Article Text

Download PDFPDF

Original article
Ventricular hypertrophy and cavity dilatation in relation to body mass index in women with uncomplicated obesity
  1. Oliver J Rider,
  2. Steffen E Petersen,
  3. Jane M Francis,
  4. Mohammed K Ali,
  5. Lucy E Hudsmith,
  6. Monique R Robinson,
  7. Kieran Clarke,
  8. Stefan Neubauer
  1. Department of Cardiovascular Medicine, University of Oxford Centre for Clinical Magnetic Resonance Research, University of Oxford, Oxford, UK
  1. Correspondence to Professor Stefan Neubauer, Department of Cardiovascular Medicine, University of Oxford Centre for Clinical Magnetic Resonance Research, John Radcliffe Hospital, Oxford OX3 9DU, UK; stefan.neubauer{at}


Objective The traditionally accepted mechanism for ventricular adaptation to obesity suggests that cavity dilatation in response to increased blood volume and elevated filling pressure results in ventricular hypertrophy as a compensatory mechanism. Our hypothesis was that, instead, initiation of ventricular hypertrophy in obesity may be explained by changes in hormonal milieu and not by cavity dilatation.

Research design and methods 88 female subjects without identifiable cardiovascular risk factors, covering a wide range of body mass indices (BMI), from normal (21.2±1.6 kg/m2) to severely obese (45.0±4.6 kg/m2), underwent cardiovascular MRI to determine left ventricular (LV) and right ventricular (RV) mass and volumes.

Results BMI correlated positively with LV and RV mass and end-diastolic volumes (EDV). However overweight is associated with a significant LV and RV hypertrophy (LV: 78±11 g vs 103±16 g, p<0.01; RV: 26±7 g vs 40±11 g, p<0.01) was observed in the absence of differences in LV and RV volumes (LV: EDV 119±15 vs 121±21 ml, p>0.99, RV: 131±17 vs 130±24 ml; p>0.99). Furthermore, significant increases of serum leptin occurred at this pre-obese stage (15.6±19 vs 36.5±22 ng/ml; p=0.013).

Conclusion In a cohort of healthy female subjects with a wide range of BMIs, ventricular hypertrophy occurs without associated cavity dilatation in overweight individuals, while in manifest obesity, both cavity dilatation and ventricular hypertrophy occur. Elevated leptin levels may have a role in this effect on ventricular mass.

  • Left ventricular hypertrophy
  • obesity

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Given the rising global epidemic of obesity, it is likely that adverse health consequences of excess adiposity will escalate in the future. In this context, several investigators have described the adverse effects of obesity on the heart.1 2 Obesity and excess adiposity have been linked to a spectrum of cardiovascular changes from a hyperdynamic circulation, through subclinical changes in cardiac structure to manifest heart failure.3

Excess adiposity imposes an increased metabolic demand on the body, and thus, both cardiac output and total blood volume are elevated in obesity. This hyperdynamic circulation causes left and right ventricular structural changes in obesity, which lead to elevated ventricular mass and cavity dilatation. Traditionally, it has been thought that these volumetric changes in the circulation initiate the sequence of events that lead to this adaptive cardiac response.1 2 4 In this model, the hypertrophic response of the ventricle is secondary to increased wall stress imposed by cavity dilatation and increased filling pressures.5 6 Although this model accounts for the eccentric hypertrophic pattern, which is most commonly seen in obesity, it is unknown whether the hypertrophic response can occur separately from cavity dilatation.

Advances in our understanding of hormonal changes in obesity may, however, suggest an alternative scenario, whereby the hypertrophic response can occur independently from the dilatory response. Increased adiposity causes higher serum leptin concentrations, and leptin has been linked to ventricular hypertrophy in both animals and humans.7–10 Thus, it is possible that with modest increases in weight, serum leptin concentrations could rise above the threshold needed to initiate left ventricular hypertrophy, while the associated increases in blood volume remain too small to cause cavity dilatation.

Therefore, we hypothesised that (1) in the setting of overweight ventricular hypertrophy can occur independently from cavity dilatation and (2) as a result of increases in serum leptin concentration, left and right ventricular hypertrophy would occur at lower levels of obesity than at those leading to ventricular cavity dilatation. In order to investigate this, we assessed left and right ventricular mass and volumes and serum leptin concentrations in a female cohort with a wide range of body mass index measures.

As the assessment of left and right ventricular geometry with echocardiography in obesity is often difficult, owing to the presence of excess subcutaneous fat limiting acoustic windows, we employed cardiovascular magnetic resonance (CMR), which yields accurate and highly reproducible information on ventricular mass and volumes in subjects with differing amounts of adipose tissue.11


Ethics and study cohort

The study was approved by the local ethics committee, and informed written consent was obtained from each patient.

Eighty-eight healthy female subjects were included into the study. All subjects were screened for the presence of identifiable cardiac risk factors and excluded if they had a history of cardiovascular disease, hypertension, diabetes, current smoking, or use of cardiac medications. All subjects had a normal 12-lead electrocardiogram and were normotensive at the time of scanning (averaged over three supine measures taken within 10 minutes) with no historical evidence of hypertension. Subjects were excluded if they had either an elevated fasting glucose level (≥6.7 mmol) or a fasting total cholesterol level ≥6.5 mmol, a history of coronary artery disease (CAD), or of cardiac chest pain or valvular heart disease. All patients were excluded if they had clinical or historical evidence of obstructive sleep apnoea. All subjects were limited to three 30-minute sessions of sweat producing exercise per week, and excluded if physical activity exceeded this level. The cohort in this crosssectional study was separated into quartiles according to body mass index; (quartile I; BMI 21.2±1.6; n=22, quartile II; BMI 28.362.2; n=22, quartile III; BMI 34.762.0; n=22 and quartile III; BMI 45.064.7; n=22).

Blood tests

Blood tests for glucose, insulin, cholesterol and leptin were taken on the day of the scan. Patients were asked to be fasted for at least 8 hours. Leptin and insulin samples were analysed using a commercially available ELISA kit (AssayMax Human Leptin ELISA Kit, Assaypro, Mercodia Insulin ELISA Kit, Mercodia). An estimate of insulin resistance was calculated using the HOMA-IR equation (fasting insulin (μU/ml)×fasting glucose (mmol/l)/22.5).12 13

MR imaging

All MR scans for the assessment of myocardial mass, volumes and ejection fraction were performed on a 1.5 Tesla MR system (Siemens Medical Solutions, Erlangen, Germany). All imaging was prospectively cardiac gated with a precordial four-lead ECG and acquired during end expiration breath-hold. Images were acquired using a steady-state free precession (SSFP) sequence with an echo time (TE) of 1.5 ms, a repetition time (TR) of 3.0 ms, temporal resolution 47.84 ms and a flip angle of 60° as previously described.14–16 SSFP cine sequences were used to acquire localisation images followed by an SSFP left and right ventricular short axis stack of contiguous images with a slice thickness of 7 mm and an interslice gap of 3 mm.

Data analysis

Image analysis for left and right ventricular volumes and mass was performed using Siemens analytical software (Argus). The short axis stack was analysed manually contouring the endocardial borders from base to apex at end-diastole and end-systole. The epicardial border was contoured at end-diastole to yield myocardial mass. Left and right ventricular mass (g) was calculated as the epicardial volume minus the endocardial volume multiplied by 1.05 (specific gravity of myocardium).

Statistical analysis

All statistics were analysed using a commercial software package (SPSS 15; SPSS). All results were separated by body mass index measures into quartiles and presented as the mean±SD. All data were subjected to Kolmogorov-Smirnov tests to establish normal distribution of the data. All data were compared using a one-way ANOVA technique with post-hoc Bonferroni correction. Any differences were considered significant at p<0.05. Leptin and insulin levels are commonly log transformed in the literature because of skewed distribution. We present data that are not log transformed as values were not statistically skewed. When log transformed, results were in a similar pattern (data not shown).


Patient characteristics

All patient groups were well matched for age and height with no significant differences between the quartiles (table 1). There were no significant differences in fasting glucose, cholesterol (figure 1), systolic blood pressure or diastolic blood pressure between the body mass index quartiles (table 1).

Table 1

Anthropometric, left ventricular (LV) and right ventricular (RV) mass and volumes for the four body mass index (kg/m2) quartiles

Figure 1

Scatterplot of fasting blood samples for the four body mass index (kg/m2) quartiles.

Ventricular characteristics

Left and right ventricular mass

Left and right ventricular mass, both in absolute terms and when indexed to height (table 1 and figure 2), increased with increasing body mass index (p<0.001). When comparing the lowest quartile (average body mass index 21.2±1.6 g/m2) to the second quartile (average body mass index 28.3±2.3 g/m2), there was a 24% greater absolute left ventricular mass (78±11 vs 103±16 g; p=0.02) and a 44% larger absolute right ventricular mass (26±7 vs 37±11 g; p=0.001). When indexed to height, height2.7 and body surface area both left and right ventricular mass measures remained statistically greater than for the lowest quartile (table 1). Left ventricular mass increased further, by 40% and 60% (compared to normal) for the third and fourth quartile. Right ventricular mass was also seen to be larger (by 85% and 127%) for the third and fourth quartile when compared to the lowest quartile.

Figure 2

Left and right ventricular end-diastolic volume index (ml/m) and mass index (g/m) for the four body mass index (kg/m2) quartiles. (Open bars, quartile I, vertical stripes, quartile II, horizontal stripes, quartile III, solid bars, quartile IV.)

Left and right ventricular volumes

Despite the greater left and right ventricular masses observed for the second overweight quartile compared to the normal weight quartile, there was no corresponding increase in left ventricular end-diastolic volume, both in absolute terms and when indexed to height, which was found to be similar between the lowest two quartiles (LV-119±15 vs 121±21 ml; p <0.99, RV-131±17 vs 130±24 ml; p<0.99, table 1 and figure 2). However, when the lowest quartile (BMI 21.2±1.6) was compared to the third quartile (average BMI 34.7±2.0), there were significant differences in right and left end-diastolic volumes (LV p=0.013, RV p=0.047). End-diastolic volumes were also greater in the severely obese upper quartile individuals (average body mass index 45.0±4.7) compared to the lowest quartile (figures 2 and 3). Left and right ventricular end-systolic volume was not significantly different between the lowest two quartiles but was found to be significantly larger in the two upper quartiles.

Left and right ventricular mass:volume ratio

When comparing quartile I with quartile II there is a significant 34% increase in LV mass:volume ratio. When comparing quartile II with quartile III there was no significant difference in LV mass:volume ratio (table 1). When comparing quartile I with quartile II there is a significant 47% increase in RV mass: volume ratio. When comparing quartile II with quartile III there was no significant difference in RV mass:volume ratio (table 1).

Left and right ventricular function

Overall, left and right stroke volumes were seen to increase with increasing body mass index (LV p=0.01, RV p=0.005). With a body mass index increase from normal to overweight there was no significant difference in LV stroke volume (p>0.99), but stroke volume was significantly larger for the higher levels of obesity (by 16% and 22% for the third and fourth quartiles respectively, p=0.02). In contrast, there was no significant difference in left or right ventricular ejection fraction (%) among any of the quartiles (figure 2, table 1).

Serum leptin

Serum leptin level was seen to correlate positively with increasing body mass index (p=0.02, figure 1). Importantly, a significantly higher serum leptin was observed in quartile II when compared to quartile I (15.6±19 ng/ml to 36.5±22 ng/ml), with further increases in obese (80.7±52 ng/ml) and severely obese individuals (155±72 ng/ml, figure 1).

Serum insulin

Serum insulin level was seen to correlate positively with increasing body mass index (p=0.02, figure 1). Importantly, however, in contrast to serum leptin, there was no significant difference in serum insulin levels between the lowest quartile and quartiles II and III (3.3±1.8 μU/ml vs 4.4±3.0 μU/ml; p<0.99 and p=3.3 μU/ml±1.8 vs 6.6±3.5 μU/ml; p=0.25, respectively). However, the highest body mass index quartile, there was a significant increase in serum insulin concentration (3.3±1.8 μU/ml vs 11.2±6 μU/ml; p<0.01, figure 1).

Homeostasis model assessment of insulin resistance (HOMA-IR)

HOMA-IR was seen to correlate positively with increasing body mass index measures, with all four quartiles remaining within the insulin sensitive range (figure 1). There was no significant difference in HOMA-IR between the lowest two quartiles (0.7±0.4 vs 0.99±0.6; p>0.99). The highest quartile however was seen to have a significantly higher HOMA-IR (0.7±0.4 to 2.58±1.6; p 0.001, figure 1).

Associations of left and right ventricular mass

In order to further separate the effects of overweight and obesity regression analysis was undertaken between quartiles I and II and also between quartiles II and III. On linear analysis between quartile I and II serum leptin concentration was related to left ventricular mass (R2 0.44, p<0.001), whereas the relation between left ventricular end-diastolic volume and left ventricular mass was not statistically significant (R2 0.08, p=0.06). On linear regression analysis between quartiles II and III both serum leptin and left ventricular end-diastolic volume were related to left ventricular mass (leptin R2 0.14, p=0.04, LV-EDV R2 0.32, p=0.01).

On linear regression analysis between quartiles I and II right ventricular mass was not related to either serum leptin concentration or right ventricular end-diastolic volume (leptin R2 0.12, p=0.07, RV-EDV; R2 0.00, p>0.99). On linear regression analysis between quartiles II and III both serum leptin and end-diastolic volume were related to right ventricular mass (leptin R2 0.18, p<0.01, RV-EDV; R2 0.48, p<0.001).


In this study, cardiovascular magnetic resonance was used to investigate the adaptive response of the left and right ventricle to increasing body mass index (figure 3). We have focused on the differing morphology of the left and right ventricle in overweight and compared this to that seen in obesity. This has allowed the separation of the effects of the volumetric changes which are less marked in the overweight cohort, but present in the obese quartile, with those of the adipocytokines which are present in both the overweight and obese quartiles.

Figure 3

Basal left ventricular short-axis images at the same level of magnitude from each of the four quartiles (A) quartile I, (B) quartile II, (C) quartile III and (D) quartile IV showing the early increase in left ventricular mass preceding the changes in end-diastolic volume.

Our results clearly demonstrate that in overweight individuals a significant biventricular hypertrophic response occurs without associated ventricular cavity dilatation. In addition to this the left and right ventricular mass:volume ratio is significantly higher in the overweight quartile II than in the normal weight quartile I, but is similar when comparing the overweight quartile II and the obese quartile III. This again points to the fact that in the overweight quartile II there is an excess hypertrophic response, not associated with end-diastolic volume changes, that is not seen in the obese quartile III, suggesting that the ventricular hypertrophic response is more sensitive to small changes in body mass index than the end-diastolic volume and can occur independently of ventricular dilatation.

This finding is in direct opposition to the generally accepted hypothesis that hypertrophy in obesity occurs as a response to increased wall stress imposed by cavity dilatation. If ventricular hypertrophy can occur independently from cavity dilatation then this raises the question which mechanism may be responsible for this.

This would have to be an obesity-induced process leading to ventricular hypertrophy without causing ventricular dilatation. In our view, by far the most likely explanation for this is the change in hormonal milieu that occurs early in obesity.

Hyperinsulinaemia, as a result of insulin resistance, is a potential candidate for the ventricular hypertrophic response seen in overweight population. Hyperinsulinaemia has been linked to ventricular hypertrophy in obesity directly via the binding of insulin to myocardial insulin-like growth factor 1 receptors which are found in abundance in the myocardium.17 However, this study has shown that in insulin sensitive overweight subjects (quartile II) in whom insulin levels, HOMA-IR and blood pressure are not significantly different from normal weight individuals a significant hypertrophic response is seen when compared to the lowest quartile. This makes hyperinsulinaemia an unlikely explanation for the hypertrophy seen in this study. However, with greater degrees of obesity, where insulin levels are seen to be significantly higher, the direct effects of insulin may well contribute to ventricular hypertrophy.

In our view the most likely candidate for the hypertrophic changes seen in the overweight patients in quartile II is the fat-derived hormone leptin. Leptin receptors were initially thought to be limited to adipose tissue,18 but recently leptin receptor isoforms have been shown to be expressed in myocardium,19 suggesting that leptin has specific myocardial effects. The most consistent finding is that leptin induces myocardial cell hypertrophy in culture,8 9 20 suggesting that leptin may be, at least in part, responsible for ventricular hypertrophy in humans. Hyperleptinaemia has also been linked to left ventricular hypertrophy in severe obesity in humans.7 The direct molecular mechanism whereby leptin induces cardiomyocyte hypertrophy is unclear, but there is evidence to suggest that, in the absence of wall stress, leptin induces eccentric ventricular hypertrophy by inducing myocardial cell elongation in vitro.10 The physiological effects of leptin are mediated via membrane-bound receptors. Recent studies have shown that the hypertrophic effects of leptin involve several signalling cascades including JAK/STAT, MAPK, protein kinase C and Rho/ROCK-dependent kinases.21–23

In line with this hypothesis, our study also showed that in the second quartile of overweight patients in whom ventricular hypertrophy occurred without cavity dilatation (when compared to the lowest quartile), there was an associated significant increase in serum leptin concentration by 130%. Thus, it is possible that, when comparing the lowest two quartiles, leptin concentrations were sufficiently higher in quartile II in comparison to quartile I, to stimulate a hypertrophic response, but blood volume increases were not sufficient to cause left ventricular volumetric changes. In addition to this, when comparing normal weight quartile I with overweight quartile II, serum leptin concentration is seen to be related to left ventricular mass on linear regression, whereas end-diastolic volume was not related to left ventricular mass. Furthermore, both serum leptin and end-diastolic volume were related to left ventricular mass on linear regression when comparing overweight quartile II and obese quartile III. This points to the fact that volumetric changes may be acting independently from endocrine changes in the adaptive response to the ventricle to adiposity.

Interestingly, our study has shown a greater hypertrophic response in the right ventricle compared to the left. This may potentially be explained by a variation in leptin receptor expression, and in the female rat ventricle there is evidence to suggest that leptin receptors are asymmetrically distributed within the myocardium with right ventricular levels being higher than left ventricular levels.19

Thus, our data suggest that the left and right ventricular adaptive changes to increasing fat mass may occur in two phases, an early, predominantly leptin mediated, hypertrophic response with modest weight gain, and a later mixed endocrine-volumetric response characterised by cavity dilatation and wall stress-induced eccentric hypertrophy.


Our study does not address the sequential changes over time within individual subjects when gaining weight, and a longitudinal follow-up study should address this question. Our work was limited to female subjects, as a similarly sized male cohort was not available to us for comparison. Thus, the effect of gender needs to be investigated.

Conclusion and clinical relevance

The cardiac phenotype of obesity includes elevated left and right ventricular mass and biventricular cavity dilatation. This study provides evidence that the hypertrophic response may occur independently from ventricular dilatation in patients presenting with a wide range of adiposity. With the growing body of literature showing a strong relation between left ventricular hypertrophy and morbidity and all-cause mortality,24–26 understanding the various mechanisms responsible for left ventricular hypertrophy is of great clinical importance. Cardiovascular mortality has been shown to be higher even in overweight pre-obese individuals than normal weight individuals,27 and leptin-induced left and right ventricular hypertrophy may be one potential mechanism for this. However, given that in patients with manifest heart failure, increased body mass index may also be advantageous prognostically,28 this issue is complex and warrants further study.



  • See Editorial, p 171

  • The authors had full access to and take full responsibility for the integrity of the data. All authors have read and agree to the manuscript as written.

  • Funding The study was supported by grants from the Wellcome Trust and the British Heart Foundation, and was supported by the Oxford Partnership Comprehensive Biomedical Research Centre with funding from the Department of Health's NIHR Biomedical Research Centres funding scheme.

  • Competing interests None.

  • Ethics approval This study was conducted with the approval of the Milton Keynes Local Research Ethics Comittee.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Linked Articles