (k50.81).

Discussion
Our study demonstrates that prospective ECG-gated high-pitch 128-slice DSCT angiography shares high diagnostic accuracy and sensitivity in the assessment of congenital extracardiac vascular anomalies in children in comparison with TTE.
With the image quality of DSCT maintained satisfactory, the radiation dose is lowered to 0.29¡0.08 mSv.
The modalities in assessment of congenital extracardiac vascular anomalies TTE is adopted usually for the initial assessment of children with CHD. TTE combined with Doppler flow imaging has the advantage in the diagnosis of intracardiac deformities. However, as it is limited by the acoustic window, small field of view (FOV) and lower spatial resolution and is obscured by overlying bone and aerated lung, TTE has relative difficulties in visualizing some extracardiac deformities.
CCA can provide both haemodynamic and anatomical information in a single procedure, with high spatial resolution on the images. The major arguments not to use this modality are the fact it is invasive and runs a 1% procedure-related mortality risk in neonates [19], the long procedure time, possible need for sedation and relatively high doses of ionizing radiation.
As a noninvasive and non-ionizing method, MR which can provide both morphologic and functional information has become an effective modality in the assessment of congenital extracardiac anomalies. However, the long acquisition time may necessitate long periods of sedation for small children which is not desirable. In addition, the spatial resolution is also a limiting factor compared to CT or CCA and can limit clear visualization of the smaller anomalies.
MSCT with high spatial and temporal resolution, various image postprocessing techniques, no anesthesia needed due to its fast acquisition, is now Table 1. Findings at prospective ECG-gated high-pitch dual-source CT (DSCT) angiography and transthoracic echocardiography (TTE) referring to surgical and/or CCA results (n575). Surgical/CCA results   TP  TN  FP  FN  TP  TN  FP  FN   Transposition of the great arteries  2  73  0  0  2  becoming one of the most valuable and potential cardiovascular examining tools [20][21][22][23].

The technique advantages of 128-slice DSCT
Based on the 64-slice DSCT, the recent 128-slice DSCT has increased the gantry rotation time to 280 ms, resulting in a temporal resolution of 75 ms. Such a fast acquisition allows satisfactory image to be acquired in children with relatively high heart rate and reduces breathing artifacts. The concern with CT imaging in pediatrics is the radiation exposure and here the high-pitch DSCT has advantages over retrospective ECG-gated or sequential techniques previously used, by limiting the exposure to a predefined window in the single cardiac cycle which saves significant dose to the patient. CT offers a high resolution data set that covers both intracardiac and extracardiac anatomies. The data can also be post processed in a variety of ways (MRP, MIP, VR, etc) making it extremely flexible for visualization and surgical planning. The location, relative size and extent of the lesion, the relationship of the lesion to the adjacent structures are better displayed with MPR, VR and MIP images than with axial images. Axial images are sufficient for assessing intracardiac deformities, and three-dimensional VR images can show great vessels in different directions and angles.   The diagnostic value of 128-slice DSCT Our study showed the diagnostic accuracy and sensitivity of high-pitch DSCT angiography in the assessment of extracardiac vascular anomalies was 99.67% and 97.53%, respectively. Compared to TTE, the deformities of aorta, pulmonary vessels, aortopulmonary collateral artery, systemic vein and coronary artery were better demonstrated on DSCT.
DSCT is of greater value compared to TTE in evaluation of congenital aortic diseases [24][25][26]. Except for 2 BAVs, other 44 aortic lesions were all identified by DSCT. Due to the short neck of children, overlying bone and aerated lung and the small FOV examined through suprasternal approach, TTE failed to identify 1 aortic dysplasia and 2 CoAs and misdiagnosed 1 CoA as IAA. Seven cases of CoA accompanied with collateral circulation including prominent internal mammary arteries, descending scapular arteries or intercostal arteries which were identified by DSCT were not diagnosed by TTE.
DSCT allows accurate characterization of pulmonary arterial disorders. It is an important tool helpful in demonstrating the connection of pulmonary artery and the right ventricle, defining anatomic relationship between pulmonary arteries and adjacent structures and calculating McGoon ratio to evaluate the development of pulmonary arteries [27]. DSCT misdiagnosed a severe pulmonary artery stenosis as pulmonary artery atresia, other 44 pulmonary arterial anomalies were all accurately diagnosed. As being easily limited by acoustic window and obscured by aerated lung, TTE is not capable enough to identify pulmonary disorders. Furthermore, the visual zone of left pulmonary artery is smaller than its right counterpart as the former vessel runs backward and is partially covered by the lung, and branches of pulmonary arteries within the lung could not been seen on TTE. In this study, TTE was unable to diagnose 5 pulmonary artery stenoses, 1 dilated pulmonary artery and 4 pulmonary slings. It misdiagnosed 1 case of pulmonary artery atresia accompanied by ventricular septal defect as truncus arteriosus, 2 cases of absence of one pulmonary artery as pulmonary sling.
DSCT is the most valuable modality in diagnosis of anomalous pulmonary venous connections by showing the number, route, location and a full view of the abnormal veins [28]. Seven APVRs in this study were all identified by DSCT. On TTE, the vessels in the proximal part are easily to be detected. However, it is difficult to track the abnormal vein far away the heart. TTE failed to identify 2 PAPVRs and misdiagnosed 1 mixed TAPVR as cardiac type.
In cases of pulmonary atresia, absence or stenosis, the pulmonary vascular bed may be supplied by aortic blood flow including MAPCA or PDA. The number, origin and route of MAPCAs can be shown on DSCT regardless of the size of the vessels [29]. Only large collateral vessels can be identified by TTE, and the ones originating from the descending aorta are difficult to be detected on TTE. Eight cases of MAPCAs were not diagnosed on TTE in our study.
Limited by the acoustic window and small FOV, TTE gives poor performance in demonstrating coronary artery anomalies, especially the middle and distal segment lesions. In 3 patients, TTE missed all of the coronary artery anomalies.
DSCT failed to identify 1 single coronary artery, as the acquisition window covered R wave, leading to a big pulsation artifact of the heart.
The diagnosis of LPSVC on TTE relies on the detection of coronary sinus dilation. In our study, one case of LPSVC missed by TTE as the left superior vena cava returned to left atrium and the size of coronary sinus was normal.
DSCT has prominent advantage over TTE in demonstrating trachea and bronchus [30]. In vascular ring and slings, anomalous anatomical relationships between vascular and tracheal structures can be shown on DSCT. Compression narrowing of the airways in 4 patients with double aortic arch and 7 patients with pulmonary sling were identified by DSCT.
We have shown that DSCT can provide low dose, diagnostic image quality data for the majority of extracardiac vascular anomalies. However, there are several anatomical anomalies that are difficult to diagnose with certainty, such as the BAV might go unnoticed because the data set of high-pitch DSCT was obtained in one cardiac cycle and the anomalies of the aortic valves could not be assessed, as happened to 2 patients in our study. In contrast, TTE was able to give accurate diagnosis of the 2 BAVs. Both TTE and DSCT are complimentary, and we found it useful to combine image data from the two compared modalities. Using TTE and DSCT in combination gave 99.4% diagnostic accuracy of the extracardiac vascular anomalies. Only one single coronary artery was not identified with both modalities. With an irregular heart rate in the patient, the starting position of the data acquisition in high-pitch DSCT did not correspond to the preselected starting phase, and the acquisition window covered the R wave, leading to severe artifacts of coronary arteries.

Limitations
Our study has some limitations. First, high-pitch DSCT could not provide hemodynamic and functional information which may be valuable in the assessment of extracardiac vascular anomalies. Second, a large spectrum of congenital vascular anomalies were enrolled in this study, future studies should be focused on a certain disease. Third, TTE was used as the comparison imaging method and future studies should also look at the diagnostic statistics for other modalities and DSCT.
In conclusion, prospective ECG-gated high-pitch 128-slice DSCT angiography with low radiation dose and high diagnostic accuracy has higher sensitivity compared to TTE in the detection of congenital extracardiac vascular anomalies in infants and children.