QUANTIFICATION OF LEFT VENTRICULAR (LV) systolic function is considered the foundation of cardiac imaging (
1). It has been shown to be a predictor of morbidity and mortality and is reported in most echocardiographic studies.
LV systolic function describes the contractility of the LV which is defined by the Frank-Starling relationship. This relationship is an intrinsic property of myocardial fibers by which increased length (ventricular volume, preload) results in enhanced performance during the subsequent contraction. The preload status at the time of the examination is reflected by LV chamber dimensions (diameter, area, or volume). LV thickness (or mass) may also be reported with systolic function and LV chamber size to complete the overall estimate of LV systolic performance.
MEASUREMENTS OF LV SIZE—WALL THICKNESS, CHAMBER SIZE, AND LV MASS
Advances in echocardiographic technology have enabled actual visualization of the blood-tissue interface and it is no longer necessary to use leading edge methodology (
1,
2). Measurements should be made over several cardiac cycles; three beats in sinus rhythm and five beats in atrial fibrillation (
1). The same range of normal values for LV chamber dimensions and volumes are recommended for transthoracic and transesophageal echocardiography (TTE and TEE) (
1).
Left Ventricular Wall Thickness
Normal values (reference range) for both posterior (inferolateral) wall and septal wall thickness: Men 0.6 to 1.0 cm; women 0.6 to 0.9 cm (
1).
Measurements of LV wall thickness are typically made in the transgastric (TG) mid short-axis (SAX) view (excluding the papillary muscles,
Fig. 3.1). Measurements can be made from 2D or M-mode recordings, although 2D is the preferred method, since M-mode measurements are subject to inaccuracies caused by nonperpendicular placement of the M-mode cursor (
1). Usually, both septal wall thickness at end diastole (SWTd) and posterior wall thickness at end diastole (PWTd) are reported. Septal wall thickness is measured
from the right septal surface to the left septal surface, whereas posterior wall thickness is measured from epicardial surface to endocardial surface (being careful not to include pericardial tissue). Wall thickness increases with age, even in normal populations (
3)
Relative Wall Thickness
Relative wall thickness (RWT) is often used in patients with LV hypertrophy. In TEE the measurements are again made in a TG SAX and may be calculated from either of the two formulae given below. RWT is expressed as a decimal and used to describe LV hypertrophy and remodeling.
Relative wall thickness (RWT) mm = (2 × PWTd)/LVIDd or (PWTd + SWTd)/LVIDd
Normal values: Men 0.24 to 0.42 cm; women 0.22 to 0.42 cm (
1,
2).
In these two formulae, LVIDd refers to the minor-axis diameter measurement (LV cavity, see below). A regional wall thickness equal to or greater than 0.42 denotes concentric hypertrophy (wall thickness is increased in the presence of a normal internal diameter) and an RWT less than 0.42 denotes eccentric hypertrophy (dilated LV). The distinction between the two forms of hypertrophy is of prognostic interest, as concentric hypertrophy is associated with a higher incidence of cardiovascular events than eccentric hypertrophy.
Chamber Size (LV Diameter)
Linear measurements of chamber size are defined by a minor- (short-) axis diameter and a major- (long-) axis diameter (
Fig. 3.2). The minor-axis diameter measurement is made at, or immediately below, the level of the mitral valve tips and perpendicular to the long-axis diameter (
1). That means it should be measured in the midesophageal (ME) two-chamber view or TG two-chamber view at the level of the papillary muscles, rather than in the TG SAX view which is created at the level of the papillary muscle bodies. The major-axis diameter (LV length) is typically made in the ME two- or four-chamber view since the apex is not well seen in the TG two-chamber view. The measurement is made from the midpoint of a line connecting the two opposite points of the mitral annulus to the endocardium at the apex. The longer of the two lengths acquired from the ME two- and four-chamber views, is the recommended measurement to use (
1).
Normal values (mean) for minor-axis diameter (LVID) (
1):
In diastole : men 50.2 ± 4.1 mm and women 45.0 ± 3.6 mm
In systole : men 32.4 ± 3.7 mm and women 28.2 ± 3.3 mm
In general, a minor-axis diameter greater than 5.4 cm during diastole is considered enlarged.
Chamber Size (LV Volume)
The calculation of left ventricular volumes from single linear measurements for example, the method of Teichholz (
4), is no longer recommended. A single measurement may not represent regional variation in LV size and shape (
1). Nevertheless, Teichholz LV volumes are still automatically reported out by some ultrasound vendors when a measurement of LVID is made (
Fig. 3.2); note that there is a cubed function in this formula which greatly confounds any measurement errors.
The most commonly used and recommended method for calculating LV volumes in 2D echocardiography is the biplane method of disk summation (Modified Simpson’s Rule).
LV volume (Modified Simpson’s Rule):
LV volume (mL) = (π/4) Σ(n = 1 – 20) (LVIDnminor (ME two chamber)) × (LVIDnminor (ME four chamber)) × LVIDmajor/20
The views required for this calculation are midesophageal four-chamber and midesophageal two-chamber views. The LV is “divided into a series of 20 disks” from the base to the apex of the LV, like a stack of coins of decreasing size. The computer software calculates the volume of each disk (area × height) and the volumes are summated to give a total LV volume (
Fig. 3.3). Foreshortening of the LV will result in underestimation of volume (
1). LV volume using the Modified Simpson’s Rule may be acquired from 3D echocardiography (3DE) images (
Fig. 3.4).
Because the areas are acquired using both the ME four- and two-chamber views, this method corrects for shape distortion to some extent, for example, in patients with regional wall motion abnormalities or an aneurysm (
1). Accuracy is dependent upon being able to identify and trace the endocardial borders with fidelity.
In cases where the endocardial border (particularly at the apex) is not well defined, the area-length method may be used (
5).
LV volume (area length formula) = 5/6 (area × length)
The area is the cross-sectional area of the LV cavity measured in the TG SAX view and length refers to the major axis measured in either the ME two-chamber or ME four-chamber views. As with the Modified Simpson’s Rule, care should be taken to avoid foreshortening the LV. Some vendors provide the calculation from a single TG SAX view, assuming that the major axis = 2 × minor axis (see above,
Fig. 3.5).
3DE is now available on most TEE probes (
Fig. 3.6;
Video 3.1). The advantage of 3DE for measuring LV volumes is that the LV can be acquired and displayed in its true shape avoiding the need for any mathematical modeling, producing a more accurate measurement. Errors do not occur because of plane positioning and foreshortening. Left ventricular volumes measured by 3DE are highly correlated with cardiac magnetic resonance (CMR) imaging which is the gold standard of imaging. There is a better agreement between 3DE and CMR with lower inter- and intraobserver variation than 2D echocardiography (2DE) and CMR in normal subjects (
6) and patients with regional wall motion abnormalities. (
7).
LV volumes acquired from 3DE data sets tend to be higher than the respective 2DE calculations (
1,
8) (
Table 3.1). Furthermore age, gender, and ethnicity have significant effects on LV volumes in both 2DE and 3DE (
1).
Left Ventricular Mass
Increased LV mass is a stronger predictor than low EF for all-cause mortality and cardiac event rates in both normal and hypertensive populations. Because LV mass increases as a function of body size (except those with morbid obesity), LV mass is preferably expressed as a function of body surface area (BSA) (
1). Normal values for LV mass are given as 67 to 162 g for women and 88 to 224 g for men. Indexed to BSA this becomes 43 to 95 g/m
2 for women and 49 to 115 g/m
2 for men (
1) (
Table 3.1). LV mass may be combined with RWT to categorize patients into various classes of hypertrophy (
1) (see following section on “Left Ventricular Hypertrophy”).
Left ventricular mass calculations are based on the subtraction of the volume of the LV cavity from the volume encompassed by the LV epicardium (
Fig. 3.7). This leaves LV myocardial volume, which is then multiplied by the density of myocardial tissue to calculate LV mass (approximately 1.05 g/mL). Left ventricular mass can be calculated from linear measurements, 2DE areas, or 3DE full-volume data sets.
For linear measurements made in 2DE the TG SAX view is required and LV mass is calculated from the following recommended formula (
1):
LV mass (g) = 0.8 [1.04 {(LVIDmajor + PWT + SWT)3 – (LVIDmajor)3}] + 0.6 g
Again, note the cubed functions which exacerbate measurement errors.
In the determination of LV mass using 2DE area measurements, either the area-length method or the truncated ellipsoid method is recommended (
1). Most current echocardiography machines include the
software to calculate LV mass by one or both of these two methods (
Fig. 3.8). The LV is acquired in the TG mid SAX view (
Fig. 3.7). An area tracing is made of the epicardial and endocardial borders. The difference between the two areas is the area occupied by the myocardium. A major-axis length is then acquired from a long-axis view (
Fig. 3.8) and the software calculates the mass of the LV according to the formulae used by the vendor. Using area-based measurements rather than linear measurements for LV mass partially corrects for LV shape distortions.
3DE-based LV mass calculations are more accurate and reproducible than linear or area-based measurements. Abnormal shape or asymmetric hypertrophy is directly measured without any mathematical assumptions (
Fig. 3.9).
QUANTITATIVE MEASURES OF LEFT VENTRICULAR SYSTOLIC PERFORMANCE
LV systolic performance may be assessed qualitatively or quantitatively with echocardiography. An echocardiographer may become quite efficient and accurate at visually estimating systolic function. However, accuracy and reproducibility are dependent upon the individual interpreter’s skill and interobserver measurements may vary considerably. Consequently, calibrated measurements are preferred, and it is recommended that even experienced echocardiographers regularly cross-check qualitative evaluations against calibrated measurements (
2).
Systolic function is expressed mathematically as an end diastolic parameter minus the corresponding end systolic parameter divided by the original end diastolic parameter, expressed as a percentage. End diastole is defined as the first frame after mitral valve closure or the frame in the cardiac cycle in which the respective measurement is the largest. End systole is best defined as the frame after aortic valve closure or the frame in which the measurement is smallest.
The parameter may be a linear measurement, an area, or a volume. For example:
{(LVEDV – LVESV)/LVEDV} × 100%
where LVEDV is LV end diastolic volume and LVESV is LV end systolic volume.
LV systolic function estimates based on linear measurements are known as endocardial fractional shortening,
Endocardial fractional shortening (%) = {(LVIDd – LVIDs)/LVIDd} × 100
Normal values: men 25% to 43%, women 27% to 45% (
1).
Linear measurements made from motion mode (M-mode) or 2DE images have the lowest interobserver variability as compared to area or volume measurements, render quite accurate estimates of systolic function in healthy subjects, but are probably the least representative of overall LV systolic function in cardiac diseases that produce regional abnormalities of the myocardium. In spite of the higher pulse rate and better temporal in M-mode tracings compared to 2DE, the recommendation is to use linear measurements from 2DE images (
Fig. 3.10) (
1).
Area measurements offer some improvements in accuracy over linear dimensions, as more of the LV is represented in the measurement. LV systolic function estimates based on area measurements are known as fractional area change (FCA).
Fractional area change (FAC) (%) = {(LVAd – LVAs)/LVAd} × 100
The area of the LV cavity is measured at end-systole (LVAs) and at end diastole (LVAd) and used to calculate FAC (
Fig. 3.11). Most commonly these measurements are made from the TG mid SAX view of the LV, but when this view is suboptimal, long-axis views can be substituted. The endocardium is manually traced around the LV cavity ignoring the papillary muscles. Fractional area change is essentially the measurement which is mentally made when a clinician “eyeballs” the left ventricular ejection fraction (LVEF). A TG SAX view is acquired and a mental estimate of FAC is made. If the apex is normal, 15% is added; if akinetic, zero is added; and if dyskinetic, 15% is subtracted. If there is hypokinesia of the apex, a value of 5% to 10% is added.
Alternatively, automated border detection software provides real time, beat-to-beat measures of LVAd, LVAs, and FAC. Since the acoustic properties of tissue and blood create significantly different backscatter and signal strength, the software can perform automated detection of the endocardial border as it moves through the cardiac cycle. Typically, the TG mid SAX view is used. Some adjustment in time compensated gain, lateral gain, and overall gain settings may be required to ensure that the displayed automated border tracks the endocardium throughout the cardiac cycle (
Fig. 3.12).
LV systolic function based on volumes is known as LVEF and is the most recognized and reported value.
A normal ejection fraction is 63 ± 5% for both men and women (
1). The range 53% to 73% is considered normal in adults. There is not a significant effect of gender, age, or body size (
1).
OTHER MEASURES OF LEFT VENTRICULAR SYSTOLIC FUNCTION
The rate of rise in left ventricular pressure (d
P/d
T) has been demonstrated to be well correlated with systolic function. The greater the contractile force exerted, the greater the rise in ventricular pressure. Previously
this could only be measured invasively with LV catheterization; however continuous-wave Doppler (CWD) determination of the velocity of a mitral regurgitant (MR) jet allows calculation of instantaneous pressure gradients between the left ventricle and the left atrium. Left atrial pressure variations in early systole can be considered to be negligible; therefore, the rising segment of the MR velocity curve should essentially reflect LV pressure increase only. If the rate of rise in ventricular pressure is reduced because of poor LV function, the rate of increase of the MR jet velocity will also be low.
To perform a d
P/d
T measurement (
Fig. 3.13), the MR jet is interrogated with CWD. The cursor is placed on the MR velocity profile at 1 m/s and then at 3 m/s and the time interval between the two points is determined (
10). Using the simplified Bernoulli equation, the pressure differential is [4(3)
2] – [4(1)
2] or 32 mm Hg.
dP/dT is therefore 32 mm Hg divided by the time interval in seconds. Normal values exceed 1,000 mm Hg/s.
Assessing Left Ventricular Systolic Function With Doppler Techniques
Tissue Doppler Imaging
Tissue Doppler imaging (TDI, or tissue Doppler echocardiography [TDE]) measures the velocity of myocardial tissue using low-pass filters to screen out higher velocities generated by blood flow. Unlike blood flow Doppler signals that are typified by high velocity and low amplitude, myocardial motion is characterized by low velocity and high amplitude. Tissue motion creates Doppler shifts that are approximately 40 dB higher than Doppler signals from blood flow and their velocities rarely exceed 20 cm/s. To record low wall motion
velocity, gain amplification is reduced and high-pass filters are bypassed. Temporal resolution is optimized by selecting as narrow an image sector as possible, which increases frame rate (
Fig. 3.14). A frame rate >140/s is recommended for all forms of TDI (
11).
In TDI, a small pulsed-wave sampling volume measures the velocities of the myocardium as it moves toward and away from the transducer. The sample volume is placed in the middle of a segment of the heart and velocities within that area are measured. A velocity against time plot is displayed, using the convention that tissue moving toward the transducer is positive and away from the transducer is negative (as in conventional Doppler). During interrogation of the basal segment of the septum in the ME four-chamber view, as the heart contracts and thickens during systole, the atrioventricular ring moves toward the apex (away from the transducer) producing a
negative deflection (
Fig. 3.15).
Because this is a Doppler technique, TDI will underestimate the myocardial velocities if the angle of interrogation is not parallel to motion (
12). Care should therefore be taken to ensure that the ultrasound beam is aligned with the direction of the motion of the myocardium. The angle of incidence should not exceed 15°,
thereby keeping the velocity underestimation to <4% (
11). Although most ultrasound platforms allow for correction of the Doppler equation for the angle of incidence, this is not recommended.
Errors in TDI can occur because of tethering. For example, in an ME four-chamber view, an akinetic segment at the basal part of the septum should by definition have a longitudinal systolic velocity of zero. However, if the midventricular segment of the septal wall moves normally, tethering will cause the akinetic basal segment to also move toward the apex.
In general, longitudinal measurements are made at the basal and midventricular segments, obtained from the ME two- and four-chamber views. A gradient of systolic velocities exists from the base of the heart to the apex (
Table 3.2). Peak systolic longitudinal velocities at the MV annulus (Sa) are greater than velocities in the midventricular segments (Sm). Sm velocities are considered to be representative of overall systolic function. Annular velocities are difficult to acquire in patients with mitral annular calcification or with a prosthetic valve or annuloplasty ring because of signal drop out. Myocardial velocities are age and gender dependent (
Table 3.2). From
transthoracic studies, patients with normal global LV function have systolic velocities greater than 7.5 cm/s (
13) whereas velocities less than or equal to 5.5 cm/s indicate LV failure (
14). Systolic velocities less than 3 cm/s are associated with a significantly increased risk of cardiac death within 2 years (
15).
(Note that these quoted values are positive because transthoracic measurements are acquired from the apex of the heart. During systole, the myocardium moves toward the transducer producing a positive deflection.)
The typical systolic TDI profile (
Fig. 3.15) has two parts with a biphasic wave during isovolumic contraction (IVCa and IVCb) and a monophasic wave during systolic ejection. IVCa corresponds to the timing of the MV closure and represents early myocardial activation at the base of the heart; occurring 20 to 30 ms earlier in the anteroseptal than the posterior free wall (
16). The initial movement of the myocardium at the annulus is inward and toward the apex. The second wave IVCb is in the opposite direction caused by subsequent contraction of the apex making the base bulge up and outward just before ejection. The monophasic negative systolic (S) wave is produced as the myocardium moves inward and toward the apex as the LV contracts during ejection.
Color Tissue Doppler
In the same way that conventional Doppler can be color coded to provide a color map of blood flow patterns, tissue Doppler can be color coded to display myocardial velocities; red depicting positive velocities (toward the transducer) and blue for negative velocities (away from the transducer). The display is of real-time 2D gray-scale images overlain by color-coded myocardial velocities (
Fig. 3.14;
Video 3.2).
Placing “markers” at various points along a ventricular wall during color tissue Doppler produces a graphic representation of velocity against time called
curved M-mode (
Fig. 3.16). This form of color TDI combines spatial resolution with high temporal resolution and can be displayed in real time.