Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Quantitative Analysis of Left Ventricular Anatomy and Systolic Function
Quantitative Analysis of LV Anatomy
LV Size
Linear Measurements
The shape of the normal left ventricle (LV) corresponds to a prolate ellipsoid. Linear measurements of the LV therefore allow limited conclusions about its size and should not be used for determining its volume. Nevertheless, the diameter of the LV provides an estimation of LV size and may be useful in patients with limited echocardiographic windows, such as ventilated patients in the intensive care unit, who often have reduced image quality in the apical views. Linear measurements of the LV are supported by outcome data. For example, the LV end-diastolic short-axis diameter is associated with death for patients with dilated cardiomyopathy.
The diameter of the LV should be determined in the parasternal long-axis view. The line used for measurements should be oriented perpendicular to the LV long axis at the level of the mitral valve leaflet tips ( Fig. 4.1 ). The calipers should be placed at the intersection points of this line with the myocardial wall on the edge of the compacted myocardium ( Table 4.1 ).
TABLE 4.1
Quantification of LV Size and Function: Checklist for Acquisition of Appropriate Views and Measurements.
| 1-Dimensional Method |
|
| 2-Dimensional Method |
|
| 3-Dimensional Method |
|
This measurement can be obtained by two-dimensional (2D) echocardiography or by a 2D-guided M-mode approach. The M-mode is technically more demanding in that it may be difficult to line up the correct orientation of the measurement line. However, M-mode has an excellent temporal resolution with a frame rate of several thousand scan lines per second with high interphase definition and consequently reproducible measurements. In most patients, the M-mode orientation can be improved by tilting the patient forward and moving the transducer toward the upper sternum.
Because the LV diameter is a linear measurement, it is representative of LV size only for ventricles with a normal shape and for the anteroposterior diameter. Alterations in global shape such as elongation or truncation of the ventricular long axis and changes in regional shape such as aneurysm formation hamper its diagnostic accuracy. Extrapolation of LV volumes from such linear measurements using the Teichholz or Quinones method, each of which assumes a prolate ellipsoid LV shape, is no longer recommended.
Volumetric Measurements
LV volumes can be determined by 2D or 3D imaging. Although the 2D method is still considered the standard approach by most clinicians, the 3D method is gaining popularity in clinical routine.
The 2D and 3D methods for determining LV volumes are based on precise tracing of the endocardial border of the compacted myocardium. The tracing points are placed on the endocardial border, and all the noncompacted structures, such as papillary muscles, muscle bundles, and trabeculations, are excluded. The tracing line is closed by connecting the two tracing points at the mitral annulus by a straight line. The long axis of the ventricle is defined by a straight line between the mitral annulus and the most apical point of the LV ( Fig. 4.2 ).
The recommended approach for 2D measurement of LV volumes is the biplane Simpson method. It is based on careful tracing of the endocardial border of the compacted myocardium in the 4-chamber and 2-chamber projections (see Fig. 4.2 ). This approach is not ideal with regard to ventricular geometry because the two views are separated by an angle of approximately 60 degrees rather than being perpendicular to each other. Moreover, the left ventricular outflow tract (LVOT) is excluded from the volume calculation because the 3-chamber projection is not included. Nevertheless, this method has acceptable reproducibility and associated outcomes.
2D measurements of LV volumes are supported by strong prognostic data. The LV end-diastolic volume index is associated with mortality for patients with advanced heart failure caused by severely reduced systolic function. The LV end-systolic volume index is associated with hospitalization for heart failure in patients with coronary artery disease ( Fig. 4.3 ).
Optimal image quality with clear endocardial border definition is essential for accomplishing an accurate and reproducible tracing line. Image depth and width should be reduced so that only the LV from apex to annulus is depicted; with this LV-centered approach, a frame rate greater than 60 fps can easily be reached, ensuring an appropriate temporal resolution for LV border tracing and strain measurement. Care should be taken to visualize the maximal length of the LV without foreshortening of the apex.
Identification of the true apex is facilitated by two observations: (1) myocardial thickness decreases toward the apex, and (2) the apex does not move in a longitudinal direction during the cardiac cycle. To ensure that the true apex is depicted, it is useful to compare the apex as imaged in 4-chamber and 2-chamber projections. The left atrium (LA) should not be considered on such images; it appears shortened in most individuals because the atrial and ventricular long axes are usually not located in the same plane. Truncation of the LA on LV-centered images has an additional advantage in that the echocardiographer is not distracted by the atrial contours.
Even for experienced echocardiographers, it can be challenging to generate a correct tracing line throughout the LV endocardial border, and the lateral wall often is not clearly seen. The process can be facilitated by observing the LV borders for a few heartbeats in real time before tracing is initiated and can be further improved by letting several frames adjacent to the end-diastolic frame run back and forth (i.e., yo-yo view). After the tracing line in 4-chamber and 2-chamber views has been completed, LV length in both views should be compared, and the measurement should be repeated if there is a difference in length of more than 5%.
If the endocardial borders of two or more myocardial segments cannot be visualized, the use of an intravenous contrast agent is recommended. Because it is easier to recognize the true border of the compacted myocardium in the presence of a contrast agent, the volumes measured under the latter conditions are somewhat larger than those measured without contrast. The normal values available for LV volumes should not be applied when contrast has been used; separate normal values have been developed for contrast-enhanced LV quantification ( Table 4.1 ).
Normal values for 2D LV volumes are well established ( Table 4.2 ). Even after indexing for body surface area, normal ranges differ for males and females. They are also influenced by age, which can be explained by the lifelong remodeling of the LV that occurs in the normal population and results in a progressive decrease in LV volumes with advancing age ( Fig. 4.4 ).
TABLE 4.2
Normal Values and Severity Partition Cutoff Values for 2D Echocardiography–Derived LV Size and Function.
| Measurement | Male | Female | ||||||
|---|---|---|---|---|---|---|---|---|
| Normal Range | Mildly Abnormal | Moderately Abnormal | Severely Abnormal | Normal Range | Mildly Abnormal | Moderately Abnormal | Severely Abnormal | |
| LV Dimension | ||||||||
| LV diastolic diameter (cm) | 4.2–5.8 | 5.9–6.3 | 6.4–6.8 | >6.8 | 3.8–5.2 | 5.3–5.6 | 5.7–6.1 | >6.1 |
| LV diastolic diameter/BSA (cm/m 2 ) | 2.2–3.0 | 3.1–3.3 | 3.4–3.6 | >3.6 | 2.3–3.1 | 3.2–3.4 | 3.5–3.7 | >3.7 |
| LV systolic diameter (cm) | 2.5–4.0 | 4.1–4.3 | 4.4–4.5 | >4.5 | 2.2–3.5 | 3.6–3.8 | 3.9–4.1 | >4.1 |
| LV systolic diameter/BSA (cm/m 2 ) | 1.3–2.1 | 2.2–2.3 | 2.4–2.5 | >2.5 | 1.3–2.1 | 2.2–2.3 | 2.4–2.6 | >2.6 |
| LV Volume | ||||||||
| LV diastolic volume (mL) | 62–150 | 151–174 | 175–200 | >200 | 46–106 | 107–120 | 121–130 | >130 |
| LV diastolic volume/BSA (mL/m 2 ) | 34–74 | 75–89 | 90–100 | >100 | 29–61 | 62–70 | 71–80 | >80 |
| LV systolic volume (mL) | 21–61 | 62–73 | 74–85 | >85 | 14–42 | 43–55 | 56–67 | >67 |
| LV systolic volume/BSA (mL/m 2 ) | 11–31 | 32–38 | 39–45 | >45 | 8–24 | 25–32 | 33–40 | >40 |
| LV Function | ||||||||
| LVEF (%) | 52–72 | 41–51 | 30–40 | <30 | 54–74 | 41–53 | 30–40 | <30 |
BSA , Body surface area; LVEF , LV ejection fraction.
From Lang RM, Badano LP, Mor-Avi V, et al. Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. J Am Soc Echocardiogr . 2015;28(1):1–39.
Calculation of LV volumes with the use of 3D imaging is gaining popularity in clinical practice. The 3D approach has several advantages over 2D. It depends less on geometric assumptions and is less prone to foreshortening, although the latter is not automatically excluded in a 3D volume data set, and it might escape the attention of the investigator that part of the LV was not included.
The LV volumes derived from 3D data sets are larger than those from 2D acquisitions and therefore closer to the volumes obtained by cardiac magnetic resonance (CMR) imaging. This observation has been confirmed by several studies of different patient populations ( Table 4.3 ). The 3D LV measurements seem to exhibit a stronger association with outcomes than the 2D parameters, suggesting that the 3D measurements are closer to reality. The latest 3D applications have good accuracy and underestimate the CMR-derived LV volumes by less than 10% ( Table 4.4 ). ,
TABLE 4.3
Comparison of 2D and 3D Echocardiography–Derived Values for LV End-Diastolic Volume, End-Systolic Volume, and Ejection Fraction.
| Study a | Study Type | LV EDV (mL) | LV ESV (mL) | LV EF (%) | |||
|---|---|---|---|---|---|---|---|
| 2D Echo | 3D Echo | 2D Echo | 3D Echo | 2D Echo | 3D Echo | ||
| Dorosz et al. | Meta-analysis 1174 patients |
ΔCMR = −48 | ΔCMR = −16 | ΔCMR = −28 | ΔCMR = −10 | ΔCMR = 0.1 | ΔCMR = 0.0 |
| Muraru et al. | Single center 226 patients |
97 ± 25 | 107 ± 25 | 35 ± 11 | 39 ± 11 | 65 ± 4 | 64 ± 4 |
| Tamborini et al. | Single center 189 patients |
142 ± 58 | 175 ± 64 | 71 ± 49 | 102 ± 59 | 54 ± 15 | 45 ± 14 |
| Medvedofsky et al. | Single center 416 patients |
166 ± 85 | 187 ± 89 | 88 ± 77 | 99 ± 84 | 52 ± 16 | 52 ± 17 |
EF , Ejection fraction; EDF , end-diastolic volume; ESV , end-systolic volume.
a The first study (Dorosz et al.) is a meta-analysis, and the difference between echocardiographic– and cardiac magnetic resonance–derived values (ΔCMR) is indicated. For all the other studies, mean values and standard deviations of 2D and 3D echocardiographic measurements are listed.
TABLE 4.4
Comparison of 3D Echocardiography Versus Cardiac Magnetic Resonance–Derived Values for LV End-Diastolic Volume, End-Systolic Volume, and Ejection Fraction.
| Study a | Study Type | LV EDV (mL) | LV ESV (mL) | LV EF (%) | ||||
|---|---|---|---|---|---|---|---|---|
| 3D Echo | CM | 3D Echo | CMR | 3D Echo | CMR | |||
| Dorosz et al. | Meta-analysis 1174 patients 3D manual or automated |
Δ Echo-CMR = −19 | Δ Echo-CMR = −10 | Δ Echo-CMR = −1 | ||||
| Tsang et al. | Single center 159 patients 3D automated + manual adjustment |
190 ± 64 | 201 ± 66 | 118 ± 60 | 122 ± 71 | 41 ± 12 | 43 ± 16 | |
| Tamborini et al. | Single center 189 patients 3D automated + manual adjustment |
175 ± 61 | 200 ± 74 | 108 ± 59 | 117 ± 73 | 42 ± 13 | 45 ± 17 | |
CMR , Cardiac magnetic resonance; EF , ejection fraction; EDF , end-diastolic volume; ESV , end-systolic volume.
a The first study (Dorosz et al.) is a meta-analysis, and the difference between echocardiographic- and CMR-derived values (ΔCMR) is indicated. For all the other studies, mean values and standard deviations of 3D echocardiographic and CMR measurements are listed.
To reach an optimal accuracy, LV volumes in a 3D data set are usually determined by a semi-automated approach involving automated border detection with manual contour adjustment. Multiple rounds of correction sometimes are required because the algorithm may not always accept the manual changes. Manual editing seems particularly important when image quality is poor because the accuracy of automated border detection is lower under such conditions. The quality of manual editing depends on operator experience, and automated border detection can be improved by experienced investigators. In addition to manual correction, the accuracy of 3D LV measurements can be increased by a higher volume rate and by the application of echocardiographic contrast.
Whereas manual measurements of 3D LV volumes exhibit moderate intraobserver and interobserver reproducibility, the latter is improved when automated border detection is combined with manual contour adjustment and improved even further when no manual correction is performed. In these circumstances, however, the accuracy of the measurements usually decreases. This situation opens the possibility of applying 3D echocardiography for determining LV volumes in a patient-tailored approach.
Automatic border detection should be corrected manually when accuracy is more important, but refraining from manual correction can be considered when changes in LV volumes or LVEF are essential for follow-up. When reproducibility is preferred, the maximal benefit of the automated approach can be achieved when images of good quality are examined by inexperienced echocardiographers, and application of echocardiographic contrast can improve interobserver reproducibility. ,
The 3D LV volumes need to be recorded very carefully to avoid data sets that cannot be quantified. All LV measurements are obtained from a single data set in the 3D approach, and this is often not possible, particularly in very dilated ventricles. Control of the image quality of the 3D data set in reconstructed imaging planes derived from the data set being registered is recommended. These imaging planes should include the equivalent of the apical 4-chamber, 2-chamber, and 3-chamber views and short-axis views at the apical, mid-ventricular, and basal levels ( Fig. 4.5 ). The 3D data set should be accepted only when the endocardial border of the compacted myocardium is visible in all segments.
3D can be applied for measurement of LV volumes in any patient in whom image quality is sufficient for appropriate acquisition of the LV. Important indications for routine clinical application include a need for frequent follow-up examinations for surveillance of LV systolic function, such as in patients undergoing cardiotoxic chemotherapy and those under optimization of medical therapy before implantation of a cardioverter-defibrillator or cardiac resynchronization therapy. Consequently, it becomes essential that the same method of LV volume quantification is maintained during serial assessment of LV function. ,
Normal values for 3D-derived LV volumes are less well established than those for 2D volumes, and there are not enough data to allow clear definition of a normal range. However, several well-performed studies providing normal values for relatively small populations are available ( Table 4.5 ). When an LV volume is determined in a 3D data set, it has to be documented in the echocardiography report because the normal values are larger than those measured in 2D due to systematic underestimation of LV volumes by the 2D approach.
TABLE 4.5
Normal Values for 3D Echocardiography–Derived LV End-Diastolic Volume Index and Ejection Fraction in Various Populations. a
| Study | Population | LV EDVI (mL/m 2 ) | LV Ejection Fraction (%) |
|---|---|---|---|
| Aune et al. | Scandinavian men ( n = 79) Scandinavian women ( n = 87) |
66 ± 10 58 ± 8 |
57 ± 4 61 ± 6 |
| Fukuda et al. | Japanese men ( n = 222) Japanese women ( n = 134) |
50 ± 12 46 ± 9 |
61 ± 4 63 ± 4 |
| Chahal et al. | European White men ( n = 338) European White women ( n = 161) |
49 ± 9 42 ± 8 |
61 ± 6 62 ± 5 |
| South Asian men ( n = 290) South Asian women ( n = 189) |
41 ± 9 39 ± 8 |
62 ± 5 62 ± 5 |
|
| Mararu et al. | White men ( n = 101) White women ( n = 125) |
63 ± 11 56 ± 8 |
62 ± 4 65 ± 4 |
| Bernard et al. | European men ( n = 187) European women ( n = 253) |
69 ± 14 60 ± 11 |
59 ± 4 60 ± 5 |
EF , Ejection fraction; EDVI, end-diastolic volume index.
a Individual data sources are indicated because there is no consensus on normal values.
The advantages and disadvantages of various methods for measuring LV volumes are summarized in Table 4.6 .
TABLE 4.6
Quantification of LV Size and Function: Advantages and Disadvantages of Various Methods.
| Method | Advantages | Disadvantages |
|---|---|---|
| 1D | High temporal resolution | Acquisition may be technically difficult |
| Very good reproducibility | Obsolete for calculation of ejection fraction | |
| Many published data, outcome data | For calculation of muscle mass, accurate only in normally shaped ventricles with constant muscle thickness | |
| 2D | Reasonably good reproducibility | Accurate endocardial border detection may be technically difficult |
| Fewer geometric assumptions | Real apex visualization may be technically difficult | |
| Many published data, outcome data | Blind to changes in shape or function not visualized in 4- and 2-chamber views | |
| 3D | Virtually no geometric assumptions | More dependent on image quality |
| Less underestimation of volumes | Lower special and temporal resolution | |
| Higher reproducibility of measurements in some studies | Fewer published data, outcome data |
LV Wall Thickness and Mass
LV mass shows a strong association with cardiovascular outcomes and independently predicts adverse events, including death. LV mass determined by echocardiography provides information on outcomes independent of cardiovascular risk factors. The Framingham Heart Study showed that LV mass measured by echocardiography was associated with total mortality, cardiovascular mortality, and cardiovascular morbidity rates.
LV mass can be calculated by M-mode, 2D, or 3D imaging. All of these methods determine myocardial volume, which is then converted into myocardial mass by multiplying the volume by the specific gravity of myocardial tissue (1.04 g/cm 3 ). Whereas myocardial volume has to be calculated when using M-mode or 2D imaging, it can be measured directly using 3D. For all the methods mentioned, the recommendation is to place the tracing points on the endocardial border of the compacted myocardium. This approach has become possible because of refinements in echocardiographic technology that allow detection of the true border of the myocardium corresponding to the blood–tissue interface. Previously, the recommended approach was to measure from leading edge to leading edge because gain settings influence the trailing edge of the endocardial signal. This method resulted in inaccurate calculations at both ends of the LV size spectrum, with overestimation of LV mass in small ventricles and underestimation of mass in large ventricles.
Use of M-mode for LV mass calculation requires measurement of the LV inner diameter and the anteroseptal and inferolateral diameters of the compacted myocardium, all in one line at the level of the mitral valve leaflet tips in end-diastole (see Fig. 4.1 ). This approach is based on the assumption that the LV is a prolate ellipsoid with a ratio of 2:1 for its long versus short axis. The following formula is applied for calculation of muscle mass and has been validated anatomically:
SWT d is the diastolic septal wall thickness; LVID d the diastolic LV inner diameter; and PWT d is the diastolic posterior wall thickness. Because these linear measurements are cubed, small measurement errors have a considerable effect on the accuracy of mass calculation.
The 2D method for LV mass calculation is based on measurement of LV length obtained from the apical 4-chamber projection and calculation of mean myocardial wall thickness at mid-ventricular level obtained from parasternal short-axis imaging in end-diastole. LV mass is then calculated by the area-length or the truncated ellipsoid method. The formula applied for calculation of muscle mass based on the area-length method is as follows:
A 1 is the LV epicardial cross-sectional area in short-axis view; A 2 is the endocardial cross-sectional area in short-axis view; a is the distance of the LV minor radius to the LV apex, d is the distance to the LV mitral valve plane, and t is the LV mean wall thickness. Muscle volume is obtained as the difference between the epicardial and the endocardial shell.
Although there are fewer geometric assumptions with this approach compared with M-mode imaging, it is still assumed that the LV is an ellipsoid. The LV mass determined by M-mode is comparable to that obtained by both 2D methods, and all three methods show a moderate correlation with true LV mass measured at autopsy.
Depending on the clinical situation, it may be indicated to give preference to a specific method. In patients with a normally shaped LV, the M-mode method is reasonable because it is simple and has low variability and good accuracy. In patients with a remodeled LV, the 2D method may be the better solution because it offers an improved analysis of LV shape and is better suited for follow-up.
The 3D method for LV mass calculation is based on tracing the endocardial borderline of the ventricular cavity and the epicardial border. This approach has the advantage of being independent of geometric assumptions because all of the ventricular borders are contained within the 3D data set. The 3D method is potentially more accurate than the 2D method and has a better agreement with CMR measurement. It also has lower intraobserver and interobserver variability under study conditions. , However, it critically depends on having very good image quality, particularly because the spatial and temporal resolution of 3D images are inferior to those of 2D.
A frequently encountered problem is subaortic septal hypertrophy (i.e., sigmoid septum), which can often be seen in elderly individuals. M-mode measurements should not be used in this situation because the septal wall thickness is measured at the site of a localized hypertrophy and LV mass would be overestimated. The 2D method is more accurate because wall thickness is determined at the mid-ventricular level; however, localized hypertrophy would be ignored and LV mass underestimated with this approach. The best method for measuring LV mass in such individuals is probably the 3D method, as long as image quality allows accurate tracing of the compacted myocardium.
LV wall thickness relative to LV diameter in end-diastole is a simple parameter for assessing LV geometry. It is determined by calculating the ratio of posterior wall thickness times 2 and the LV inner diameter obtained in end-diastole with the transducer in the parasternal long-axis position:
PWT d is the diastolic posterior wall thickness, and LVID d is the diastolic LV inner diameter. A cutoff value of 0.42 separates normal geometry from concentric remodeling when LV mass is in the normal range and eccentric from concentric hypertrophy when LV mass is increased ( Table 4.7 ). Relative wall thickness is a ratio, so it should always be considered in the context of the absolute values for LV diameter and wall thickness ( Fig. 4.6 ).
TABLE 4.7
Descriptors of LV Remodeling with Cutoff Values for Geometry and Hypertrophy.
| Type of Remodeling | Population | LV Mass Index (g/m 2 ) | Relative Wall Thickness |
|---|---|---|---|
| Normal | Men | ≤115 | ≤0.42 |
| Women | ≤95 | ≤0.42 | |
| Concentric remodeling | Men | ≤115 | >0.42 |
| Women | ≤95 | >0.42 | |
| Eccentric hypertrophy | Men | >115 | ≤0.42 |
| Women | >95 | ≤0.42 | |
| Concentric hypertrophy | Men | >115 | >0.42 |
| Women | >95 | >0.42 |
Data from Lang RM, Badano LP, Mor-Avi V, et al. Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. J Am Soc Echocardiogr . 2015;28(1):1–39.
LV mass should be indexed to body surface area (BSA) because LV mass depends on body size. Because of the difference between males and females in LV mass, separate reference ranges are required for the indexed values. Although indexing to body surface area is reasonable, it is not accurate for all groups of individuals and underestimates LV hypertrophy in obese patients. Like other echocardiographic parameters, indexing to BSA is particularly important for individuals with small body size. An alternative approach is indexing to height, but this is less well accepted than BSA. ,
Normal values for 3D-derived LV mass are more difficult to define than for 2D because fewer studies are available. , Normal values for LV mass depend on body size, body mass, gender, age, and ethnic group. The normal values for the M-mode and 2D methods are provided in Table 4.8 , and those for the 3D approach are provided in Table 4.9 . Advantages and disadvantages of the various methods are summarized in Table 4.6 .
TABLE 4.8
Normal Values and Severity Partition Cutoff Values for 2D Echocardiography–Derived LV Mass.
| Male | Female | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Normal Range | Mildly Abnormal | Moderately Abnormal | Severely Abnormal | Normal Range | Mildly Abnormal | Moderately Abnormal | Severely Abnormal | ||
| LV Mass by Linear Method | |||||||||
| Septal wall thickness (cm) | 0.6–1.0 | 1.1–1.3 | 1.4–1.6 | >1.6 | 0.6–0.9 | 1.0–1.2 | 1.3–1.5 | >1.5 | |
| Posterior wall thickness (cm) | 0.6–1.0 | 1.1–1.3 | 1.4–1.6 | >1.6 | 0.6–0.9 | 1.0–1.2 | 1.3–1.5 | >1.5 | |
| LV mass (g) | 88–224 | 225–258 | 259–292 | >292 | 67–162 | 163–186 | 187–210 | >210 | |
| LV mass/BSA (g/m 2 ) | 49–115 | 116–131 | 132–148 | >148 | 43–95 | 96–108 | 109–121 | >121 | |
| LV Mass by 2D Method | |||||||||
| LV mass (g) | 96–200 | 201–227 | 228–254 | >254 | 66–150 | 151–171 | 172–193 | >193 | |
| LV mass/BSA (g/m 2 ) | 50–102 | 103–116 | 117–130 | >130 | 44–88 | 89–100 | 101–112 | >112 | |
BSA , Body surface area.
From Lang RM, Badano LP, Mor-Avi V, et al. Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. J Am Soc Echocardiogr . 2015;28(1):1–39.
TABLE 4.9
Normal Values for 3D Echocardiography–Derived LV Mass in Various Populations. a
| Study | Population | 3D LV Mass (g/m 2 ) |
|---|---|---|
| Fukuda et al. | Japanese men ( n = 222) Japanese women ( n = 134) |
64 ± 12 56 ± 11 |
| Mararu et al. | White men ( n = 101) White women ( n = 125) |
77 ± 10 74 ± 8 |
| Mizukoshi et al. | Japanese men ( n = 121) Japanese women ( n = 109) |
70 ± 8 61 ± 8 |
| American men ( n = 78) American women ( n = 82) |
70 ± 10 60 ± 8 |
a Individual data sources are indicated because there is no consensus on normal values.
LV Shape and Its Implications for Measurements
The normally shaped LV is a prolate ellipsoid with a ratio of the long axis to short axis of approximately 2:1. The linear method for LV mass calculation is based on the assumption that the LV fulfills this criterion. In many individuals with cardiac disease, however, the LV undergoes remodeling, resulting in a shape that diverges from the ideal ellipsoid in a global or a regional manner.
Global aberration from the standard shape consists of an atypical ratio of LV short and long axes. In normal individuals, this ratio ranges from 0.45 to 0.62. , In hypertensive patients with normal LV ejection fraction (LVEF), the ratio tends to be higher, with values ranging from 0.52 ± 0.04 for those with concentric remodeling to 0.63 ± 0.03 for those with hypertrophy. Similarly, the LV undergoes a spherical remodeling in cardiomyopathies such as dilated cardiomyopathy or noncompaction cardiomyopathy. In these patients, the remodeling seems to be particularly pronounced in the apical and septal regions of the LV.
Spherical remodeling has also been observed in patients with mitral or aortic regurgitation. In patients with acute myocardial infarction, LV sphericity predicts the extent of ventricular remodeling. In a population with various cardiac diseases, global and regional alterations of LV shape are associated with cardiovascular mortality. When LV systolic function is supported by a ventricular assist device, LV volume and sphericity progressively decrease with increasing speed of the assist device, and the LV regains a more conical shape. Patients with mild heart failure with reduced ejection fraction (EF) undergoing cardiac resynchronization therapy develop an increase in LVEF accompanied by a decrease in sphericity, and improved sphericity is associated with fewer heart failure hospitalizations and lower mortality rates. A similar reverse remodeling can be observed after surgical or percutaneous repair of mitral regurgitation.
The LV may also undergo elongation during a global remodeling process such as occurs in individuals with an athlete’s heart. It is essential to recognize these alterations and to choose the method for calculation of LV sphericity, volumes, and mass accordingly. The linear method can be used for LV size in such situations, but the LV diameter should be reported as a diameter, and neither LV volume nor LV mass should be calculated from it. The area-length method is more useful for these patients, but it is still based on geometric assumptions. The Simpson method is not greatly affected by alterations in LV shape and is useful in these instances. The 3D method is certainly the best approach in patients with appropriate echocardiographic windows.
Regional deviations from the standard LV shape can occur for various reasons. Apical hypoplasia is a rare congenital abnormality, and endomyocardial fibrosis is an acquired pathology associated with a disturbed LV shape in the apical region. , Aneurysms usually occur from regional loss of myocardial tissue in coronary artery disease or myocarditis but may be congenital. Similarly, large diverticula represent a problem for accurate quantification of LV volume.
Whenever there are regional abnormalities of LV shape or regional thinning of the LV wall, neither LV volume nor LV mass should be calculated from the linear measurements. The area-length method and the Simpson rule likewise may not be accurate for quantification of LV volume. 3D echocardiography is the best method of calculating volumes and mass in these patients when they have appropriate echocardiographic windows.
Quantitative Analysis of LV Systolic Function
You're Reading a Preview
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here