Conventional spin echo (CSE) T1-weighted sequence with ECG-triggering and respiratory compensation (Bailes et al. 1985) has traditionally been the mainstay for black blood imaging since the early days of cardiac MRI (Herfkens et al. 1983). This CSE sequence is no longer used primarily due to the long scan time. Acquisition time for spin echo sequences can be shortened by using segmented κ-space technique as in TSE or fast spin echo (FSE), by using echo planar imaging (EPI) readout, and by using parallel imaging techniques in combination. Parallel imaging allows for the use of higher number of signal averages (NSA) to compensate for respiratory motion (Chung and Muthupillai 2004) without significant increases in scan time. Respiratory triggering as a means of respiratory compensation can be used to reduce the number of signal averages needed, and in combination with thin sections of 2 mm or less, can be used to evaluate coarctation, vascular rings, vessel wall thickness, and airway compromise (Fig. 2). The general disadvantage with spin echo imaging is an incomplete “black blood” appearance with slow flow artifacts. Double or triple inversion recovery TSE sequence has been shown to produce excellent quality black blood images (Simonetti et al. 1996). However, breath-holding is necessary, which limits its utility in the pediatric population. This sequence can be used on freely-breathing patients with either multiple NSAs or with respiratory triggering, although the scan duration is longer. Clinically, this sequence is used when better nulling of the blood signal is needed, to minimize metallic artifacts from endovascular stents, and for myocardial tissue characterization to improve conspicuity of lesions such as cardiac tumors. A triple inversion recovery sequence provides additional fat suppression and can be used in edema imaging with T2-weighted sequences in acute myocarditis (Friedrich et al. 2009). With some modifications of this sequence, and in combination with respiratory navigator gating, black blood coronary artery imaging can be accomplished (Stuber et al. 2001). The dielectric shading artifacts on 3T MRI scanner has been minimized by multi-transmit radiofrequency (RF) technology such that reproducible diagnostic quality double inversion recovery black blood images can be produced routinely on 3T MRI scanners. (Mueller et al. 2012).

Noninvasive quantitative blood flow analysis is yet another powerful tool that MRI adds to the diagnostic armamentarium. The most widely used pulse sequence is a retrospective ECG-triggered cine phase contrast (PC), also known as velocity-encoded cine MRI (Brenner et al. 1992; Caputo et al. 1991; Helbing et al. 1996; Hundley et al. 1995; Powell and Geva 2000; Rebergen et al. 1993a, b, 1995; Sieverding et al. 1992; Steffens et al. 1994). The accuracy of this technique has been validated both in vitro and in vivo (Bogren et al. 1989; Evans et al. 1993; Firmin et al. 1987; Frayne et al. 1995; Kondo et al. 1991a, b; Powell et al. 2000; Greil et al. 2002a, b). Scan time can be shortened using segmented k-space technique, with the penalty of decreased temporal resolution (Kellenberger et al. 2005), and by combining parallel imaging techniques (Beerbaum et al. 2003; Prakash et al. 2006). Real-time phase contrast has also been applied clinically (Korperich et al. 2004). The main clinical applications include estimation of regurgitant fraction (Fig. 6) in patients with pulmonary regurgitation after repair of right ventricular outflow obstructive lesions such as tetralogy of Fallot (Helbing and de Roos 2000; Oosterhof et al. 2006), differential pulmonary flow to the right and left lungs (Roman et al. 2005; Sridharan et al. 2006), and systemic-to-pulmonary shunts (Powell and Geva 2000). Phase contrast techniques have also been used as a sensitive means of detecting flow dephasing to locate the presence of atrial and ventricular septal defects, to detect flow restrictive conditions (Fig. 3), and also for quantification of the gradient across a stenosis (Oshinski et al. 1996). The Q_P:Q_S is calculated by phase contrast imaging across the main pulmonary artery and ascending aorta, and provides an important decision-making tool in the presence of an atrial septal defect (ASD), ventricular septal defect (VSD), or anomalous pulmonary veins. 3D whole-heart volume acquisition of phase contrast data (4D flow) has been reported widely, although there are compromises related to decreased temporal resolution (Markl et al. 2011; Sorensen et al. 2005). More recently, exciting research work has been presented that combines compressed sensing on the acquisition side to speed up scan time and sophisticated post-processing tools to allow for simultaneous flow quantification, 3D flow visualization, and ventricular functional evaluation using the same 4D flow dataset (Hsiao et al. 2012). This has great potential to allow for operator-independent acquisition of volumetric data for congenital heart disease evaluation, and possibly even improve on flow quantification accuracy (Nordmeyer et al. 2013).

Besides using cine SSFP, as previously described, for evaluation of global and regional myocardial function and wall motion, techniques for perfusion imaging (Schwitter 2006) with fast gradient echo (fGRE) or hybrid EPI-FGRE pulse sequences, (Jahnke et al. 2006), and viability imaging with an inversion recovery T1-weighted FGRE sequence (Kim et al. 2000) have been well established in adults for evaluation of ischemic heart disease. These techniques have also been applied to the pediatric population in the setting of treated congenital heart disease, especially after coronary reimplantation (Buechel et al. 2009a, b; Fratz et al. 2011), prolonged bypass procedures, tumors (Fig. 7), thrombus, right ventricular dysfunction, systemic right ventricle (Babu-Narayan et al. 2005; Fratz et al. 2006), single left ventricle (Rathod et al. 2010; Robinson et al. 2010), cardiomyopathy, myocarditis, and vasculitis (Prakash et al. 2004; Taylor et al. 2005a, b). With the widespread availability of phase-sensitive inversion recovery viability imaging (Kellman et al. 2002), robust viability imaging can be acquired in combination with respiratory navigator for sedated freely-breathing infants and young children. The acquisition can be performed either in end-systole or end-diastole. The right ventricular free wall can be more easily assessed in end-systole as the myocardium is thicker. An extension of this technique is quantitative T1-mapping of the myocardium, which may have potential for diagnosis of microscopic fibrosis in cardiomyopathy or myocarditis (Messroghli et al. 2007).

Another technique of myocardial tissue characterization is T2* mapping. T2* mapping for iron quantification is a well-established clinical cardiac MRI examination in patients with thalassemia and sickle cell disease (Anderson et al. 2001; Westwood et al. 2005a, b; Wood 2009). Typically, a multi-echo fast gradient echo sequence with multiple TE’s is used and the T2* decay is measured in the septal wall after background noise correction (Westwood et al. 2003).

Myocardial tissue tagging was first applied clinically in the late 1980s (Zerhouni et al. 1988; Axel and Dougherty 1989). The basic principle is to “tag” the myocardium with either one-dimensional RF saturation lines or two-dimensional RF saturation grid at the beginning of a cardiac cycle. As the myocardium contracts, the “tag lines or grids” will deform. By analyzing the deformation of these tags, the mechanical properties of the myocardium such as stress and strain can be estimated. With the introduction of harmonic phase (HARP) MRI that allows for a faster and automated method of tagging analysis, there is more widespread application of tagging clinically (Castillo et al. 2005). On the hardware side, with 3T MRI, the tags persist much longer throughout the cardiac cycle compared to that on 1.5T, thereby improving the results of myocardial tagging (Valeti et al. 2006). It has been successfully applied in congenital heart disease (Fogel et al. 1998; Menteer et al. 2005), as well as in pediatric cardiology for the evaluation of cardiomyopathy due to Duchenne muscular dystrophy (Hor et al. 2009, 2011).

Systemic venous anomalies frequently coexist with congenital heart disease, especially in the setting of heterotaxy (Fig. 11). The most common anomaly is a persistent left superior vena cava, which drains into the coronary sinus. A connecting vein between the two superior vena cavae is typically absent in this setting. Interruption of the IVC with azygos continuation, and/or aberrant hepatic venous drainage is frequently present in heterotaxy. These anomalies significantly alter surgical management in single ventricle repair, in which the cavopulmonary connections have to be modified to accommodate the aberrant venous drainage.

Systemic venous obstruction is common after congenital heart surgery or prolonged central vascular catheter usage, and MR venography (MRV) offers an accurate means of diagnosing the extent of thrombosis, the acuity of the process, and the degree and adequacy of collateral venous flow. Simultaneous imaging of the brain should be considered in the setting of SVC or bilateral jugular vein thrombosis to assess for elevated intracranial pressure, which may manifest with ventriculomegaly or intracranial hemorrhage.

Viscero-atrial situs abnormalities and cardiac malpositions are usually easily identified by echo. However, in the presence of situs ambiguous, atrioventricular or ventriculoarterial discordance, or anomalous pulmonary or systemic venous connections (Fig. 11), difficulties may arise in defining the topographic relation of the major cardiac segments (Araoz et al. 2002). MRI is an excellent technique for defining the morphologic features of each atrium and ventricle. MRI depicts anatomical details that are easily related to the surrounding structures of the body, and thus provides reliable diagnoses in heterotaxy (Sorensen et al. 2004; Kersting-Sommerhoff et al. 1990a, b). In patients with complex anomalies, especially in older patients, MRI may be the primary imaging technique so as to maximize noninvasive information prior to catheterization.

The sensitivity of echo for diagnosing secundum-type ASD is close to 100 % in infancy, but drops to 85–90 % in older children and adults. Transesophageal echocardiography (TEE) is a trusted method of sizing an ASD prior to surgery or percutaneous device closure, but is invasive, uncomfortable, and may carry a small risk of morbidity and mortality. MRI may be a useful noninvasive alternative in patients who refuse or are unable to tolerate TEE and may provide additional information on the shape and location of the ASD. ASD sizing by MRI using bFFE and phase contrast (PC) protocols correlates well with TEE estimations, and PC-MRI provides additional information on ASD shape and proximity to adjacent structures (Piaw et al. 2006). MRI guidance has also been used to navigate endovascular catheters and deliver ASD closure devices (Henk et al. 2005).

The superior sinus venosus defect is the most difficult form of atrial septal defect to detect echocardiographically due to the extreme rightward and superior position of this type of defect. The inferior type is also difficult to depict by transthoracic echocardiography because of its infero-posterior location relative to the fossa ovalis. MRI has become the gold standard for depiction of sinus venosus defects (Fig. 13). The coronary sinus septal defect involves partial absence of the atrial septum between the coronary sinus and the left atrium due to incomplete development of the left atrio-venous fold. It may be associated with left SVC to coronary sinus, unroofing of the coronary sinus, and other complex congenital heart defects, and its evaluation by MRI can overcome the limitations of echocardiography. MRI may also identify ASD or partial anomalous pulmonary venous connection in adults with right-sided chamber enlargement, hypertrophy, or dysfunction of unknown etiology. In all forms of septal defects, quantification of shunt size (pulmonary to systemic flow ratio, otherwise known as the Qp:Qs) by flow velocity mapping compares favorably to other imaging techniques, and enables decision-making regarding conservative therapy versus surgery (Hundley et al. 1995).