A Deep Learning Pipeline for Assessing Ventricular Volumes from a Cardiac MRI Registry of Patients with Single Ventricle Physiology.

Purpose To develop an end-to-end deep learning (DL) pipeline for automated ventricular segmentation of cardiac MRI data from a multicenter registry of patients with Fontan circulation (Fontan Outcomes Registry Using CMR Examinations [FORCE]). Materials and Methods This retrospective study used 250 cardiac MRI examinations (November 2007-December 2022) from 13 institutions for training, validation, and testing. The pipeline contained three DL models: a classifier to identify short-axis cine stacks and two U-Net 3+ models for image cropping and segmentation. The automated segmentations were evaluated on the test set (n = 50) by using the Dice score. Volumetric and functional metrics derived from DL and ground truth manual segmentations were compared using Bland-Altman and intraclass correlation analysis. The pipeline was further qualitatively evaluated on 475 unseen examinations. Results There were acceptable limits of agreement (LOA) and minimal biases between the ground truth and DL end-diastolic volume (EDV) (bias: -0.6 mL/m2, LOA: -20.6 to 19.5 mL/m2) and end-systolic volume (ESV) (bias: -1.1 mL/m2, LOA: -18.1 to 15.9 mL/m2), with high intraclass correlation coefficients (ICCs > 0.97) and Dice scores (EDV, 0.91 and ESV, 0.86). There was moderate agreement for ventricular mass (bias: -1.9 g/m2, LOA: -17.3 to 13.5 g/m2) and an ICC of 0.94. There was also acceptable agreement for stroke volume (bias: 0.6 mL/m2, LOA: -17.2 to 18.3 mL/m2) and ejection fraction (bias: 0.6%, LOA: -12.2% to 13.4%), with high ICCs (>0.81). The pipeline achieved satisfactory segmentation in 68% of the 475 unseen examinations, while 26% needed minor adjustments, 5% needed major adjustments, and in 0.4%, the cropping model failed. Conclusion The DL pipeline can provide fast standardized segmentation for patients with single ventricle physiology across multiple centers. This pipeline can be applied to all cardiac MRI examinations in the FORCE registry. Keywords: Cardiac, Adults and Pediatrics, MR Imaging, Congenital, Volume Analysis, Segmentation, Quantification Supplemental material is available for this article. © RSNA, 2023.

Real-Time Strain-Encoding Cardiovascular MRI for Assessment of Regional Heart Function in Tetralogy of Fallot Patients.

BACKGROUND: Tetralogy of Fallot (ToF) is the most common form of cyanotic congenital heart disease, where right ventricular (RV) function is an important determinant of subsequent intervention. OBJECTIVE: In this study, we evaluate the feasibility of fast strain-encoding (fastSENC; a one-heartbeat sequence) magnetic resonance imaging (MRI) for assessing regional cardiac function in ToF. METHOD: FastSENC was implemented to characterize regional circumferential (Ecc) and longitudinal (Ell) strains in the left ventricle (LV) and RV in post-repair ToF. Data analysis was conducted to compare strain measurements in the RV to those in the LV, as well as to those generated by the MRI Tissue-Tracking (MRI-TT) technique, and to assess the relationship between strain and ejection fraction (EF). RESULTS: Despite normal LVEF (55±8.5%), RVEF was borderline (46±6.4%), but significantly lower than LVEF. RV strains (RV-Ell=-20.2±2.9%, RV-Ecc=-15.7±6.4%) were less than LV strains (LV-Ell=-21.7±3.7%, LV-Ecc=-18.3±4.7%), and Ell was the dominant strain component. Strain differences between fastSENC and MRI-TT were less significant in RV than in LV. There existed moderate and weak correlations for RV-Ecc and RV-Ell, respectively, against RVEF. Compared to LV strain, RV strain showed regional heterogeneity with a trend for reduced strain from the inferior to anterior regions. Inter-ventricular strain delay was larger for Ell (64±47ms) compared to Ecc (36±40ms), reflecting a trend for contraction dyssynchrony. CONCLUSION: FastSENC allows for characterizing subclinical regional RV dysfunction in ToF. Due to its sensitivity for evaluating regional myocardial contractility patterns and real-time imaging capability without the need for breath-holding, fastSENC makes it more suitable for evaluating RV function in ToF.

Cardiac Magnetic Resonance Imaging to Determine Single Ventricle Function in a Pediatric Population is Feasible in a Large Trial Setting: Experience from the Single Ventricle Reconstruction Trial Longitudinal Follow up.

The Single Ventricle Reconstruction (SVR) Trial was a randomized prospective trial designed to determine survival advantage of the modified Blalock-Taussig-Thomas shunt (BTTS) vs the right ventricle to pulmonary artery conduit (RVPAS) for patients with hypoplastic left heart syndrome. The primary aim of the long-term follow-up (SVRIII) was to determine the impact of shunt type on RV function. In this work, we describe the use of CMR in a large cohort follow up from the SVR Trial as a focused study of single ventricle function. The SVRIII protocol included short axis steady-state free precession imaging to assess single ventricle systolic function and flow quantification. There were 313 eligible SVRIII participants and 237 enrolled, ages ranging from 10 to 12.5 years. 177/237 (75%) participants underwent CMR. The most common reasons for not undergoing CMR exam were requirement for anesthesia (n = 14) or ICD/pacemaker (n = 11). A total of 168/177 (94%) CMR studies were diagnostic for RVEF. Median exam time was 54 [IQR 40-74] minutes, cine function exam time 20 [IQR 14-27] minutes, and flow quantification time 18 [IQR 12-25] minutes. There were 69/177 (39%) studies noted to have intra-thoracic artifacts, most common being susceptibility artifact from intra-thoracic metal. Not all artifacts resulted in non-diagnostic exams. These data describe the use and limitations of CMR for the assessment of cardiac function in a prospective trial setting in a grade-school-aged pediatric population with congenital heart disease. Many of the limitations are expected to decrease with the continued advancement of CMR technology.

Congenitally corrected transposition: not correct at all.

PURPOSE OF REVIEW: Congenitally corrected transposition of the great arteries is a rare congenital defect with several management options. Disagreement continues on strategies, such as anatomic repair, physiologic repair or observation-only. This review discusses recent data that provide further guidance for clinical decision-making. RECENT FINDINGS: New data provide greater insights into practice patterns and outcomes. Recent data from high-volume centers show progressively high rates of systemic right ventricle dysfunction over time with lower rates of systemic left ventricle dysfunction following anatomic repair; there is a statistical trend towards better survival of anatomic repair patients. Data comparing anatomic repair to observation showed that anatomic repair patients had a lower hazard of reaching a composite adverse outcome. These complex operations are predominantly performed at a small subset of congenital heart surgery centers. SUMMARY: Anatomic repair compared with physiologic repair may have better outcomes, although there are relatively high rates of morbidity for both approaches. In the patient without associated lesions, nonsurgical management can have excellent outcomes but is complicated by right ventricular failure over time. Multicenter research will help determine risk factors for bad outcomes; management at high volume, experienced centers will probably be beneficial for this complex patient population.

Effects of Mitral Valve Prolapse on Quantification of Mitral Regurgitation and Ejection Fraction Using Cardiac MRI.

Purpose: To determine the impact of prolapsed volume on regurgitant volume (RegV), regurgitant fraction (RF), and left ventricular ejection fraction (LVEF) in patients with mitral valve prolapse (MVP) using cardiac MRI. Materials and Methods: Patients with MVP and mitral regurgitation who underwent cardiac MRI from 2005 to 2020 were identified retrospectively from the electronic record. RegV is the difference between left ventricular stroke volume (LVSV) and aortic flow. Left ventricular end-systolic volume (LVESV) and LVSV were obtained from volumetric cine images, with prolapsed volume inclusion (LVESVp, LVSVp) and exclusion (LVESVa, LVSVa) providing two estimates of RegV (RegVp, RegVa), RF (RFp, RFa), and LVEF (LVEFa, LVEFp). Interobserver agreement for LVESVp was assessed using intraclass correlation coefficient (ICC). RegV was also calculated independently using measurements from mitral inflow and aortic net flow phase-contrast imaging as the reference standard (RegVg). Results: The study included 19 patients (mean age, 28 years ± 16 [SD]; 10 male patients). Interobserver agreement for LVESVp was high (ICC, 0.98; 95% CI: 0.96, 0.99). Prolapsed volume inclusion resulted in higher LVESV (LVESVp: 95.4 mL ± 34.7 vs LVESVa: 82.4 mL ± 33.8; P < .001), lower LVSV (LVSVp: 100.5 mL ± 33.8 vs LVSVa: 113.5 mL ± 35.9; P < .001), and lower LVEF (LVEFp: 51.7% ± 5.7 vs LVEFa: 58.6% ± 6.3; P < .001). RegV was larger in magnitude when prolapsed volume was excluded (RegVa: 39.4 mL ± 21.0 vs RegVg: 25.8 mL ± 22.8; P = .02), with no evidence of a difference when including prolapsed volume (RegVp: 26.4 mL ± 16.4 vs RegVg: 25.8 mL ± 22.8; P > .99). Conclusion: Measurements that included prolapsed volume most closely reflected mitral regurgitation severity, but inclusion of this volume resulted in a lower LVEF.Keywords: Cardiac, MRI© RSNA, 2023See also commentary by Lee and Markl in this issue.

Computational modelling of multi-temporal ventricular-vascular interactions during the progression of pulmonary arterial hypertension.

A computational framework is developed to consider the concurrent growth and remodelling (G&R) processes occurring in the large pulmonary artery (PA) and right ventricle (RV), as well as ventricular-vascular interactions during the progression of pulmonary arterial hypertension (PAH). This computational framework couples the RV and the proximal PA in a closed-loop circulatory system that operates in a short timescale of a cardiac cycle, and evolves over a long timescale due to G&R processes in the PA and RV. The framework predicts changes in haemodynamics (e.g. 68.2% increase in mean PA pressure), RV geometry (e.g. 38% increase in RV end-diastolic volume) and PA tissue microstructure (e.g. 90% increase in collagen mass) that are consistent with clinical and experimental measurements of PAH. The framework also predicts that a reduction in RV contractility is associated with long-term RV chamber dilation, a common biomarker observed in the late-stage PAH. Sensitivity analyses on the G&R rate constants show that large PA stiffening (both short and long term) is affected by RV remodelling more than the reverse. This framework can serve as a foundation for the future development of a more predictive and comprehensive cardiovascular G&R model with realistic heart and vascular geometries.

Data-driven computational models of ventricular-arterial hemodynamics in pediatric pulmonary arterial hypertension.

Pulmonary arterial hypertension (PAH) is a complex disease involving increased resistance in the pulmonary arteries and subsequent right ventricular (RV) remodeling. Ventricular-arterial interactions are fundamental to PAH pathophysiology but are rarely captured in computational models. It is important to identify metrics that capture and quantify these interactions to inform our understanding of this disease as well as potentially facilitate patient stratification. Towards this end, we developed and calibrated two multi-scale high-resolution closed-loop computational models using open-source software: a high-resolution arterial model implemented using CRIMSON, and a high-resolution ventricular model implemented using FEniCS. Models were constructed with clinical data including non-invasive imaging and invasive hemodynamic measurements from a cohort of pediatric PAH patients. A contribution of this work is the discussion of inconsistencies in anatomical and hemodynamic data routinely acquired in PAH patients. We proposed and implemented strategies to mitigate these inconsistencies, and subsequently use this data to inform and calibrate computational models of the ventricles and large arteries. Computational models based on adjusted clinical data were calibrated until the simulated results for the high-resolution arterial models matched within 10% of adjusted data consisting of pressure and flow, whereas the high-resolution ventricular models were calibrated until simulation results matched adjusted data of volume and pressure waveforms within 10%. A statistical analysis was performed to correlate numerous data-derived and model-derived metrics with clinically assessed disease severity. Several model-derived metrics were strongly correlated with clinically assessed disease severity, suggesting that computational models may aid in assessing PAH severity.

SCMR expert consensus statement for cardiovascular magnetic resonance of acquired and non-structural pediatric heart disease.

Cardiovascular magnetic resonance (CMR) is widely used for diagnostic imaging in the pediatric population. In addition to structural congenital heart disease (CHD), for which published guidelines are available, CMR is also performed for non-structural pediatric heart disease, for which guidelines are not available. This article provides guidelines for the performance and reporting of CMR in the pediatric population for non-structural ("non-congenital") heart disease, including cardiomyopathies, myocarditis, Kawasaki disease and systemic vasculitides, cardiac tumors, pericardial disease, pulmonary hypertension, heart transplant, and aortopathies. Given important differences in disease pathophysiology and clinical manifestations as well as unique technical challenges related to body size, heart rate, and sedation needs, these guidelines focus on optimization of the CMR examination in infants and children compared to adults. Disease states are discussed, including the goals of CMR examination, disease-specific protocols, and limitations and pitfalls, as well as newer techniques that remain under development.

mRNA Coronavirus Disease 2019 Vaccine-Associated Myopericarditis in Adolescents: A Survey Study.

In this survey study of institutions across the US, marked variability in evaluation, treatment, and follow-up of adolescents 12 through 18 years of age with mRNA coronavirus disease 2019 (COVID-19) vaccine-associated myopericarditis was noted. Only one adolescent with life-threatening complications was reported, with no deaths at any of the participating institutions.

Ventricular Function in Physiologically Repaired and Unrepaired Congenitally Corrected Transposition of the Great Arteries.

In patients with congenitally corrected transposition of the great arteries (ccTGA) and hemodynamically significant concomitant lesions, physiologic repair may be undertaken, in which the circulation is septated but the morphologic right ventricle (RV) remains the systemic ventricle. Patients without significant concomitant lesions may be observed without surgery, with a similar physiologic result. We compared cardiovascular magnetic resonance measures of ventricular size and function in patients with physiologically repaired and unrepaired ccTGA. Patients with ccTGA who underwent cardiovascular magnetic resonance at our center between September 2007 and July 2019 were analyzed. In 38 patients identified (12, physiologically repaired; 26, unrepaired; mean age 34.5 [18.7 to 52.0] years), there was a higher proportion of RV ejection fraction ≤45% in physiologically repaired (75% vs unrepaired 35%, p = 0.02). Physiologically repaired patients had worse left ventricle global longitudinal strain (-14.9% ± 5.0% vs unrepaired patients -18.4% ± 2.7%, p = 0.04). The difference in tricuspid regurgitant fraction between groups did not achieve statistical significance (physiologically repaired 27.4 ± 11.1% vs unrepaired patients 19.2 ± 13.0%, p = 0.08). Evaluation for late gadolinium enhancement was more commonly undertaken in physiologically repaired patients (8 of 12 vs unrepaired 7 of 26, p = 0.03) and present more frequently in the left ventricle in physiologically repaired patients in patients evaluated (6 of 8 vs unrepaired 0 of 7, p = 0.01). In conclusion, ventricular function is decreased in patients with ccTGA undergoing physiologic repair compared with those without previous surgery. These cohorts should be considered separately when using ventricular function as an outcome. RV dysfunction is concerning for long-term outcomes following physiologic repair.

Neo-aortic Root Dilatation, Aortic Stiffness, and Ventricular interactions in Long-Term Follow-Up After the Ross Procedure in Childhood.

Patients after the Ross procedure are at risk for right (RV) and left ventricular (LV) dysfunction due to neo-aortic and pulmonary dysfunction. While neo-aortic root dilatation has been related to LV dysfunction, the potential contributions of aortic stiffness and ventricular interactions have not been evaluated. Patients status post Ross procedure up to age 18 years with cardiac magnetic resonance (CMR) exam from 2007 to 2018 were retrospectively reviewed. Aortic pulse wave velocity (PWV) was calculated from phase contrast and angiogram images. RV and LV peak global longitudinal (GLS) and circumferential strain (GCS) were measured using tissue tracking software. Multivariable regression was performed for variables associated with parameters of LV function. In 58 patients (median age 20.5 years at CMR exam), male gender, longer time since Ross procedure, aortic root dilatation, and lower RV ejection fraction (EF) were associated with decreased LV EF. There was no association with LV late gadolinium enhancement or neo-aortic or conduit regurgitation. LV GCS and GLS also correlated with RV GCS, RV GLS and PWV. In multivariable analysis, the relation of RV and LV systolic function, but not aortic measurements, remained significant. In conclusion, in long-term follow-up after pediatric Ross procedure, RV function rather than aortic root size or aortic stiffness most closely relates to LV function. Ventricular interactions may impact decision-making on timing of conduit intervention, which could differ from established criteria in populations with only aortic or pulmonary valve disease. Further study is warranted to evaluate possible association with clinical outcome.

Differences in Pulmonary and Systemic Flow Measurements by Cardiac Magnetic Resonance vs Cardiac Catheterization and Relation to Collateral Flow in Single Ventricle Patients.

Both cardiac magnetic resonance (CMR) and cardiac catheterization (cath) may assess patients with single ventricle physiology prior to stage II or Fontan palliation. However, development of significant aortopulmonary collaterals may invalidate assumptions of the Fick method. We compared CMR and cath flow measurements and evaluated the relation to collateral flow. This single-center study included all pre-stage II and pre-Fontan patients between 2010 and 2017 with CMR and cath within 1 month. Pulmonary (Qp) and systemic flow (Qs) by cath were calculated by Fick method. CMR Qp was calculated by total pulmonary venous flow, and Qs by total vena caval flow. Collateral flow by CMR was the difference of pulmonary vein and pulmonary artery flow. In 26 studies (16 pre-stage II and 10 pre-Fontan) in 21 patients, collateral flow was higher in pre-Fontan patients (1.8 ± 0.6 vs 0.9 ± 0.8 L/min/m2, p = 0.01). Overall, CMR and cath had good agreement for Qs and Qp:Qs, with moderate correlation (r = 0.44, p = 0.02 for Qs, r = 0.48, p = 0.02 for Qp:Qs). In pre-Fontan but not in pre-stage II patients, CMR had higher Qp (mean difference - 1.71 L/min/m2) and Qp:Qs (mean difference - 0.36). The underestimation of cath Qp correlated with amount of collateral flow (r = - 0.47, p = 0.02). Neither cath nor CMR flow measurements correlated with outcomes in this small cohort. In conclusion, collaterals lead to systematically higher Qp and Qp:Qs measurements by CMR vs cath in single ventricle patients. Measurements may not be used interchangeably, with potential clinical significance in estimating pulmonary vascular resistance. Further study is necessary to evaluate possible relation to clinical outcomes.

Identifying Risk Factors for Massive Right Ventricular Dilation in Patients With Repaired Tetralogy of Fallot.

In repaired tetralogy of Fallot (rTOF), pulmonary insufficiency results in varying degrees of right ventricle (RV) dilation. A subset of patients is diagnosed at initial cardiac magnetic resonance imaging (CMR) with a massively dilated RV, far beyond pulmonary valve replacement (PVR) criteria, which is unlikely to return to normal size after PVR. This study aimed to identify risk factors for massive RV dilation at initial CMR. This nested case-control study included all patients at our institution with rTOF and massive RV dilation (indexed RV end-diastolic volume [RVEDVi] ≥200 ml/m2) on initial CMR. Patients were matched by age at first CMR, gender, and type of repair with rTOF controls with RVEDVi<200 ml/m2. In 39 cases (median RVEDVi 227 ml/m2, interquartile range [IQR] 213 to 250) and 73 controls (median RVEDVi 155 ml/m2, IQR 130 to 169), repair at >6 months of age, longer QRS duration, and non-Caucasian race were significantly associated with massive RV dilation on univariate analysis. In multivariate analysis, repair at >6 months of age (adjusted odds ratio [AOR] 2.90, 95% confidence interval [CI] 1.12 to 7.55, p = 0.03), longer QRS duration (AOR = 1.03, 95% CI 1.01 to 1.05, p = 0.005), and non-Caucasian race (AOR = 7.84, 95% CI 1.76 to 34.8, p = 0.01) remained independently associated with massive RV dilation. Era of repair, history of systemic to pulmonary shunt palliation, genetic anomaly, and additional cardiac lesions did not differ between groups. In conclusion, these risk factors identify a subset of patients who may benefit from earlier CMR evaluation to avoid massive irreversible RV dilation.

Biatrial Drainage of the Right Superior Vena Cava: Imaging Findings.

Biatrial drainage of the right superior vena cava (SVC) is a rare form of interatrial shunting that can have substantial clinical consequences. Cross-sectional imaging techniques (CT and MRI) are well suited for evaluation and surgical planning. This review article focuses on the embryologic development, hemodynamics, and imaging features to enable a timely diagnosis. Biatrial drainage of the right SVC has important clinical implications, and knowledge of its imaging appearance and hemodynamics is essential in diagnosis and treatment planning. © RSNA, 2020.

Differential Myocardial Mechanics in Volume and Pressure Loaded Right Ventricles Demonstrated by Cardiac Magnetic Resonance.

In patients with D-looped transposition of the great arteries (D-TGA) status post atrial switch operation, the systemic right ventricle (RV) shifts to predominantly circumferential (CS) rather than longitudinal strain (LS), which may represent adaptation or dysfunction. We aimed to evaluate myocardial mechanics in pressure loaded, volume-loaded, and normal RVs by cardiac magnetic resonance (CMR). Patients with D-TGA post atrial switch operation with CMR from 2008 to 2015 were matched 1:1 for age and RV ejection fraction (EF) with repaired tetralogy of Fallot (TOF) patients (volume-loaded RVs), and 1:1 for age with control patients. RV free wall LS and CS were measured using feature tracking software (TomTec, Unterscleissheim, Germany). A total of 32 D-TGA (median age 32 years, 56% male), 32 TOF, and 32 control patients were included. D-TGA patients had less dilatation than TOF patients (125 ± 35 ml/m2 vs. 149 ± 44 ml/m2, p = 0.02) and lower RVEF than controls (42.9 ± 7.7% vs. 56.3 ± 5.6%, p < 0.0001). RV LS was similar in D-TGA and TOF ( - 13.2 ± 4.5% vs. - 14.5 ± 5.9%, p = 0.32), both decreased compared to controls. However, CS in D-TGA was higher than controls ( - 14.1 ± 4.1% vs. - 11.4 ± 4.4%, p = 0.01), with a higher CS:LS ratio (1.2 ± 0.7 vs. 0.6 ± 0.3, p < 0.0001), while CS in TOF and controls did not differ. RVEF in D-TGA correlated closely with CS (r = - 0.85, p < 0.0001) but not LS (r = 0.10, p = 0.58). I n conclusion, CMR can differentiate strain patterns in pressure- and volume-loaded RVs, with decreased LS in both conditions, while systemic RVs compensate with supra-normal CS. CS may be a more clinically relevant measure of RV function in this population.