Impact of training data composition on the generalizability of CNN aortic cross section segmentation in 4D Flow MRI.

BACKGROUND: Time-resolved, three-dimensional phase-contrast magnetic resonance imaging (4D flow MRI) plays an important role in assessing cardiovascular diseases. However, the manual or semi-automatic segmentation of aortic vessel boundaries in 4D flow data introduces variability and limits reproducibility of aortic hemodynamics visualization and quantitative flow-related parameter computation. This paper explores the potential of deep learning to improve 4D flow MRI segmentation by developing models for automatic segmentation and analyzes the impact of the training data on the generalization of the model across different sites, scanner vendors, sequences, and pathologies. METHODS: The study population consists of 260 4D flow MRI datasets, including subjects without known aortic pathology, healthy volunteers, and patients with bicuspid aortic valve (BAV) examined at different hospitals. The dataset was split to train segmentation models on subsets with different representations of characteristics such as pathology, gender, age, scanner model, vendor, and field strength. An enhanced 3D U-net convolutional neural network (CNN) architecture with residual units was trained for 2D+t aortic cross-sectional segmentation. The model performance was evaluated using Dice score, Hausdorff distance, and average symmetric surface distance on test data, datasets with characteristics not represented in the training set (model-specific), and an overall evaluation set. Standard diagnostic flow parameters were computed and compared with manual segmentation results using Bland-Altman analysis and interclass correlation. RESULTS: The representation of technical factors such as scanner vendor and field strength in the training dataset had the strongest influence on the overall segmentation performance. Age had a greater impact than gender. Models solely trained on BAV patients' datasets performed well on datasets of healthy subjects but not vice versa. CONCLUSION: This study highlights the importance of considering a heterogeneous dataset for the training of widely applicable automatic CNN segmentations in 4D flow MRI, with a particular focus on the inclusion of different pathologies and technical aspects of data acquisition.

Ambulatory sedation for children under 6 years with CHD in MRI and CT.

INTRODUCTION: In infants and young children, good image quality in MRI and CT requires sedation or general anesthesia to prevent motion artefacts. This study aims to determine the safety of ambulatory sedation for children with CHD in an outpatient setting as a feasible alternative to in-hospital management. METHODS: We recorded 91 consecutive MRI and CT examinations of patients with CHD younger than 6 years with ambulatory sedation. CHD diagnoses, vital signs, applied sedatives, and adverse events during or after ambulatory sedation were investigated. RESULTS: We analysed 91 patients under 72 months (6 years) of age (median 26.0, range 1-70 months; 36% female). Sixty-eight per cent were classified as ASA IV, 25% as ASA III, and 7% as ASA II (American Society of Anesthesiologists Physical Status Classification). Ambulatory sedation was performed by using midazolam, propofol, and/or S-ketamine. The median sedation time for MRI was 90 minutes (range 35-235 minutes) and 65 minutes for CT (range 40-280 minutes). Two male patients (age 1.5 months, ASA II, and age 17 months, ASA IV) were admitted for in-hospital observation due to unexpected severe airway obstruction. The patients were discharged without sequelae after 1 and 3 days, respectively. All other patients were sent home on the day of examination. CONCLUSION: In infants and young children with CHD, MRI or CT imaging can be performed under sedation in an outpatient setting by a well-experienced team. In-hospital backup should be available for unexpected events.

Improved Tricuspid Valve Function, Preload Recruitment and Ventricular Efficiency During Submaximal Exercise in Patients with Unoperated Ebstein's Anomaly: An MRI Study.

BACKGROUND: Adolescents and adults with native Ebstein's anomaly (EA) are at the benign part of the Ebstein spectrum, having survived infancy without surgery. In this population, surgical indication and timing remain objects of controversy and depend, among other factors, on exercise capacity. PURPOSE: To better understand the pathophysiology of exercise adaptation in native EA. STUDY TYPE: Retrospective. POPULATION: Ten patients with unoperated EA (age range 18-61 years) and 13 healthy subjects as controls. FIELD STRENGTH/SEQUENCE: Balanced steady-state free precession cine and phase contrast flow sequences at 1.5 T. ASSESSMENT: We measured volumes and flows at rest and during submaximal exercise. Hemodynamic parameters including stroke volume (SV), cardiac index (CI), ejection fraction (EF), and tricuspid regurgitation (TR) were calculated. STATISTICAL TESTS: We used nonparametric Mann-Whitney U-test and Wilcoxon signed-rank test. A P-value of <0.05 was considered statistically significant. RESULTS: Rest CI and SV were significantly higher in controls; rest heart rate (HR) was similar in the two groups (median 71 bpm by patients and 65 bpm by controls, P = 0.448). During exercise, CI increased significantly in both groups: from 2.40 to 3.35 L/min/m2 in the patient group and from 3.60 to 4.20 L/min/m2 in controls; HR increased significantly in both groups. SV increased significantly in the patient group, whereas it remained stable in controls (P = 0.5284). Patients' median TR decreased significantly: median 42% at rest and 30% during exercise; concomitantly, left ventricular (LV) preload increased significantly (+3% indexed LV end-diastolic volume) as did LVEF (median 59% at rest vs. 65% during exercise). DATA CONCLUSION: During submaximal exercise, patients with mild to moderate EA improved their cardiovascular system's total efficiency by increasing CI; this was obtained by an increase in HR and by the recruitment of volume, as shown by an increased LV end-diastolic volume and SV, with simultaneous decrease in TR. This was different from healthy subjects in which CI increased only due to HR increase. LEVEL OF EVIDENCE: 3 TECHNICAL EFFICACY STAGE: 3.

Different CMR Imaging Modalities for Native and Patch-Repaired Right Ventricular Outflow Tract Sizing: Impact on Percutaneous Pulmonary Valve Replacement Planning.

Percutaneous pulmonary valve replacement (PPVI) in native or patched right ventricular outflow tract (RVOT) has proven to be feasible. The procedure is highly dependent on the size of the RVOT. Several methods exist to evaluate the size of the RVOT by cardiovascular magnetic resonance (CMR). We evaluated different CMR modalities for measuring RVOT diameters. Thirty-one consecutive patients with native or patched RVOT were retrospectively evaluated. CMR was part of follow-up of patients with corrected Tetralogy of Fallot or pulmonary stenosis with significant pulmonary regurgitation (PR). CMR included 3D-SSFP whole-heart in systole, diastole, and contrast-enhanced MR angiography (ceMRA). Diameters of the RVOT were assessed by the three sequences. Additionally, in patients who underwent cardiac catheterization (n = 11) for PPVI, vessel diameters assessed by cine-angiography were compared to CMR. Systolic diameters of RVOT were significantly larger compared to the diameters taken in diastole and ceMRA (median difference 5.0 mm and 3.8 mm). Diastolic and ceMRA diameters did not differ significantly. CMR diameters taken in systole showed no statistical difference to systolic diameters taken by cine-angiography, while diastolic and ceMRA diameters were significantly smaller. PPVI was feasible to a maximal CMR diameter of 31 mm measured by SSFP whole-heart sequence in systole. Absolute diameters of native RVOT differ depending on the CMR sequence and timing of acquisition (systolic vs diastolic gating). Diameters taken during heart catheterization by cine-angiography best correlate to systolic CMR values. Data may help to select RVOTs suitable for PPVI.

Magnetic Resonance Imaging Risk Factors for Ventricular Arrhythmias in Tetralogy of Fallot.

Sudden cardiac death (SCD) is the most common cause of late mortality in tetralogy of Fallot (TOF). Pulmonary regurgitation (PR) was previously found to be the most common hemodynamic abnormality associated with ventricular arrhythmias (VA), but cardiovascular magnetic resonance (CMR)-based studies did not show this association. The aim of this study is to investigate the risk factors for VA in TOF using CMR. Electronic records of TOF patients and their CMR studies between July 2006 and October 2018 in one center were retrospectively reviewed. Demographic, clinical and CMR data of patients were collected. Outcome was defined as sustained ventricular tachycardia (VT), aborted SCD and SCD. From a total of 434 TOF patients with complete CMR studies, 19 (4.4%) patients developed a positive outcome (12 sustained VT, 4 aborted SCD, 3 SCD) at a median age of 24 years. The number of surgical interventions was significantly greater in patients who developed VA. Right ventricular volumes were significantly larger in patients who suffered a positive outcome. Odds ratio for developing VA was 6.905 for RVEDVI ≥ 160 ml/m2 and 6.141 for RVESVI ≥ 80 ml/m2 (P = 0.0014 and 0.0012, respectively). Event-free survival was longer in patients with smaller right ventricular volumes. In conclusion, right ventricular dimensions are the most significant factors associated with the development of VA in TOF. The number of surgical interventions is also related to an increased risk.

Elevated diastolic wall shear stress in regurgitant semilunar valvular lesions.

BACKGROUND: Alterations in wall shear stress (WSS) assessed using 4D flow MRI have been shown to play a role in various vascular pathologies, such as bicuspid aortic valve aortopathy. Most studies have focused on systolic WSS, whereas altered diastolic hemodynamics in regurgitant semilunar valvular lesions have not so far been well characterized. PURPOSE: To investigate diastolic WSS in aortic and pulmonary regurgitation. STUDY TYPE: Retrospective data analysis. POPULATION: Thirty tetralogy of Fallot patients, 19 bicuspid aortic valve patients, 11 healthy volunteers. FIELD STRENGTH/SEQUENCE: 5 T, 3D time-resolved phase-contrast MRI with 3D velocity encoding. ASSESSMENT: Estimation of WSS and its axial and circumferential vector components along cardiac cycle timeframes in the proximal main pulmonary artery in pulmonary regurgitation (PR) and in the proximal ascending aorta in aortic regurgitation (AR) as well as in healthy volunteers. STATISTICAL TESTS: Wilcoxon matched pairs test was used for intra-group comparisons and Mann-Whitney test for intergroup comparisons. Correlations were assessed using Spearman correlation. RESULTS: WSS along the entire cardiac cycle was higher in PR and AR in comparison with controls (mean WSS 0.381 ± 0.070 vs. 0.220 ± 0.018, P < 0.0001; 0.361 ± 0.099 vs. 0.212 ± 0.030, P < 0.0001; respectively). Peak diastolic WSS was significantly higher than the mean WSS in AR and PR (P < 0.0001-0.005). The severity of PR correlated with the peak diastolic axial WSS (Spearman's r s = 0.454, P = 0.018), whereas the severity of AR correlated with both peak systolic and diastolic tangential WSS (Spearman's r s = 0.458, P = 0.049; r s = 0.539, P = 0.017, respectively). DATA CONCLUSION: Elevated diastolic WSS is a component of the altered flow hemodynamics in AR and PR. This may give more insight into the pathophysiologic role of WSS in vascular remodeling in AR and PR. LEVEL OF EVIDENCE: 4 Technical Efficacy Stage: 1 J. Magn. Reson. Imaging 2019;50:763-770.

Non-invasive Hemodynamic CMR Parameters Predicting Maximal Exercise Capacity in 54 Patients with Ebstein's Anomaly.

BACKGROUND: Exercise capacity is a well-defined marker of outcome in congenital heart disease. We analyzed seventeen cardiovascular magnetic resonance (CMR) derived parameters and their correlation to exercise capacity in patients with Ebstein's anomaly (EA). METHODS: Fifty-four surgery free patients, age 5 to 69 years (median 30 years) prospectively underwent CMR examination and cardiopulmonary exercise testing (CPET). The following volume/flow parameters were compared with peak oxygen uptake as the percentage of normal (peakVO2%) using univariate and multivariate analysis: right and left ventricular ejection fraction (RVEF and LVEF), the indexed end-diastolic and end-systolic volumes (RVEDVi, RVESVi, LVEDVi, and LVESVi), the indexed stroke volumes (RVSVi and LVSVi), the total normalized right and left heart volumes; the total right to left heart volume ratio (R/L-ratio). The indexed antegrade flow (ante), indexed net flow (net) as well as cardiac index (CI) in the aorta (Ao) and pulmonary artery (PA) were used. RESULTS: RVEF (R2 0.2788), indexed flow PA net (R2 0.2330), and PA ante (R2 0.1912) showed the best correlation with peakVO2% (all p < 0.001) in the univariate model. Further significant correlation could also be demonstrated with CI-PA, LVEF, LVSVi, Aorta net, RVESVi, and Aorta ante. Multivariate analysis for RVEF and indexed net flow PA revealed a R2 of 0.4350. CONCLUSION: Functional CMR parameters as RVEF and LVEF and flow data of cardiac forward flow correlate to peakVO2%. Evaluation of the indexed net flow in the pulmonary artery and the overall function of the right ventricle best predicts the maximal exercise capacity in patients with EA.

Aortopulmonary collateral flow quantification by MR at rest and during continuous submaximal exercise in patients with total cavopulmonary connection.

BACKGROUND: Aortopulmonary collateral flow is considered to have significant impact on the outcome of patients with single ventricle circulation and total cavopulmonary connection (TCPC). There is little information on collateral flow during exercise. PURPOSE: To quantify aortopulmonary collateral flow at rest and during continuous submaximal exercise in clinical patients doing well with TCPC. STUDY TYPE: Prospective, case controlled. POPULATION: Thirteen patients with TCPC (17 (11-37) years) and 13 age and sex-matched healthy controls (18 (11-38) years). FIELD STRENGTH: 1.5T; free breathing; phase sensitive gradient echo sequence. ASSESSMENT: Blood flow in the ascending and descending aorta and superior vena cava were measured at rest and during continuous submaximal physical exercise in patients and controls. Systemic blood flow (Qs ) was assumed to be represented by the sum of flow in the superior caval vein (Qsvc ) and the descending aorta (QAoD ) at the diaphragm level. Aortopulmonary collateral flow (Qcoll ) was calculated by subtracting Qs from flow in the ascending aorta (QAoA ). STATISTICS: Mann-Whitney U-test and Wilcoxon test for comparison between groups and between rest and exercise. RESULTS: Absolute collateral flow in TCPC patients at rest was 0.4 l/min/m2 (-0.1-1.2), corresponding to 14% (-2-42) of Qs . Collateral flow did not change during exercise (difference -0.01 (-0.7-1.0) l/min/m2 , P = 0.97). TCPC patients had significantly lower Qs at rest (2.5 (1.6-4.1) vs. 3.5 (2.6-4.8) l/min/m2 , P = 0.001) and during submaximal exercise (3.2 (2.0-6.0) vs. 4.8 (3.3-6.9) l/min/m2 , P = 0.001), compared to healthy controls. The increase in Qs with exercise was also significantly lower in patients than in healthy controls (median 0.6 vs. 1.2 l/min/m2 , P < 0.02). DATA CONCLUSION: Clinical patients doing well with TCPC have significant aortopulmonary collateral flow at rest (14% of Qs ) compared to healthy controls, which does not change during submaximal exercise. LEVEL OF EVIDENCE: 2 Technical Efficacy: Stage 3 J. Magn. Reson. Imaging 2018;47:1509-1516.

Wall shear stress estimation in the aorta: Impact of wall motion, spatiotemporal resolution, and phase noise.

BACKGROUND: Wall shear stress (WSS) presents an important parameter for assessing blood flow characteristics and evaluating flow-mediated lesions in the aorta. PURPOSE: To investigate the robustness of WSS and oscillatory shear index (OSI) estimation based on 4D flow MRI against vessel wall motion, spatiotemporal resolution, and velocity encoding (VENC). STUDY TYPE: Simulated and prospective. POPULATION: Synthetic 4D flow MRI data of the aorta, simulated using the Lattice-Boltzmann method; in vivo 4D flow MRI data of the aorta from healthy volunteers (n = 11) and patients with congenital heart defects (n = 17). FIELD STRENGTH/SEQUENCE: 1.5T; 4D flow MRI with PEAK-GRAPPA acceleration and prospective electrocardiogram triggering. ASSESSMENT: Predicated upon 3D cubic B-splines interpolation of the image velocity field, WSS was estimated in mid-systole, early-diastole, and late-diastole and OSI was derived. We assessed the impact of spatiotemporal resolution and phase noise, and compared results based on tracked-using deformable registration-and static vessel wall location. STATISTICAL TESTS: Bland-Altman analysis to assess WSS/OSI differences; Hausdorff distance (HD) to assess wall motion; and Pearson's correlation coefficient (PCC) to assess correlation of HD with WSS. RESULTS: Synthetic data results show systematic over-/underestimation of WSS when different spatial resolution (mean ± 1.96 SD up to -0.24 ± 0.40 N/m2 and 0.5 ± 1.38 N/m2 for 8-fold and 27-fold voxel size, respectively) and VENC-depending phase noise (mean ± 1.96 SD up to 0.31 ± 0.12 N/m2 and 0.94 ± 0.28 N/m2 for 2-fold and 4-fold VENC increase, respectively) are given. Neglecting wall motion when defining the vessel wall perturbs WSS estimates to a considerable extent (1.96 SD up to 1.21 N/m2 ) without systematic over-/underestimation (Bland-Altman mean range -0.06 to 0.05). DATA CONCLUSION: In addition to sufficient spatial resolution and velocity to noise ratio, accurate tracking of the vessel wall is essential for reliable image-based WSS estimation and should not be neglected if wall motion is present. LEVEL OF EVIDENCE: 2 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2018.

Long-Standing Cyanosis in Congenital Heart Disease Does not Cause Diffuse Myocardial Fibrosis.

The assumption of the presence of diffuse myocardial fibrosis in long-standing cyanotic congenital heart disease (CHD) inspired us to noninvasively determine the myocardial extracellular volume (ECV) using contrast CMR. T1 maps were measured pre and 10 min after the injection of 0.15 mmol/kg of gadolinium in 25 subjects. Seven patients with long-standing cyanotic CHD and no previous cardiac surgery (aged 16-53 years and oxygen saturations of 69-90%), nine normal subjects (aged 14-49 years), and nine patients with previously cyanotic CHD, who had been corrected by open heart surgery (aged 2 months-58 years, mean 9 years). Late gadolinium enhancement was performed to exclude scar areas. The T1 values were measured in the interventricular septum and in the left lateral or inferior ventricular wall, such that same areas were assessed in every patient in the pre- and post-contrast T1 scan. ECV was calculated according to ΔR1myocardium/ΔR1blood * (1 - hematocrit). Cyanotic patients had significantly lower ECV percentage than the previous cyanotic patients (septum: 22 ± 2.7% vs 35 ± 4.6%, p = 0.002; LV wall: 22 ± 2.2% vs 30 ± 3.7%, p = 0.01, respectively). No significant differences were found between cyanotic patients and normal controls (septum: 22 ± 2.7% vs 24 ± 1.4%, p = 0.44; LV wall: 22 ± 2.2% vs 24 ± 2%, p = 0.57, respectively). Long-standing cyanosis in CHD without cardiac surgery does not cause diffuse myocardial fibrosis or expansion of the myocardial ECV.

Comparison of MR flow quantification in peripheral and main pulmonary arteries in patients after right ventricular outflow tract surgery: A retrospective study.

PURPOSE: To compare the quantification of pulmonary stroke volume (SV) by phase contrast magnetic resonance (PC-MR) in the main pulmonary artery (MPA) to the sum of SVs in both peripheral pulmonary arteries (PPA) in different right ventricular (RV) outflow pathologies. MATERIALS AND METHODS: Pulmonary SV was determined by PC-MR in the MPA and the PPA in healthy individuals (H, n = 54), patients after correction for tetralogy of Fallot with significant pulmonary regurgitation and without pulmonary or RV outflow tract stenosis (PR, n = 50), and in patients with RV outflow tract or pulmonary valve stenosis (PS, n = 50). Resulting SVs were compared to aortic SV in the ascending aorta. RESULTS: Mean age was similar between the groups: H 28 ± 17 vs. PR 24 ± 11 vs. PS 22 ± 10 years. Bland-Altman analyses revealed in all groups a relatively small systemic (bias) but large random error (limits of agreement) for pulmonary SV determined in the MPA as compared to summed SVs in the PPA. The largest limits of agreement were present in PS patients: H: MPA 3.9% (-11, + 19) vs. PPA 0.4% (-15, + 15); PR: MPA 5.2% (-25, + 36) vs. PPA 0.6% (-24, + 26); PS: MPA 5% (-36; + 46), PPA -0.03% (-34, + 35). CONCLUSION: The accuracy of PC-MR in the MPA is reasonable; however, a large random error (precision) is observed that is most pronounced in PS patients. This potential error should be taken into consideration when interpreting MPA flow measurements. LEVEL OF EVIDENCE: 3 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2017;46:1839-1845.

Importance of Non-invasive Right and Left Ventricular Variables on Exercise Capacity in Patients with Tetralogy of Fallot Hemodynamics.

Good quality of life correlates with a good exercise capacity in daily life in patients with tetralogy of Fallot (ToF). Patients after correction of ToF usually develop residual defects such as pulmonary regurgitation or stenosis of variable severity. However, the importance of different hemodynamic parameters and their impact on exercise capacity is unclear. We investigated several hemodynamic parameters measured by cardiovascular magnetic resonance (CMR) and echocardiography and evaluated which parameter has the most pronounced effect on maximal exercise capacity determined by cardiopulmonary exercise testing (CPET). 132 patients with ToF-like hemodynamics were tested during routine follow-up with CMR, echocardiography and CPET. Right and left ventricular volume data, ventricular ejection fraction and pulmonary regurgitation were evaluated by CMR. Echocardiographic pressure gradients in the right ventricular outflow tract and through the tricuspid valve were measured. All data were classified and correlated with the results of CPET evaluations of these patients. The analysis was performed using the Random Forest model. In this way, we calculated the importance of the different hemodynamic variables related to the maximal oxygen uptake in CPET (VO2%predicted). Right ventricular pressure showed the most important influence on maximal oxygen uptake, whereas pulmonary regurgitation and right ventricular enddiastolic volume were not important hemodynamic variables to predict maximal oxygen uptake in CPET. Maximal exercise capacity was only very weakly influenced by right ventricular enddiastolic volume and not at all by pulmonary regurgitation in patients with ToF. The variable with the most pronounced influence was the right ventricular pressure.

Quantifying aortic valve regurgitation in patients with congenital aortic valve disease by 2D and 4D flow magnetic resonance analysis.

BACKGROUND: In congenital aortic valve disease, quantifying aortic regurgitation (AR) varies by the measurement site. Our study aimed to identify the optimal site for AR assessment using 2D and 4D MR flow measurements, with a focus on vortices. METHODS: We retrospectively analysed 31 patients with congenital aortic valve disease, performing 2D and 4D MR flow measurements at the aortic valve, sinotubular junction (STJ), ascending aorta (AAo), and using midpulmonary artery measurements as a reference. We assessed percentage AR and net forward volumes, calculated linear correlations, and plotted Bland-Altman plots. Net forward flow at all aortic sites were correlated with the main pulmonary artery. Differences in AR between 2D and 4D flows were linked to vortices detected by 4D streamlines. RESULTS: The best agreement in % AR between 2D and 4D flows was at the aortic valve (mean difference 4D2D -2.9%, limits of agreement 8.7% to -14.3%; r2 = 0.7). Correlations weakened at STJ and AAo. Vortices in the ascending aorta led to AR overestimation in 2D measurements. Net forward flow at the aortic valve by 4D flow correlated closer with main pulmonary artery than did 2D flow. (Mean difference for 2D and 4D MR flow 7.5 ml and 4.2 ml, respectively). CONCLUSIONS: For congenital aortic valve disease, the most accurate AR quantification occurs at the aortic valve using 2D and 4D MR flow. Notably, vortices in the ascending aorta can result in AR overestimation with 2D MR flow.

Demonstration of value of optimizing ECG triggering for cardiovascular magnetic resonance in patients with congenital heart disease.

BACKGROUND: Optimal ECG triggering is of paramount importance for correct blood flow quantification during cardiovascular magnetic resonance (CMR). However, optimal ECG triggering and therefore blood flow quantification is impaired in many patients with congenital heart disease (CHD) due to complex QRS patterns. Therefore, a new ECG-trigger algorithm was developed to address triggering problems due to complex QRS patterns.The aim of this study was to test this new ECG-trigger algorithm in routine patients with CHD and its impact on blood flow quantification. METHODS: 35 consecutive routine patients with CHD undergoing CMR were included in the study. (40% Fallot's Tetralogy, 20% aortic arch pathology, 14% transposition of the great arteries, 26% others; age 26+/-11 yrs).In all patients, blood flow in the ascending aorta was quantified using the old ECG-trigger algorithm and the new ECG-trigger algorithm in random order. Blood flow quantified using the old or new ECG-trigger algorithm was compared by Bland-Altman analysis.Three blinded investigators evaluated the vector clouds and trigger points of both ECG-trigger methods. Evaluation criteria were false positive and false negative triggered QRS complexes (specificity and sensitivity), and accuracy of detection. Accuracy of detection was defined as time scatter of the trigger around the correct trigger point. RESULTS: Specificity, sensitivity, and accuracy of detection significantly increased using the new ECG-trigger algorithm compared to the old ECG-trigger algorithm.Blood flow quantification using the old or new ECG-trigger algorithm differed more than 5% in 31% of the cases. CONCLUSIONS: Our results suggest that optimizing ECG triggering during CMR using our new algorithm can avoid errors of >5% in approximately 1/3 of routine patients with congenital heart disease (CHD). We furthermore suggest that incorrect ECG triggering appears to be problematic for blood flow quantification of many patients with CHD undergoing routine CMR.

Junctional rhythm produces retrograde flow in the whole Fontan system: cardiac magnetic resonance-derived flow documentation-a case report.

Background: Arrhythmias after palliation of congenital malformations with functional monoventricle by different Fontan modifications are very common. Sinus node dysfunction and junctional rhythm are known to have a high prevalence and a detrimental impact on the optimal functioning of Fontan circulations. Maintaining sinus node function has high prognostic significance, and some cases have even been described where atrial pacing with restoring of atrioventricular synchrony was able to reverse protein-losing enteropathy with overt failure of the Fontan. Case summary: A 12-year-old boy with a complex congenital malformation (double outlet right ventricle, transposition of the great arteries, pulmonary stenosis, and straddling atrioventricular valve) palliated through a modified Fontan (total cavopulmonary connection with a fenestrated extracardiac 18 mm Gore-Tex conduit) presented for cardiac magnetic resonance evaluation for mild asthenia and worsening of exercise tolerance. Flow profiles in all the regions of the Fontan (both caval veins and right and left pulmonary arteries) showed a small amount of retrograde flow; a four-chamber cine sequence clearly showed contraction of the atria against closed atrioventricular valve; this haemodynamic condition can be caused either by retro-conducted junctional rhythm (previously demonstrated in our patient) or by isorhythmic dissociation of sinus rhythm. Discussion: Our finding directly demonstrates the profound impact of retro-conducted junctional rhythm on the haemodynamic of a Fontan circulation in which, with each cardiac beat, the pressure rise in the atria and pulmonary veins due to atrial contraction with closed atrioventricular valves is able to stop and invert the passive flow of the systemic venous return towards the lungs.

Aortic wall shear stress in bicuspid aortic valve disease-10-year follow-up.

Background: Bicuspid aortic valve (BAV) disease leads to deviant helical flow patterns especially in the mid-ascending aorta (AAo), potentially causing wall alterations such as aortic dilation and dissection. Among others, wall shear stress (WSS) could contribute to the prediction of long-term outcome of patients with BAV. 4D flow in cardiovascular magnetic resonance (CMR) has been established as a valid method for flow visualization and WSS estimation. The aim of this study is to reevaluate flow patterns and WSS in patients with BAV 10 years after the initial evaluation. Methods: Fifteen patients (median age 34.0 years) with BAV were re-evaluated 10 years after the initial study from 2008/2009 using 4D flow by CMR. Our particular patient cohort met the same inclusion criteria as in 2008/2009, all without enlargement of the aorta or valvular impairment at that time. Flow patterns, aortic diameters, WSS and distensibility were calculated in different aortic regions of interest (ROI) with dedicated software tools. Results: Indexed aortic diameters in the descending aorta (DAo), but especially in the AAo did not change in the 10-year period. Median difference 0.05 cm/m2 (95% CI: 0.01 to 0.22; P=0.06) for AAo and median difference -0.08 cm/m2 (95% CI: -0.12 to 0.01; P=0.07) for DAo. WSS values were lower in 2018/2019 at all measured levels. Aortic distensibility decreased by median 25.6% in the AAo, while stiffness increased concordantly (median +23.6%). Conclusions: After a ten years' follow-up of patients with isolated BAV disease, indexed aortic diameters did not change in this patient cohort. WSS was lower compared to values generated 10 years earlier. Possibly a drop of WSS in BAV could serve as a marker for a benign long-term course and implementation of more conservative treatment strategies.

Does the amplatzer septal occluder device alter ventricular contraction pattern? A ventricular motion analysis by MR tagging.

PURPOSE: To assess by cardiovascular magnetic resonance (CMR) and CMR tagging if the Amplatzer Septal Occluder affects right ventricular (RV) and left ventricular (LV) motion pattern. MATERIALS AND METHODS: Sixteen consecutive patients with significant atrial septal defect (ASD) and nine consecutive patients with persistent foramen ovale (PFO) as controls were studied before and a median of 14 days after defect closure by an Amplatzer occluder. By CMR end-diastolic (EDV) and end-systolic (ESV) RV and LV volumes were determined. Aortic and pulmonary artery flow was measured for assessment of left-to-right shunt (Qp/Qs). By CMR tagging circumferential strain and radial shortening, maximal rotation and torsion were measured, RESULTS: In ASD patients RV-EDV and RV-ESV decreased (P < 0.05). LV-EDV and LV-ESV increased after ASD closure (P < 0.005). Qp/Qs dropped from 1.8 to 1.0 (P < 0.001). PFO patients showed no ventricular volume change after PFO closure. In ASD patients circumferential strain and radial shortening and maximal rotation of the RV decreased by ASD closure (P < 0.01). In LV only maximal rotation at the base and apex decreased significantly (P < 0.05). Torsion remained constant. In PFO patients no tagging parameter changed after defect closure. CONCLUSION: The Amplatzer occluder itself does not change the ventricular contraction pattern. All volume and myocardial deformation changes were caused by ventricular loading shifts.

Safety of intrahospital transport for MR or CT scans in ventilated pediatric intensive care patients with congenital heart disease.

Background: MR or CT scans are often required in the treatment of pediatric intensive care patients. During the therefore needed intrahospital transport continuous ventilation of the patient must be maintained. Intrahospital transport is considered safe regarding changes in hemodynamics or adverse events (AEs). As those studies cover inhomogeneous patient groups, we analyzed the safety for ventilated pediatric patients with congenital heart disease (CHD) focusing on differences between manual and mechanical ventilation during transport and examination. Methods: Retrospective monocentric case-control study covering a 10 years' period in a tertiary cardiac center for CHD. Sixty-three critically ill ventilator-dependent patients, median 2 (0-37) months, were included. Fifty-one patients got ventilated manually and 12 patients got ventilated with a mobile ventilator. The data include vital parameters and blood gas data, as well as catecholamine support, occurrence of AEs and total duration of transport and examination. Results: In both groups we found minor changes of vital parameters or blood gas data. Regarding the HR we found a drop from median 135/min before leaving the ICU to 130/min after returning to the ICU (P=0.0072) in the manually ventilated group and a drop from median 133/min to 123/min (P=0.0703) in the mechanically ventilated group. The median transport time including scan was higher in the manually ventilated group (P=0.0098) with a median duration of transport and scan time of 100 minutes (30-245 minutes) in the manually ventilated group and of 97.5 minutes (60-224 minutes) in the mechanically ventilated group. A total of 9 AEs was recorded, 7 (13.7%) of them in the manually and 2 (16.7%) in the mechanically ventilated group with a drop of the mean arterial pressure (MAP) and an increase in catecholamine support. Conclusions: We consider both manual ventilation and mechanical ventilation for intrahospital transport safe for pediatric intensive care patients with CHD. Using a mechanical ventilator might have the advantage to react faster to changes in hemodynamics.

Detection of early signs of right ventricular systolic impairment in unoperated Ebstein's anomaly by cardiac magnetic resonance feature tracking.

Background: Cardiovascular magnetic resonance feature-tracking analysis (CMR-FT) provides a quantitative assessment of myocardial contraction with potential for diagnostic and prognostic ability in a wide spectrum of diseases. Ebstein's anomaly (EA) is a rare congenital heart disease characterized by apical displacement of the tricuspid valve. However, it is also considered a disorder of development affecting the global right ventricular myocardium. Aim of our study is to describe the complex contractile mechanics of the functional right ventricle (RV) in patients affected by EA through CMR-FT. Methods: Fifty surgery-free EA patients who had undergone a complete CMR protocol at our institution between January 2017 and December 2020 were selected for the retrospective study. A historical control group of twenty-five healthy subjects was also included. CMR-FT analysis was performed at a dedicated workstation by manually tracing RV endo- end epicardial borders on steady-state-free-precession (SSFP) cine images. Strain values were calculated. Apical displacement of the tricuspid valve (TV) was measured on a 4-chamber cine image from the right atrio-ventricular junction to the functional annulus of the TV. Results: EA patients presented significantly impaired RV global radial strain (GRS) and global circumferential strain (GCS) compared to controls (P<0.0001 and P=0.0008, respectively). In a subgroup analysis, GRS was significantly compromised in patients with a severely displaced TV (>16 mm/m2) compared to milder forms (P=0.03) and to controls (P<0.0001). Among EA patients with a preserved ejection fraction, 12 (48%) vs. 6 (24%) controls had reduced both GRS and GCS. Conclusions: The contractile pattern of the functional RV in EA is characterised by prevalent alterations in the short-axis direction as indicated by reduced GRS and GCS. Strain values might be reduced prior to routine used functional parameters like RV ejection fraction (RVEF) and can possibly serve as an early predictor of myocardial dysfunction in EA patients.

Magnetic resonance imaging of myocardial injury and ventricular torsion after marathon running.

Recent reports provide indirect evidence of myocardial injury and ventricular dysfunction after prolonged exercise. However, existing data is conflicting and lacks direct verification of functional myocardial alterations by CMR [cardiac MR (magnetic resonance)]. The present study sought to examine structural myocardial damage and modification of LV (left ventricular) wall motion by CMR imaging directly after a marathon. Analysis of cTnT (cardiac troponin T) and NT-proBNP (N-terminal pro-brain natriuretic peptide) serum levels, echocardiography [pulsed-wave and TD (tissue Doppler)] and CMR were performed before and after amateur marathon races in 28 healthy males aged 41 ± 5 years. CMR included LGE (late gadolinium enhancement) and myocardial tagging to assess myocardial injury and ventricular motion patterns. Echocardiography indicated alterations of diastolic filling [decrease in E/A (early transmitral diastolic filling velocity/late transmitral diastolic filling velocity) ratio and E' (tissue Doppler early transmitral diastolic filling velocity)] postmarathon. All participants had a significant increase in NT-proBNP and/or cTnT levels. However, we found no evidence of LV LGE. MR tagging demonstrated unaltered radial shortening, circumferential and longitudinal strain. Myocardial rotation analysis, however, revealed an increase of maximal torsion by 18.3% (13.1 ± 3.8 to 15.5 ± 3.6 °; P=0.002) and maximal torsion velocity by 35% (6.8 ± 1.6 to 9.2 ± 2.5 °·s-1; P<0.001). Apical rotation velocity during diastolic filling was increased by 1.23 ± 0.33 °·s-1 after marathon (P<0.001) in a multivariate analysis adjusted for heart rate, whereas peak untwist rate showed no relevant changes. Although marathon running leads to a transient increase of cardiac biomarkers, no detectable myocardial necrosis was observed as evidenced by LGE MRI (MR imaging). Endurance exercise induces an augmented systolic wringing motion of the myocardium and increased diastolic filling velocities. The stress of marathon running seems to be better described as a burden of myocardial overstimulation rather than cardiac injury.