Secondary Logo

Journal Logo

Obesity and left ventricular hypertrophy: does my heart look big on this?

Jennings, Garry

doi: 10.1097/HJH.0b013e3283401fad
Editorial commentaries

Baker IDI Heart and Diabetes Institute, Melbourne, Australia

Correspondence to Garry Jennings, MD, FRACP, Baker IDI Heart and Diabetes Institute, Melbourne, Australia E-mail:

Larger people have larger hearts. Furthermore, for any given body size, men have larger hearts than women, athletes may have larger hearts compared to nonathletes and just about anyone with cardiac disease has cardiac enlargement of some form. This is especially important in those with hypertension in whom the presence of left ventricular hypertrophy (LVH) is an important prognostic marker. Even within the normotensive population, LVH carries some increased risk, although it is not known how important are changes that occur with ageing or interventions within the normal range of left ventricular mass (LVM) in nonhypertensive individuals. Quantification of LVH, therefore, always introduces the challenges of taking into account sex, body size, and physical activity, factors that potentially could result in people being misclassified as to whether they have LVH or not [1].

The issue of large body size becomes more complicated when obesity is the cause. LVM relates better to lean body than overall body mass [2,3]. Obesity is associated with disproportionate accumulation of body fat, not a big factor in myocardial structure, but clearly manifest in the pericardium. Increase in the cardiac outline due to myocardial hypertrophy or increase due to enlargement of pericardial fat pads may be indistinguishable with coarser methods of assessing cardiac structure, but are readily identifiable with contemporary imaging techniques. Obesity is also associated with haemodynamic changes, including increased stroke volume and cardiac output that may make the heart appear larger.

There are qualitative differences in the patterns of change seen in athletes and various forms of cardiac disease in that eccentric hypertrophy with dilatation of the cavity of the ventricle, good contractility and normal or near normal wall thickness is the rule with the former, whereas concentric hypertrophy is more common in hypertension [4]. However, strength training can cause concentric hypertrophy. A proportion of hypertensive patients have eccentric hypertrophy, particularly those with decompensated ventricles. In fact, almost every variation of thick and thin ventricular wall, dilated or reduced ventricular cavity dimension is found in hypertensive populations [5]. Different intracellular signalling pathways associated with these patterns have been identified and in future it may be possible to use biomarkers to classify different forms of alteration in cardiac structure and their different implications for prognosis [6]. In the meantime, we must use imaging to derive structural and functional information about the heart and face the challenges outlined above of correcting for body size and other confounding factors.

Of the plethora of methods described so far calculation of left ventricular mass index (LVMI) from two-dimensional echocardiographic measurements of left ventricular wall thickness (LVWT) and left ventricular internal diameter during diastole (LVIDD) has received the most attention. The classical formula of Devereux and Reichek [7], with slight amendments and subsequent conventions, has been widely used and to a degree anatomically validated against post-mortem left ventricular weights [8]. It involves a number of assumptions about the shape of the ventricle, contribution of the epicardium and other aspects of the heart. Although these mean that estimation of LVM in this way can be an approximation at best, the method has stood clinical research in good stead as an excellent predictor of outcome in hypertension (better than clinic blood pressure itself), reversible to some extent with effective therapy for elevated blood pressure. It is, however, strongly influenced by body size and systematically different in men and women. Because of this and other contributions to individual variation, including genetics, the variance in both the normotensive and the hypertensive populations is huge, causing difficulties in defining cut-offs for normal and abnormal LVM and underestimating the frequency of LVH in hypertension. Variables that quantify cardiac dimension with less variance necessarily give higher prevalence of LVH in hypertension than LVM on statistical considerations alone [9].

It has been appreciated since the late nineteenth century that across mammalian species, the weight of the heart and other organs except the brain is related to body weight through power laws. This field, called allometry, was the subject of a classic book by Julian Huxley in which numerous body measurements were plotted on log-log paper against total body weight, height or length and shown to be linear [10]. In 1965, Stahl [11] summarized the previous observations and found the relationship between heart and total body weight [y = axb, where y is heart (or other organ weight) in g, x is body weight in kg and a and b are coefficients of the allometric equations] to be robust across a huge range of body size from tree shrews weighing about 10 g to humans and large primates. There was a strikingly good fit statistically for the equation:

However, as with human echocardiographic data, there was good fit across a wide range, but the standard error was too large to accurately predict heart mass in an individual.

Despite reasonable anatomical validation of absolute cardiac mass, early correction of LVM determined from two-dimensional echocardiography by indexing for BMI was unsuccessful due to wide variation in any given population and an inability to correct for sex, racial and other differences between individuals. Indexing LVM to height or lean body mass was found to be a little better. Eventually, however, allometric corrections were applied and index LVM to height 2.7 has been a recent standard [12]. We have found this to be useful in eliminating apparent ethnic differences in LVM between Asian and Europid populations. However, Chirinos et al.[13] recently reported that the standard allometric power (2.7) does not account for sex. Furthermore, it was proposed that any correction of LVM for height incidentally collects prognostic information associated with height, but that indexing to the power 1.2 was less subject to sex differences, applied equally well to echocardiographic or MRI measurements and was a better indicator of future outcome than LVM indexed to height2.7.

Of course in assessing concentric left ventricular hypertrophy, the simple ratio of wall thickness: left ventricular internal diameter has considerable merit (Fig. 1). It avoids the assumptions made in calculation of LVM, is dimensionless and, therefore, needs no correction for body size or sex, and has smaller variance in a healthy population [1]. The latter mitigates the underestimation of the prevalence of LVH in hypertension found with indices that have greater variance (i.e. 2 SD from the mean has a smaller spread making outliers more frequent in a disease population). It is also more biologically meaningful in terms of the mechanisms, whereby LVH exerts a negative effect on outcomes [14].

Fig. 1

Fig. 1

There are grounds for assuming that methods that detect the highest frequency of LVH in a hypertensive population are most accurate. LVH is largely a compensatory response to increased cardiac afterload. It is found uniformly in wild strain animal models of hypertension. People with increased afterload due to hypertension without some LVH beg the question why not? What protective mechanism against compensatory LVH is involved, would this be a good thing or a bad thing? It is our contention that some degree of hypertrophic response probably occurs in all hypertensive patients, but this is not always detectable with contemporary methodology.

Of the present commonly used methods, the electrocardiogram designated as LVH when assessed using Cornell criteria, Sokolow–Lyon or other methods has a low prevalence in hypertensive populations and does not necessarily reflect anatomical LVH [15]. It is not necessarily anatomical LVH because the R-wave measurement can be altered minute to minute (when no great change in cardiac mass can occur) by changing cardiac volume with administration of an arterial vasodilator or a diuretic. However, ECG LVH is an outstanding predictor of outcome, identifying a group with a poor outcome matching those with previous myocardial infarction, or some cancers.

LVH on echocardiogram has its problems as outlined above, but is also established as a good risk marker, better than any one of the classical risk factors. MRI assessment of LVM is the present gold standard of noninvasive methods of assessing the presence of LVH. As multiple slices contribute to the measurement, no impossible assumptions about the shape of the ventricle need be made, unlike the single slice technique with the standard echocardiographic method. The estimations of LVM are systematically less than with two-dimensional echocardiography [16]. Edge definition is also better. MRI studies have not been performed on the large scale of those reported that have used electrocardiograms or echocardiograms, but it is a reasonable assumption that the value of an individual investigation will be enhanced by the better fidelity of the method compared to previous techniques.

Obesity introduces another order of complexity in assessing LVH, especially using echocardiography in which image quality is generally diminished. There have been many studies examining the influence of body mass on LVM in a cross-sectional manner. There have also been many studies reporting reduction in LVM in normotensive and hypertensive individual with weight loss interventions, so it is reasonable to hypothesize that if LVM falls when body mass falls, it should increase when people gain weight [17]. However, there are few reports in which body mass and LVM changes have been tracked together over some years to gather longitudinal data on the relationship with time. Age, sex, BMI, systolic blood pressure, antihypertensive treatment, smoking, and diabetes correlated with LVM during follow-up of the Framingham offspring study [18].

In a useful contribution to the subject, Monsuez et al.[19] report a study of 344 individuals with two-dimensional echocardiography in the present issue. This was a convenience sample of healthy individuals who participated in a vitamin supplementation intervention trial. Of these, 64 had recordings that were unsuitable for quantitative analysis, an inherent sampling issue with studies of this kind. Over 6 years, BMI increased by 0.6 kg/m2 and waist circumference by an average of 2.3 cm. These small changes in body size were associated with equally small increases in LVM by 2.3 and 0.4 g/m2.7 for a 1 kg/m2 increase in BMI or a 1 cm increase in waist circumference, respectively. From a clinical perspective, the question arises as to whether these changes in LVM are important as a case can be made that most of the prognostic implications after 6 years are merely the result of ageing of a middle-aged cohort and the incremental information provided by a small increase of LVM over this period is minimal.

Ageing itself is associated with progressive increase in LVM. However, it is also associated with stiffening of the ventricle and in this regard, it is of note that early transmitral flow velocity, E did not change during the course of the observation period. On the other hand, atrial size did increase in concert with the weight changes, so perhaps other more precise measures of diastolic filling than E might have revealed progressive stiffening of the ventricle, increasing the left atrial afterload.

The other conclusion, favoured by the authors, is that the association of increased body weight with time with increased LVM emphasizes the potential benefit of a healthy diet and lifestyle to maintain body weight and, in turn, cardiac geometry and function in ageing adults. This assumes that the increase in LVM accounted for by increased body mass is as important prognostically as any other contributor to LVH in hypertensive patients and not a surrogate for ageing itself and the usual increase in blood pressure that occurs with age in the population of a developed country. One finding from the study by Monsuez et al.[19] that suggests it was not increased body fat per se that directly caused increase in LVM was that total body fat was related to LVM at baseline but did not track with it during follow-up.

In conclusion, body size and LVM are related at any age and an increase in the former is associated with a small increase in the latter. The implications of this for outcome are presently unclear, but the general conclusion that weight gain during middle age is to be avoided is valid, not only because of a possible influence on outlook due to increased blood pressure and LVM, but because obesity, which has multiple adverse consequences, is looming as the major public health problem in developed and developing countries of the present century.

Back to Top | Article Outline


The study received grant support from National Health and Medical Research Council of Australia and author would like to thank Paul Korner for providing a version of Fig. 1, redrawn using data from the following reference: Paul Korner, Essential Hypertension and its Causes, Oxford University Press 2007.

Back to Top | Article Outline


1 Laufer E, Jennings GL, Korner PI, Dewar E. Prevalence of cardiac structural and functional abnormalities in untreated primary hypertension. Hypertension 1989; 13:151–162.
2 Whalley GA, Gamble GD, Doughty RN, Culpan A, Plank L, MacMahon S, Sharpe N. Left ventricular mass correlates with fat-free mass but not fat mass in adults. J Hypertens 1999; 17:569–574.
3 Savage DD, Levy D, Dannenberg AL, Garrison RJ, Castelli WP. Association of echocardiographic left ventricular mass with body size, blood pressure and physical activity (the Framingham Study). Am J Cardiol 1990; 65:371–376.
4 Jennings G, Dart A, Meredith I, Korner P, Laufer E, Dewar E. Effects of exercise and other nonpharmacological measures on blood pressure and cardiac hypertrophy. J Cardiovasc Pharmacol 1991; 17(Suppl 2):S70–S74.
5 de Simone G, Palmieri V, Koren MJ, Mensah GA, Roman MJ, Devereux RB. Prognostic implications of the compensatory nature of left ventricular mass in arterial hypertension. J Hypertens 2001; 19:119–125.
6 McMullen JR, Jennings GL. Differences between pathological and physiological cardiac hypertrophy: novel therapeutic strategies to treat heart failure. Clin Exp Pharmacol Physiol 2007; 34:255–262.
7 Devereux RB, Reichek N. Echocardiographic determination of left ventricular mass in man. Anatomic validation of the method. Circulation 1977; 55:613–618.
8 Devereux RB, Alonso DR, Lutas EM, Gottlieb GJ, Campo E, Sachs I, Reichek N. Echocardiographic assessment of left ventricular hypertrophy: comparison to necropsy findings. Am J Cardiol 1986; 57:450–458.
9 Korner PI, Jennings GL. Assessment of prevalence of left ventricular hypertrophy in hypertension. J Hypertens 1998; 16:715–723.
10 Huxley J. Problems of relative growth. London: Methuen; 1932.
11 Stahl W. Organ weights in primates and other mammals. Science 1965; 150:1039–1042.
12 de Simone G, Devereux RB, Maggioni AP, Gorini M, de Divitiis O, Verdecchia P. Different normalizations for body size and population attributable risk of left ventricular hypertrophy: the MAVI study. Am J Hypertens 2005; 18:1288–1293.
13 Chirinos JA, Segers P, De Buyzere ML, Kronmal RA, Raja MW, De Bacquer D, et al. Left ventricular mass: allometric scaling, normative values, effect of obesity, and prognostic performance. Hypertension 2010; 56:91–98.
14 Zhang J, Merkle H, Hendrich K, Garwood M, From AH, Ugurbil K, Bache RJ. Bioenergetic abnormalities associated with severe left ventricular hypertrophy. J Clin Invest 1993; 92:993–1003.
15 Devereux RB, Bella J, Boman K, Gerdts E, Nieminen MS, Rokkedal J, et al. Echocardiographic left ventricular geometry in hypertensive patients with electrocardiographic left ventricular hypertrophy: the LIFE study. Blood Press 2001; 10:74–82.
16 Bluemke DA, Kronmal RA, Lima JAC, Liu K, Olson J, Burke GL, et al. The relationship of left ventricular mass and geometry to incident cardiovascular events: the MESA (Multi-Ethnic Study of Atherosclerosis) study. J Am Coll Cardiol 2008; 52:2148–2155.
17 Reid C, Dart AM, Dewar EM, Jennings GL. Interactions between the effects of exercise and weight loss on risk factors, cardiovascular haemodynamics and left ventricular structure in overweight subjects. J Hypertens 1994; 12:291–301.
18 Lieb W, Xanthakis V, Sullivan LM, Aragam J, Pencina MJ, Larson MG, et al. Longitudinal tracking of left ventricular mass over the adult life course: clinical correlates of short- and long-term change in the Framingham offspring study. Circulation 2009; 119:3085–3092.
19 Monsuez JJ, Kesse-Guyot E, Fezeu L, Blacher J, Galan P, Sebbane G. Impact of 6-year body weight change on cardiac geometry and function in ageing adults: the SU.VI.MAX 2 cardiovascular ultrasound substudy. J Hypertens 2010; 28:2309–2315.
© 2010 Lippincott Williams & Wilkins, Inc.