Accelerated cardiac magnetic resonance imaging using deep learning for volumetric assessment in children.

BACKGROUND: Ventricular volumetry using a short-axis stack of two-dimensional (D) cine balanced steady-state free precession (bSSFP) sequences is crucial in any cardiac magnetic resonance imaging (MRI) examination. This task becomes particularly challenging in children due to multiple breath-holds. OBJECTIVE: To assess the diagnostic performance of accelerated 3-RR cine MRI sequences using deep learning reconstruction compared with standard 2-D cine bSSFP sequences. MATERIAL AND METHODS: Twenty-nine consecutive patients (mean age 11 ± 5, median 12, range 1-17 years) undergoing cardiac MRI were scanned with a conventional segmented 2-D cine and a deep learning accelerated cine (three heartbeats) acquisition on a 1.5-tesla scanner. Short-axis volumetrics were performed (semi-)automatically in both datasets retrospectively by two experienced readers who visually assessed image quality employing a 4-point grading scale. Scan times and image quality were compared using the Wilcoxon rank-sum test. Volumetrics were assessed with linear regression and Bland-Altman analyses, and measurement agreement with intraclass correlation coefficient (ICC). RESULTS: Mean acquisition time was significantly reduced with the 3-RR deep learning cine compared to the standard cine sequence (45.5 ± 13.8 s vs. 218.3 ± 44.8 s; P < 0.001). No significant differences in biventricular volumetrics were found. Left ventricular (LV) mass was increased in the deep learning cine compared with the standard cine sequence (71.4 ± 33.1 g vs. 69.9 ± 32.5 g; P < 0.05). All volumetric measurements had an excellent agreement with ICC > 0.9 except for ejection fraction (EF) (LVEF 0.81, RVEF 0.73). The image quality of deep learning cine images was decreased for end-diastolic and end-systolic contours, papillary muscles, and valve depiction (2.9 ± 0.5 vs. 3.5 ± 0.4; P < 0.05). CONCLUSION: Deep learning cine volumetrics did not differ significantly from standard cine results except for LV mass, which was slightly overestimated with deep learning cine. Deep learning cine sequences result in a significant reduction in scan time with only slightly lower image quality.

Feasibility of Non-Gated Dynamic Fetal Cardiac MRI for Identification of Fetal Cardiovascular Anatomy.

INTRODUCTION: The aim of this study was to evaluate the feasibility of identifying the fetal cardiac and thoracic vascular structures with non-gated dynamic balanced steady-state free precession (SSFP) MRI sequences. METHODS: We retrospectively assessed the visibility of cardiovascular anatomy in 60 fetuses without suspicion of congenital heart defect. Non-gated dynamic balanced SSFP sequences were acquired in three anatomic planes of the fetal thorax. The images were analyzed following a segmental approach in consensus reading by an experienced pediatric cardiologist and radiologist. An imaging score was defined by giving one point to each visualized structure, yielding a maximum score of 21 points. Image quality was rated from 0 (poor) to 2 (excellent). The influence of gestational age (GA), field strength, placenta position, and maternal panniculus on image quality and imaging score were tested. RESULTS: 30 scans were performed at 1.5T, 30 at 3T. Heart position, atria, and ventricles could be seen in all 60 fetuses. Basic diagnosis (>12 points) was achieved in 54 cases. The mean imaging score was 16.8+/-3.8. Maternal panniculus (r = -0.3; p = 0.015) and GA (r = 0.6; p < 0.001) correlated with imaging score. Field strength influenced image quality, with 1.5T being better than 3T images (p = 0.012). Imaging score or quality was independent of placenta position. CONCLUSION: Fetal cardiac MRI with non-gated SSFP sequences enables recognition of basic cardiovascular anatomy.

Retraining Convolutional Neural Networks for Specialized Cardiovascular Imaging Tasks: Lessons from Tetralogy of Fallot.

Ventricular contouring of cardiac magnetic resonance imaging is the gold standard for volumetric analysis for repaired tetralogy of Fallot (rTOF), but can be time-consuming and subject to variability. A convolutional neural network (CNN) ventricular contouring algorithm was developed to generate contours for mostly structural normal hearts. We aimed to improve this algorithm for use in rTOF and propose a more comprehensive method of evaluating algorithm performance. We evaluated the performance of a ventricular contouring CNN, that was trained on mostly structurally normal hearts, on rTOF patients. We then created an updated CNN by adding rTOF training cases and evaluated the new algorithm's performance generating contours for both the left and right ventricles (LV and RV) on new testing data. Algorithm performance was evaluated with spatial metrics (Dice Similarity Coefficient (DSC), Hausdorff distance, and average Hausdorff distance) and volumetric comparisons (e.g., differences in RV volumes). The original Mostly Structurally Normal (MSN) algorithm was better at contouring the LV than the RV in patients with rTOF. After retraining the algorithm, the new MSN + rTOF algorithm showed improvements for LV epicardial and RV endocardial contours on testing data to which it was naïve (N = 30; e.g., DSC 0.883 vs. 0.905 for LV epicardium at end diastole, p < 0.0001) and improvements in RV end-diastolic volumetrics (median %error 8.1 vs 11.4, p = 0.0022). Even with a small number of cases, CNN-based contouring for rTOF can be improved. This work should be extended to other forms of congenital heart disease with more extreme structural abnormalities. Aspects of this work have already been implemented in clinical practice, representing rapid clinical translation. The combined use of both spatial and volumetric comparisons yielded insights into algorithm errors.

Additional value and new insights by four-dimensional flow magnetic resonance imaging in congenital heart disease: application in neonates and young children.

Cardiovascular MRI has become an essential imaging modality in children with congenital heart disease (CHD) in the last 15-20 years. With use of appropriate sequences, it provides important information on cardiovascular anatomy, blood flow and function for initial diagnosis and post-surgical or -interventional monitoring in children. Although considered as more sophisticated and challenging than CT, in particular in neonates and infants, MRI is able to provide information on intra- and extracardiac haemodynamics, in contrast to CT. In recent years, four-dimensional (4-D) flow MRI has emerged as an additional MR technique for retrospective assessment and visualisation of blood flow within the heart and any vessel of interest within the acquired three-dimensional (3-D) volume. Its application in young children requires special adaptations for the smaller vessel size and faster heart rate compared to adolescents or adults. In this article, we provide an overview of 4-D flow MRI in various types of complex CHD in neonates and infants to demonstrate its potential indications and beneficial application for optimised individual cardiovascular assessment. We focus on its application in clinical routine cardiovascular workup and, in addition, show some examples with pathologies other than CHD to highlight that 4-D flow MRI yields new insights in disease understanding and therapy planning. We shortly review the essentials of 4-D flow data acquisition, pre- and post-processing techniques in neonates, infants and young children. Finally, we conclude with some details on accuracy, limitations and pitfalls of the technique.

Real-World Experience Measurement of Liver Iron Concentration by R2 vs. R2 Star MRI in Hemoglobinopathies.

BACKGROUND: Non-invasive determination of liver iron concentration (LIC) is a valuable tool that guides iron chelation therapy in transfusion-dependent patients. Multiple methods have been utilized to measure LIC by MRI. The purpose of this study was to compare free breathing R2* (1/T2*) to whole-liver Ferriscan R2 method for estimation of LIC in a pediatric and young adult population who predominantly have hemoglobinopathies. METHODS: Clinical liver and cardiac MRI scans from April 2016 to May 2018 on a Phillips 1.5 T scanner were reviewed. Free breathing T2 and T2* weighted images were acquired on each patient. For T2, multi-slice spin echo sequences were obtained. For T2*, a single mid-liver slice fast gradient echo was performed starting at 0.6 ms with 1.2 ms increments with signal averaging. R2 measurements were performed by Ferriscan analysis. R2* measurements were performed by quantitative T2* map analysis. RESULTS: 107 patients underwent liver scans with the following diagnoses: 76 sickle cell anemia, 20 Thalassemia, 9 malignancies and 2 Blackfan Diamond anemia. Mean age was 12.5 ± 4.5 years. Average scan time for R2 sequences was 10 min, while R2* sequence time was 20 s. R2* estimation of LIC correlated closely with R2 with a correlation coefficient of 0.94. Agreement was strongest for LIC < 15 mg Fe/g dry weight. Overall bias from Bland-Altman plot was 0.66 with a standard deviation of 2.8 and 95% limits of agreement -4.8 to 6.1. CONCLUSION: LIC estimation by R2* correlates well with R2-Ferriscan in the pediatric age group. Due to the very short scan time of R2*, it allows imaging without sedation or anesthesia. Cardiac involvement was uncommon in this cohort.

Creating three dimensional models of the right ventricular outflow tract: influence of contrast, sequence, operator, and threshold.

The use of 3D printed models of the right ventricular outflow tract (RVOT) for surgical and interventional planning is growing and often requires image segmentation of cardiac magnetic resonance (CMR) images. Segmentation results may vary based on contrast, image sequence, signal threshold chosen by the operator, and manual post-processing. The purpose of this study was to determine potential biases and post-processing errors in image segmentation to enable informed decisions. Models of the RVOT and pulmonary arteries from twelve patients who had contrast enhanced CMR angiography with gadopentetate dimeglumine (GPD), gadofosveset trisodium (GFT), and a post-GFT inversion-recovery (IR) whole heart sequence were segmented, trimmed, and aligned by three operators. Geometric agreement and minimal RVOT diameters were compared between sequences and operators. To determine the contribution of threshold, interoperator variability was compared between models created by the same two operators using the same versus different thresholds. Geometric agreement by Dice between objects was high (intraoperator: 0.89-0.95; interoperator: 0.95-0.97), without differences between sequences. Minimal RVOT diameters differed on average by - 1.9 to - 1.3 mm (intraoperator) and by 0.4 to 1.4 mm (interoperator). The contribution of threshold to interoperator geometric agreement was not significant (same threshold: 0.96 ± 0.06, different threshold: 0.93 ± 0.05; p = 0.181), but minimal RVOT diameters were more variable with different versus constant thresholds (- 9.12% vs. 2.42%; p < 0.05). Thresholding does not significantly change interoperator variability for geometric agreement, but does for minimal RVOT diameter. Minimal RVOT diameters showed clinically relevant variation within and between operators.

Newborn girl with coarctation of the aorta and anomalous left coronary artery from pulmonary artery, with retrograde perfusion of left circumflex artery: a case report.

BACKGROUND: Anomalous left coronary artery from pulmonary artery (ALCAPA) is a rare coronary abnormality. Although it exists usually as an isolated abnormality, ALCAPA has been described with aortic pathologies like coarctation or aortopulmonary window. CASE SUMMARY: An 18-day-old female was admitted to the paediatric intensive care unit because of a heart murmur and weak femoral pulses. A transthoracic two-dimensional echocardiography was performed and confirmed suspected diagnosis of aortic coarctation. In addition, a total retrograde perfusion of the left circumflex coronary artery (LCX) was found, without visible flow through the ostium of the left coronary artery (LCA) into the aorta. A coronary angiography was performed, showing a single right coronary artery with a normal right posterior descending artery (RPD). Supplied by collaterals from the RPD, the LCX was perfused retrogradely, passing by the lateral wall of the ascending aorta without flowing into it, but into the right pulmonary artery. At 23 days of age, surgery was performed with resection of the aortic coarctation and reimplantation of the LCA into the posterior aortic wall. DISCUSSION: This case demonstrates that coronary artery anomalies like ALCAPA may occur together with other cardiac malformations. Despite concomitant cardiac lesions, careful assessment of the coronary arteries is mandatory, including cardiac catheterization in case of doubt.

Timely Pulmonary Valve Replacement May Allow Preservation of Left Ventricular Circumferential Strain in Patients with Tetralogy of Fallot.

INTRODUCTION: Patients with Tetralogy of Fallot (TOF) and pulmonary insufficiency and a dilated right ventricle (RV) may suffer from a reduction in left ventricular (LV) performance. It is not clear whether timely pulmonary valve replacement (PVR) preserves LV mechanics. METHODS: Ten TOF patients who underwent PVR were identified from hospital records, and pre- and postoperative cardiac magnetic resonance images were post-processed with a semi-automatic tissue tracking software. LV circumferential strain, time to peak strain, and torsion were compared before and after PVR. A control group of 10 age-matched normal volunteers was assessed as a comparison. RESULTS: LV circumferential strain did not change before vs. after PVR (basal -18.3 ± 3.7 vs. -20.5 ± 3%, p = 0.082; mid-ventricular -18.4 ± 3.6 vs. -19.1 ± 2%, p = 0.571; apical -22.7 ± 5.2 vs. -22.1 ± 4%; p = 0.703). There was also no difference seen between the baseline strain and normal controls (control basal -18.2 ± 3.3%, p = 0.937; mid -18 ± 3.2%, p = 0.798; apex -24.1 ± 5%, p = 0.552). LV torsion remained unchanged from baseline to post PVR [systolic 2.75 (1.23-9.51) °/cm vs. 2.3 ± 1.2°/cm, p = 0.285; maximum 5.5 ± 3.5°/cm vs. 2.34 (1.37-8.07) °/cm, p = 0.083]. There was no difference in time to measured peak LV circumferential strain before vs. after PVR (basal 0.44 ± 0.1 vs. 0.43 ± 0.05, p = 0.912; mid-ventricular 0.42 ± 0.08 vs. 0.38 ± 0.06, p = 0.186; apical 0.40 ± 0.08 vs. 0.40 ± 0.06, p = 0.995). At the same time, pulmonary regurgitation and RV end-diastolic and end-systolic volume indices decreased and LV end-diastolic volume increased after PVR. RV and LV ejection fractions remained constant. CONCLUSION: PVR allows for favorable remodeling of both ventricular volumes for TOF patients with significant pulmonary regurgitation. In this cohort, LV myocardial functional parameters such as circumferential strain, time to peak strain, and LV torsion were normal at baseline and remain unchanged after PVR.

Prophylactic milrinone for the prevention of low cardiac output syndrome and mortality in children undergoing surgery for congenital heart disease.

BACKGROUND: Children with congenital heart disease often undergo heart surgery at a young age. They are at risk for postoperative low cardiac output syndrome (LCOS) or death. Milrinone may be used to provide inotropic and vasodilatory support during the immediate postoperative period. OBJECTIVES: This review examines the effectiveness of prophylactic postoperative use of milrinone to prevent LCOS or death in children having undergone surgery for congenital heart disease. SEARCH METHODS: Electronic and manual literature searches were performed to identify randomised controlled trials. We searched CENTRAL, MEDLINE, EMBASE and Web of Science in February 2014 and conducted a top-up search in September 2014 as well as clinical trial registries and reference lists of published studies. We did not apply any language restrictions. SELECTION CRITERIA: Only randomised controlled trials were selected for analysis. We considered studies with newborn infants, infants, toddlers, and children up to 12 years of age. DATA COLLECTION AND ANALYSIS: Two review authors independently extracted data according to a pre-defined protocol. We obtained additional information from all study authors. MAIN RESULTS: Three of the five included studies compared milrinone versus levosimendan, one study compared milrinone with placebo, and one compared milrinone verus dobutamine, with 101, 242, and 50 participants, respectively. Three trials were at low risk of bias while two were at higher risk of bias. The number and definitions of outcomes were non-uniform as well. In one study comparing two doses of milrinone and placebo, there was some evidence in an overall comparison of milrinone versus placebo that milrinone lowered risk for LCOS (risk ratio (RR) 0.52, 95% confidence interval (CI) 0.28 to 0.96; 227 participants). The results from two small studies do not provide enough information to determine whether milrinone increases the risk of LCOS when compared to levosimendan (RR 1.22, 95% CI 0.32 to 4.65; 59 participants). Mortality rates in the studies were low, and there was insufficient evidence to draw conclusions on the effect of milrinone compared to placebo or levosimendan or dobutamine regarding mortality, the duration of intensive care stay, hospital stay, mechanical ventilation, or maximum inotrope score (where available). Numbers of patients requiring mechanical cardiac support were also low and did not allow a comparison between studies, and none of the participants of any study received a heart transplantation up to the end of the respective follow-up period. Time to death within three months was not reported in any of the included studies. A number of adverse events was examined, but differences between the treatment groups could not be proven for hypotension, intraventricular haemorrhage, hypokalaemia, bronchospasm, elevated serum levels of liver enzymes, or a reduced left ventricular ejection fraction < 50% or reduced left ventricular fraction of shortening < 28%. Our analysis did not prove an increased risk of arrhythmias in patients treated prophylactically with milrinone compared with placebo (RR 3.59, 95% CI 0.83 to 15.42; 238 participants), a decreased risk of pleural effusions (RR 1.78, 95% CI 0.92 to 3.42; 231 participants), or a difference in risk of thrombocytopenia on milrinone compared with placebo (RR 0.86, 95% CI 0.39 to 1.88; 238 participants). Comparisons of milrinone with levosimendan or with dobutamine, respectively, did not clarify the risk of arrhythmia and were not possible for pleural effusions or thrombocytopenia. AUTHORS' CONCLUSIONS: There is insufficient evidence of the effectiveness of prophylactic milrinone in preventing death or low cardiac output syndrome in children undergoing surgery for congenital heart disease, compared to placebo. So far, no differences have been shown between milrinone and other inodilators, such as levosimendan or dobutamine, in the immediate postoperative period, in reducing the risk of LCOS or death. The existing data on the prophylactic use of milrinone has to be viewed cautiously due to the small number of small trials and their risk of bias.

Stenting of the obstructed ductus venosus as emergency and bridging strategy in a very low birth weight infant with infradiaphragmatic total anomalous pulmonary venous connection.

A very low birth weight neonate (1.2 kg) with total anomalous pulmonary venous connection and obstructed infracardiac pulmonary venous drainage was treated by stenting of the ductus venosus via a transjugular approach. With one reintervention to dilate an in-stent stenosis, the palliative stented situation stabilized the patient for as long as three months until definitive surgical correction at a weight of 2 kg. © 2014 Wiley Periodicals, Inc.

Iron insufficiency and hypovitaminosis D in adolescents with chronic fatigue and orthostatic intolerance.

OBJECTIVES: More than 10% of adolescents suffer from severe fatigue and/or orthostatic intolerance. Adult studies show correlations between iron insufficiency and fatigue as well as between hypovitaminosis D and non-specific pain. We sought to determine whether there were correlations between nutritional factors (iron status, and serum vitamin D levels) and chronic ill health. METHODS: We reviewed records of 188 adolescents with symptoms of fatigue and/or orthostatic intolerance and who underwent autonomic reflex screening. RESULTS: Of the 188 patients, 130 patients (69%) had excessive postural tachycardia (PT) with a heart rate (HR) change of ≥30 bpm. 62 patients (47%, n = 131) had iron insufficiency with low iron stores, and 29 patients (22%, n = 131) were iron deficient. HR change did not correlate to ferritin level (P = 0.15). 21 patients (22%, n = 95) had hypovitaminosis D (25-hydroxyvitamin D ≤20 ng/mL). There was a significant association with hypovitaminosis D and orthostatic intolerance (P = 0.024). CONCLUSION: In patients presenting with chronic fatigue and/or orthostatic intolerance, low ferritin levels and hypovitaminosis D are common, especially in patients with PT.

Exercise performance in adolescents with autonomic dysfunction.

OBJECTIVE: To test the hypothesis that excessive postural tachycardia is associated with deconditioning rather than merely being an independent sign of autonomic dysfunction in patients with postural orthostatic tachycardia syndrome (POTS). STUDY DESIGN: We retrospectively analyzed records from 202 adolescents who underwent both head up-tilt and maximal exercise testing. Patients were classified as POTS if they had ≥ 30 min(-1) rise in heart rate (HR) after tilt-table test; and deconditioned if peak O(2) uptake was < 80% predicted. Changes in HR during exercise and recovery were compared between groups. RESULTS: Two-thirds of patients were deconditioned, irrespective of whether they fulfilled diagnostic criteria for POTS, but peak O(2) uptake among patients with POTS was similar to patients without POTS. HR was higher at rest and during exercise; whereas stroke volume was lower during exercise, and HR recovery was slower in patients with POTS compared with patients without POTS. CONCLUSIONS: Most patients who presented with chronic symptoms of dizziness, fatigue, or pre-syncope, were deconditioned, but, because the proportion of deconditioned patients was similar in POTS vs non-POTS groups, we conclude that HR changes in POTS are not solely because of inactivity resulting in deconditioning.