Effect of Obesity on Left Ventricular Diastolic Function in Children
?ocuklarda Obezitenin Sol Ventrik?l Diyastolik Fonksiyonlar? ?zerine Etkisi
?eref Alpsoy1, Ayd?n Aky?z1, Dursun ?ayan Akkoyun1, Bur?in Nalbanto?lu2, Birol Top?u3,
Hasan De?irmenci1, Mustafa Metin Donma2
1 Department of Cardiology, Tekirdag Namik Kemal University Research and Practice Hospital, Tekirdag, Turkey
1 Tekirda? Nam?k Kemal ?niversitesi Ara?t?rma ve Uygulama Hastanesi, Kardiyoloji Klini?i, Tekirda?, T?rkiye
2 Department of Pediatrics, Tekirdag Namik Kemal University Research and Practice Hospital, Tekirdag, Turkey
2 Tekirda? Nam?k Kemal ?niversitesi Ara?t?rma ve Uygulama Hastanesi, ?ocuk Hastal?klar? Klini?i,
3 Department of Statistics, Tekirdag Namik Kemal University Research and Practice Hospital, Tekirdag, Turkey
3 Tekirda? Nam?k Kemal ?niversitesi Ara?t?rma ve Uygulama Hastanesi, ?statistik Anabilim Dal?,
Introduction: We aimed to asses two dimensional, M-Mode, conventional and tissue Doppler echocardiographic parameters of left ventricle and to evaluate the effects of antropometric and biochemical factors on these parameters in obese and normal weight children.
Patients and Methods: This observational cross-sectional study included 60 obese and 60 non-obese children aged 8-15 years. Echocardiographic parameters of left ventricle were measured in all subjects and investigated the relationship with anthropometric and biochemical factors.
Results: Waist circumference, body mass index-Z (BMI-Z) score, systolic and diastolic blood pressure, triglyceride, fasting glucose, insulin concentrations, and homeostatic model assessment (HOMA) index were significantly higher, whereas high density lipoprotein-cholesterol (HDL-C) concentration was significantly lower in the obese than in the non-obese group. Mitral annulus septal E' and E'/A', mitral annulus lateral E' and E'/A', anterior E' and E'/A', inferior E' and E'/A' values were significantly lower, whereas mitral E/septal E' and mitral E/lateral E' values were significantly higher in obese than in non-obese subjects. Most of the tissue Doppler parameters were correlated with waist circumference, BMI-Z score, fasting blood glucose, HOMA, systolic and diastolic blood pressure, serum insulin levels, triglyceride and HDL. Waist circumference and BMI-Z score were decisive for some tissue Doppler parameters.
Conclusion: Left ventricular diastolic functions are impaired in obese children compared to that of their peers. Although there were correlation between diastolic dysfunction parameters and risk factors, obesity itself is related with reduction of left ventricular diastolic functions.
Key Words: Child, pulsed Doppler echocardiography, left ventricle, obesity.
Received: 29.07.2013 • Accepted: 06.08.2013
Ama?: Bu ?al??mada obez ve normal kilolu ?ocuklarda iki boyutlu, M-Mod, geleneksel ve doku Doppler ekokardiyografiyle sol ventrik?l diyastolik fonksiyon parametrelerini ?l?meyi ve bu parametreler ?zerine antropometrik ve biyokimyasal fakt?rlerin etkilerini de?erlendirmeyi ama?lad?k.
Hastalar ve Y?ntem: Bu olgu-kontroll? kesitsel ?al??maya, 8-15 ya? aras? 60 obez ve 60 normal kilolu ?ocuk dahil edildi. ?ocuklar?n antropometrik ve biyokimyasal de?erleriyle sol ventrik?l ekokardiyografik parametreleri ?l??ld? ve aralar?ndaki ili?ki ara?t?r?ld?.
Bulgular: Obez grupta bel ?evresi ve beden kitle indeksi (BK?-Z) skoru, sistolik ve diyastolik kan bas?nc?, trigliserid, a?l?k kan ?ekeri, ins?lin ve homeostatik modeli de?erlendirmesi (HOMD) indeksi anlaml? olarak y?ksekken y?ksek yo?unluklu lipoprotein kolesterol (YYL-K) konsantrasyonu anlaml? derecede d???k bulundu. Mitral annulus septal E' ve E'/A', mitral annulus lateral E' ve E'/A', anteriyor E' ve E'/A', inferiyor E' ve E'/A' de?erleri anlaml? derecede d???k bulundu, mitral E'/septal E' ve mitral E/lateral E de?erleri anlaml? olarak daha y?ksekti. ?o?u doku Doppler parametreleri ile bel ?evresi, BK?-Z skoru, a?l?k kan ?ekeri, HOMD, serum ins?lin d?zeyi sistolik ve diyastolik kan bas?nc?, trigliserid ve YYL-K aras?nda korelasyon vard?. Bel ?evresi ve BK?-Z skoru baz? doku Doppler parametreleri i?in belirleyiciydi.
Sonu?: Obez ?ocuklarda akranlar?na g?re sol ventrik?l diyastolik fonksiyonlar? bozulmu?tur. Risk fakt?rleriyle diyastolik fonksiyon parametreleri aras?nda korelasyon olsa da, obezitenin kendisi sol ventrik?l diyastolik fonksiyonlar?nda azalmayla ili?kilidir.
Anahtar Kelimeler: ?ocuk, at?ml? Doppler ekokardiyografi, sol ventrik?l, obezite.
Geli? Tarihi: 29.07.2013 • Kabul Tarihi: 06.08.2013
Obesity, epidemic, a significant public health problem, is one of the complex clinical syndromes affecting both children and adults over the world(1). Eleven percent of children in the US are obese. Childhood obesity is associated with hypertension, dyslipidemia and abnormal glucose metabolism(2,3). The prevalence of metabolic syndrome is higher in obese than in normal-weight children(4). In children, the numbers of cardiovascular risk factors are increase in parallel with increasing body mass index (BMI)(5).
Obesity has been shown to be an important risk factor for heart failure in adults(6). Subclinical diastolic dysfunction is reported to be correlated with BMI in isolated adult obesity(7). Left ventricular diastolic dysfunction (LVDD) is a predictor of future development of heart failure(8). In addition to risk of early atherosclerosis, childhood obesity is leading to cardiac structural and functional changes. As a result of changes in the heart associated with obesity in childhood, cardiac function may be impaired, and the risk of heart failure may increase in early adulthood(9). Tissue Doppler imaging is a suitable method for the detection of subclinical LVDD(10).
Various publications showed that LVDD related to childhood obesity by conventional and tissue Doppler echocardiography. These publications found a relationship between LVDD, BMI, waist circumference, insulin, blood pressure and diastolic function(11,12,13,14).
In the present study, we aimed to asses two dimensional, M-Mode, conventional and tissue Doppler echocardiographic parameters of left ventricle and to evaluate the effects of anthropometric and biochemical factors on these parameters in obese children.
PATIENTS and METHODS
The study included children aged between 8 and 15 years who presented to the pediatric outpatient clinic at our institution without concomitant diseases. Informed consent was obtained from the children's parents. Children with congenital cardiac disease, hypertension, any systemic disease, thyroidal disorders, kidney or renal diseases, other endocrine disease such as diabetes, Cushing's disease, hypoparathyroidism or acromegaly, the children with macrosomic or low birth weight, rheumatic heart disease and arrhythmia on electrocardiogram (ECG) were excluded. This case-control study consisted of 60 obese children with a mean age of 11.5 ? 2.1 years, 34 (56.7%) girls and 26 (43.3%) boys; and 60 non-obese children with a mean age of 11.3 ? 2.2 years, 35 (58.3%) girls and 25 (41.7%) boys. After weighed and their heights were measured, BMI-Z score were calculated. Childhood obesity was defined as a BMI equal to or greater than the 95th percentile for children of the same age and gender (Z score ≥ 1.645)(15). The study groups were matched for age and gender. The protocol of the study was approved by the Institutional Ethics Committee.
While the subjects were in the left lateral decubitus position, two-dimensional, pulsed-wave Doppler transmitral recordings and tissue Doppler imaging (TDI) recordings of the mitral annulus diastolic velocities were performed with conventional views (parasternal long and short axes, apical 4-chamber) by a cardiologist with a 3.5-MHz sector transducer (Esaote MyLab 50, Genoa, Italy) and with simultaneously ECG recordings according to the recommendations of the American Society of Echocardiography (ASE) proposals(16). Ejection fraction and left ventricle mass index (LVMI) were determined as described previously(17,18). We performed pulsed-wave Doppler to measure mitral inflow velocity in apical 4-chamber view during diastole. Peak early (E) and late (A) mitral inflow velocity, E/A ratio and deceleration time (DT) of E velocity were obtained, and left ventricular ejection time (LVET), left ventricular isovolumetric contraction time (IVCT) and left ventricular isovolumetric relaxation time (IVRT) were measured as described previously(19). For tissue Doppler measurements, velocities were recorded from the two- and four-chamber views after a 3-5 mm pulsed wave Doppler volume was placed. Early (E') and late (A') diastolic velocities, peak systolic velocity (S') were obtained subsequently. All of these measurements were performed on the anterior, inferior, septal and lateral mitral annulus in accordance with the recommendations of the ASE(19).
Diastolic function parameters of 20 children (12 obese, 8 normal) were re-examined after 10 days to determine the coefficients of intraobserver variation between visits. The intraobserver variability was 2.9% for mitral E, 3.0% for mitral E/A, 3.2% for septal E', 3.1% for septal E'/A', 3.4% for lateral E', 3.4% for lateral E'/A', 2.8% for anterior E', 3.0% for anterior E'/A', 2.9% for inferior E', 2.7% for inferior E'/A', 2.9% for mitral E/septal E' and 3.2% for mitral E/lateral E' respectively.
Data were analyzed using Predictive Analysis Software (PASW) Statistics 18 (SPSS Inc., Chicago, IL, USA). Descriptive statistics were expressed as mean ? standard deviation or median (min-max). Normally distributed variables were compared by using independent sample t-tests, and non-normally distributed variables were compared using Mann-Whitney U tests. A p value < 0.05 was considered statistically significant. Correlations between tissue Doppler parameters and anthropometric, biochemical findings were calculated by using Spearman's rank correlation. The variables detected as a significant difference between obese and non-obese group were entered stepwise multiple linear regression analysis.
Anthropometric and Biochemical Characteristics
The study group, obese group, consisted of 60 children with a mean BMI-Z score of 2.1 (1.74-2.74), 34 (56.7%) girls and 26 (43.3%) boys, mean age 11.5 ? 2.1 years and mean obesity duration of 8.75 ? 1.8 years; the control group, non-obese group, consisted of 60 children with a mean BMI-Z score of 0.5 (0.4-0.7), 35 (58.3%) girls and 25 (41.7%) boys, mean age 11.3 ? 2.2 years (Table 1). Waist circumference, BMI-Z score, BMI and systolic and diastolic blood pressures, fasting blood glucose, insulin concentrations and HOMA index, and triglyceride were significantly higher in the obese than in the non-obese group (all p< 0.001). HDL-C concentration was significantly lower (p= 0.003) in the obese than in the non-obese group, and total cholesterol and low density lipoprotein cholesterol (LDL-C) concentrations were similar in the two groups (all p> 0.05).
Left Ventricular M-Mode and Conventional Doppler Flow Parameters
Left ventricular systolic and diastolic diameter, interventricular septum and posterior wall thicknesses, LVMI, left atrial and aortic root diameters were significantly higher (all p< 0.001), and mitral peak E and E/A ratio (all p< 0.05) were significantly lower in the obese than in the non-obese group. Ejection fraction and other conventional diastolic parameters were similar in the two groups (all p> 0.05, Table 2).
Left Ventricle Tissue Doppler Parameters
Compared to controls, septal E', septal E'/A', lateral E', lateral E'/A', anterior E', anterior E'/A', inferior E'and inferior E'/A' were all significantly lower (all p= 0.001, Table 3) and mitral E/ septal E' and mitral E/ lateral E' values were significantly higher in the obese group (both p< 0.001, Table 3).
Association Between Diastolic Function Parameters and Anthropometric and Biochemical Variables
Septal E' and septal E'/A'were found to be negatively correlated with waist circumference (r: -0.719; p< 0.001, r: -0.582; p< 0.001), BMI-Z score (r: -0.356; p= 0.005, r: -0.359; p= 0.005), fasting glucose (r: -0.320; p= 0.013, r: -0379; p= 0.003), insulin levels (r: -0.671; p< 0.001, r: -0583; p< 0.001), HOMA (r: -0.644; p≤ 0.001, r: -0.568; p< 0.001), triglyceride (r: -0.615; p< 0.001, r: -0.511; p< 0.001), but positively with HDL-C (r: 0.617; p< 0.001, r: 0.517; p< 0.001), (Table 4,5).
Lateral E' was inversely correlated with waist circumference, BMI-Z score, insulin, HOMA, triglyceride and positively with HDL (r: -0.579; p< 0.001, r: -0.341; p= 0.008, r: -0.541; p< 0.001, r: -0.516; p< 0.001, r: -0.475; p< 0.001, r: 0.479; p< 0.001, respectively, Table 4). Lateral E'/A' was negatively correlated with waist circumference, insulin, HOMA, and positively with HDL-C (r: -0430; p= 0.001, r: -0.444; p< 0.001, r: -0.433; p= 0.001, r: 0.276; p= 0.033; respectively, Table 5).
Anterior E' was negatively correlated with waist circumference, BMI-Z score, insulin levels, HOMA, systolic blood pressure, triglyceride and positively with HDL-C (r: -0.398; p= 0.002, r: -0.408; p= 0.001, r: -0.405; p= 0.001, r: -0.367; p= 0.004, r: -0.285; p= 0.027, r: -0.317; p= 0.014, r: 0.327; p= 0.011; respectively, Table 4), in addition anterior E'/A' were negatively correlated with waist circumference, BMI-Z score, and insulin levels (r: -0.258; p= 0.046, r: -0.322; p= 0.012, r: -0.266; p= 0.040, respectively, Table 5).
Inferior E' was negatively correlated with waist circumference, insulin levels, HOMA and systolic blood pressure (r: -0.279; p= 0.031, r: -0.285; p= 0.027, r: -0.266; p= 0.040, r: -0.363; p= 0.004, respectively, Table 4) and inferior E'/A' was negatively correlated with systolic and diastolic blood pressures (-0.299; p= 0.020, r: -0.394; p= 0.002 respectively, Table 5).
Mitral E/septal E' was positively correlated with waist circumference (Figure 1), BMI-Z score, fasting glucose, insulin levels, HOMA and triglyceride, and negatively with HDL-C (r: 0.674; p< 0.001, r: 0.421; p= 0.001, r: 0.383; p= 0.003, r: 0.574; p< 0.001, r: 0.561; p< 0.001, r: 0.332; p= 0.010, r: -0.354; p= 0.006, respectively, Table 6 and Figure 2).
There was a positive correlation between mitral E/lateral E' and BMI-Z score, (r: 0.537; p< 0.001, Table 6).
Age and obesity duration did not correlate with TDI parameters in obese group (all p values > 0.05; tables 4-6).
According to multivariate linear regression analysis, waist circumference was associated with septal E' (b? SE: -0.197 ? 0.043), septal E'/A' (b ? SE: -0.024 ? 0.006), lateral E' (b ? SE: -0.178 ? 0.043), lateral E'/A' (b ? SE: -0.016 ? 0.006) and mitral E/ septal E' (b ? SE: -0.173 ? 0.033), respectively (all p< 0.001, Table 7). BMI-Z score was related to anterior E' (b ? SE: -2.432 ? 1.143, p= 0.039) and mitral E/lateral E' (b ? SE: 4.880 ? 1.060, p< 0.001) (Table 7).
Obesity is often associated with cardiac changes such as increased left ventricular (LV) mass, LV hypertrophy, LV dilatation, left atrial dilation and LV diastolic, and systolic dysfunction(20). We compared two dimensional, M-mode, conventional pulsed Doppler and tissue Doppler echocardiographic parameters as well as conventional anthropometric parameters in groups of obese and non-obese children. We found that BMI, BMI-Z score, waist circumference, systolic and diastolic blood pressures were significantly higher in the obese group than in the non-obese group. In addition, glucose, HOMA index, insulin and triglyceride serum levels were significantly higher and HDL-cholesterol concentrations were significantly lower in the obese group.
Studies showed that obese children had higher absolute chamber dimensions, volumes, wall thickness, and left ventricular mass than their age and gender-matched normal controls(12,21,22). Similarly, we found that left ventricular wall dimensions, interventricular septum, posterior wall, LVMI, left atrial and aortic diameters were higher in the obese than in the non-obese group.
Our findings potentially suggest that both increased adipose tissue and elevated serum insulin levels are associated with the increased heart's workload, LV dimension and LVMI in obese children. Previous studies found that there were significant differences in conventional and tissue Doppler diastolic parameters between obese and non-obese groups(11,12,13,14). The importance of insulin resistance in obesity is well recognized and insulin increase heart rate, blood pressure and the workload of the heart by activating the sympathetic nervous system as well as in increased LV size and mass as a result of increased intravascular volume, changed myocardial structures and functions(23,24,25). Furthermore, myocardial structural changes have been associated with insulin resistance and volume excess that predispose to heart failure in obese subjects(24). Insulin resistance and hyperinsulinemia also relate to glucose intolerance, an increase in plasma triglyceride and a decrease in HDL cholesterol concentrations, high blood pressure, hyperuricemia, and smaller denser LDL particles(26). Some hormones and cytokines such as leptin, nonesterified free fatty acids, tumor necrosis factor-a (TNF-α), interleukin-6 (IL-6), C-reactive protein (CRP), are released from adipose tissue, change the sympathetic activity and insulin sensitivity(27). In addition, myocardial energy metabolism shifts to the use of free fatty acids in obesity, and then occur cardiac lipotoxicity leading to programmed cell death. In addition, fatty infiltration can induce structural and functional changes in the heart as well as diastolic dysfunction occurs(28).
We observed deterioration in conventional and tissue Doppler parameters in the obese than in the non-obese group. Although waist circumference, BMI-Z score, fasting blood glucose, HOMA, systolic and diastolic blood pressure, serum insulin levels, triglyceride and HDL were found to be correlated with the number of tissue Doppler parameters in the study. In some previous studies, both BMI(11,12) and waist circumferencewere found to be independent predictors for various diastolic function parameters(13). In our study, while waist circumference was the most important variable affecting septal E, lateral E', septal E'/A', lateral E'/A' and mitral E/septal E', BMI-Z score was seemed to be affect anterior E'and mitral E/lateral E'. These findings suggested that waist circumference and BMI-Z score in obesity were the most important determinants for the deterioration of diastolic function.
The main limitation of this study was its small sample size. If heart rate, sympathetic activity level, cytokine levels and adipose tissue derived hormones were measured and morbid obese children could be included as different group in the study, our results would have been validated.
Left ventricular diastolic functions are impaired in obese children compared to that of their peers. Although there were correlation between diastolic dysfunction parameters and risk factors, waist circumference and BMI-Z score as markers of obesity were associated with changes in cardiac structures and diastolic functions in childhood. Consequently, obesity itself is related with reduction of left ventricular diastolic functions. Further observational prospective studies concerning adipose tissue related cytokines and hormones are needed to determine the effects of these changes in adulthood.
Conflict of interest
Dr. ?eref Alpsoy
Tekirda? Nam?k Kemal ?niversitesi Ara?t?rma ve
Uygulama Hastanesi, Kardiyoloji Klini?i,