Left ventricular mass and geometry

Clinical background

Left ventricular mass and geometry are central measures in the assessment of hypertensive heart disease, valvular disease and cardiomyopathy. Increased left ventricular mass is an independent risk marker for cardiovascular morbidity and mortality, and the geometric pattern — concentric or eccentric — indicates the type of overload to which the ventricle has been subjected. The decision the tool serves is primarily whether left ventricular hypertrophy is present and, if so, which remodelling mechanism predominates. This influences the choice of antihypertensive treatment, the assessment of prognosis and the indication for further investigation, for example cardiac MRI when asymmetric hypertrophy is suspected.

Calculating left ventricular mass and geometry

Left ventricular mass is calculated with the corrected Devereux formula, derived from autopsy data:

$$\text{LVM (g)} = 0{.}8 \times \left{1{.}04 \times \left[(\text{LVIDd} + \text{PWTd} + \text{SWTd})^3 - \text{LVIDd}^3\right]\right} + 0{.}6$$

where LVIDd is the left ventricular internal diameter in diastole, PWTd is the posterior wall thickness in diastole and SWTd is the septal wall thickness in diastole, all in centimetres. The formula rests on the assumption that the ventricle is a prolate spheroid with symmetrical wall thickness. Left ventricular mass is indexed to body surface area to give the mass index (g/m²).

Relative wall thickness (RWT) is calculated as:

$$\text{RWT} = \frac{2 \times \text{PWTd}}{\text{LVIDd}}$$

RWT quantifies the relationship between wall thickness and cavity size, independently of total mass.

The derivation cohort consisted of 55 patients who underwent blinded echocardiographic measurements before death, with left ventricular mass measured at autopsy as the reference [1]. The Penn cube formula correlated strongly with autopsy mass (r = 0.92) and overestimated by only 6 per cent, whereas the ASE cube formula overestimated by an average of 25 per cent. The correction $\text{LVM} = 0{.}80 \times \text{ASE cube} + 0{.}6$ g — that is, the Devereux formula used by this calculator — reduced the systematic error. Sensitivity for left ventricular hypertrophy was 100 per cent (18 of 18) and specificity 86 per cent (29 of 34) in this cohort [1]. The patients were, however, a selected autopsy series with a wide range of chamber sizes, and the results reflect performance under ideal measurement conditions rather than routine clinical practice.

Interpretation in practice

The calculator classifies geometry along two dimensions: mass index (normal or increased) and relative wall thickness (≤ 0.42 or > 0.42). The upper limit of normal for the mass index is 95 g/m² in women and 115 g/m² in men according to the ASE/EACVI.

Concentric remodelling is an early sign of hypertensive cardiac involvement and should prompt intensified blood pressure control, even when the mass index is still normal. Concentric hypertrophy indicates marked pressure overload and is associated with an unfavourable prognosis. Eccentric hypertrophy, particularly with a dilated ventricle, has in observational studies shown a stronger association with reduced ejection fraction than concentric hypertrophy, with a relative risk of 2.3 for the development of impaired left ventricular function [2].

It should be noted that the traditional four-pattern classification has been questioned. In a study of 400 asymptomatic hypertensive patients examined with cardiac MRI, concentric and eccentric hypertrophy carried a similar prognosis (log-rank P = 0.62), suggesting that the geometric pattern in itself adds no prognostic information beyond that given by mass [3]. The clinical value of the geometric classification lies rather in mechanistic understanding and choice of treatment strategy than in risk stratification.

Validation and performance

The Devereux formula was originally validated against autopsy mass in a small cohort with good results [1]. In subsequent studies, M-mode-based linear methods have shown poorer agreement with reference methods, particularly in disease that alters ventricular geometry. In an animal study of 254 pigs, in which various models of heart disease simulated ischaemic and non-ischaemic heart failure, the M-mode method exceeded actual left ventricular mass by an average of 21.7 g, with progressive overestimation as heart weight increased [4]. The correlation with actual weight was moderate (r = 0.68) compared with the 2D area-length method (r = 0.82). Ischaemic heart disease and ventricular dilatation were the strongest predictors of overestimation by the M-mode method [4]. In animals without ischaemia or dilatation the correlation was, by contrast, strong (r = 0.89), confirming that the formula performs well when geometry is symmetrical.

In the Framingham Heart Study, echocardiographic measures including the mass index were evaluated in 1,497 participants (mean age 65 years, 55.4 per cent women) over a mean follow-up of 8.3 years [5]. The mass index was associated with the composite outcome of cardiovascular disease or death, but left atrial volume and function were stronger prognostic markers. An abnormal mass index was present in 17 per cent of participants [5]. The study confirms that the mass index carries prognostic information in a general population, but also that other echocardiographic measures may outperform it.

Limitations

The linear Devereux formula presupposes a symmetrically thickened ventricle with a circular cross-section. In asymmetric hypertrophy, as in hypertrophic cardiomyopathy, or with regional wall motion abnormalities after myocardial infarction, the estimate becomes unreliable [4]. In such cases cardiac MRI or 3D echocardiography should be used.

The measurements are operator-dependent. LVIDd, PWTd and SWTd are measured in M-mode or from 2D images at the papillary muscle level, and small measurement errors are amplified by the cubing in the formula. A systematic measurement error of 1 mm in wall thickness can change the calculated mass by as much as 15 per cent, particularly in small ventricles. This limits reproducibility across repeated measurements by different operators.

The formula does not apply to right ventricular overload or to lung disease in which septal motion is paradoxical, nor to patients with a pacemaker or conduction disease that alters the motion pattern of the septum. In dilated cardiomyopathy the formula overestimates mass because the spheroid geometry no longer holds [4].

Indexing to body surface area can mislead at extremes of body weight. Obesity gives a higher body surface area and hence a lower mass index, which may mask hypertrophy. Alternative indexing methods, for example indexing to height^2.7, have been proposed to address this, but the calculator uses body surface area indexing in accordance with current guidelines.

References

1. Devereux RB et al. Echocardiographic assessment of left ventricular hypertrophy: comparison to necropsy findings. Am J Cardiol. 1986;57(6):450–8. PMID: 2936235
2. Mutai SM et al. Hypertension and left ventricular geometry: Diagnosis, prognosis, and management. EXCLI J. 2026;25:756–775. PMID: 42376431
3. Le TT et al. The remodelling index risk stratifies patients with hypertensive left ventricular hypertrophy. Eur Heart J Cardiovasc Imaging. 2021;22(6):670–679. PMID: 32255186
4. Miyashita S et al. Echocardiographic Left Ventricular Mass Estimation: Two-Dimensional Area-Length Method is Superior to M-Mode Linear Method in Swine Models of Cardiac Diseases. J Cardiovasc Transl Res. 2020;13(4):648–658. PMID: 31828537
5. von Jeinsen B et al. Prognostic Significance of Echocardiographic Measures of Cardiac Remodeling. J Am Soc Echocardiogr. 2020;33(1):72–81. PMID: 31624026