- Email: [email protected]
Impact of body composition, fat distribution and sustained weight loss on cardiac function in obesity
International Journal of Cardiology 159 (2012) 128–133
Contents lists available at ScienceDirect
International Journal of Cardiology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / i j c a r d
Impact of body composition, fat distribution and sustained weight loss on cardiac function in obesity Dimitris Kardassis a,d,⁎, Odd Bech-Hanssen a,d, Marie Schönander b, Lars Sjöström b,d, Max Petzold c, Kristjan Karason a,d a
Department of Cardiology, Sahlgrenska University Hospital, Gothenburg, Sweden Department of Medicine, Sahlgrenska University Hospital, Gothenburg, Sweden Nordic School of Public Health, Gothenburg, Sweden d Sahlgrenska Academy at University of Gothenburg, Gothenburg, Sweden b c
a r t i c l e
i n f o
Article history: Received 20 June 2010 Received in revised form 7 December 2010 Accepted 10 February 2011 Available online 2 March 2011 Keywords: Obesity Weight loss Body composition Fat distribution echocardiography Left ventricular function
a b s t r a c t Background: Obesity is associated with alterations in left ventricular function varying along with the degree of fatness, but the mechanisms underlying this co-variation are not clear. In a case–control study we examined how sustained weight losses affect cardiac function and report on how body composition and fat distribution relate to the left ventricular performance. Methods: At the 10-year follow-up of the Swedish obese subjects (SOS) study cohort we identified 44 patients with sustained weight losses after bariatric surgery (surgery group) and 44 matched obese control patients who remained weight stable (obese group). We also recruited 44 matched normal weight subjects (lean group). Dual-energy X-ray absorptiometry, computed tomography and echocardiography were performed to evaluate body composition, fat distribution and cardiac function. Results: BMI was 42.5 kg/m2, 31.5 kg/m2 and 24.4 kg/m2 for the obese, surgery and lean groups respectively. Increasing degree of obesity was associated with larger left ventricular volumes (p b 0.001), higher cardiac output (p b 0.001), reduced systolic myocardial velocity (p b 0.001) and impaired ventricular relaxation (p = 0.015). In multivariate analyses, left ventricular volume, stroke volume and cardiac output primarily associated with lean body mass, whereas blood pressure, heart rate and variables reflecting cardiac dysfunction were more related to total body fat and visceral adiposity. Conclusion: Obesity is associated with discrete but distinct disturbances in the left ventricular performance appearing to be related to both the total amount of body fat and degree of visceral adiposity. Patients with sustained weight losses display superior left ventricular systolic and diastolic functions as compared with their obese counterparts remaining weight stable. © 2011 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Obesity is associated with increased risk of heart failure, in part due to associated co-morbidities that promote the development of coronary artery disease [1,2]. In addition, obesity has a more direct adverse effect on cardiac function through the rise in hemodynamic load that occurs along with body fat accumulation [3]. As body weight increases, both blood volume and cardiac output rise in order to fulfill the requirements of a higher metabolic rate [3,4]. Although the expansion of circulating blood volume is followed by a lowering of the peripheral vascular resistance this is often insufficient to prevent a rise in blood pressure [5,6]. Consequently, the heart in people with obesity is burdened with both volume and pressure
⁎ Corresponding author at: Department of Cardiology, Sahlgrenska University Hospital, Sweden. Tel.: +46 313427720. E-mail address: [email protected] (D. Kardassis). 0167-5273/$ – see front matter © 2011 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ijcard.2011.02.036
overload, which may lead to various degrees of left ventricular hypertrophy [7]. Such structural changes may in turn interfere with ventricular filling and lead to diastolic dysfunction. In long-standing morbid obesity an impairment of myocardial contractility may also supervene and give rise to overt heart failure, a clinical condition referred to as “obesity cardiomyopathy” [8]. Both lean body mass and adipose tissue increase as obesity develops, but the hemodynamic effects of these body compartments may differ. Furthermore, the distribution of body fat is of importance, since adverse cardiovascular risk is associated with abdominal obesity in particular [9]. In this respect, the separate effects of different body compartments and fat distribution on cardiac function are of interest. The optimal treatment for cardiac dysfunction in obesity would, apparently, be sustained weight loss. However, this is difficult to achieve with conventional methods, which offer results that, for the most part, are only modest and temporary [10]. Bariatric surgery, on the other hand, induces weight losses that are both large and maintained over time [11].
D. Kardassis et al. / International Journal of Cardiology 159 (2012) 128–133
Little is known about the influence of different body compartments and fat distribution on left ventricular performance and, although short-term weight reduction has been reported to improve cardiac function, the effects of sustained weight loss have not been described. The aim of this study was to investigate how body composition and fat distribution relate to variables reflecting left ventricular contractility and filling and to assess the effects of sustained weight loss on systolic and diastolic function.
129
Lean body mass and total body fat were measured with a whole-body dual-energy X-ray absorptiometry (DXA) scanner (DPXL, Lunar Radiation, Madison, WI) using software version 1.35. The coefficient of variance was 0.7% and 1.7% for lean body mass and total body fat, respectively, assessed from double examinations in 10 females. Intra-abdominal and subcutaneous adipose tissue areas were measured at the level of the fourth lumbar vertebra using a single slice Computed Tomography scan (HSA, GE Medical Systems, Milwaukee, WI, version RP2) as previously described [14]. Precision errors, calculated from double determinations, were 1.2% and 0.5% for intra-abdominal and subcutaneous adipose tissues, respectively. 2.3. Echocardiography
2. Methods 2.1. Study groups The SOS study is a prospective, controlled, surgical interventional trial, which enrolled 4047 obese subjects at 25 surgical departments and 480 primary health care centers in Sweden [12]. The study protocol is described in detail elsewhere [11,12]. Briefly, the surgery group comprised 2010 eligible subjects desiring surgery and, at the same time, a matched control group of 2037 obese subjects. Inclusion criteria were age ranging from 37 to 60 years and BMI of ≥34 for men and of ≥38 or more for women. The exclusion criteria, aiming to obtain operable patients, were the same for both study groups. Patients included in the present case control study were recruited consequently during a two-year period (2004–2006) among participants of the SOS study residing in Gothenburg that had been monitored for 10 years. According to pre-specified weight change criteria, we identified 44 surgery patients who, after 10 years, displayed a weight loss of greater than 15% and 44 obese control patients, in which the weight had changed less than 5%. To ensure comparability prior to intervention, the two groups were carefully matched with respect to baseline data from the SOS study. Matching variables included age, gender, BMI, hypertension, hyperlipidemia, diabetes and smoking status (Table 1). In addition, we included 44 healthy normal weight patients, recruited from a randomly selected sample of adults living in the municipality of Mölndal [13], who matched the surgery and obese groups at the 10-year follow-up with respect to age, height and smoking status. In total, 132 subjects were included in the study, comprising 69 women and 63 men with ages ranging from 44 to 71 years. The three study groups underwent a cross-sectional examination with respect to body composition, fat distribution and left ventricular function. The study protocol was approved by the ethics committee at the University of Gothenburg, and all study subjects gave their informed consent to participate.
2.2. Clinical measurements and body composition Measurements of weight and height were obtained and BMI was calculated. Blood pressure was measured in the supine position after 10 min of rest and the mean of two recordings was registered. Blood samples were obtained in the morning after a fast of 12 h and analyzed at the Central Laboratory of Sahlgrenska University Hospital (accredited according to European norm 45001).
Echocardiographic examinations were performed by use of a commercially available ultrasound system. All measurements were performed under resting conditions with the subject in the left lateral decubitus position at end-expiration. Recordings were performed by experienced investigators and analyzed off-line by a single observer blinded with respect to the subject's treatment. All measurements were performed according to recommendations from the American Society of Echocardiography [15]. Two-dimensional measurements of the left ventricular diastolic and systolic volumes were estimated from the apical four- and two chamber view and LV ejection fraction was calculated according to the Simpson's rule [16]. Planimetry of the left and right atriums was performed from a late systolic stop frame showing the maximum atrial size. The left ventricular mass was determined using the truncated ellipsoid model according to Byrd et al. [17]. Blood flow velocity in the left ventricular outflow tract was estimated by pulsewave Doppler from an apical 4-chamber view with a sample size of 5 mm. Stroke volume (SV) was calculated as the product of the cross-sectional area of the left ventricular outflow tract and the velocity time integral. Cardiac output (CO) was estimated by multiplying stroke volume by heart rate (HR). Isovolumetric relaxation time (IVRT) was measured as the time between the aortic valve closure and the mitral valve opening. Mitral flow was recorded between the mitral leaflets in the 4-chamber view. Measurements of early flow velocity (E), E-wave deceleration time (DT), and peak velocity during atrial systole (A) were obtained and the E/A ratio calculated. Pulmonary venous flow velocities were acquired from the upper right pulmonary vein. Peak velocities during systole (S) and diastole (D) were recorded and the S/D ratio calculated. Continuous-wave Doppler was used to measure the peak velocity of tricuspid regurgitation when present. The pressure gradient was calculated according to the simplified Bernoulli equation and used, together with an inferior vena cava estimate of the right atrial pressure, to approximate pulmonary artery pressure. The presence or absence of diastolic dysfunction was evaluated by using an integrated assessment of the left ventricular filling patterns. For each study subject, the expected E/A ratio, DT and S/D ratio were predicted by a regression equation derived from a healthy control population. The observed values in study subjects were regarded as being abnormal if they differed by 1.96 SD from the predicted value using the Z-score [18,19]. Using a previously described criteria [20], subjects were classified to have either normal diastolic function or impaired left ventricular relaxation. Myocardial tissue Doppler imaging was performed from the apical 4-chamber view. Peak systolic (SMV) and early diastolic (Ea) tissue velocities were recorded at the septal corner of the mitral annulus and the E/Ea ratio was calculated.
Table 1 Clinical characteristics of the obese and surgery groups at baseline in the SOS study and changes at a 10-year follow-up. Values are given as means (SD) or absolute numbers (% of total).
BMI, kg/m2 Systolic BP, mm Hg Diastolic BP, mm Hg Cholesterol, mmol/L Triglycerids, mmol/L Glucose, mmol/L Insulin, mU/L No (%) on antihypertensives No (%) with diabetes No (%) current smoker ⁎ p b 0.05 as compared with the obese group. ⁎⁎ p b 0.01 as compared with the obese group. ⁎⁎⁎ p b 0.001 as compared with the obese group.
Baseline change at 10 years Baseline change at 10 years Baseline change at 10 years Baseline change at 10 years Baseline change at 10 years Baseline change at 10 years Baseline change at 10 years Baseline change at 10 years Baseline change at 10 years Baseline change at 10 years
Obese group (N = 44)
Surgery group (N = 44)
41.2 (4.0) + 2.1 (5.4) 139.5 (16.3) − 3.2 (18.0) 87.7 (8.9) − 6.0 (11.0) 6.0 (1.1) − 0.8 (0.9) 2.2 (0.95) − 0.3 (2.0) 5.2 (1.4) + 1.8 (1.9) 19 (9.8) 0.7 (14.8) 5 (11.4) + 25 12 (27.3) +1 14 (31.2) −6
40.0 (4.1) − 9.7 (4.4)⁎⁎⁎ 144.4 (18.0) − 15.1 (18.3)⁎⁎ 93.8 (11.6)⁎ − 16.8 (11.4)⁎⁎ 5.8 (1.9) − 0.8 (1.2) 2.4 (0.9) − 1.1 (0.8) 5.2 (1.8) + 0.4 (1.2)⁎⁎⁎ 19.4 (8.1) − 11.5 (8.0)⁎⁎⁎ 5 (11.4) + 10 12 (27.3) −8 15 (34.1) +1
130
D. Kardassis et al. / International Journal of Cardiology 159 (2012) 128–133
Two-dimensional echocardiographic evaluation of left ventricular volumes can be technically difficult, especially in subjects with obesity. Despite this, satisfactory recordings were available for 64%, 72% and 89% of subjects in the obese, surgery and lean groups, respectively. On the other hand, Doppler measurements were easier to obtain and corresponding numbers for adequate readings were 93%, 98% and 100%. The coefficient of variance, based on double determination in 15 patients, was 11.2% and 5.0% for two-dimensional volumes and Doppler measurements, respectively. 2.4. Statistical analysis Statistical analyses were performed using the SPSS for Windows version 16 (SPSS, Chicago, IL). Descriptive statistical results are given as the mean (standard deviation) or proportions (%). Variables that did not display a normal distribution were transformed logarithmically prior to analysis. Comparisons between obese and surgery groups with respect to SOS baseline data and change from baseline were performed with unpaired t-tests. Overall comparisons between the three study groups with respect to crosssectional data were carried out with ANOVA and post-hoc comparisons between obese vs. surgery groups and surgery vs. lean groups were performed with the Bonferroni test. After pooling data from the three study groups, Pearson's correlation coefficients were calculated in order to estimate associations between body composition and measures of left ventricular performance. Forward stepwise multiple regression analyses were then used to examine how lean body mass, total body fat, visceral adipose tissue and subcutaneous adipose tissue related to indices of the left ventricular function. A P-value of less than 0.05 was considered significant.
3. Results
Table 3 Cross-sectional body composition assessments (DXA and CT at the 4th lumbar vertebra) in the three study groups. Values are given as means (SD). Obese group (n = 44)
Surgery group (n = 44)
Lean group (n = 44)
Lean body mass, 61.7 (11.2) 52.5 (11.6)⁎ 49.9 (10) kg (DXA) ⁎ Total body fat, kg (DXA) 52 (11) 34.8 (11) 19.6 (7)# Visceral adipose tissue 313.2 (121.4) 169.4 (79.4)⁎ 112.5 (50.6)# area cm2 (CT) 623.7 (201) 394 (144.6)⁎ 200.9 (79.9)# Subcutaneus adipose tissue area cm2 (CT)
P-value b0.001 b0.001 b0.001 b0.001
Multiple comparisons were adjusted for according to the Bonferoni procedure. ⁎ p b 0.001 as compared with the obese group. # p b 0.001 as compared with the surgery group.
higher HDL than the obese group. Surgery patients smoked more often but had less diabetes and hypertensive treatment than the obese patients. As compared with the surgery group, the lean group showed lower body weight, BMI, glucose and insulin. None of the lean patients had diabetes or were treated with antihypertensive medication. The smoking frequency was comparable in the two groups.
3.1. Clinical characteristics Table 1 shows baseline data from the SOS study by means of which the obese and surgery groups were matched and changes at a 10-year follow-up. At baseline the two groups were quite comparable with respect to clinical parameters apart from diastolic blood pressure, which was slightly higher in the surgery group. At SOS 10-year followup the surgery group displayed substantial reductions in BMI and significant improvements in several clinical variables as compared with the obese group. Table 2 shows cross-sectional clinical characteristics of the obese and surgery groups after 10-years of follow-up in the SOS study and for the matched lean group. The lean group was somewhat younger than the obese and surgery group but of comparable height. As compared with the obese group, the surgery group showed lower body weight, body mass index (BMI) and diastolic blood pressure. The surgery group also had lower triglycerides, glucose and insulin and Table 2 Cross-sectional clinical characteristics for the obese, surgery and lean groups. Values are given as means (SD) or absolute numbers (% of total).
Gender, male/female Age, years Height, cm Weight, kg BMI, kg/m2 Systolic BP, mm Hg Diastolic BP, mm Hg Cholesterol, mmol/L Triglycerids, mmol/L Glucose, mmol/L Insulin, mU/L No (%) on antihypertensives No (%) with diabetes No (%) current smoker
Obese group (N = 44)
Surgery group (N = 44)
Lean group (N = 44)
P-value
21/23 58.9 (4.7) 1.71 (0.1) 127.1 (17.8) 42.5 (6.1) 136.3 (17.7) 81.7 (11.7) 5.2 (1.1) 1.9 (1.3) 7.0 (2.3) 19.7 (14) 30 (68)
21/23 59.6 (5.1) 1.73 (0.1) 90.6 (13)⁎⁎⁎ 31.5 (4.9)⁎⁎⁎ 129.3 (17.1) 77.0 (8.7)⁎
21/23 55 (5.3)# 1.73 (0.1) 73.0 (10)### 24.4 (3.7)### 122.9 (20.3) 76.7 (10.9) 5.4 (0.85) 1.5 (1.1) 5.0 (0.6)# 6.4 (3.5)## 0###
0.923 b0.001 0.157 b0.001 b0.001 0.003 0.021 0.237 0.062 b0.001 b0.001 b0.001
0
b0.001 0.104
13 (29.5) 8 (18.2)
5.0 (1.1) 1.3 (0.5)⁎⁎ 5.6 (1.4)⁎⁎ 7.9 (3.7)⁎⁎⁎ 15 (34)⁎⁎ 4 (9)⁎ 16 (36.4)
11 (25)
Multiple comparisons were adjusted for according to the Bonferoni procedure. ⁎ p b 0.05 as compared with the obese group. ⁎⁎ p b 0.01 as compared with the obese group. ⁎⁎⁎ p b 0.001 as compared with the obese group. # p b 0.05 as compared with the surgery group. ## p b 0.01 as compared with the surgery group. ### p b 0.001 as compared with the surgery group.
3.2. Body composition Cross-sectional measurements of lean body mass and total body fat with DXA and of intra-abdominal and subcutaneous adipose tissue area at the level of L4 with CT are shown in Table 3. All body
Table 4 Cross-sectional echocardiographic measurements in the three study groups. Values are given as means (SD) or absolute numbers.
Structure LVM (g) LVM/BSA (g/m2) LVEDV (mL) LVEDV/BSA (mL/m2) LVESV(mL) LVESV/BSA (mL/m2) Hemodynamics Stroke volume (mL) Heart rate (bpm) Cardiac output (L/min) Pulmonary pressure (mm Hg) Systolic function Left ventricular ejection fraction (%) SMV (cm/s) Diastolic function Normal/abnormal (n)
Obese group (n = 44)
Surgery group (n = 44)
201.4 87.9 113.8 48.5 42.5 17.8
(21.6) (11.8) (23.8) (8.9) (14.1) (4.7)
157.7 79.1 87.0 43.8 31.0 15.5
(20.5)⁎⁎ (10.9)⁎⁎ (12.2)⁎⁎⁎ (6.0) (7.8)⁎⁎
88.3 (1.6) 72.8 (11.6) 6.4 (1.3) 24.3 (3)
83.1 64.5 5.3 20.9
(1.5) (9.0)⁎⁎⁎ (0.8)⁎⁎⁎ (3)⁎⁎
62.5 (6.2) 9.3 (1.6)
37/7
Lean group (n = 44)
P-value
(21.7)# (11.4)# (19.3) (9.1) (10.1) (4.8)
b0.001 b0.001 b0.001 0.173 0.001 0.257
(1.5) (9.8) (1.0) (4)
0.158 b0.001 b0.001 b0.001
64.6 (5.6)
66.0 (6.2)
0.019
10.6 (1.0)⁎⁎
11.2 (1.7)
b0.001
44/0⁎
(3.5)
133.9 71.8 85.3 46.1 29.8 16.1
80.0 65.0 5.2 20.7
44/0
b0.001
Multiple comparisons were adjusted for according to the Bonferoni procedure. LVM = left ventricular mass; BSA = body surface area; LVEDV = left ventricular end diastolic volume; LVESV = left ventricular end systolic volume; SMV = peak systolic velocity at the basal septal segment; E/A = ratio of early (E) to late (A) peak diastolic transmitral flow velocity; S/D = ratio of peak pulmonary venous flow velocity during ventricular systole (S) to peak pulmonary venous flow velocity during ventricular diastole (D); E/Ea = ratio of mitral early peak velocity (E) to mitral annulus early peak velocity (Ea). ⁎ p b 0.05 as compared with the obese group. ⁎⁎ p b 0.01 as compared with the obese group. ⁎⁎⁎ p b 0.001 as compared with the obese group. # p b 0.05 as compared with the surgery group.
D. Kardassis et al. / International Journal of Cardiology 159 (2012) 128–133
composition parameters decreased significantly across the groups in the order from obese to surgery to lean groups. 3.3. Echocardiography Cross-sectional echocardiography measurements are shown in Table 4. Left ventricular mass, and volumes decreased across the three groups from obese to surgery to lean. These structural measures were significantly different in the obese group as compared with the surgery group and, except for left ventricular volumes, in the surgery group as compared with the lean group. Heart rate, cardiac output and pulmonary pressures also decreased across the groups in a similar manner with significant differences between the obese group and the surgery group. Stroke volume did not show a significant linear trend across the three groups. Left ventricular ejection fraction and systolic myocardial velocity increased with decreasing BMI, and systolic myocardial velocity was significantly different in the obese group as compared with the surgery group. Evaluation on the basis of integrated transmitral and pulmonary venous flow velocity revealed diastolic dysfunction in 16% of the obese group (all mild), whereas none of the patients in the surgery or lean group displayed abnormal filling patterns. Other Dopplerechocardiographic measurements of diastolic function are shown in Fig. 1. E/A ratio increased, while S/D ratio, E/Ea ratio, IVRT and deceleration time decreased along with a declining degree of obesity with significant differences between the obese group and the surgery group. Furthermore, the left atrial area decreased significantly across the three groups. 3.4. Association between body composition and left ventricular function Correlation coefficients and stepwise multiple regression analysis for associations between body composition and selected left ventricular measurements can be seen in Tables 5 and 6. Left ventricular enddiastolic volume, stroke volume and cardiac output were positively and independently related to lean body mass, whereas heart rate,
131
blood pressure and IVRT were positively and independently associated with the visceral adipose tissue. A lower E/A ratio, higher S/D ratio and longer deceleration time were independently associated with total body fat, while reduced SMV and higher E/Ea ratio were independently related to both total body fat and visceral adipose tissue. 4. Discussion In the present study, we confirm that obesity is associated with discrete but distinct disturbances in the left ventricular systolic and diastolic performances. Cardiac dysfunction appears to be related to both the total amount of body fat and the degree of visceral adiposity. Persistent weight loss, on the other hand, was associated with increased myocardial contractility and enhanced left ventricular relaxation. Our findings indicate that long-term sustained weight loss in obesity is beneficial for the left ventricular systolic and diastolic functions. 4.1. Hemodynamics It is well documented that obesity is associated with an augmentation in blood flow in order to meet the requirements of increased metabolic rate [3,4]. In agreement, we observed increased cardiac output across the study groups in the order from lean to surgery to obese. Previously, weight-related variations in cardiac output have been assumed to be linked to changes in stroke volume [5,21], but in the present study obesity-related rise in blood flow was also due to a higher heart rate. We therefore propose that adaptation of cardiac output to variations in body fatness is mediated, not only by altered stroke volume, but also by changes in heart rate. In fact, previous studies have shown that sympathetic nervous system activity may rise along with increasing obesity and thus contribute to a higher heart rate [22]. Left ventricular volume, stroke volume and cardiac output, which are crude markers of preload, were primarily associated with lean body mass and similar observations have been reported by Collis et al. [23]. Thus, it appears that a variation in preload is more dependent on the obesity-associated increase in lean body mass than on the actual
P<0.001
P=0.002 P=0.015
*
***
Surgery
Lean
P<0.001
IVRT ms
**
Obese
Surgery
Lean
E/Ea ratio Obese
Surgery
Lean
Obese
Surgery
Lean
P<0.001
P<0.001
Deceleration time (ms)
Obese
**
Obese
**
Left atrial area (cm2)
E/A ratio
S/D ratio
*
Surgery
Lean
***
Obese
Surgery
Lean
Fig. 1. Scatter plots with diamond means (group mean and 95% CI) for echocardiographic measures of the diastolic function by study group. Results from overall comparisons with ANOVA (P-value) and post hoc analyses with Bonferroni (asterisk) are shown. *P b 0.05, **P b 0.01, ***P b 0.001.
132
D. Kardassis et al. / International Journal of Cardiology 159 (2012) 128–133
Table 5 Pearson's correlation coefficients (r) and multivariate stepwise linear regression analyses of hemodynamic and systolic function on body composition. LVEDV
Stroke volume
Heart rate
Cardiac output
Systolic BP
Systolic mean velocity
Correlation
r
P
r
P
r
P
r
P
r
P
r
P
Lean body mass Total body fat Visceral adipose tissue Subcutaneous adipose tissue
0.49 − 0.03 0.26 0.164
b 0.001 0.755 0.013 0.120
0.36 − 0.14 − 0.02 − 0.01
0.001 0.240 0.858 0.931
0.01 0.35 0.39 0.33
0.913 b 0.001 b 0.001 b 0.001
0.33 0.17 0.25 0.23
0.003 0.154 0.021 0.036
0.05 0.29 0.35 0.29
0.554 b 0.001 b 0.001 b 0.001
0.06 − 0.43 − 0.31 − 0.37
0.913 b0.001 b0.001 b0.001
Multiple regression
β
P
β
P
β
P
β
P
β
P
β
P
b0.001
− 0.18
0.024
0.03
0.007 − 0.01
b0.001
3.08
b 0.001
3.64
b 0.001
Lean body mass Total body fat Visceral adipose tissue Subcutaneous adipose tissue Adjusted R2 (%)
b 0.001
1.18
0.62
47
32
19
13
12
20
LVEDV = left ventricular end-diastolic volume; BP = blood pressure.
body fat accumulation. Systolic blood pressure, a simple estimate of afterload, and heart rate were, on the other hand, mainly related to the extent of visceral adipose tissue. The disparate influence of separate body compartments on loading conditions and heart rate is of interest. While lean body mass appears to be a main determinant of blood flow, visceral obesity may contribute to a rise in blood pressure by interfering with regulation of heart rate and vascular resistance. Indeed, paracrine and hormonal factors derived from adipose tissues have been suggested to play a role in the pathophysiology of obesityrelated hypertension [24,25].
4.2. Systolic function Although overt systolic heart failure in prolonged morbid obesity has been reported, this appears to be relatively uncommon. In the present study, there was a trend towards lower left ventricular ejection fraction with increasing degree of obesity, but post hoc analyses did not reveal significant differences between study groups. On the other hand, systolic myocardial tissue velocity was significantly lower in the obese group as compared with the surgery group. Further, this measure of contractility was inversely related to the extent of body fat and visceral adiposity independent of other variables. Our findings are consistent with previous studies on otherwise healthy obese subjects [26,27], suggesting that subclinical systolic dysfunction may be a prevalent condition in obesity. Still, it should be emphasized that loading conditions can influence measures of systolic function, although tissue Doppler imaging is probably less affected than other estimates of myocardial contractility.
4.3. Diastolic function An integrated evaluation of the left ventricular filling patterns revealed definite relaxation disturbances in 16% of the obese group, whereas none of the patients in the surgery or lean group displayed overt diastolic dysfunction. Although mean values for variables describing left ventricular filling were within normal ranges for all three study groups, the patterns were consistently less favorable in the obese group as compared to the surgery group, which in turn did not differ from the lean group. In addition, the obese group displayed a larger left atrium and a higher pulmonary artery pressure, also suggesting impaired left ventricular filling. Our findings are in line with previous studies reporting a correlation between obesity and disturbances in the left ventricular diastolic function [26,28]. Moreover, we observed that adverse filling patterns were, aside from total body fat, independently related to the extent of visceral adiposity. This suggests a connection between diastolic dysfunction and the metabolic syndrome, possibly mediated by related hemodynamic, metabolic and hormonal aberrations interfering with ventricular relaxation [29–31]. 4.4. Weight loss Short-term weight loss has been reported to have favorable effects on myocardial contractility and ventricular filling [22] and our present observations suggest that such improvements persist long-term following bariatric surgery. In fact, the surgery group did not differ significantly from the lean group with respect to variables describing systolic and diastolic functions. In a previous report from the SOS
Table 6 Pearson's correlation coefficients (r) and multivariate stepwise linear regression analyses of variables of left ventricular function on body composition. Deceleration time
Left atrial area
Correlation
E/A ratio r
P
r
P
r
P
r
P
r
P
r
P
Lean body mass Total body fat Visceral adipose tissue Subcutaneous adipose tissue
− 0.02 − 0.34 − 0.31 − 0.21
0.829 b0.001 b0.001 0.018*
0.18 0.24 0.20 0.18
0.05 b0.006 b0.029 0.049
0.21 0.42 0.44 0.42
0.023 b0.001 b0.001 b0.001
0.16 0.34 0.34 0.11
0.116 b 0.001 b 0.001 0.251
0.19 0.33 0.32 0.20
0.040 b0.001 b0.001 0.027
0.56 0.33 0.31 0.27
b0.001 b0.001 b0.001 0.002
Multiple regression
β
P
β
P
β
P
β
P
β
P
β
P
b0.001
0.29 0.26 − 1.49
b0.001 b0.001 0.017
Lean body mass Total body fat Visceral adipose tissue Subcutaneous adipose tissue Adjusted R2 (%)
− 0.01
14
S/D ratio
b0.001
0.01
8
E/Ea ratio
0.007
− 0.05 0.96 20
E/A ratio = ratio of early (E) to late (A) peak diastolic transmitral flow velocity. S/D ratio = ratio of peak systolic (S) to diastolic (D) pulmonary flow velocity. E/Ea = ratio of mitral early peak velocity (E) to mitral annulus early peak velocity (Ea). IVRT = isovolumetric relaxation time.
IVRT
b0.001 0.043
0.31 0.56 − 4.17 19
b 0.001 0.031 13
43
D. Kardassis et al. / International Journal of Cardiology 159 (2012) 128–133
study [32], weight loss in obese subjects was associated with a marked relief in breathlessness and increased physical activity, which could, in part, be related to improvement in cardiac function. However, whether sustained weight loss may reduce the risk of overt cardiac failure is still unknown. Further, the feasibility of surgical obesityintervention, as a treatment option for chronic heart failure, remains to be determined. 4.5. Limitations The main limitation of the present study was that echocardiographic examinations for the obese and surgery groups were not performed prior to inclusion in the SOS study. This weakens our conclusions with respect to the effects of weight loss on the left ventricular function. On the other hand, these two groups were carefully matched with respect to important SOS baseline characteristics, resulting in two almost identical groups prior to intervention. It is therefore reasonable to assume that the left ventricular function did not differ between obese and surgery groups at the SOS baseline. Further, our study is unique with respect to the long-term follow-up of obese subjects with large and sustained weight losses. Such patient populations have rarely been studied previously due to the difficulties in attaining reliable long-term weight reductions with conventional therapy. 4.6. Conclusion Obesity is associated with discrete but distinct disturbances in the left ventricular systolic and diastolic performances. Whereas obesityrelated increase in lean body mass appears to be the main determinant of stroke volume and cardiac output, accumulation of body fat is more related to disturbances in ventricular contractility and relaxation. Visceral adiposity may further impair the left ventricular function by interfering with heart rate and blood pressure. Surgical obesity intervention appears to have long-term favorable effects on the left ventricular systolic and diastolic functions and could be considered as a treatment option in chronic heart failure. Acknowledgment The authors of this manuscript have certified that they comply with the Principles of Ethical Publishing in the International Journal of Cardiology [33]. References [1] Gordon T, Kannel WB. Obesity and cardiovascular diseases: the Framingham study. Clin Endocrinol Metab 1976;5:367–75. [2] Haslam DW, James WP. Obesity. Lancet 2005;366:1197–209. [3] de Divitiis O, Fazio S, Petitto M, et al. Obesity and cardiac function. Circulation 1981;64:477–82. [4] Vasan RS. Cardiac function and obesity. Heart (British Cardiac Society) 2003;89: 1127–9. [5] Alexander JK, Dennis EW, Smith WG, et al. Blood volume, cardiac output, and distribution of systemic blood flow in extreme obesity. Cardiovasc Res Cent Bull 1962;1:39–44. [6] Lauer MS, Anderson KM, Levy D. Separate and joint influences of obesity and mild hypertension on left ventricular mass and geometry: the Framingham heart study. J Am Coll Cardiol 1992;19:130–4. [7] Messerli FH, Sundgaard-Riise K, Reisin ED, et al. Dimorphic cardiac adaptation to obesity and arterial hypertension. Ann Intern Med 1983;99:757–61.
133
[8] Alexander JK. The cardiomyopathy of obesity. Prog Cardiovasc Dis 1985;27: 325–34. [9] Eckel RH, Grundy SM, Zimmet PZ. The metabolic syndrome. Lancet 2005;365: 1415–28. [10] de las Fuentes L, Waggoner AD, Mohammed BS, et al. Effect of moderate diet-induced weight loss and weight regain on cardiovascular structure and function. J Am Coll Cardiol 2009;54:2376–81. [11] Sjostrom L, Narbro K, Sjostrom CD, et al. Effects of bariatric surgery on mortality in Swedish obese subjects. N Engl J Med 2007;357:741–52. [12] Sjostrom L, Larsson B, Backman L, et al. Swedish obese subjects (SOS). Recruitment for an intervention study and a selected description of the obese state. Int J Obes Relat Metab Disord 1992;16:465–79. [13] Larsson I, Berteus Forslund H, Lindroos AK, et al. Body composition in the SOS (Swedish Obese Subjects) reference study. Int J Obes Relat Metab Disord 2004;28: 1317–24. [14] Chowdhury B, Sjostrom L, Alpsten M, et al. A multicompartment body composition technique based on computerized tomography. Int J Obes Relat Metab Disord 1994;18:219–34. [15] Lang RM, Bierig M, Devereux RB, et al. Recommendations for chamber quantification: a report from the American Society of Echocardiography's Guidelines and Standards Committee and the Chamber Quantification Writing Group, developed in conjunction with the European Association of Echocardiography, a branch of the European Society of Cardiology. J Am Soc Echocardiogr 2005;18:1440–63. [16] Starling MR, Crawford MH, Sorensen SG, et al. Comparative accuracy of apical biplane cross-sectional echocardiography and gated equilibrium radionuclide angiography for estimating left ventricular size and performance. Circulation 1981;63:1075–84. [17] Byrd III BF, Finkbeiner W, Bouchard A, et al. Accuracy and reproducibility of clinically acquired two-dimensional echocardiographic mass measurements. Am Heart J 1989;118:133–7. [18] Gjertsson P, Caidahl K, Bech-Hanssen O. Left ventricular diastolic dysfunction late after aortic valve replacement in patients with aortic stenosis. Am J Cardiol 2005;96:722–7. [19] Gjertsson P, Caidahl K, Farasati M, et al. Preoperative moderate to severe diastolic dysfunction: a novel Doppler echocardiographic long-term prognostic factor in patients with severe aortic stenosis. J Thorac Cardiovasc Surg 2005;129:890–6. [20] Rakowski H, Appleton C, Chan KL, et al. Canadian consensus recommendations for the measurement and reporting of diastolic dysfunction by echocardiography: from the Investigators of Consensus on Diastolic Dysfunction by Echocardiography. J Am Soc Echocardiogr 1996;9:736–60. [21] Alexander JK, Peterson KL. Cardiovascular effects of weight reduction. Circulation 1972;45:310–8. [22] Karason K, Wallentin I, Larsson B, et al. Effects of obesity and weight loss on cardiac function and valvular performance. Obes Res 1998;6:422–9. [23] Collis T, Devereux RB, Roman MJ, et al. Relations of stroke volume and cardiac output to body composition: the strong heart study. Circulation 2001;103:820–5. [24] Hall JE, da Silva AA, do Carmo JM, et al. Obesity-induced hypertension: role of sympathetic nervous system, leptin and melanocortins. J Biol Chem 2010;285: 17271–6. [25] Sowers JR. Obesity as a cardiovascular risk factor. Am J Med 2003;115(Suppl 8A): 37S–41S. [26] Peterson LR, Waggoner AD, Schechtman KB, et al. Alterations in left ventricular structure and function in young healthy obese women: assessment by echocardiography and tissue Doppler imaging. J Am Coll Cardiol 2004;43: 1399–404. [27] Wong CY, O'Moore-Sullivan T, Leano R, et al. Alterations of left ventricular myocardial characteristics associated with obesity. Circulation 2004;110:3081–7. [28] Chinali M, de Simone G, Roman MJ, et al. Impact of obesity on cardiac geometry and function in a population of adolescents: the strong heart study. J Am Coll Cardiol 2006;47:2267–73. [29] Mureddu GF, Greco R, Rosato GF, et al. Relation of insulin resistance to left ventricular hypertrophy and diastolic dysfunction in obesity. Int J Obes Relat Metab Disord 1998;22:363–8. [30] Peterson LR, Herrero P, Schechtman KB, et al. Effect of obesity and insulin resistance on myocardial substrate metabolism and efficiency in young women. Circulation 2004;109:2191–6. [31] Engeli S, Negrel R, Sharma AM. Physiology and pathophysiology of the adipose tissue renin-angiotensin system. Hypertension 2000;35:1270–7. [32] Karason K, Lindroos AK, Stenlof K, et al. Relief of cardiorespiratory symptoms and increased physical activity after surgically induced weight loss: results from the Swedish Obese Subjects study. Arch Intern Med 2000;160:1797–802. [33] Shewan LG, Coats AJ. Ethics in the authorship and publishing of scientific articles. Int J Cardiol 2010;144:1–2.
Contents lists available at ScienceDirect
International Journal of Cardiology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / i j c a r d
Impact of body composition, fat distribution and sustained weight loss on cardiac function in obesity Dimitris Kardassis a,d,⁎, Odd Bech-Hanssen a,d, Marie Schönander b, Lars Sjöström b,d, Max Petzold c, Kristjan Karason a,d a
Department of Cardiology, Sahlgrenska University Hospital, Gothenburg, Sweden Department of Medicine, Sahlgrenska University Hospital, Gothenburg, Sweden Nordic School of Public Health, Gothenburg, Sweden d Sahlgrenska Academy at University of Gothenburg, Gothenburg, Sweden b c
a r t i c l e
i n f o
Article history: Received 20 June 2010 Received in revised form 7 December 2010 Accepted 10 February 2011 Available online 2 March 2011 Keywords: Obesity Weight loss Body composition Fat distribution echocardiography Left ventricular function
a b s t r a c t Background: Obesity is associated with alterations in left ventricular function varying along with the degree of fatness, but the mechanisms underlying this co-variation are not clear. In a case–control study we examined how sustained weight losses affect cardiac function and report on how body composition and fat distribution relate to the left ventricular performance. Methods: At the 10-year follow-up of the Swedish obese subjects (SOS) study cohort we identified 44 patients with sustained weight losses after bariatric surgery (surgery group) and 44 matched obese control patients who remained weight stable (obese group). We also recruited 44 matched normal weight subjects (lean group). Dual-energy X-ray absorptiometry, computed tomography and echocardiography were performed to evaluate body composition, fat distribution and cardiac function. Results: BMI was 42.5 kg/m2, 31.5 kg/m2 and 24.4 kg/m2 for the obese, surgery and lean groups respectively. Increasing degree of obesity was associated with larger left ventricular volumes (p b 0.001), higher cardiac output (p b 0.001), reduced systolic myocardial velocity (p b 0.001) and impaired ventricular relaxation (p = 0.015). In multivariate analyses, left ventricular volume, stroke volume and cardiac output primarily associated with lean body mass, whereas blood pressure, heart rate and variables reflecting cardiac dysfunction were more related to total body fat and visceral adiposity. Conclusion: Obesity is associated with discrete but distinct disturbances in the left ventricular performance appearing to be related to both the total amount of body fat and degree of visceral adiposity. Patients with sustained weight losses display superior left ventricular systolic and diastolic functions as compared with their obese counterparts remaining weight stable. © 2011 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Obesity is associated with increased risk of heart failure, in part due to associated co-morbidities that promote the development of coronary artery disease [1,2]. In addition, obesity has a more direct adverse effect on cardiac function through the rise in hemodynamic load that occurs along with body fat accumulation [3]. As body weight increases, both blood volume and cardiac output rise in order to fulfill the requirements of a higher metabolic rate [3,4]. Although the expansion of circulating blood volume is followed by a lowering of the peripheral vascular resistance this is often insufficient to prevent a rise in blood pressure [5,6]. Consequently, the heart in people with obesity is burdened with both volume and pressure
⁎ Corresponding author at: Department of Cardiology, Sahlgrenska University Hospital, Sweden. Tel.: +46 313427720. E-mail address: [email protected] (D. Kardassis). 0167-5273/$ – see front matter © 2011 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ijcard.2011.02.036
overload, which may lead to various degrees of left ventricular hypertrophy [7]. Such structural changes may in turn interfere with ventricular filling and lead to diastolic dysfunction. In long-standing morbid obesity an impairment of myocardial contractility may also supervene and give rise to overt heart failure, a clinical condition referred to as “obesity cardiomyopathy” [8]. Both lean body mass and adipose tissue increase as obesity develops, but the hemodynamic effects of these body compartments may differ. Furthermore, the distribution of body fat is of importance, since adverse cardiovascular risk is associated with abdominal obesity in particular [9]. In this respect, the separate effects of different body compartments and fat distribution on cardiac function are of interest. The optimal treatment for cardiac dysfunction in obesity would, apparently, be sustained weight loss. However, this is difficult to achieve with conventional methods, which offer results that, for the most part, are only modest and temporary [10]. Bariatric surgery, on the other hand, induces weight losses that are both large and maintained over time [11].
D. Kardassis et al. / International Journal of Cardiology 159 (2012) 128–133
Little is known about the influence of different body compartments and fat distribution on left ventricular performance and, although short-term weight reduction has been reported to improve cardiac function, the effects of sustained weight loss have not been described. The aim of this study was to investigate how body composition and fat distribution relate to variables reflecting left ventricular contractility and filling and to assess the effects of sustained weight loss on systolic and diastolic function.
129
Lean body mass and total body fat were measured with a whole-body dual-energy X-ray absorptiometry (DXA) scanner (DPXL, Lunar Radiation, Madison, WI) using software version 1.35. The coefficient of variance was 0.7% and 1.7% for lean body mass and total body fat, respectively, assessed from double examinations in 10 females. Intra-abdominal and subcutaneous adipose tissue areas were measured at the level of the fourth lumbar vertebra using a single slice Computed Tomography scan (HSA, GE Medical Systems, Milwaukee, WI, version RP2) as previously described [14]. Precision errors, calculated from double determinations, were 1.2% and 0.5% for intra-abdominal and subcutaneous adipose tissues, respectively. 2.3. Echocardiography
2. Methods 2.1. Study groups The SOS study is a prospective, controlled, surgical interventional trial, which enrolled 4047 obese subjects at 25 surgical departments and 480 primary health care centers in Sweden [12]. The study protocol is described in detail elsewhere [11,12]. Briefly, the surgery group comprised 2010 eligible subjects desiring surgery and, at the same time, a matched control group of 2037 obese subjects. Inclusion criteria were age ranging from 37 to 60 years and BMI of ≥34 for men and of ≥38 or more for women. The exclusion criteria, aiming to obtain operable patients, were the same for both study groups. Patients included in the present case control study were recruited consequently during a two-year period (2004–2006) among participants of the SOS study residing in Gothenburg that had been monitored for 10 years. According to pre-specified weight change criteria, we identified 44 surgery patients who, after 10 years, displayed a weight loss of greater than 15% and 44 obese control patients, in which the weight had changed less than 5%. To ensure comparability prior to intervention, the two groups were carefully matched with respect to baseline data from the SOS study. Matching variables included age, gender, BMI, hypertension, hyperlipidemia, diabetes and smoking status (Table 1). In addition, we included 44 healthy normal weight patients, recruited from a randomly selected sample of adults living in the municipality of Mölndal [13], who matched the surgery and obese groups at the 10-year follow-up with respect to age, height and smoking status. In total, 132 subjects were included in the study, comprising 69 women and 63 men with ages ranging from 44 to 71 years. The three study groups underwent a cross-sectional examination with respect to body composition, fat distribution and left ventricular function. The study protocol was approved by the ethics committee at the University of Gothenburg, and all study subjects gave their informed consent to participate.
2.2. Clinical measurements and body composition Measurements of weight and height were obtained and BMI was calculated. Blood pressure was measured in the supine position after 10 min of rest and the mean of two recordings was registered. Blood samples were obtained in the morning after a fast of 12 h and analyzed at the Central Laboratory of Sahlgrenska University Hospital (accredited according to European norm 45001).
Echocardiographic examinations were performed by use of a commercially available ultrasound system. All measurements were performed under resting conditions with the subject in the left lateral decubitus position at end-expiration. Recordings were performed by experienced investigators and analyzed off-line by a single observer blinded with respect to the subject's treatment. All measurements were performed according to recommendations from the American Society of Echocardiography [15]. Two-dimensional measurements of the left ventricular diastolic and systolic volumes were estimated from the apical four- and two chamber view and LV ejection fraction was calculated according to the Simpson's rule [16]. Planimetry of the left and right atriums was performed from a late systolic stop frame showing the maximum atrial size. The left ventricular mass was determined using the truncated ellipsoid model according to Byrd et al. [17]. Blood flow velocity in the left ventricular outflow tract was estimated by pulsewave Doppler from an apical 4-chamber view with a sample size of 5 mm. Stroke volume (SV) was calculated as the product of the cross-sectional area of the left ventricular outflow tract and the velocity time integral. Cardiac output (CO) was estimated by multiplying stroke volume by heart rate (HR). Isovolumetric relaxation time (IVRT) was measured as the time between the aortic valve closure and the mitral valve opening. Mitral flow was recorded between the mitral leaflets in the 4-chamber view. Measurements of early flow velocity (E), E-wave deceleration time (DT), and peak velocity during atrial systole (A) were obtained and the E/A ratio calculated. Pulmonary venous flow velocities were acquired from the upper right pulmonary vein. Peak velocities during systole (S) and diastole (D) were recorded and the S/D ratio calculated. Continuous-wave Doppler was used to measure the peak velocity of tricuspid regurgitation when present. The pressure gradient was calculated according to the simplified Bernoulli equation and used, together with an inferior vena cava estimate of the right atrial pressure, to approximate pulmonary artery pressure. The presence or absence of diastolic dysfunction was evaluated by using an integrated assessment of the left ventricular filling patterns. For each study subject, the expected E/A ratio, DT and S/D ratio were predicted by a regression equation derived from a healthy control population. The observed values in study subjects were regarded as being abnormal if they differed by 1.96 SD from the predicted value using the Z-score [18,19]. Using a previously described criteria [20], subjects were classified to have either normal diastolic function or impaired left ventricular relaxation. Myocardial tissue Doppler imaging was performed from the apical 4-chamber view. Peak systolic (SMV) and early diastolic (Ea) tissue velocities were recorded at the septal corner of the mitral annulus and the E/Ea ratio was calculated.
Table 1 Clinical characteristics of the obese and surgery groups at baseline in the SOS study and changes at a 10-year follow-up. Values are given as means (SD) or absolute numbers (% of total).
BMI, kg/m2 Systolic BP, mm Hg Diastolic BP, mm Hg Cholesterol, mmol/L Triglycerids, mmol/L Glucose, mmol/L Insulin, mU/L No (%) on antihypertensives No (%) with diabetes No (%) current smoker ⁎ p b 0.05 as compared with the obese group. ⁎⁎ p b 0.01 as compared with the obese group. ⁎⁎⁎ p b 0.001 as compared with the obese group.
Baseline change at 10 years Baseline change at 10 years Baseline change at 10 years Baseline change at 10 years Baseline change at 10 years Baseline change at 10 years Baseline change at 10 years Baseline change at 10 years Baseline change at 10 years Baseline change at 10 years
Obese group (N = 44)
Surgery group (N = 44)
41.2 (4.0) + 2.1 (5.4) 139.5 (16.3) − 3.2 (18.0) 87.7 (8.9) − 6.0 (11.0) 6.0 (1.1) − 0.8 (0.9) 2.2 (0.95) − 0.3 (2.0) 5.2 (1.4) + 1.8 (1.9) 19 (9.8) 0.7 (14.8) 5 (11.4) + 25 12 (27.3) +1 14 (31.2) −6
40.0 (4.1) − 9.7 (4.4)⁎⁎⁎ 144.4 (18.0) − 15.1 (18.3)⁎⁎ 93.8 (11.6)⁎ − 16.8 (11.4)⁎⁎ 5.8 (1.9) − 0.8 (1.2) 2.4 (0.9) − 1.1 (0.8) 5.2 (1.8) + 0.4 (1.2)⁎⁎⁎ 19.4 (8.1) − 11.5 (8.0)⁎⁎⁎ 5 (11.4) + 10 12 (27.3) −8 15 (34.1) +1
130
D. Kardassis et al. / International Journal of Cardiology 159 (2012) 128–133
Two-dimensional echocardiographic evaluation of left ventricular volumes can be technically difficult, especially in subjects with obesity. Despite this, satisfactory recordings were available for 64%, 72% and 89% of subjects in the obese, surgery and lean groups, respectively. On the other hand, Doppler measurements were easier to obtain and corresponding numbers for adequate readings were 93%, 98% and 100%. The coefficient of variance, based on double determination in 15 patients, was 11.2% and 5.0% for two-dimensional volumes and Doppler measurements, respectively. 2.4. Statistical analysis Statistical analyses were performed using the SPSS for Windows version 16 (SPSS, Chicago, IL). Descriptive statistical results are given as the mean (standard deviation) or proportions (%). Variables that did not display a normal distribution were transformed logarithmically prior to analysis. Comparisons between obese and surgery groups with respect to SOS baseline data and change from baseline were performed with unpaired t-tests. Overall comparisons between the three study groups with respect to crosssectional data were carried out with ANOVA and post-hoc comparisons between obese vs. surgery groups and surgery vs. lean groups were performed with the Bonferroni test. After pooling data from the three study groups, Pearson's correlation coefficients were calculated in order to estimate associations between body composition and measures of left ventricular performance. Forward stepwise multiple regression analyses were then used to examine how lean body mass, total body fat, visceral adipose tissue and subcutaneous adipose tissue related to indices of the left ventricular function. A P-value of less than 0.05 was considered significant.
3. Results
Table 3 Cross-sectional body composition assessments (DXA and CT at the 4th lumbar vertebra) in the three study groups. Values are given as means (SD). Obese group (n = 44)
Surgery group (n = 44)
Lean group (n = 44)
Lean body mass, 61.7 (11.2) 52.5 (11.6)⁎ 49.9 (10) kg (DXA) ⁎ Total body fat, kg (DXA) 52 (11) 34.8 (11) 19.6 (7)# Visceral adipose tissue 313.2 (121.4) 169.4 (79.4)⁎ 112.5 (50.6)# area cm2 (CT) 623.7 (201) 394 (144.6)⁎ 200.9 (79.9)# Subcutaneus adipose tissue area cm2 (CT)
P-value b0.001 b0.001 b0.001 b0.001
Multiple comparisons were adjusted for according to the Bonferoni procedure. ⁎ p b 0.001 as compared with the obese group. # p b 0.001 as compared with the surgery group.
higher HDL than the obese group. Surgery patients smoked more often but had less diabetes and hypertensive treatment than the obese patients. As compared with the surgery group, the lean group showed lower body weight, BMI, glucose and insulin. None of the lean patients had diabetes or were treated with antihypertensive medication. The smoking frequency was comparable in the two groups.
3.1. Clinical characteristics Table 1 shows baseline data from the SOS study by means of which the obese and surgery groups were matched and changes at a 10-year follow-up. At baseline the two groups were quite comparable with respect to clinical parameters apart from diastolic blood pressure, which was slightly higher in the surgery group. At SOS 10-year followup the surgery group displayed substantial reductions in BMI and significant improvements in several clinical variables as compared with the obese group. Table 2 shows cross-sectional clinical characteristics of the obese and surgery groups after 10-years of follow-up in the SOS study and for the matched lean group. The lean group was somewhat younger than the obese and surgery group but of comparable height. As compared with the obese group, the surgery group showed lower body weight, body mass index (BMI) and diastolic blood pressure. The surgery group also had lower triglycerides, glucose and insulin and Table 2 Cross-sectional clinical characteristics for the obese, surgery and lean groups. Values are given as means (SD) or absolute numbers (% of total).
Gender, male/female Age, years Height, cm Weight, kg BMI, kg/m2 Systolic BP, mm Hg Diastolic BP, mm Hg Cholesterol, mmol/L Triglycerids, mmol/L Glucose, mmol/L Insulin, mU/L No (%) on antihypertensives No (%) with diabetes No (%) current smoker
Obese group (N = 44)
Surgery group (N = 44)
Lean group (N = 44)
P-value
21/23 58.9 (4.7) 1.71 (0.1) 127.1 (17.8) 42.5 (6.1) 136.3 (17.7) 81.7 (11.7) 5.2 (1.1) 1.9 (1.3) 7.0 (2.3) 19.7 (14) 30 (68)
21/23 59.6 (5.1) 1.73 (0.1) 90.6 (13)⁎⁎⁎ 31.5 (4.9)⁎⁎⁎ 129.3 (17.1) 77.0 (8.7)⁎
21/23 55 (5.3)# 1.73 (0.1) 73.0 (10)### 24.4 (3.7)### 122.9 (20.3) 76.7 (10.9) 5.4 (0.85) 1.5 (1.1) 5.0 (0.6)# 6.4 (3.5)## 0###
0.923 b0.001 0.157 b0.001 b0.001 0.003 0.021 0.237 0.062 b0.001 b0.001 b0.001
0
b0.001 0.104
13 (29.5) 8 (18.2)
5.0 (1.1) 1.3 (0.5)⁎⁎ 5.6 (1.4)⁎⁎ 7.9 (3.7)⁎⁎⁎ 15 (34)⁎⁎ 4 (9)⁎ 16 (36.4)
11 (25)
Multiple comparisons were adjusted for according to the Bonferoni procedure. ⁎ p b 0.05 as compared with the obese group. ⁎⁎ p b 0.01 as compared with the obese group. ⁎⁎⁎ p b 0.001 as compared with the obese group. # p b 0.05 as compared with the surgery group. ## p b 0.01 as compared with the surgery group. ### p b 0.001 as compared with the surgery group.
3.2. Body composition Cross-sectional measurements of lean body mass and total body fat with DXA and of intra-abdominal and subcutaneous adipose tissue area at the level of L4 with CT are shown in Table 3. All body
Table 4 Cross-sectional echocardiographic measurements in the three study groups. Values are given as means (SD) or absolute numbers.
Structure LVM (g) LVM/BSA (g/m2) LVEDV (mL) LVEDV/BSA (mL/m2) LVESV(mL) LVESV/BSA (mL/m2) Hemodynamics Stroke volume (mL) Heart rate (bpm) Cardiac output (L/min) Pulmonary pressure (mm Hg) Systolic function Left ventricular ejection fraction (%) SMV (cm/s) Diastolic function Normal/abnormal (n)
Obese group (n = 44)
Surgery group (n = 44)
201.4 87.9 113.8 48.5 42.5 17.8
(21.6) (11.8) (23.8) (8.9) (14.1) (4.7)
157.7 79.1 87.0 43.8 31.0 15.5
(20.5)⁎⁎ (10.9)⁎⁎ (12.2)⁎⁎⁎ (6.0) (7.8)⁎⁎
88.3 (1.6) 72.8 (11.6) 6.4 (1.3) 24.3 (3)
83.1 64.5 5.3 20.9
(1.5) (9.0)⁎⁎⁎ (0.8)⁎⁎⁎ (3)⁎⁎
62.5 (6.2) 9.3 (1.6)
37/7
Lean group (n = 44)
P-value
(21.7)# (11.4)# (19.3) (9.1) (10.1) (4.8)
b0.001 b0.001 b0.001 0.173 0.001 0.257
(1.5) (9.8) (1.0) (4)
0.158 b0.001 b0.001 b0.001
64.6 (5.6)
66.0 (6.2)
0.019
10.6 (1.0)⁎⁎
11.2 (1.7)
b0.001
44/0⁎
(3.5)
133.9 71.8 85.3 46.1 29.8 16.1
80.0 65.0 5.2 20.7
44/0
b0.001
Multiple comparisons were adjusted for according to the Bonferoni procedure. LVM = left ventricular mass; BSA = body surface area; LVEDV = left ventricular end diastolic volume; LVESV = left ventricular end systolic volume; SMV = peak systolic velocity at the basal septal segment; E/A = ratio of early (E) to late (A) peak diastolic transmitral flow velocity; S/D = ratio of peak pulmonary venous flow velocity during ventricular systole (S) to peak pulmonary venous flow velocity during ventricular diastole (D); E/Ea = ratio of mitral early peak velocity (E) to mitral annulus early peak velocity (Ea). ⁎ p b 0.05 as compared with the obese group. ⁎⁎ p b 0.01 as compared with the obese group. ⁎⁎⁎ p b 0.001 as compared with the obese group. # p b 0.05 as compared with the surgery group.
D. Kardassis et al. / International Journal of Cardiology 159 (2012) 128–133
composition parameters decreased significantly across the groups in the order from obese to surgery to lean groups. 3.3. Echocardiography Cross-sectional echocardiography measurements are shown in Table 4. Left ventricular mass, and volumes decreased across the three groups from obese to surgery to lean. These structural measures were significantly different in the obese group as compared with the surgery group and, except for left ventricular volumes, in the surgery group as compared with the lean group. Heart rate, cardiac output and pulmonary pressures also decreased across the groups in a similar manner with significant differences between the obese group and the surgery group. Stroke volume did not show a significant linear trend across the three groups. Left ventricular ejection fraction and systolic myocardial velocity increased with decreasing BMI, and systolic myocardial velocity was significantly different in the obese group as compared with the surgery group. Evaluation on the basis of integrated transmitral and pulmonary venous flow velocity revealed diastolic dysfunction in 16% of the obese group (all mild), whereas none of the patients in the surgery or lean group displayed abnormal filling patterns. Other Dopplerechocardiographic measurements of diastolic function are shown in Fig. 1. E/A ratio increased, while S/D ratio, E/Ea ratio, IVRT and deceleration time decreased along with a declining degree of obesity with significant differences between the obese group and the surgery group. Furthermore, the left atrial area decreased significantly across the three groups. 3.4. Association between body composition and left ventricular function Correlation coefficients and stepwise multiple regression analysis for associations between body composition and selected left ventricular measurements can be seen in Tables 5 and 6. Left ventricular enddiastolic volume, stroke volume and cardiac output were positively and independently related to lean body mass, whereas heart rate,
131
blood pressure and IVRT were positively and independently associated with the visceral adipose tissue. A lower E/A ratio, higher S/D ratio and longer deceleration time were independently associated with total body fat, while reduced SMV and higher E/Ea ratio were independently related to both total body fat and visceral adipose tissue. 4. Discussion In the present study, we confirm that obesity is associated with discrete but distinct disturbances in the left ventricular systolic and diastolic performances. Cardiac dysfunction appears to be related to both the total amount of body fat and the degree of visceral adiposity. Persistent weight loss, on the other hand, was associated with increased myocardial contractility and enhanced left ventricular relaxation. Our findings indicate that long-term sustained weight loss in obesity is beneficial for the left ventricular systolic and diastolic functions. 4.1. Hemodynamics It is well documented that obesity is associated with an augmentation in blood flow in order to meet the requirements of increased metabolic rate [3,4]. In agreement, we observed increased cardiac output across the study groups in the order from lean to surgery to obese. Previously, weight-related variations in cardiac output have been assumed to be linked to changes in stroke volume [5,21], but in the present study obesity-related rise in blood flow was also due to a higher heart rate. We therefore propose that adaptation of cardiac output to variations in body fatness is mediated, not only by altered stroke volume, but also by changes in heart rate. In fact, previous studies have shown that sympathetic nervous system activity may rise along with increasing obesity and thus contribute to a higher heart rate [22]. Left ventricular volume, stroke volume and cardiac output, which are crude markers of preload, were primarily associated with lean body mass and similar observations have been reported by Collis et al. [23]. Thus, it appears that a variation in preload is more dependent on the obesity-associated increase in lean body mass than on the actual
P<0.001
P=0.002 P=0.015
*
***
Surgery
Lean
P<0.001
IVRT ms
**
Obese
Surgery
Lean
E/Ea ratio Obese
Surgery
Lean
Obese
Surgery
Lean
P<0.001
P<0.001
Deceleration time (ms)
Obese
**
Obese
**
Left atrial area (cm2)
E/A ratio
S/D ratio
*
Surgery
Lean
***
Obese
Surgery
Lean
Fig. 1. Scatter plots with diamond means (group mean and 95% CI) for echocardiographic measures of the diastolic function by study group. Results from overall comparisons with ANOVA (P-value) and post hoc analyses with Bonferroni (asterisk) are shown. *P b 0.05, **P b 0.01, ***P b 0.001.
132
D. Kardassis et al. / International Journal of Cardiology 159 (2012) 128–133
Table 5 Pearson's correlation coefficients (r) and multivariate stepwise linear regression analyses of hemodynamic and systolic function on body composition. LVEDV
Stroke volume
Heart rate
Cardiac output
Systolic BP
Systolic mean velocity
Correlation
r
P
r
P
r
P
r
P
r
P
r
P
Lean body mass Total body fat Visceral adipose tissue Subcutaneous adipose tissue
0.49 − 0.03 0.26 0.164
b 0.001 0.755 0.013 0.120
0.36 − 0.14 − 0.02 − 0.01
0.001 0.240 0.858 0.931
0.01 0.35 0.39 0.33
0.913 b 0.001 b 0.001 b 0.001
0.33 0.17 0.25 0.23
0.003 0.154 0.021 0.036
0.05 0.29 0.35 0.29
0.554 b 0.001 b 0.001 b 0.001
0.06 − 0.43 − 0.31 − 0.37
0.913 b0.001 b0.001 b0.001
Multiple regression
β
P
β
P
β
P
β
P
β
P
β
P
b0.001
− 0.18
0.024
0.03
0.007 − 0.01
b0.001
3.08
b 0.001
3.64
b 0.001
Lean body mass Total body fat Visceral adipose tissue Subcutaneous adipose tissue Adjusted R2 (%)
b 0.001
1.18
0.62
47
32
19
13
12
20
LVEDV = left ventricular end-diastolic volume; BP = blood pressure.
body fat accumulation. Systolic blood pressure, a simple estimate of afterload, and heart rate were, on the other hand, mainly related to the extent of visceral adipose tissue. The disparate influence of separate body compartments on loading conditions and heart rate is of interest. While lean body mass appears to be a main determinant of blood flow, visceral obesity may contribute to a rise in blood pressure by interfering with regulation of heart rate and vascular resistance. Indeed, paracrine and hormonal factors derived from adipose tissues have been suggested to play a role in the pathophysiology of obesityrelated hypertension [24,25].
4.2. Systolic function Although overt systolic heart failure in prolonged morbid obesity has been reported, this appears to be relatively uncommon. In the present study, there was a trend towards lower left ventricular ejection fraction with increasing degree of obesity, but post hoc analyses did not reveal significant differences between study groups. On the other hand, systolic myocardial tissue velocity was significantly lower in the obese group as compared with the surgery group. Further, this measure of contractility was inversely related to the extent of body fat and visceral adiposity independent of other variables. Our findings are consistent with previous studies on otherwise healthy obese subjects [26,27], suggesting that subclinical systolic dysfunction may be a prevalent condition in obesity. Still, it should be emphasized that loading conditions can influence measures of systolic function, although tissue Doppler imaging is probably less affected than other estimates of myocardial contractility.
4.3. Diastolic function An integrated evaluation of the left ventricular filling patterns revealed definite relaxation disturbances in 16% of the obese group, whereas none of the patients in the surgery or lean group displayed overt diastolic dysfunction. Although mean values for variables describing left ventricular filling were within normal ranges for all three study groups, the patterns were consistently less favorable in the obese group as compared to the surgery group, which in turn did not differ from the lean group. In addition, the obese group displayed a larger left atrium and a higher pulmonary artery pressure, also suggesting impaired left ventricular filling. Our findings are in line with previous studies reporting a correlation between obesity and disturbances in the left ventricular diastolic function [26,28]. Moreover, we observed that adverse filling patterns were, aside from total body fat, independently related to the extent of visceral adiposity. This suggests a connection between diastolic dysfunction and the metabolic syndrome, possibly mediated by related hemodynamic, metabolic and hormonal aberrations interfering with ventricular relaxation [29–31]. 4.4. Weight loss Short-term weight loss has been reported to have favorable effects on myocardial contractility and ventricular filling [22] and our present observations suggest that such improvements persist long-term following bariatric surgery. In fact, the surgery group did not differ significantly from the lean group with respect to variables describing systolic and diastolic functions. In a previous report from the SOS
Table 6 Pearson's correlation coefficients (r) and multivariate stepwise linear regression analyses of variables of left ventricular function on body composition. Deceleration time
Left atrial area
Correlation
E/A ratio r
P
r
P
r
P
r
P
r
P
r
P
Lean body mass Total body fat Visceral adipose tissue Subcutaneous adipose tissue
− 0.02 − 0.34 − 0.31 − 0.21
0.829 b0.001 b0.001 0.018*
0.18 0.24 0.20 0.18
0.05 b0.006 b0.029 0.049
0.21 0.42 0.44 0.42
0.023 b0.001 b0.001 b0.001
0.16 0.34 0.34 0.11
0.116 b 0.001 b 0.001 0.251
0.19 0.33 0.32 0.20
0.040 b0.001 b0.001 0.027
0.56 0.33 0.31 0.27
b0.001 b0.001 b0.001 0.002
Multiple regression
β
P
β
P
β
P
β
P
β
P
β
P
b0.001
0.29 0.26 − 1.49
b0.001 b0.001 0.017
Lean body mass Total body fat Visceral adipose tissue Subcutaneous adipose tissue Adjusted R2 (%)
− 0.01
14
S/D ratio
b0.001
0.01
8
E/Ea ratio
0.007
− 0.05 0.96 20
E/A ratio = ratio of early (E) to late (A) peak diastolic transmitral flow velocity. S/D ratio = ratio of peak systolic (S) to diastolic (D) pulmonary flow velocity. E/Ea = ratio of mitral early peak velocity (E) to mitral annulus early peak velocity (Ea). IVRT = isovolumetric relaxation time.
IVRT
b0.001 0.043
0.31 0.56 − 4.17 19
b 0.001 0.031 13
43
D. Kardassis et al. / International Journal of Cardiology 159 (2012) 128–133
study [32], weight loss in obese subjects was associated with a marked relief in breathlessness and increased physical activity, which could, in part, be related to improvement in cardiac function. However, whether sustained weight loss may reduce the risk of overt cardiac failure is still unknown. Further, the feasibility of surgical obesityintervention, as a treatment option for chronic heart failure, remains to be determined. 4.5. Limitations The main limitation of the present study was that echocardiographic examinations for the obese and surgery groups were not performed prior to inclusion in the SOS study. This weakens our conclusions with respect to the effects of weight loss on the left ventricular function. On the other hand, these two groups were carefully matched with respect to important SOS baseline characteristics, resulting in two almost identical groups prior to intervention. It is therefore reasonable to assume that the left ventricular function did not differ between obese and surgery groups at the SOS baseline. Further, our study is unique with respect to the long-term follow-up of obese subjects with large and sustained weight losses. Such patient populations have rarely been studied previously due to the difficulties in attaining reliable long-term weight reductions with conventional therapy. 4.6. Conclusion Obesity is associated with discrete but distinct disturbances in the left ventricular systolic and diastolic performances. Whereas obesityrelated increase in lean body mass appears to be the main determinant of stroke volume and cardiac output, accumulation of body fat is more related to disturbances in ventricular contractility and relaxation. Visceral adiposity may further impair the left ventricular function by interfering with heart rate and blood pressure. Surgical obesity intervention appears to have long-term favorable effects on the left ventricular systolic and diastolic functions and could be considered as a treatment option in chronic heart failure. Acknowledgment The authors of this manuscript have certified that they comply with the Principles of Ethical Publishing in the International Journal of Cardiology [33]. References [1] Gordon T, Kannel WB. Obesity and cardiovascular diseases: the Framingham study. Clin Endocrinol Metab 1976;5:367–75. [2] Haslam DW, James WP. Obesity. Lancet 2005;366:1197–209. [3] de Divitiis O, Fazio S, Petitto M, et al. Obesity and cardiac function. Circulation 1981;64:477–82. [4] Vasan RS. Cardiac function and obesity. Heart (British Cardiac Society) 2003;89: 1127–9. [5] Alexander JK, Dennis EW, Smith WG, et al. Blood volume, cardiac output, and distribution of systemic blood flow in extreme obesity. Cardiovasc Res Cent Bull 1962;1:39–44. [6] Lauer MS, Anderson KM, Levy D. Separate and joint influences of obesity and mild hypertension on left ventricular mass and geometry: the Framingham heart study. J Am Coll Cardiol 1992;19:130–4. [7] Messerli FH, Sundgaard-Riise K, Reisin ED, et al. Dimorphic cardiac adaptation to obesity and arterial hypertension. Ann Intern Med 1983;99:757–61.
133
[8] Alexander JK. The cardiomyopathy of obesity. Prog Cardiovasc Dis 1985;27: 325–34. [9] Eckel RH, Grundy SM, Zimmet PZ. The metabolic syndrome. Lancet 2005;365: 1415–28. [10] de las Fuentes L, Waggoner AD, Mohammed BS, et al. Effect of moderate diet-induced weight loss and weight regain on cardiovascular structure and function. J Am Coll Cardiol 2009;54:2376–81. [11] Sjostrom L, Narbro K, Sjostrom CD, et al. Effects of bariatric surgery on mortality in Swedish obese subjects. N Engl J Med 2007;357:741–52. [12] Sjostrom L, Larsson B, Backman L, et al. Swedish obese subjects (SOS). Recruitment for an intervention study and a selected description of the obese state. Int J Obes Relat Metab Disord 1992;16:465–79. [13] Larsson I, Berteus Forslund H, Lindroos AK, et al. Body composition in the SOS (Swedish Obese Subjects) reference study. Int J Obes Relat Metab Disord 2004;28: 1317–24. [14] Chowdhury B, Sjostrom L, Alpsten M, et al. A multicompartment body composition technique based on computerized tomography. Int J Obes Relat Metab Disord 1994;18:219–34. [15] Lang RM, Bierig M, Devereux RB, et al. Recommendations for chamber quantification: a report from the American Society of Echocardiography's Guidelines and Standards Committee and the Chamber Quantification Writing Group, developed in conjunction with the European Association of Echocardiography, a branch of the European Society of Cardiology. J Am Soc Echocardiogr 2005;18:1440–63. [16] Starling MR, Crawford MH, Sorensen SG, et al. Comparative accuracy of apical biplane cross-sectional echocardiography and gated equilibrium radionuclide angiography for estimating left ventricular size and performance. Circulation 1981;63:1075–84. [17] Byrd III BF, Finkbeiner W, Bouchard A, et al. Accuracy and reproducibility of clinically acquired two-dimensional echocardiographic mass measurements. Am Heart J 1989;118:133–7. [18] Gjertsson P, Caidahl K, Bech-Hanssen O. Left ventricular diastolic dysfunction late after aortic valve replacement in patients with aortic stenosis. Am J Cardiol 2005;96:722–7. [19] Gjertsson P, Caidahl K, Farasati M, et al. Preoperative moderate to severe diastolic dysfunction: a novel Doppler echocardiographic long-term prognostic factor in patients with severe aortic stenosis. J Thorac Cardiovasc Surg 2005;129:890–6. [20] Rakowski H, Appleton C, Chan KL, et al. Canadian consensus recommendations for the measurement and reporting of diastolic dysfunction by echocardiography: from the Investigators of Consensus on Diastolic Dysfunction by Echocardiography. J Am Soc Echocardiogr 1996;9:736–60. [21] Alexander JK, Peterson KL. Cardiovascular effects of weight reduction. Circulation 1972;45:310–8. [22] Karason K, Wallentin I, Larsson B, et al. Effects of obesity and weight loss on cardiac function and valvular performance. Obes Res 1998;6:422–9. [23] Collis T, Devereux RB, Roman MJ, et al. Relations of stroke volume and cardiac output to body composition: the strong heart study. Circulation 2001;103:820–5. [24] Hall JE, da Silva AA, do Carmo JM, et al. Obesity-induced hypertension: role of sympathetic nervous system, leptin and melanocortins. J Biol Chem 2010;285: 17271–6. [25] Sowers JR. Obesity as a cardiovascular risk factor. Am J Med 2003;115(Suppl 8A): 37S–41S. [26] Peterson LR, Waggoner AD, Schechtman KB, et al. Alterations in left ventricular structure and function in young healthy obese women: assessment by echocardiography and tissue Doppler imaging. J Am Coll Cardiol 2004;43: 1399–404. [27] Wong CY, O'Moore-Sullivan T, Leano R, et al. Alterations of left ventricular myocardial characteristics associated with obesity. Circulation 2004;110:3081–7. [28] Chinali M, de Simone G, Roman MJ, et al. Impact of obesity on cardiac geometry and function in a population of adolescents: the strong heart study. J Am Coll Cardiol 2006;47:2267–73. [29] Mureddu GF, Greco R, Rosato GF, et al. Relation of insulin resistance to left ventricular hypertrophy and diastolic dysfunction in obesity. Int J Obes Relat Metab Disord 1998;22:363–8. [30] Peterson LR, Herrero P, Schechtman KB, et al. Effect of obesity and insulin resistance on myocardial substrate metabolism and efficiency in young women. Circulation 2004;109:2191–6. [31] Engeli S, Negrel R, Sharma AM. Physiology and pathophysiology of the adipose tissue renin-angiotensin system. Hypertension 2000;35:1270–7. [32] Karason K, Lindroos AK, Stenlof K, et al. Relief of cardiorespiratory symptoms and increased physical activity after surgically induced weight loss: results from the Swedish Obese Subjects study. Arch Intern Med 2000;160:1797–802. [33] Shewan LG, Coats AJ. Ethics in the authorship and publishing of scientific articles. Int J Cardiol 2010;144:1–2.