Assessment of LV Mass by Echocardiography
To the Editor:
The data presented by Vasan et al1 have clear and important implications in the way echocardiographic data should be reported and understood in routine practice because they make it possible to readily identify (hypertensive) patients with a “normal sized” heart in whom left ventricular (LV) mass is truly increased. In addition, it is likely that these data would help distinguish the dilated and thickened but otherwise healthy heart of an athlete from cardiomyopathy.
The general utility of the data, however, would be enhanced by displaying it as shown here in the Figure⇓ , which illustrates the relation between LV mass, LV end-diastolic internal dimension (ID), and combined LV wall thickness, superimposed on the relation between the 95th percentile for LV ID and LV mass for males (upper line) and females (lower line) reported in Tables 2 and 3 of the article by Vasan et al.1
In this way, it is readily seen, for example, that an individual with an LV ID of 5.0 cm who has a combined LV wall thickness of 2.0 cm has an LV mass beyond the 95th percentile range of normal even though LV hypertrophy would not ordinarily be considered present on the basis of accepted normal values. In contrast, an otherwise healthy but taller individual with an LV ID of 6.0 cm would be recognized as having an LV mass within the 95% CI of normal even in the presence of a combined LV wall thickness of 2.3 cm, which by standard criteria would be reported as LV hypertrophy.
Furthermore, the Framingham data1 also indicate that there is a strong relation between LV ID and wall thickness, so that in general, when LV ID is ≤5.6 cm, LV mass will fall within the 95% confidence limits of normal when the ratio of LV ID to combined wall thickness is >2.7. In subjects with an LV ID >5.6 cm, however, this ratio should be >2.6. This rule of thumb may simplify the distinction between left ventricles with a normal and abnormal mass in the absence of reference tables or charts at the time of reporting.
- Copyright © 1999 by American Heart Association
Vasan RS, Larson MG, Levy D, Evans JC, Benjamin EJ. Distribution and categorization of echocardiographic measurements in relation to reference limits: the Framingham Heart Study: formulation of a height- and sex-specific classification and its prospective validation. Circulation. 1997;96:1863–1873.
We thank Drs Nidorf and McQuillan for their insightful comments on our work.R1 We have developed a figure to display the relations between LVM, LVIDed, and LVWT (left ventricular mass, left ventricular internal diameter end diastole, and sum of the left ventricular wall thicknesses [septum plus posterior wall], respectively, using the Penn corrected American Society of Echocardiography LVMR2 ). For the Figure⇑ , the lower set of symbols (•) represents the 95th percentile values of LVM in women (height, 54 to 72 in) and the upper set (□) represents the 95th percentile values of LVM in men (height, 60 to 78 in). To use this figure correctly, it is necessary to locate the intersection of LVIDed with a specific LVWT line (or interpolate); the ordinate is the calculated LVM. Then it is necessary to compare the calculated LVM with a sex- and height-specific LVM “dot.” If the calculated LVM lies above the “dotted” value, then the individual exceeds the sex-height–specific LVM reference limit. Consider a woman who has the relatively “normal” echocardiographic dimensions of LVIDed=5.0 and LVWT=1.6, giving LVM=140 g and a LVIDed/LVWT “ratio” of 3.1. If this woman is <61″ tall, the LVM exceeds the 95% reference value for her height; in contrast, for anyone >61″ tall, the LVM is below the reference value. Thus, one should not interpret LVIDed, LVWT, or LVM, without reference to height and sex.
It should be noted that there is not a strong relation between LVIDed and LVWT. Although the 95th percentile values of LV mass, LVIDed, and LVWT seem to plot along a straight line, this does not suggest a strong relation between LVIDed and LVWT. Framingham data, on the contrary, suggest a very poor correlation of LVIDed and LVWT (r=0.042 [men] and −0.079 [women]).
We greatly appreciate Drs Nidorf and McQuillan’s efforts to increase the utility of our data by prompting us to provide a graphical representation of the relations between LV mass, LVIDed, and combined LVWT. We hope that the figure will enhance efforts to standardize the clinical interpretation of echocardiographic measurements.
Vasan RS, Larson MG, Levy D, Evans DW, Benjamin EJ. Distribution and categorization of echocardiographic measurements in relation to reference limits: the Framingham Heart Study: formulation of a height- and sex-specific classification and its prospective validation. Circulation. 1997;96:1863–1873.