7 Cardiac walls

The echocardiographic pitfalls relating to imaging of the cardiac walls are concerned with the diagnosis of:

PITFALLS WHEN DIAGNOSING WALL HYPERTROPHY

In current practice, echocardiography is the method of choice for identifying and quantifying left ventricular hypertrophy (LVH). However, the method has certain limitations relating to:

• measuring the wall thicknesses

• calculating the myocardial mass of the left ventricle (LV)

• distinguishing between physiological and pathological LVH

• diagnosing primary hypertrophic cardiomyopathy.

The echocardiographic prevalence of LVH varies according to the population studied, the calculation method used and the criteria used to define it. LVH is present in about 35% of hypertensive patients. Echocardiography with Doppler tissue imaging (DTI) makes it possible to identify the early myocardial anomalies in hypertensive patients without ventricular hypertrophy. In fact, the diastolic myocardial velocities are significantly reduced in hypertensive patients compared with in normotensive patients, in the absence of LVH.

Pitfalls in measuring the wall thicknesses

M-mode measurements of the LV are vital in the diagnosis of left ventricular wall hypertrophy. The interpretation of the measured values (end-diastolic thickness of the septum and of the posterior wall of the LV) is only reliable if the strict quality criteria for M-mode recording are applied (Box 7.1). Conventionally, the M-mode examination is guided by the two-dimensional (2D) mode and begins in the left parasternal window. This makes it possible to obtain M-mode measurements of the base of the longitudinal cross-section. Other echocardiographic projections may be used (the short-axis parasternal view at the level of the papillary muscles below the mitral valve, or the subcostal view) if the parasternal long-axis view is not usable.

The causes of errors in the measurement of wall thicknesses are summarized in Box 7.2.

Inclusion of the tricuspid apparatus, mitral chordae or false septal tendon in the measurement of the wall thicknesses (Fig. 7.5)

When measuring the septum, care must be taken to distinguish between the tricuspid subvalvular apparatus and the anterior septal endocardial border. Between these two structures there is normally a clear, echo-free space. Likewise, the inclusion of the mitral chordae may lead to incorrect positioning of the measuring cursor when detecting the anterior border of the posterior wall of the LV. However, the M-mode slope of the posterior wall is greater than that of the chordae.

Figure 7.5 Anatomical causes of echocardiographic overestimation of the wall thicknesses. Possible mistaken inclusion in the M-mode measurement of (a) the tricuspid apparatus, (b) the mitral chordae or (c) the false septal tendon (FT). In these images, the measurements have been made correctly. AO, aorta; LA, left atrium; LV, left ventricle; RA, right atrium; RV, right ventricle.

The presence of a false septal tendon may be responsible for an overestimation of the septal thickness. A false tendon, which is composed of muscular tissue, is transformed in 2D echocardiography into a linear structure attached to the left surface of the septum in the form of a bridge. It should be identified by careful echocardiographic examination of the septum in order to exclude it from the M-mode measurement of the septal thickness. In order to eliminate these anatomical sources of error, it is imperative to use multiple echocardiographic projections (lower left parasternal, minor axis, apical, subcostal) in order to obtain a clear, measurable and interpretable M-mode trace.

Presence of a subaortic septal rim or a muscular band

The subaortic septal rim is defined as a hypertrophy located exclusively on the basal part of the interventricular septum, the diastolic thickness of which is > 13 mm. This rim is identifiable in the 2D parasternal, long-axis or apical cross-section (the two left chambers and the aorta). The rim may lead to an overestimation of the septal thickness because of the use of an inappropriate M-mode projection. In order to avoid this pitfall, the M-mode line should be positioned below the septal rim (Figs 7.6 and 7.7). Use of the subcostal projection is sometimes preferred. A subaortic septal rim should be distinguished from a septal hypertrophy arising from hypertrophic obstructive cardiomyopathy (HOCM), which is an echocardiographic diagnostic pitfall to beware of (see Table 7.4).

Figure 7.6 Subaortic septal rim: (a) optimum M-mode projection; (b) M-mode projection passing through the rim and overestimating the septal thickness. AO, aorta; IVS, interventricular septum; LA, left atrium; LV, left ventricle; PW, posterior wall.

Figure 7.7 Anatomical causes of echocardiographic overestimation of the septal thickness: (a) subaortic septal rim; (b) septal kinking with an aortoseptal angle of 80°. AO, aorta; LA, left atrium; LV, left ventricle; RV, right ventricle.

Table 7.4 Differential diagnosis between the subaortic septal rim and hypertrophic obstructive cardiomyopathy (HOCM)

Parameter	Septal rim	HOCM
Clinical context	Older, often hypertensive subject	Younger subject, familial primary HOCM
Septal hypertrophy	Located on the basal septum	More diffuse, surrounding the middle segment of the septum
Diastolic septal thickness	13–20 mm	Often > 20 mm
Septal echostructure	Normal	Hyperechoic, dense appearance
Septal motion	Normal	Hypokinetic
Left ventricle	Normal size	Small, deformed
Dynamic obstruction	Rare (around 18% of cases)	Frequent (> 50% of cases)
Diastolic function	Conserved	Altered

Likewise, the presence of an abnormal muscular band of the right ventricle may obstruct the exact measurement of the septal thickness. The moderator band passes from the lower part of the interventricular septum to the anterior part of the right ventricle, where it is attached to the anterior papillary muscle. It may be identified by imaging multiple echocardiographic cross-sections of the right heart.

Presence of septal kinking (angulation)

Septal kinking is a particular morphology of the septum that is identifiable in the parasternal, long-axis cross-section (see Fig. 7.7). Normally, the angulation of the septum in relation to the axis of the aorta falls between 140° and 125°. In the case of septal kinking (aortoseptal angle £ 90°), a poorly directed ultrasound beam approaches the septum obliquely, giving an appearance of false septal hypertrophy (Fig. 7.8).

Figure 7.8 Septal kinking. Abnormal aortoseptal angle (< 90°) leading to an exaggerated measurement of the septal thickness (IVST₁). Normal septal thickness (IVST₂) reflecting the anatomical reality. AO, aorta; LA, left atrium; LV, left ventricle.

Thus it is often the case that the probe has to be repositioned in order to avoid the septal kinking that may lead to overestimation of the thickness of the septum. As far as possible, the M-mode line should be perpendicular to the septum and the posterior wall. In certain cases, the presence of septal kinking can invalidate the measurements.

Pitfalls when calculating the left ventricular mass

LVH translates anatomically into an increase in the left ventricular mass (LVM). The LVM may be calculated using echocardiography:

• either with the M-mode measurements according to two recommended techniques (ASE and PENN; see Fig. 7.1 and Box 7.3)

• or by resorting to the 2D mode, which makes it possible to partially avoid the heterogeneous nature of the LV, but at a reproducibility that is too low to justify its complexity.

Box 7.3 The ASE and PENN formulae for calculating the left ventricular mass (LVM) based on the M-mode measurement of the diameter of the left ventricle (EDD), the interventricular septal thickness (IVST) and the posterior wall thickness (PWT) at end-diastole (see Fig. 7.1)

American Society of Echocardiography (ASE) formula:

LVM = 0.8[1.04 × (EDD + IVST + PWT)³ – EDD³] + 0.6

Pennsylvania (PENN)/Devereux formula:

LVM = 1.04[(EDD + IVST + PWT)³ – EDD³] – 13.6

The most commonly used method for measuring the LVM is based on M-mode echocardiography. This method is relatively reliable for quantifying LVH, as it takes into account the wall thickness and the diameter of the ventricular chamber. It has been validated by anatomical correlations.

In reality, the size of the LV often varies, due to the variations in blood volume and load conditions, which renders the isolated measurement of the wall thicknesses insufficient for the quantification of hypertrophy.

Two formulae are commonly used to calculate the LVH in M-mode echocardiography: the ASE formula and the PENN/Devereux formula (see Box 7.3). These formulae are based on the common hypothesis of a LV that is geometrically a rotational ellipsoid truncated at one of its poles, and the major axis of which is double the length of the minor axis. In order to calculate the LVM, mathematical formulae based on the principle of cubes are used. It is nonetheless vital to match the chosen formula with the corresponding measurement convention being used.

In order to be reliable, the measurement of the M-mode parameters (thicknesses and diameter) used in calculating the mass must be rigorous and precise (see Fig. 7.2). However, echocardiographic measurement of the LVM may run into technical and methodological difficulties, and the examiner should be familiar with these in order to make an informed view of the final result (Box 7.4).

The pitfalls involved in calculating the LVM using M-mode echocardiography are described below.

Inappropriate use of the formulae for calculating the left ventricular mass

The formulae for calculating the LVM are valid only for elliptically shaped, morphologically homogeneous LVs. Moreover, it is vital to check which of the mathematical formulae is integrated in the calculation software of the ultrasound machine being used in order to apply the correct corresponding measurement technique (ASE or PENN convention). This makes it possible to avoid errors in the diagnosis of LVH.

Incorrect calculation of the myocardial mass

When the examination conditions are difficult, or in certain pathological situations, the elliptical model of the LV becomes inapplicable (see Box 7.4). The technical difficulties associated with obtaining a high-quality M-mode trace in patients who have a low echogenicity constitute an overwhelming limitation on the precise measurement of the wall thicknesses.

Failure to respect the conditions of M-mode measurements (see Box 7.1)

In short, an echocardiographic methodology that is not rigorous gives rise to an unreliable result.

Errors due to ‘cubing’ of the ventricle

In practice, even a small error in measurement can have significant consequences, as the ventricular thicknesses and diameters are cubed. For example, for wall thicknesses of 12 mm and a diameter of 55 mm, a 1 mm overestimation of the thickness of the septum and the posterior wall lead to an error of 40 g in an LVM of 351 g.

Poor reproducibility of the left ventricular mass calculation

The coefficient of variation in the LVM calculation is around 20 g, or even 30 g. This generates an apparent spontaneous regression of the LVH over repeated examinations, linked to the phenomenon of regression to the mean.

Significant intra- and interobserver variability

Intra- and interobserver variability limits the reproducibility of the examination, and should cause examiners to be particularly careful in their interpretation of individual results. In the same patient, the comparison of two successive examinations or even readings made at two different moments during the same examination may lead to notable variation in the calculated LVM. For good reproducibility, in terms of intraobserver variability, differences of less than 1 mm in the measured wall thicknesses and of less than 2 mm in the measured chambers are vital. It is, therefore, sensible to make measurements over at least three cardiac cycles to obtain good reproducibility. The definitive values will then be the mean values.

Variability of threshold values used for the diagnosis of left ventricular hypertrophy

The parameters that significantly modify the LVM are the gender and the body habitus of the patient. As a matter of routine, the LVM is corrected according to the patient’s body surface area, with an average threshold of 134 g/m² for men and 110 g/m² for women usually being applied. However, the threshold values proposed in the literature are somewhat variable: 110–134 g/m² for men and 100–125 g/m² for women.

Uncertainty over the choice of the threshold value has major consequences for the final diagnosis of LVH. By applying low thresholds, the overall prevalence of LVH is increased considerably. Therefore the choice of the thresholds for defining LVH has a significant influence on the estimated value of this prevalence. In contrast, an underestimation of LVH is observed in obese patients when relating the LVM to the body surface area. For this reason, indexing the LVM by patient size, preferably raised to the power of 2.7, has been proposed. In a normal subject, this index of the LVM should be lower than 50 g/m^2.7 in men and 47 g/m^2.7 in women. Generally, the thresholds determining the LVH are lower in women (Table 7.1).

Table 7.1 Normal values of the left ventricular mass (LVM) indexed according to the body surface area and size of the patient

Indexed LVM	Men	Women
To body surface area (g/m²)	< 134	< 110
To size (g/m^2.7)	< 50	< 47

Finally, in order to define the left ventricular geometry, the calculation of the LVM should be complemented by the corresponding measurement of the relative wall thickness (RWT). The RWT is expressed as:

This echocardiographic parameter, interpreted according to the corrected LVM, makes it possible to specify the concentric or eccentric character of the LVH, which is often a source of confusion (Table 7.2 and Fig. 7.9). In fact, even if the LVM is normal, an RWT of > 0.45 makes it possible to identify concentric remodelling of the LV. The advantage of a morphological analysis of the geometry of the LV lies in its prognostic implications (stratification of the cardiovascular risk in hypertensive patients).

Table 7.2 Three morphological types of abnormal left ventricular geometry

Abnormal geometry	Left ventricular mass (LVM)	Relative wall thickness (RWT)
Concentric remodelling	Normal	> 0.45
Eccentric LVH	Increased (↑)	< 0.45
Concentric LVH	Increased (↑↑)	> 0.45

LVH, left ventricular hypertrophy.

Figure 7.9 Concentric hypertrophy of the left ventriclular walls in 2D and M-mode echocardiography (diastolic wall thicknesses = 13 mm, relative wall thickness = 0.54). AO, aorta; LA, left atrium; LV, left ventricle; RV, right ventricle.

Finally, quantification of the LVM in 2D echocardiography on the basis of the myocardial contours has never been proved superior to M-mode measurement.

Preliminary data on the calculation of the LVM by means of three-dimensional echocardiography appear promising, particularly in cases of non-homogeneous hypertrophy.

DISTINCTION BETWEEN PHYSIOLOGICAL AND PATHOLOGICAL LEFT VENTRICULAR HYPERTROPHY

The regular practice of a sustained sporting activity is accompanied by morphological modifications of the LV, in particular an increase in the wall thicknesses. It is sometimes difficult to trace the exact border between ‘physiological’ hypertrophy linked to sport and pathological hypertrophy. The echocardiographic data that make it possible to distinguish between these two types of LVH are summarized in Table 7.3, where measurements of the heart of an athlete are compared with measurements found in hypertrophic cardiomyopathy (HCM). The application of DTI is very useful in this differential diagnosis.

Table 7.3 Differential diagnosis between physiological hypertrophy (athlete’s heart) and pathological hypertrophy (hypertrophic cardiomyopathy (HCM))

Parameter	Athlete’s heart	HCM
Wall thickness	< 13 mm (13–15 mm in 5% of cases)	≥ 13 mm
IVST/PWT	< 1.3	> 1.3
RWT	Variable	> 0.45
Left ventricular mass	Increased (↑)	Increased (↑↑)
EDD of the LV	≥ 55 mm (> 60 mm in 14% of cases)	< 50 mm
LV systolic function	Normal	Normal
Left ventricular diastolic function
Mitral flow	E/A > 1 DT > 150 ms IVRT ≥ 90 ms	E/A < 1 or > 1 DT > 220 ms, IVRT > 90 ms
PV (colour M-mode)	> 45 cm/s	< 45 cm/s
Ea (annular DTI)	> 8 cm/s	< 8 cm/s
Filling pressure	Normal	Normal or increased
Myocardial DTI
Myocardial velocities	Normal	Reduced
Transparietal gradient	Normal	Increased
Intra-LV dynamic obstruction	Absent	Often present
Mitral regurgitation	Absent or small	Moderate
Pulmonary arterial pressures	Normal	Normal or increased
Evolution	Regression of the hypertrophy with loss of fitness	Persistence of hypertrophy

DT, deceleration time; DTI, Doppler tissue imaging; Ea, velocity of the E wave in DTI applied to the mitral annulus; E/A, ratio of the E and A waves of the mitral flow; IVRT, isovolumic relaxation time; IVST/PWT, ratio of the septal diastolic thicknesses to the posterior wall thickness; LV, left ventricle; PV, propagation velocity of the mitral flow in colour M-mode; RWT, relative wall thickness; EDD, end-diastolic diameter of the LV.