Insulin resistance (IR) and inflammation are associated with an increased risk of cardiovascular disease and may contribute to obesity cardiomyopathy. The earliest sign of obesity cardiomyopathy is impaired left ventricular (LV) diastolic function, which may be evident in obese children and adolescents. However, the precise metabolic basis of the impaired LV diastolic function remains unknown. The aims of this study were to evaluate cardiac structure and LV diastolic function by tissue Doppler imaging in overweight and obese (OW) youth and to assess the relative individual contributions of adiposity, IR, and inflammation to alterations in cardiac structure and function. We studied 35 OW (body mass index standard deviation score 2.0 ± 0.8; non-IR n = 19, IR n = 16) and 34 non-OW youth (body mass index standard deviation score 0.1 ± 0.7). LV diastolic function was reduced in OW youth compared with non-OW controls, as indicated by lower peak myocardial relaxation velocities (p <0.001) and greater filling pressures (p <0.001). OW youth also had greater LV mass index (p <0.001), left atrial volume index, and LV interventricular septal thickness (LV-IVS; both p = 0.02). IR-OW youth had the highest LV filling pressures, LV-IVS, and relative wall thickness (all p <0.05). Homeostasis model of assessment–insulin resistance and C-reactive protein were negative determinants of peak myocardial relaxation velocity and positive predictors of filling pressure. Adiponectin was a negative determinant of LV-IVS, independent of obesity. In conclusion, OW youth with IR and inflammation are more likely to have adverse changes to cardiovascular structure and function which may predispose to premature cardiovascular disease in adulthood.
Obesity is one of the main factors contributing to the heart failure epidemic that is predicted to increase by 25% by 2030. The earliest manifestation of obesity cardiomyopathy and heart failure is impaired diastolic function which is characterized by a reduced peak myocardial relaxation velocity (septal e′) and elevated left ventricular (LV) filling pressure (E/e′). Studies in obese adolescents and children have shown a relative reduction in diastolic function and altered cardiac geometry. However, the contribution of early reversible metabolic changes such as insulin resistance (IR) and inflammation has not been fully examined in this population. Excess white adipose tissue secretes cytokines and hormones, termed adipokines which promote chronic inflammation, IR, and cardiovascular remodeling. IR has been associated with increased cardiac mass, fibrosis, and mitochondrial dysfunction which attenuate diastolic function. The adverse effects of IR and obesity are further supported by the improvements in diastolic function observed after reductions in weight and IR through lifestyle interventions. The aim of our study was to determine whether obese adolescents had reduced diastolic function and cardiovascular remodeling and whether IR, inflammation, and adipokines were associated with these cardiac changes.
From March 2010 to January 2013, we prospectively recruited 35 children and adolescents who were overweight and obese (OW) and 34 non-OW control children, as defined by the International Obesity Task Force reference ranges for children. The age range of the participants was from 10 to 19 years, and for the purposes of this study, this age group will be referred to as youth. OW youth were recruited through advertisements or by referral from the Mater Children’s Hospital obesity and endocrine outpatient clinics. Normal weight control youth were recruited as a sample of convenience and were often associated with the project, through the participants or researchers. IR was determined by calculating the homeostasis model of assessment–insulin resistance [ HOMA − IR = ( fasting insulin in U / mL × fasting glucose in mmol / L ) / 22.5 ] . OW participants were then subdivided into those who were IR as defined by a HOMA-IR ≥3.0 (n = 16) and those who were not IR (NIR; n = 19). Exclusion criteria included type 1 diabetes mellitus, congenital heart disease, cardiovascular conditions, infection, use of medications that affect insulin secretion and action such as glucocorticoids and β blockers, Cushing disease, hypothyroidism, and pregnancy. Participants gave written informed assent, and parents provided written informed consent before participating in the study. This study was conducted in accordance with the principles of the Declaration of Helsinki, and all methods were approved by the Mater Children’s Hospital and Children’s Health Queensland institutional human research ethics committee before commencement of the study.
Anthropometric measurements included height, weight, body mass index (BMI), and waist circumference (WC). The ratio of waist circumference to height (WC/Ht) was used as a surrogate measure of visceral adiposity. Pubertal status was assessed according to Tanner staging by a pediatric endocrinologist. Cardiovascular measurements included blood pressure and resting heart rate which were measured and recorded as an average of 3 measurements after the participant rested in the supine position for 15 minutes as per recommendations of the Fourth Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure.
Glucose, insulin, cholesterol, triglycerides, high-density lipoprotein, low-density lipoprotein (LDL), C-reactive protein (CRP), and adipocytokines were obtained after fasting overnight for 10 to 12 hours. Glucose was analyzed on a Vitros 5600 system using an in vitro glucose oxidase method (Ortho-Clinical Diagnostics, Rochester, New York). Insulin was assayed using a chemiluminescent microparticle immunoassay performed on the Abbott Architect i2000SR (Gibbco Scientific, Inc. Coon Rapids, Minnesota) using an immunoenzymometric assay. CRP was measured by immunonephelometry using the Siemens Cardiophase CRP reagent and measured on a Siemens BNII instrument. The sensitivity of this assay was 0.2 mg/L. Intra-assay and interassay coefficients of variation for all immunoassays were <9%. Serum adiponectin was measured using a Procarta kit (Panomics, Fremont, California) was measured using a high-sensitivity cytokine Milliplex kit (Millipore, Billerica, Massachusetts) as per the manufacturer’s recommendations. Serum leptin was measured using an in-house multiplex assay and measured on the Luminex 100 platform. Antibodies to leptin (R&D Systems Inc., Minneapolis, Minnesota) were coupled to xMAP beads and run in a standard multiplex reaction according to Luminex protocols (Luminex, Austin, Texas) as described previously.
Transthoracic echocardiography was performed as per American Society of Echocardiography guidelines. Left atrial (LA) and LV volume, LV end-diastolic and end-systolic diameter, and septal and LV posterior wall thicknesses in diastole and systole were obtained using standard M-mode recordings and 2-dimensional (2D) imaging. The LV ejection fraction was derived using modified biplane Simpson’s method from the apical 2- and 4-chamber views. LV mass was determined using the method described by Devereux, and LV mass index (LVMI) was calculated by dividing LV mass by height to the power 2.7, which has been validated for pediatric obese participants. Measurements were obtained in end-systole from the frame preceding mitral valve opening, and the LA volume was also indexed for height to the power 2.7 to maintain consistency with LVMI. Conventional Doppler tracings of the mitral inflow were obtained from an apical 4-chamber view. Tissue Doppler velocities of the septal mitral annulus were obtained from the apical 4-chamber view by placing the sample volume at the medial mitral annulus. The ratio of early mitral flow velocity (E) to early diastolic velocity of the mitral annulus (e′) was calculated (E/e′). All Doppler signals were recorded at a speed of 50 mm/s. For each parameter, the average of 3 cycles was used. All the measurements were made offline by the same observer blinded to the patients’ clinical details.
Cardiac geometry was defined according to cutoffs for relative wall thickness (RWT) >0.43, a measure of LV wall thickness relative to LV cavity size. For LV hypertrophy in youth, LVMI (ht 2.7 ) of >40 g/m 2.7 for girls and >45 g/m 2.7 for boys was used as this corresponds to the ninety-fifth percentile in children and adolescents. Concentric remodeling was defined by an RWT >0.43 but normal LVMI. Eccentric hypertrophy was demonstrated with RWT ≤0.43 and LV hypertrophy; and concentric hypertrophy was defined by an RWT >0.43 and LV hypertrophy. Concentric remodeling is the most benign change in cardiac geometry. Eccentric and concentric hypertrophies are both associated with increased cardiovascular morbidity and mortality in adults.
Data were summarized using means and SDs if they had a parametric distribution or median and interquartile range if they displayed a nonparametric distribution. Gender and pubertal stage were compared with a chi-square test. Means of non-OW, OW, NIR-OW, and IR-OW participants were compared using a one-way analysis of variance test for parametric data, whereas medians and interquartile ranges were analyzed using a Mann–Whitney Test for nonparametric data. The Fisher’s exact test was used to compare the frequency of the different classes of cardiac remodeling. To investigate the relative influence of obesity and metabolic factors on cardiac structure, Spearman correlations were performed. BMI standard deviation score (BMI-SDS), WC, WC/Ht, age, gender, BP, HOMA-IR, CRP, adiponectin, and leptin were a priori considered as possible predictor variables. Univariate regression models were constructed to examine the associations of predictor variables with the dependent echocardiographic variables. Stepwise multiple regression analysis was then used to determine which predictor variables independently explained a significant fraction of the variance of the dependent variables. Nonparametric values were analyzed as log-transformed values. Residual analysis was performed to check the validity of model assumptions, and data were adjusted for BMI-SDS, HOMA-IR, or SBP. STATA version 12.1 was used for all analyses. A p value of <0.05 was considered statistically significant.
Clinical characteristics are listed in Table 1 . There were no significant intergroup differences for age, gender, or pubertal status. The OW adolescents had greater systolic blood pressures than controls but similar diastolic blood pressures. As expected, all measurements of adiposity, including BMI-SDS, WC, and WC/ht, were significantly greater in obese compared with lean adolescents. IR-OW youth had greater WCs than their NIR-OW counterparts.
|Variables||Non-overweight||Overweight||p value||Non-insulin resistant||Insulin resistant||p value|
|Male||21 (61%)||14 (40%)||0.08||8 (42%)||6 (37.5%)||0.79|
|Systolic Blood Pressure (mm Hg) †||114±11||120±10||0.02||119+10||120+10||1.0|
|Diastolic Blood Pressure (mm Hg)||66±9||69±11||0.1||70±11||68±12||1.0|
|Weight (kg) ∗ †||60.8 (10.6)||91.3 (21.2)||<0.001||80.9 (35.3)||96.5 (15.3)||0.07|
|Body Mass Index||0.03 (0.7)||2.0 (0.8)||<0.001||1.89 (0.7)||2.25 (0.6)||0.11|
|Standard Deviation Score ∗ †|
|Waist Circumference (cm) ∗ †||72 (10)||103 (17)||<0.001||98 (18)||110 (10)||0.03|
|Waist Circumference/Height ∗ †||0.4 (0.04)||0.6 (0.07)||<0.001||0.59 (0.07)||0.65 (0.04)||0.17|
|Homeostatic model assessment of insulin †||1.7 (0.6)||3.0 (2.4)||<0.001||2.0 (0.6)||4.3 (1.2)||<0.001|
|Triglycerides (mmol/L) ∗ †||0.8 (0.3)||1.2 (0.5)||<0.001||1.1 (0.5)||1.3 (0.5)||0.32|
|Cholesterol (mmol/L) ∗||4.0±0.6||4.8±1.0||<0.001||5.1±0.9||4.6±1.0||0.35|
|High density lipoprotein (mmol/L)||1.3 (0.4)||1.1 (0.2)||0.02||1.2 (0.3)||1.1 (0.1)||0.08|
|Low density lipoprotein (mmol/L) ∗||2.4±0.6||3.2±1.0||0.001||3.5±1.0||3.0±1.0||0.52|
|High sensitivity C-reactive protein (mg/L) †||0.3 (0.3)||1.3 (3.2)||<0.001||1.3 (3.1)||1.7 (3.4)||0.96|
|Leptin (ng/mL) ∗ †||2.5 (5.8)||19.5 (27.5)||<0.001||18.2 (18.9)||24.3 (40.3)||0.15|
|Adiponectin (μg/mL) ∗ †||11.8 (5.8)||5.3 (5.7)||<0.001||7.2 (5.8)||3.2 (3.9)||0.04|
In the metabolic profile ( Table 1 ), compared with non-OW controls, OW youth had greater fasting HOMA-IR, triglycerides, cholesterol, and LDL and lower high-density lipoprotein. OW adolescents had elevated levels of CRP and leptin and decreased adiponectin levels. Adiponectin was further reduced in IR-OW compared with non–IR-OW youth.
With regard to measurements of cardiac structure and LV diastolic function ( Table 2 ), results from non-OW controls were consistent with historic controls. None of the participants met the criteria for diastolic dysfunction. However, LV diastolic function was significantly reduced in OW compared with non-OW youth with lower septal e′, a measure of myocardial relaxation, and higher E/e′ ( Figure 1 ), an echocardiographic estimate of LV filling pressure. Septal e′ was significantly greater in the controls compared to the NIR-OW and IR-OW subgroups ( Figure 1 ). Systolic function was normal with no significant difference in the ejection fraction between groups. Structural differences were greater in the OW youth who displayed a higher LA volume index, LV-IVS, and LVMI than the non-OW controls. Furthermore, IR-OW youth had increased mitral A, E/e′, LV-IVS ( Figure 1 ) and RWT and LV-PW in contrast to NIR-OW and non-OW participants. With regard to cardiac geometry, the frequency of concentric remodeling did not differ between the groups. Eccentric hypertrophy was not observed in non-OW controls but was present in 11% of OW youth. Concentric hypertrophy occurred only in the IR-OW youth ( Table 2 ).
|Variable||Non-overweight||Overweight||p value||Non-insulin resistant||Insulin resistant||p value|
|Mitral A (m/s) †||0.37 (0.12)||0.40 (0.21)||0.2||0.35 (0.2)||0.50 (0.17)||0.01|
|Septal e’ (m/s)||0.15±0.02||0.12±0.02||<0.001||0.13±0.02||0.11±0.02||0.38|
|Ejection Fraction (%)||65 (2)||66.5 (2)||0.16||67 (2)||65 (2)||0.65|
|Left atrial volume index (mL/m 2.7 ) ∗||12.5±2.4||13.5±3.5||0.02||13.7±3.3||13.2±3.9||0.9|
|Interventricular septal wall||0.81±0.11||0.89±0.14||0.02||0.83+0.11||0.95+0.14||0.02|
|Relative wall thickness †||0.35±0.05||0.37±0.06||0.1||0.35±0.05||0.40±0.07||0.02|
|Left ventricular mass index (g/m 2.7 ) ∗ †||28.4±4.9||36.6±5.8||<0.001||36.0±6.0||37.4±5.6||1.0|
|Concentric Remodeling||2 (6%)||4 (11%)||0.08||1 (5%)||3 (19%)||0.40|
|Eccentric Hypertrophy||0 (0%)||4 (11%)||0.04||3 (16%)||1 (6%)||0.04|
|Concentric Hypertrophy †||0 (0%)||2 (6%)||0.16||0 (0%)||2 (13%)||0.05|