Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Knowledge of the cardiac anatomy is essential for identifying and understanding cardiovascular disease in patients and is therefore important in clinical practice. To date, various imaging techniques such as conventional chest radiography, cardiac magnetic resonance (CMR), computed tomography (CT) and echocardiography are all used to assess aspects of cardiac and vascular anatomy. This chapter contains an overview of the techniques used and provides examples of the anatomy identified with these techniques.
Postero-anterior and lateral chest radiographs are commonly obtained in patients with cardiovascular disease. The chest radiograph provides an impression of the size of cardiovascular structures and the lung parenchyma. Specific cardiac chambers and large vessel anatomy can be appreciated on chest radiography. Evaluation of the lung parenchyma and lung vasculature is helpful for assessing and grading heart failure. Valvular calcifications may be recognised as a clue for specific valvular disease. Fig. 12.1 illustrates the normal chest radiograph, noting normal cardiovascular structures.
Cardiovascular structures and lung parenchymaCardiac Magnetic Resonance
Various imaging planes (body axis- and cardiac planes)
Anatomy of thoracic vascular structures
Cardiac anatomy and function
An advantage of choosing magnetic resonance for cardiac imaging is the free choice in obtaining imaging planes of cardiovascular anatomy in any arbitrary view, since this technique is not hampered by the limited availability of acoustic windows, as with ultrasound. This benefit is especially advantageous when imaging the morphology of the right ventricle (RV), which is excellently delineated by CMR, whereas in echocardiography the assessment of RV geometry and function is challenging because of the particular crescentic shape of the RV as it wraps around the left ventricle (LV). Furthermore, the unrestricted field of view of CMR allows superior visualisation of extracardiac and large vessel anatomy.
Single-plane two-dimensional (2-D) or multiple-plane 2-D or three-dimensional (3-D) imaging is possible with CMR. Dynamic functional information can be obtained by synchronising image acquisition to the interval of the R-waves on the electrocardiogram, using either prospective triggering or retrospective gating. With prospective triggering, the operator needs to set the expected heart rate before the acquisition and triggering will then be performed according to this defined heart rate. With retrospective gating, imaging is performed continuously and the ECG signal is stored additionally. In retrospect, image reconstruction is synchronised to the stored ECG, providing time-resolved imaging in multiple phases of the cardiac cycle, which can then be presented in cine mode.
Imaging planes in CMR are usually obtained in the orientation to the axes of the heart, or oriented to the major axes of the body. Therefore, the standard CMR planes of the heart are comparable to the standard cardiac views, well-known and established in other non-invasive imaging techniques such as echocardiography, cardiac CT, x-ray LV angiography and nuclear medicine.
The choice for a specific CMR protocol is mainly determined by the clinical questions that need to be answered. Standardised nomenclature for cross-sectional anatomy has been described, facilitating comparison between different techniques and proper communication between imaging specialists.
Another important issue in clinical CMR imaging is the ability of the patient to collaborate during the examination and to perform breath-holding repeatedly and consistently. If a patient is capable of performing breath-holding, successive imaging planes are obtained with accelerated imaging, with the patient usually performing breath-holding in end expiration, as the anatomical level may be more reproducible than planes that are examined in inspiration.
CMR techniques for anatomical evaluation include bright-blood and black-blood imaging, which essentially determines the contrast in signal intensity between myocardium and the intracardiac blood pool. For the assessment of left and right ventricular function, fast-gradient echo sequences are usually performed in combination with steady-state free-precession (SSFP) technique (balanced-Turbo Field Echo [TFE], True-Fast Imaging Steady state Precession [FISP], Fast Imaging Employing Steady-state Acquisition [FIESTA]) for optimal contrast. On these bright-blood images, the blood pool is presented with bright signal, whereas the myocardium is represented dark with low signal. This results in an excellent definition of the left ventricular endocardial and epicardial borders, which is required for accurate image segmentation during cardiac volume and function quantification. Typically, SSFP images should be acquired with slice thickness of 6–8 mm and temporal resolution less than 45 ms to obtain optimal accuracy in ventricular function assessment.
Additionally, cardiac morphology can be evaluated by double-inversion, black-blood, spin-echo sequences with fat suppression, providing gated, static images of the heart with high spatial resolution (optimally, in-plane-acquired resolution of less than 2 × 2 mm 2 and slice thickness of 5–8 mm) in the orientation of the heart or the patient's body axes. Multiple other magnetic resonance imaging techniques are available for tissue characterisation (e.g. T 2 weighted sequences, delayed gadolinium-based contrast enhancement), extending the capabilities of CMR beyond anatomical and functional evaluation of the cardiovascular system.
To acquire imaging planes in the direction of the cardiac axes, SSFP scout views are used for planning. If available, free-breathing images obtained during real-time imaging can be used instead. Perpendicular to an anatomical transverse image, which displays the heart's four chambers, an acquisition plane is chosen through the middle of the atrioventricular junction at the level of the mitral valve and running through the apex ( Figs 12.2A and B ). This plane is the so-called vertical long-axis (VLA) plane (see Fig. 12.2C ). On this VLA view, a plane is defined intersecting the apex and the middle of the mitral valve, resulting in the horizontal long-axis (HLA) view (see Fig. 12.2D ). This HLA view is almost comparable to the four-chamber view; however, often only a part of the left ventricular outflow tract is visualised in this HLA view. On the acquired HLA plane, the short-axis (SA) views (see Figs 12.2E and F ) covering the entire LV are planned parallel to the ring of the mitral valve and perpendicular to the line intersecting the apex.
For reproducibility and comparison purposes, the true two- and four-chamber views can still be obtained (see Fig. 12.2G and H ). The two-chamber view is planned perpendicular to the anterior and inferior wall of the LV through the centre of the left ventricular cavity on a mid-ventricular SA image intersecting the apex. On the two-chamber view the apex, anterior and inferior wall of the LV, the mitral valve and left atrium can be evaluated. The four-chamber view is also planned on a mid-ventricular SA image by a plane through the centre of the left ventricular cavity and the acute margin of the RV, also intersecting the apex (see Fig. 12.2E ). The four-chamber view depicts the interventricular septum, the lateral wall of the LV, the free wall of the RV, the left and right atrium as well as the interatrial septum and both the mitral and tricuspid valves.
Routinely, the three-chamber or so-called left ventricular outflow tract (see Fig. 12.2I ) view is planned on a basal SA plane (see Fig. 12.2F ), and also intersects the apex. The left ventricular outflow tract view depicts the apex, the anteroseptal interventricular wall, the left ventricular outflow tract, the inferolateral wall, and the aortic and mitral valve.
The standard SSFP cine CMR protocol for assessing left ventricular function should include the two-, three and four-chamber views in combination with SA images covering the entire LV, resulting in images covering all described 17 left ventricular segments in two directions.
Additionally, the right ventricular outflow tract (RVOT) can be obtained ( Fig. 12.3 ). This view can be planned on a coronal image, depicting the outflow tract of the RV. Alternatively, an optimised view of the RVOT view can be obtained from a plane outlining the tricuspid valve plane and the outflow tract. On this plane the outflow tract, pulmonary valve, tricuspid valve and the basal (diaphragmatic) part of the right ventricular wall are all visualised.
For the evaluation of cardiac morphology, the pericardium, the great thoracic vessels and (para-) cardiac masses, imaging planes oriented to the main body axes are obtained. Also, the transverse (or axial), coronal (frontal) and sagittal planes are well known to clinicians, as these anatomical orientations are similar to clinical (cardiac) CT. Black- and bright-blood sequence approaches ( Fig. 12.4 ) can be used in optimally adjusted planes to answer specific clinical questions. Black-blood images provide only static information in a single phase and are not suitable for quantification of left or right ventricular dimensions. For this analysis, SSFP multiphase images with appropriate temporal resolution are necessary.
Transversely orientated planes ( Fig. 12.5 ) are especially useful for the evaluation of thoracic vascular structures including the ascending and descending thoracic aorta, the superior and inferior vena cava, the pulmonary trunk and right and left pulmonary artery. The right and left pulmonary veins entering the left atrium are also well depicted. Images in transverse orientation through the heart allow the evaluation of morphology of the ventricles and atria. Also, the right ventricular free wall, the RVOT, the pericardium and mediastinum are well depicted. It has been suggested that right ventricular volume and function quantification by planimetry can be performed more accurately on transversely oriented images instead of SA images.
Coronal or frontal anatomical views can be instructive for analysing the connection between the heart and the great vessels. An advantage of the frontal view is the similarity to the well-known anatomy from chest radiography. On sagittal images, the RVOT in relation to the pulmonary valve is well outlined and the connection of the right atrium with the superior and inferior vena cava can be studied.
CMR images present distinct anatomical features of both atria and ventricles. For evaluating anatomy, either cardiac axes ( Fig. 12.6 ) or body axes imaging planes (see Figs 12.4 and 12.5 ) can be chosen. The pericardial sac encloses the heart and the roots of the great vessels. The pericardial cavity is outlined by the parietal and visceral layer of the inner pericardium. Normal pericardium has a longer T 1 than fat tissue and, therefore, yields low-signal intensity on T 1 weighted MR images and can be well visualised due to the surrounding epicardial and pericardial fat. Normally, the thickness of the pericardium measures less than 4 mm on CMR images.
In normal cardiac anatomy, the right atrium can be recognised by identifying the corresponding broad-based triangular appendage. At the base, the tricuspid valve, positioned between the right atrium and the RV, is located closer to the apex compared with the mitral valve.
The right atrium receives venous blood from the superior and inferior vena cava and the coronary sinus. The coronary sinus enters the right atrium in the posterior atrioventricular groove. The appendage of the morphological left atrium has a narrow attachment to the atrium and is more tubular shaped. Characteristically, the left atrium receives four pulmonary veins in total—two on either side—although several variations occur. To date, imaging the venous anatomy of the heart is becoming more relevant. For example, during pre-ablation work-up for supraventricular arrhythmias, the clinician needs to be informed about the exact anatomy of the left atrial morphology and number of the pulmonary ostia, as the left atrium and pulmonary veins are used to guide the interventional procedure. The interatrial septum separates the two atria. As part of the interatrial septum, the fossa ovalis is very thin and can hardly be depicted on CMR images due to the limited spatial acquisition resolution.
The RV is normally triangular in shape and anteriorly located relative to the LV, directly behind the sternum. Morphologically, the RV has typical features that can be depicted on CMR images. The RV shows a muscular moderator band ( Fig. 12.7 ) carrying branches of the conducting system. Furthermore, the RV contains a muscular outflow tract (infundibulum or conus arteriosus) and typically, the RV wall is more trabeculated than the left. In normal anatomy, the LV is positioned posteriorly and to the left. The septum is smooth with no trabeculae and the left ventricular outflow tract lacks a muscular part. The interventricular septum consists of a muscular and a membranous part. In particular, the membranous part is very thin and is sometimes not depicted on CMR images. It is important to recognise these normal anatomical features of atrial and ventricular morphology because the position of the atria and ventricles may be inversed in complex congenital heart disease.
At the outlet of each of the heart's four chambers, one-way valves are positioned to ensure that blood flows in the proper direction. The blood flow through the atria into the ventricles is regulated by the atrioventricular valves (the tricuspid valve is related to the morphological RV, the mitral valve to the morphological LV). The pulmonary valve connects the outflow tract of the RV to the pulmonary trunk and the aortic valve connects the left ventricular outflow tract to the thoracic aorta. The normal tricuspid valve consists of three cusps, whereas the mitral valve consists of two cusps. Both the normal pulmonary and the aortic valve ( Fig. 12.8 ) normally consist of three cusps. Opening of the atrioventricular valves is predominantly determined by pressure differences between the atria and ventricles, which are the result of the isovolumetric relaxation of the ventricles during diastole. Furthermore, the motion of the valves is regulated by papillary muscles, which originate from the inferolateral and anterolateral left ventricular myocardial wall and are connected to the valve leaflets by chordae tendineae. During contraction of the ventricle, the papillary muscles also contract, pulling on the chordae tendineae, closing the valves and preventing blood flow from the ventricles into the atria (i.e. regurgitation). Normally, in the RV three small papillary muscles can be depicted: the anterior, posterior and septal papillary muscle. The LV reveals two larger papillary muscles, the anterior and posterior papillary muscle; however, there is quite a wide variation in this standard morphology.
Cine SSFP long-axis and SA images, as well as transverse images, are all well suited for depicting morphology and function of the valvular apparatus. The valve leaflets can be depicted if spatial resolution is adequate. Dedicated acquisitions of specific valvular planes are used to image the valve area, which is especially useful when studying aortic valve stenosis or incompetence. Both SSFP and fast gradient-echo sequences are used for valvular imaging. Papillary muscles are well visualised on both cine bright- and black-blood sequences. Chordae tendineae, on the other hand, are difficult to visualise on CMR due to the limited spatial resolution (see Fig. 12.6 ).
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here