At our institution, imaging begins with acquisition of an abbreviated set of ungated localizer images in all three planes. From these images, the imaging planes for the left ventricle are determined. Vector ECG-gated steady-state free precession (SSFP) pulse sequences are used to assess the heart as a stack of contiguous images in the short-axis oblique, four-chamber, vertical long-axis, and LVOT imaging planes. Fast cine phase-contrast imaging is then performed in the LVOT plane. After that sequence, gadolinium is administered and delayed imaging of the myocardium is performed in the short-axis oblique, four-chamber, and vertical long-axis planes.

Left ventricular wall thickness should be calculated during end-diastole, preferably in the short axis for assessment of the mid to basal thirds of the left ventricular myocardium. The vertical and horizontal long-axis planes are useful in evaluating wall thickness at the left ventricular apex. Measurements of thickness in these planes, however, should be made with care. Images chosen for measurement should be cross-referenced with another view to avoid obliquity because only a slice that perpendicularly transects the left ventricular wall will yield accurate measurements [13]. Care should also be taken to exclude free muscle bundles when measuring wall thickness.

Measurements of left ventricular volumes, function, and mass are derived from an SSFP pulse sequence in the left ventricular short-axis oblique plane. Techniques for performing these measurements have been well described previously [14–16]. The decision to either include or exclude papillary muscles in measurements of left ventricular end-diastolic volume, end-systolic volume, or mass differs among institutions.

Assessment of regional wall motion abnormalities is generally made by visual inspection alone. In some centers, research protocols may include complementary spatial modulation of magnetization (CSPAMM), which is otherwise known as “myocardial tagging,” to assess for regional wall motion. Although fractional shortening can be formally measured, these measurements are not performed routinely in most clinical practices [10].

Asymmetric disease is the most common morphologic presentation of HCM with the anteroseptal myocardium being the most commonly hypertrophied segment [10, 12]. Clinically, patients may be asymptomatic or may present with dyspnea, presyncope or syncope, or angina [5].

Asymmetric septal wall hypertrophy causes obstruction of the LVOT in 20–30% of the cases [3, 4, 17]. This obstruction may be present when the patient is at rest or may be latent (provocable) or labile (variable) [5]. The severity of the obstruction can be affected by ventricular afterload (systolic blood pressure), preload, and contractility [5]. Sudden standing, eating a large meal, or drinking alcohol, for example, may cause a drop in afterload [5].

Although previously subject to periodic controversy, a gradient of 30 mm Hg or more is now generally recognized to be of pathophysiologic and prognostic importance [18, 19]. Peak outflow tract gradients can be estimated using phase-contrast MRI; however, these gradients are currently better assessed using echocardiography because of its higher temporal resolution when measuring peak instantaneous velocities. Efforts to improve flow measurements on MRI are focused on both improving the temporal resolution of phase-contrast imaging and developing new techniques, such as real-time Fourier velocity encoding [20].

Evaluation of left ventricular functional parameters shows that mass measurements may not necessarily be greater in patients with asymmetric HCM than in those with a normal heart. Although the left ventricular ejection fraction (LVEF) is typically preserved in asymmetric HCM, subtle evidence of regional wall hypocontractility that can be quite variable may be present [10]. The results of a study by Sipola et al. [10] suggest that impaired contractility may precede findings of hypertrophy, with a progressive decline in systolic function accompanying more marked LVH.

Delayed enhancement of the left ventricular myocardium occurs in up to 80% of all patients with HCM and is usually patchy and mid wall in location [21, 22]. We discuss this topic separately later in this article.

Abnormalities of the mitral valve may occur because of a primary abnormality of the valve itself or as a result of an LVOT obstruction. Systolic anterior motion of the mitral valve is due to the portion of the anterior mitral leaflet distal to the coaptation point being displaced into the LVOT by venturi or drag forces (Figs. 1, 2A, 2B, 2C, 2D, 3A, 3B, 3C, 3D). Over time, the systolic anterior motion of the mitral valve leads to a subaortic mitral impact lesion on the septum, which undergoes fibrosis; thickening of the anterior mitral leaflet and chordae from the resultant trauma; a posteriorly directed mitral regurgitant jet into the left atrium; and a systolic gradient along the LVOT [3–5]. If the mitral regurgitant jet is either anteriorly or centrally placed, other structural abnormalities such as prolapse or leaflet restriction should be excluded [3, 5]. These other abnormalities of the mitral valve occur in approximately 20% of patients with subaortic obstruction and include abnormal insertion of the papillary muscles, mitral valve prolapse, or excessive fibrotic thickening of the anterior mitral valve leaflet [5].

Diastolic dysfunction is thought to occur secondary to abnormal dissociation of actin and myosin filaments during the active phase of relaxation in early diastolic filling and to left ventricular myocardial properties that affect passive characteristics of late diastole, such as the widespread increased volume of the intercellular collagen network [23]. This dysfunction may also result in a restrictive picture. Mitral regurgitation from systolic anterior motion or from morphologically abnormal mitral valve leaflets may also contribute to left atrial dilatation [24]. Patients with left atrial enlargement may then subsequently develop atrial fibrillation [5]. Stretch and distention of the atrial appendage have been shown to perhaps be related to the onset of atrial fibrillation due to an increase in the dispersion of the atrial effective refractory period [25].

Tani et al. [24] found that an increase in left atrial volume and, to a lesser extent, in left atrial diameter correlated well with the presence of paroxysmal atrial fibrillation. They concluded that left atrial enlargement may be used to help predict which patients with HCM will likely develop atrial fibrillation. This distinction among HCM patients is important because the presence of atrial fibrillation reduces left ventricular pre-load and may lead to cardiac failure and systemic thromboembolism [24].

Symmetric HCM is said to occur in up to 42% of HCM cases [4]. This entity should be evaluated closely to differentiate it from other causes of symmetric left ventricular thickening, which we discuss in more detail in Part 2.

Apical HCM was originally described in individuals of Asian descent but is now being increasingly diagnosed in Western populations [4, 5]. Reports of the rate of occurrence of apical HCM vary in the literature, ranging from 25% of all patients with HCM in Japan to fewer than 2% of HCM patients in Western countries [26]. Wigle [5] reported an incidence of 9% in an HCM population in a clinic in Toronto.

ECG findings of giant negative T waves in the precordial leads raise initial suspicion of apical HCM in up to 50% of cases in some studies [5, 26] (Fig. 4). The diagnosis of apical HCM can occasionally be difficult on echocardiography. MRI plays an important role in the diagnosis and exclusion of other diseases and abnormalities that mimic apical HCM on echocardiography [5].

Diagnostic criteria for apical HCM reported in the literature include an absolute apical wall thickness of > 15 mm or a ratio comparing apical left ventricular and basal left ventricular wall thicknesses of ≥ 1.3–1.5 [27, 28]. More subjective criteria for the diagnosis of apical HCM include obliteration of the left ventricle apical cavity in systole and failure to identify a normal progressive reduction in left ventricular wall thickness toward the apex (Figs. 5A, 5B and 6A, 6B).

Moon et al. [27] found that 10 patients with a history of ECG abnormalities (i.e., antero-lateral T wave inversion) and negative findings for apical HCM on echocardiograms had positive MRI findings for apical HCM, with one patient having a left ventricular wall thickness of 28 mm on MRI. They found no evidence of a correlation between the extent of repolarization abnormality and the degree of morphologic severity [27].

The characteristic spadelike configuration of the left ventricular cavity initially described in angiographic studies is well appreciated on left ventricular vertical long-axis views on MRI [26, 28] (Fig. 5A, 5B).

Concomitant apical involvement of the right ventricle is also commonly seen (Fig. 6A, 6B).

Other common findings include apical aneurysm formation with delayed enhancement, which is sometimes referred to as a “burned out apex” [4] (Fig. 7A, 7B, 7C, 7D). This process is thought to be due to ischemia that results from reduced capillary density, hyperplasia of the arterial media, increased perivascular fibrosis, and myocardial bridging [23]. This process usually occurs in the presence of normal epicardial coronary arteries.

This variant of asymmetric HCM, predominantly involving the middle third of the left ventricle, may result in severe midventricular narrowing and obstruction. HCM with midventricular obstruction may be associated with the formation of an apical aneurysm, which is thought to result from the generation of increased systolic pressures within the cardiac apex from the midventricular obstruction. A similar appearance may be due to a burned out apex, as is seen in the apical variant [4, 5] (Fig. 7A, 7B, 7C, 7D).

Patients with HCM occasionally present with a masslike thickening of the left ventricle. This scenario may cause difficulties in diagnosis. In theory, a true mass will have no contractile portion to it, whereas HCM will show variable degrees of contractility. MR tagging by spatial modulation of magnetization has been used to assess for tag displacement and deformation. Tag deformation should be identified in hearts with HCM but should not be noted in neoplastic masses [29]. This technique has yet to be thoroughly investigated, however, and care should be made to distinguish minimal contraction from scant residual normal myocardium in the setting of infiltrative processes such as lymphoma, amyloid, or sarcoid.

Perhaps more useful than this tag deformation is signal characterization of the myocardium with double inversion recovery or spin-echo imaging using varying tissue weighting and observing perfusion of the area with first-pass perfusion and delayed enhancement techniques. Masslike HCM more precisely parallels the homogeneous signal characteristics and perfusion of adjacent normal myocardium (Fig. 8A, 8B, 8C, 8D, 8E, 8F), whereas tumors often show varying degrees of signal heterogeneity before and after gadolinium administration and show perfusion characteristics that differ from those of the remainder of the left ventricle.

The burned out phase occurs in up to 10% of patients with HCM and severe cardiac symptoms. It is characterized by left ventricular dilatation and by loss of myocardium and replacement fibrosis, which are thought to be due to small-vessel ischemia [4, 30, 31].

The burned out phase may present as a localized process with an appearance similar to that of a myocardial infarct but with normal epicardial coronary arteries, as seen in Figure 7A, 7B, 7C, 7D, or as a more diffuse end-stage process with an appearance similar to that of a dilated cardiomyopathy. Little has been written in the literature regarding the cardiac MRI appearances of diffusely burned out end-stage HCM.

It has been well established that delayed enhancement of the myocardium using a segmented inversion recovery pulse sequence can differentiate between normal and infarcted myocardium or fibrosis [32]. Delayed enhancement is thought to occur in areas of abnormal myocardium secondary to expansion of the extracellular space, alterations in the extracellular matrix composition, and altered distribution kinetics [32–34].

Delayed enhancement occurs in up to 80% of patients with HCM [21, 22]. The most common form of hyperenhancement in HCM is patchy and mid wall in location. Teraoka et al. [35] found delayed enhancement in 75% of patients with HCM, 89% of whom had patchy mid-wall-type enhancement (Figs. 2A, 2B, 2C, 2D and 9).

The various proposed histologic correlates of enhancement in HCM include plexiform fibrosis (fibrosis seen in HCM in areas of myocyte disarray), expanded interstitial spaces, and microscopic replacement fibrosis due to ischemia [22, 27–39]. However, Mahrholdt et al. [40] have suggested that areas of visually detected delayed enhancement are most likely explained by macroscopic replacement fibrosis and that enhancement secondary to diffuse interstitial processes, such as plexiform fibrosis, is less likely to cause recognizable differences in regional signal intensity.

In a case study, Moon et al. [37] compared the MRI findings and histology results of an explanted heart with HCM and found that an increased percentage of histologic collagen correlated directly with increased delayed enhancement of the myocardium. Those researchers also found that increased enhancement was associated with an increased incidence of regional wall motion abnormalities [37]. Other studies have also shown a significant decrease in global and regional cardiac function in patients with increasing volumes of delayed enhancement [21, 35, 41, 42].

Delayed enhancement tends to involve the interventricular septum, particularly the anteroseptal mid to basal segments. These segments are also the most commonly thickened segments in patients with asymmetric HCM [35, 42, 43]. If abnormal enhancement occurs elsewhere outside the interventricular septum, it also tends to occur in areas of maximal left ventricular thickness.

As we mentioned earlier, delayed enhancement that occurs in areas of hypertrophy tends to be mid wall and patchy in distribution. Another interesting feature of delayed enhancement in HCM is a predilection for enhancement to occur at the anterior and posterior right ventricular insertion points (Figs. 2A, 2B, 2C, 2D and 9). An exception to this is in areas of burned out myocardium where the left ventricular wall is typically thinned and enhancement is full thickness [41] (Fig. 7A, 7B, 7C, 7D).

Delayed enhancement in areas of burned out myocardium bares striking similarity to delayed enhancement in patients with chronic myocardial infarction secondary to coronary artery disease. This type of enhancement however may not necessarily conform to a particular vascular territory. The postulated causes of ischemia in patients with HCM and normal epicardial coronary arteries include reduced capillary density; hyperplasia of the arterial media; increased perivascular fibrosis, leading to reduced coronary vasodilator reserve; and myocardial bridging [23].

Investigators have noted that in patients without LVH there is usually no enhancement. Lack of enhancement in this population suggests that regions of delayed enhancement occur only after the development of LVH, although this theory has yet to be proven across all the many genotypic variants of LVH [21].

Despite the low yield for discovering delayed enhancement in patients with only minimally thickened left ventricular myocardium, administration of gadolinium remains useful in differentiating other causes of thickened left ventricular myocardium that show characteristic patterns of enhancement. Examples include cardiac involvement by amyloid and Fabry's disease, both of which are discussed in Part 2 [1].

The presence of delayed enhancement and its relationship to patients with an increased risk of sudden cardiac death as established by traditional risk factors [21, 22, 34, 44] are elaborated on in Part 2 [1] also.