Arterial Destiffening With Weight Loss in Overweight and Obese Middle-Aged and Older Adults
We tested the hypothesis that weight loss via a hypocaloric diet would reduce arterial stiffness in overweight and obese middle-aged and older adults. Thirty-six individuals were randomly assigned to a weight loss (n=25; age: 61.2±0.8 years; body mass index: 30.0±0.6 kg/m2) or a control (n=11; age: 66.1±1.9 years; body mass index: 31.8±1.4 kg/m2) group. Arterial stiffness was measured via carotid artery ultrasonography combined with applanation tonometry and carotid-femoral pulse wave velocity via applanation tonometry at baseline and after the 12-week intervention. Body weight, body fat, abdominal adiposity, blood pressure, β-stiffness index, and carotid-femoral pulse wave velocity were similar in the 2 groups at baseline (all P>0.05). Body weight (−7.1±0.7 versus −0.7±1.1 kg), body fat, and abdominal adiposity decreased in the weight loss group but not in the control group (all P<0.05). Brachial systolic and diastolic blood pressures declined (P<0.05) only in the weight loss group. Central systolic and pulse pressures did not change significantly in either group. β-Stiffness index (−1.24±0.22 versus 0.52±0.37 U) and carotid-femoral pulse wave velocity (−187±29 versus 15±42 cm/s) decreased in the weight loss group but not in the control group (all P<0.05). The reductions in carotid-femoral pulse wave velocity were correlated with reductions in total body and abdominal adiposity (r=0.357–0.602; all P<0.05). However, neither total body nor abdominal adiposity independently predicted reductions in arterial stiffness indices. In summary, our findings indicate that weight loss reduces arterial stiffness in overweight/obese middle-aged and older adults, and the magnitudes of these improvements are related to the loss of total and abdominal adiposity.
Advancing age is associated with stiffening of the large elastic arteries of the cardiothoracic region.1,2 Total body and abdominal adiposity increase with advancing age,3 and excess fat accumulation, particularly in the abdominal visceral region, is associated with accelerated large artery stiffening in middle-aged and older adults.4,5 Importantly, large artery stiffening contributes to the age-related rise in systolic blood pressure (BP; SBP)6 and is an independent predictor of total and cardiovascular mortality among older adults.7
The results of several studies suggest that weight loss may be efficacious in reducing large artery stiffness in the cardiothoracic region.8–12 To date, only 1 randomized, controlled trial has been conducted to address this issue.8 Balkestein et al8 reported that weight loss increased carotid artery distensibility and that exercise did not result in an additive effect. However, the lack of an adequate control group, small sample size, and focus on primarily young and middle-aged males limits generalizability and precludes a clear understanding of the impact of weight loss on arterial stiffness. This is a critical void given that weight loss is the cornerstone of obesity management and there are currently few strategies available for reducing arterial stiffness. Accordingly, we tested the hypothesis that weight loss via a hypocaloric diet alone would reduce arterial stiffness in overweight and obese middle-aged and older adults. We further hypothesized that the reduction in arterial stiffness with weight loss, if observed, would be associated with the magnitude of reduction in total body or abdominal adiposity.
Thirty-six men (n=15) and women (n=21) 55 to 75 years of age volunteered to participate in the study. All of the subjects were sedentary to recreationally active and free of overt disease. None of the subjects were smokers or taking medications that affect body weight or appetite. The Virginia Polytechnic Institute and State University Institutional Review Board approved the protocol. The nature, purpose, risks, and benefits were explained to each subject before obtaining informed consent.
After baseline testing, subjects were randomly assigned to a weight loss (n=25) intervention or a control group (n=11). The subjects randomized to the weight loss intervention followed a hypocaloric diet (1200 to 1500 kcal) based on the US Department of Agriculture food guide pyramid guidelines10 and were instructed to maintain their habitual physical activity level. The control group was instructed to maintain their current body weight, habitual physical activity level, and dietary intake. During the 12-week weight loss intervention period, subjects met with a dietitian and had their body weight measured weekly. Each subject was weight stable for ≥2 weeks before follow-up testing. All of the measurements were performed between 8:00 am and 11:00 am after a 12-hour fast and having performed no vigorous physical activity for the previous 48 hours. All of the subjects reported being free of acute illness during the week before testing.
Body weight was measured to the nearest 0.1 kg on a digital scale (Scale-Tronix model 5002). Height was measured using a stadiometer. Waist circumference was measured at the umbilical level with a spring-loaded Gulick measuring tape. Body composition was determined via dual energy x-ray absorptiometry (GE Lunar Prodigy Advance, software version 8.10e). Abdominal fat distribution was measured using computed tomography (HiSpeed CT/I, GE Medical), as described previously.13 Total, subcutaneous, and visceral fat areas were quantified using commercially available software (SliceOmatic, version 4.3, Tomovision). Resting heart rate was obtained from lead II of an ECG. Habitual dietary intake and physical activity were assessed via self-reported 4-day food intake records and accelerometry (GT1M, Actigraph Inc), respectively. Energy and macronutrient intake were assessed using nutritional analysis software (NDS-R 6.0, University of Minnesota). Plasma lipid and lipoprotein concentrations were measured in a commercial laboratory using conventional methods. Plasma glucose concentration was measured using a YSI glucose analyzer (model 2300, Yellow Springs Instruments). Plasma insulin concentration was quantified using a commercially available ELISA kit (Linco Research, Inc).
β-Stiffness index (β-SI)14 was measured as described previously.15,16 Briefly, left common carotid artery diameters were obtained 1 to 2 cm from the carotid bulb with an ultrasound unit equipped with a high-resolution linear array transducer (3 to 11 MHz). Systolic and diastolic carotid diameters were quantified offline using commercially available software (Vascular Research Tools 5, Medical Imaging Applications, LLC). Carotid waveforms were obtained using applanation tonometry of the contralateral common carotid artery and calibrated to brachial diastolic BP (DBP) and mean arterial BP obtained by automated sphygmomanometry in the supine posture. Brachial pulse pressure (PP) was calculated as the difference between SBP and DBP. Central SBP and PP were obtained from the peak and the difference between the peak and nadir of the calibrated waveform, respectively. Amplifications of SBP and PP were calculated as the difference between central and brachial SBP and PP, respectively.
Carotid-femoral (C-F) pulse wave velocity (PWV) measurements were obtained after 20 minutes of quiet rest in the supine position, as described previously.16 Pulse waveforms were obtained via applanation tonometry (Probe SPT-301, Millar Instruments) and recorded simultaneously at the right C-F arteries. The linear distance between the carotid and femoral arteries at the highest point on the patient between the recording sites was measured with a tape measure to the nearest 0.5 cm. The C-F distance did not change (P>0.05) with the weight loss intervention, and the baseline and postinterventions distances were highly correlated (Spearman ρ=0.886; P<0.05). In addition, the effects of weight loss on C-F PWV were virtually identical when the postintervention distance was substituted for the baseline distance in the calculation of baseline C-F PWV (data not shown). C-F recordings of waveforms over 10 to 20 cardiac cycles were analyzed using signal processing software (Windaq, Dataq Instruments). PWV was calculated by dividing travel distance by travel time from foot to foot of the pulse waves. Although C-F PWV is a close surrogate for aortic PWV, we also report estimated aortic PWV using the equation developed by Vermeersch et al17 for comparison.
Independent sample t tests were used to compare subject characteristics and dependent variables at baseline in the weight loss and control groups. χ2 analysis was used to compare the overall frequency of medication use between the 2 groups. Repeated-measures ANOVA was used to compare changes in subject characteristics and dependent variables over time between the 2 groups. Our study was not powered to test sex differences. As such, the pooled data are presented. Simple correlations were performed to assess relations among variables of interest. Linear regression was used to test the independent effects of measures of total body and abdominal adiposity on the reduction in arterial stiffness (C-F PWV). Furthermore, a test of mediation using linear regression was used to determine whether the change in C-F PWV with treatment was mediated by the reduction in mean BP.18 Mediation would be supported if weight loss significantly predicted the change in C-F PWV, weight loss significantly predicted the change in mean BP, and mean BP significantly predicted C-F PWV. This approach was also used to determine whether the change in C-F PWV with treatment was mediated by the reduction in sodium intake. All of the data are expressed as mean±SE. The significance level was set a priori at the P<0.05 level.
Subject characteristics at baseline and after the intervention are shown in Table 1. There were no significant differences in body weight, body composition, abdominal fat distribution, BP, or lipid and lipoprotein concentrations between the 2 groups at baseline. Plasma glucose and insulin concentrations were also similar (P>0.05) in the 2 groups. Body weight decreased (−7.1±0.7 versus −0.7±0.4 kg; P<0.05) after the intervention in the weight loss compared with the control group because of reductions in both body fat (≤4.6±0.6 versus −0.1±0.4 kg; P<0.05) and fat free mass (−1.5±0.3 versus 0.1±0.4 kg; P<0.05; Table 1). Total abdominal fat decreased (−101±14 versus −8±13 cm2; P<0.05) in the weight loss group because of reductions in abdominal subcutaneous (−55±9 versus −9±14 cm2; P<0.05) and abdominal visceral fat (−44±8 versus −3±7 cm2; P<0.05). Brachial SBP and DBP, as well as heart rate, decreased (P<0.05) in the weight loss group, but no significant changes were observed in the control group (Table 1). Central SBPs and PPs declined in both groups, but the magnitude of reduction did not differ significantly. SBP and PP amplification did not change in either group (data not shown). In addition, total cholesterol, low-density lipoprotein cholesterol, and triglycerides decreased only in the weight loss group (all P<0.05). High-density lipoprotein cholesterol did not change (P>0.05) in either group. Glucose (P<0.05) and insulin (P=0.05) concentrations decreased in the weight loss group but not in the control group.
Medication use in the 2 groups is shown in Table 2. All of the individuals had been on their current regiment for ≥8 months, and no changes were made during the weight loss intervention. None of the individuals were taking >1 medication. The frequency of medication use did not differ between the 2 groups (P>0.05).
Habitual physical activity and dietary intake at baseline and after the intervention are shown in Table 3. Habitual physical activity and dietary intake were similar (P>0.05) at baseline in the weight loss and control groups. There was no significant change in habitual physical activity after the intervention in either group. Energy intake decreased (P<0.05) after the intervention in the weight loss group but not the control group. There was no significant reduction in the percentage of fat or carbohydrate intake after the intervention. However, the percentage of protein intake increased (P<0.05) in the weight loss group but not the control group. Saturated, monounsaturated, polyunsaturated, and trans-fatty acid intake, as well as cholesterol intake, declined (all P<0.05) in the weight loss group, but there were no such significant changes observed in the control group after the intervention. Alcohol intake did not change (P<0.05) after the intervention in either group. Sodium and potassium intakes decreased (P<0.05) after the intervention in both groups; the reduction tended to be greater in the weight loss group (both P<0.05). Magnesium intake did not change (P>0.05) after the intervention in either group.
β-Stiffness index, arterial compliance, and C-F PWV (and aortic PWV) at baseline and after the intervention are shown in Table 4. There were no differences in these arterial stiffness indices between the 2 groups at baseline. The reductions in β-SI (−1.24±0.22 versus 0.52±0.37 U; Figure 1A) and C-F PWV (-187±29 versus 15±42 cm/s; Figure 1C) and increases in arterial compliance (0.0125±0.0038 versus −0.0056±0.0061 mm2/mm Hg ×10−1; Figure 1B) after the intervention were greater (all P<0.05) in the weight loss compared with the control group, respectively.
In the pooled sample, the magnitude of change in β-SI was correlated with the percentage of initial weight loss (Figure 2A), change in body mass index, and change body fat percentage (Table 5). The magnitude of change in C-F PWV was correlated with the percentage of initial weight loss (Figure 2B), the magnitude of reduction in visceral fat (Figure 2C) and waist circumference (Figure 2D), and the change in the absolute amount of weight loss, body mass index, body fat percentage, total fat mass, total abdominal fat, and subcutaneous abdominal fat (Table 5). However, linear regression analysis revealed that only the changes in body mass index (model 1: β=68.4; P<0.05), total weight loss (model 2: β=21.6; P<0.05), percentage of weight loss (model 3: β=19.4; P<0.05), and total fat loss (model 4: β=23.6; P<0.05) independently predicted the reduction in C-F PWV when indices of total and abdominal adiposity were considered for inclusion. In addition, the magnitude of reduction in C-F PWV with weight loss was correlated with the changes in triglyceride, total cholesterol, and low-density lipoprotein cholesterol concentrations (Table 5). The magnitude of reduction in arterial compliance was correlated with the change in fat free mass. There were no other significant correlates in the magnitudes of reduction in β-SI, arterial compliance, or C-F PWV.
In an attempt to gain further insight into the potential impact of BP lowering, we compared β-SI and C-F PWV in subjects above and below the median reduction in mean BP with weight loss (n=12 per group for β-SI and n=11 per group for C-F PWV). The median reduction in mean BP for the weight loss group was −7 mm Hg. The reductions in β-SI (−1.20±0.28 versus −1.29±0.41 U) and C-F PWV (−198±43 versus −175±50 cm/s) were similar in individuals with larger (−11±1 versus −4±1 mm Hg) compared with smaller reductions in mean BP, respectively.
In our mediation analysis (see the Statistical Analysis section), weight loss predicted the change in C-F PWV (β=201.68; P<0.05). However, weight loss was not a predictor of the change in mean BP (β=4.53; P>0.05). Therefore, mediation of the reduction in C-F PWV with weight loss by changes in mean BP was not supported.
The relation between change in mean arterial pressure and change in C-F PWV in the weight loss and control groups is shown in Figure 3. There was no significant correlation between change in mean arterial pressure and change in C-F PWV in the pooled sample (r=0.129) or among the individual groups (r=−0.17, P>0.05 and r=0.405, P>0.05 for weight loss and control groups, respectively).
We also compared the reduction in β-SI and C-F PWV in subjects above (n=10 for β-SI; n=11 for C-F PWV) and below (n=12 for β-SI; n=11 for C-F PWV) the median reduction in dietary sodium intake. The median reduction in dietary sodium was −733 mg/d. There were no differences in the magnitude of reduction in either β-SI (−1.01±0.32 versus −1.44±0.39 U; P>0.05) or C-F PWV (−192±51 versus −157±38 cm/s; P>0.05) in the groups. Mediation of the change in C-F PWV with weight loss by the change in sodium intake was not supported by linear regression analysis.
The major finding of the present study is that intentional weight loss via hypocaloric diet alone reduces arterial stiffness in overweight and obese middle-aged and older adults. The magnitude of this improvement in arterial stiffness was related to the magnitude of reduction in total body and abdominal adiposity. However, only reductions in indices of weight loss or total fat loss independently predicted reductions in arterial stiffness. Importantly, the improvement in arterial stiffness was observed with a modest amount of weight loss, that is, the 5% to 10% weight loss believed necessary to reduce the risk of cardiovascular diseases.19
Although previous studies have attempted to address this issue,8–12 our study is the first to demonstrate with a randomized, controlled design that intentional weight loss via hypocaloric diet alone (ie, without increases in physical activity) reduces large artery stiffness. Importantly, our findings extend the results of previous studies to overweight and obese middle-aged and older adults, a population with accelerated arterial stiffening and at increased risk of adverse cardiovascular events.
In the present study, only reductions in indices of weight loss or total fat loss independently predicted reduction in arterial stiffness. As such, our results suggest that weight loss, irrespective of reductions in abdominal adiposity, is associated with favorable reductions in arterial stiffness.
The mechanisms responsible for the reduction in arterial stiffness with weight loss observed are not clear. However, several possibilities exist. First, it is possible that changes in the elastic material content of the arterial wall may have occurred and contributed to the reductions in arterial stiffness with weight loss. Future studies in animal models will be necessary to address this important issue.
Second, collagen cross-linking that occurs in large elastic arteries as a consequence of nonenzymatic glycation of collagen is a major cause of age-related arterial stiffening. Whether short-term weight loss might also reduce arterial stiffness by reducing collagen cross-links is unclear.
Finally, the ability to reduce arterial stiffness over a relatively short time frame, such as occurred in the present study, is thought to be the result of changes in local, humoral, or neural modulation of smooth muscle tone.20 Thus, improved nitric oxide bioavailability, reductions in angiotensin II, reductions in sympathetic neural activity, and/or other factors may contribute to the favorable changes in arterial stiffness observed with weight loss.
There are some limitations of the present study that should be acknowledged. First, our sample size was small, and the age range of our subjects was restricted to 55 to 75 years. Thus, we were not able to test sex or age differences in the arterial destiffening response to our intervention. However, given the known age and sex differences in body fat distribution and potential depot specific responses to weight loss, future studies will be necessary.
Second, our weight loss intervention was relatively short in duration. Thus, whether the reductions in arterial stiffness would be sustained over time is unclear.
Third, we cannot exclude the possibility that the observed reductions in arterial stiffness were the result of the BP-lowering effects of weight loss and/or the reduction in sodium intake accompanying calorie restriction. However, we observed significant reductions in β-SI, a presumed BP-independent measure of arterial stiffness.14 In addition, changes in SBP, DBP, or mean BP (or sodium intake) were not related to changes in any of our measures of arterial stiffness. Furthermore, there were no significant differences in the magnitude of reduction in either β-SI or C-F PWV in individuals above compared with below the median reduction in mean BP or sodium intake. Taken together, our findings suggest that the reduction in arterial stiffness with weight loss is, at least in part, independent of the reductions in BP and sodium intake.
Finally, our study was not designed to address the hemodynamic consequences of arterial destiffening with weight loss. However, this will be an important objective for the future.
We recently reported that high-dose atorvastatin (80 mg once daily) reduces arterial stiffness in overweight and obese middle-aged and older adults.16 The post-treatment levels of arterial stiffness achieved in both our previous study16 and the present study suggest that arterial stiffness remains elevated compared with healthy young individuals.15 Whether weight loss (or therapeutic lifestyle change in general) combined with statin therapy (or other drugs) can reduce arterial stiffness in an additive or synergistic manner is unclear. Future studies are needed to address this important issue, because greater reductions in arterial stiffness should translate into superior cardiovascular risk reduction.
In summary, the findings from the present study suggest that modest weight loss by hypocaloric diet alone is efficacious in reducing arterial stiffness in overweight and obese middle-aged and older adults. The reductions in arterial stiffness appear to be determined, at least in part, by the magnitude of total and abdominal fat loss. Future studies will be necessary to determine the impact of long-term weight maintenance and the efficacy of combined arterial destiffening therapies.
Large artery stiffening is a potent risk factor for cardiovascular mortality among older adults. The findings of our present study suggest that weight loss by hypocaloric diet alone is efficacious in reducing large artery stiffness in overweight and obese middle-aged and older adults. The observed reductions in C-F PWV (ie, ≈150 to 200 cm/s) would translate into a reversal of age-related arterial stiffening by ≈15 to 20 years.1,2 These observations provide additional support for recommending weight loss to middle-aged and older adults who are overweight or obese. The reductions in arterial stiffness observed in the present study are similar in magnitude to the arterial destiffening effect that we observed previously with high-dose atorvastatin therapy in a similar population.16 Taken together, these observations suggest that a combination of these therapies could be particularly efficacious in reversing arterial aging. Future studies will be necessary to address this important issue.
We thank the participants for their time, effort, and commitment to the study.
Sources of Funding
This study was supported by an American Heart Association Grant-in-Aid (to K.P.D.).
- Received November 17, 2009.
- Revision received December 1, 2009.
- Accepted February 3, 2010.
Avolio A, Chen S, Wang R, Zhang C, Li M, O'Rourke M. Effects of aging on change arterial compliance and left ventricular load in a northern Chinese urban community. Circulation. 1983; 68: 50–58.
Vaitkevicius PV, Fleg JL, Engel JH, O'Connor FC, Wright JG, Lakatta LE, Yin FC, Lakatta EG. Effects of age and aerobic capacity on arterial stiffness in healthy adults. Circulation. 1993; 88: 1456–1462.
Resnick L, Militianu D, Cunnings A, Pipe J, Evelhoch J, Soulen R. Direct magnetic resonance determination of aortic distensibility in essential hypertension: relation to age, abdominal visceral fat, and in situ intracellular free magnesium. Hypertension. 1997; 30: 654–659.
Sutton-Tyrrell K, Newman A, Simonsick EM, Havlik R, Pahor M, Lakatta E, Spurgeon H, Vaitkevicius P. Aortic stiffness is associated with visceral adiposity in older adults enrolled in the study of health, aging, and body composition. Hypertension. 2001; 38: 429–433.
O'Rourke M. Arterial stiffness, systolic blood pressure, and logical treatment of arterial hypertension. Hypertension. 1990; 15: 339–347.
Sutton-Tyrrell K, Najjar SS, Boudreau RM, Venkitachalam L, Kupelian V, Simonsick EM, Havlik R, Lakatta EG, Spurgeon H, Kritchevsky S, Pahor M, Bauer D, Newman A. Elevated aortic pulse wave velocity, a marker of arterial stiffness, predicts cardiovascular events in well-functioning older adults. Circulation. 2005; 111: 3384–3390.
Barinas-Mitchell E, Kuller LH, Sutton-Tyrrell K, Hegazi R, Harper P, Mancino J, Kelley DE. Effect of weight loss and nutritional intervention on arterial stiffness in type 2 diabetes. Diabetes Care. 2006; 29: 2218–2222.
US Department of Health and Human Services. Dietary Guidelines for Americans 2005. 6th ed. Washington, DC: US Department of Agriculture; 2005.
Alvarez GE, Beske SD, Ballard TP, Davy KP. Sympathetic neural activation in visceral obesity. Circulation. 2002; 106: 2533–2536.
Hirai T, Sasayama S, Kawasaki T, Yagi S. Stiffness of systemic arteries in patients with myocardial infarction: a noninvasive method to predict severity of coronary atherosclerosis. Circulation. 1989; 80: 78–86.
Orr JS, Gentile CL, Davy BM, Davy KP. Large artery stiffening with weight gain in humans: role of visceral fat accumulation. Hypertension. 2008; 51: 1519–1524.
Orr JS, Dengo AL, Rivero JM, Davy KP. Arterial destiffening with atorvastatin in overweight and obese middle-aged and older adults. Hypertension. 2009; 54: 763–768.
National Institutes of Health. Clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults: the evidence report. 1998; 6: 51S–209S.
Zieman SJ, Melenovsky V, Kass DA. Mechanisms, pathophysiology, and therapy of arterial stiffness. Arterioscler Thromb Vasc Biol. 2005; 25: 932–943.