Author + information
- Received April 7, 2005
- Revision received July 26, 2005
- Accepted August 25, 2005
- Published online January 17, 2006.
- Timothy E. Meyer, PhD⁎,†,
- Sándor J. Kovács, PhD, MD†,
- Ali A. Ehsani, MD⁎,†,
- Samuel Klein, MD⁎,
- John O. Holloszy, MD⁎ and
- Luigi Fontana, MD, PhD⁎,‡,⁎ ()
- ↵⁎Reprint requests and correspondence:
Dr. Luigi Fontana, Division of Geriatrics, Washington University School of Medicine, 4566 Scott Avenue, Box 8113, St. Louis, Missouri 63110
Objectives We determined whether caloric restriction (CR) has cardiac-specific effects that attenuate the established aging-associated impairments in diastolic function (DF).
Background Caloric restriction retards the aging process in small mammals; however, no information is available on the effects of long-term CR on human aging. In healthy individuals, Doppler echocardiography has established the pattern of aging-associated DF impairment, whereas little change is observed in systolic function (SF).
Methods Diastolic function was assessed in 25 subjects (age 53 ± 12 years) practicing CR for 6.5 ± 4.6 years and 25 age- and gender-matched control subjects consuming Western diets. Diastolic function was quantified by transmitral flow, Doppler tissue imaging, and model-based image processing (MBIP) of E waves. C-reactive protein (CRP), tumor necrosis factor-alpha (TNF-α), and transforming growth factor-beta1(TGF-β1) were also measured.
Results No difference in SF was observed between groups; however, standard transmitral Doppler flow DF indexes of the CR group were similar to those of younger individuals, and MBIP-based, flow-derived DF indexes, reflecting chamber viscoelasticity and stiffness, were significantly lower than in control subjects. Blood pressure, serum CRP, TNF-α, and TGF-β1levels were significantly lower in the CR group (102 ± 10/61 ± 7 mm Hg, 0.3 ± 0.3 mg/l, 0.8 ± 0.5 pg/ml, 29.4 ± 6.9 ng/ml, respectively) compared with the Western diet group (131 ± 11/83 ± 6 mm Hg, 1.9 ± 2.8 mg/l, 1.5 ± 1.0 pg/ml, 35.4 ± 7.1 ng/ml, respectively).
Conclusions Caloric restriction has cardiac-specific effects that ameliorate aging-associated changes in DF. These beneficial effects on cardiac function might be mediated by the effect of CR on blood pressure, systemic inflammation, and myocardial fibrosis.
Normal aging causes a decline in cardiac performance, manifesting as an age-related impairment in diastolic function (DF) with little or no change in left ventricular (LV) systolic function (SF) (1–2). Specifically, LV relaxation slows, and peak early, suction-initiated LV filling decreases with advancing age, whereas the relative contribution of the atrial component of LV filling (atrial systole) increases (1–2).
Caloric restriction (CR) slows aging and increases maximal lifespan in drosophila, C. elegans, mice, and rats (3). In humans, risk factors for atherosclerosis are markedly improved by long-term CR (4); however, the effect of CR on aging in humans is unknown.
The purpose of the present study was to investigate whether long-term CR with optimal nutrition is associated with an attenuation of the age-related decline in DF. To assess DF we used echocardiography and evaluated serum C-reactive protein (CRP), tumor necrosis factor-alpha (TNF-α), and transforming growth factor-beta1(TGF-β1) concentrations.
Twenty-five subjects practicing CR for 3 to 15 years (mean age 53 ± 12 years; range 35 to 82 years; 21 men, 4 women) were compared with 25 age- and gender-matched subjects eating typical Western diets (WD) using a cross-sectional design. None of the subjects had evidence of chronic disease or smoked. None of the subjects were taking lipid-lowering or antihypertensive agents or other medications. All participants were weight stable (<2 kg weight change in the preceding six months) and did not perform more than 20 min of vigorous exercise twice per week. The Human Studies Committee of Washington University School of Medicine approved the study, and all subjects gave informed consent before their participation.
Anthropometrics and body composition
Body weight was obtained in the morning after an overnight fast. Body fat mass was determined by dual-energy X-ray absorptiometry (DXA) (QDR 1000/w, Hologic, Waltham, Massachusetts).
Subjects recorded all food and beverages consumed and portion sizes for seven consecutive days. Food records were analyzed using the NDS-R program (version 4.03_31), which is the Nutrition Data System for research from the Nutrition Coordinating Center at the University of Minnesota (5).
Commercial ELISA kits were used to measure serum TNF-α (Quantakine High Sensitive, R&D Systems, Minneapolis, Minnesota), high-sensitivity CRP and TGF-β1(American Laboratory Products Company Diagnostics, Windham, New Hampshire).
A standard two-dimensional Doppler echocardiographic examination was performed with an echocardiographic imaging system (Hewlett-Packard SONOS 5500, Andover, Massachusetts) (6). Doppler transmitral flow velocity images were acquired in the apical four-chamber view according to American Society of Echocardiography criteria (6). Mitral annular velocities were recorded via pulsed wave tissue Doppler imaging (TDI) technique (7,8). For technical reasons, Doppler flow and TDI measurements were not made in all subjects.
Left ventricular SF was determined from fractional shortening (systolic shortening) calculated via the conventional equation (8). Standard DF indexes were measured from the transmitral Doppler E- and A-wave contours (8). Additional DF indexes were obtained by subjecting E- and A-waves to model-based image processing (MBIP) on the basis of a kinematic model of filling that was previously validated via simultaneous cardiac catheterization and echocardiography (9–11). Three mathematically independent parameters (xo, c, k) were determined from the clinical Doppler E-wave by MBIP. The parameters account for the amplitude (load), decay rate (viscosity/relaxation), and width (average chamber stiffness) of the E-wave contour, respectively (Fig. 1)(9–11).
A Student ttest was used to evaluate the significance of differences between groups when variables were normally distributed with approximately equal standard deviations. For variables not normally distributed or with unequal standard deviations, the Mann-Whitney test was used. Statistical significance was set a priori at p < 0.05 for all tests. All data were analyzed with a commercially available statistical package (SPSS FOR WINDOWS version 10.0, SPSS Inc., Chicago, Illinois). Data are expressed as means ± SD.
The characteristics of the study subjects are shown in Table 1.Values for body mass index (BMI) in the WD group were similar to the mean range for middle-aged people in the U. S. (12). Values for body weight, BMI, and blood pressure for 18 CR subjects before they began CR are shown along with their current values in Table 2.
The WD group ate a typical WD providing approximately 2,445 ± 470 kcal/day, approximately 17% protein, approximately 52% carbohydrates, approximately 31% fat, approximately 11% saturated fat. The CR subjects ate a nutritionally balanced diet (4) (at least 100% of the Recommended daily intake for each nutrient) providing approximately 1,671 ± 294 kcal/day (p = 0.0001 vs. WD group), approximately 23% protein, (p = 0.0001 vs. WD group), approximately 49% complex carbohydrates, approximately 28% fat, approximately 6% saturated fat. Daily salt intake was lower in the CR group compared with the WD group (2.6 ± 1.3 g vs. 3.4 ± 1.0 g; p = 0.007).
No difference in LV fractional shortening was detected between the two groups (Table 3).Although E-wave peak was not different between groups, early filling fraction was greater in the CR group (Table 3). A waves of the CR group had less evidence of aging-related changes than those of the WD group. As a result, the peak E-wave to peak A-wave (E/A) ratio was greater in the CR group than in the WD group. Deceleration time was increased in the WD group, trending toward the “delayed relaxation pattern” of filling. The increased E-wave duration was solely due to increased deceleration time, because acceleration time was not different between groups.
The velocity assessed at the mitral annulus from both the lateral and septal aspects during early LV filling (E′) was significantly higher in the CR group than in the WD group (Table 3), and cand kwere lower in the CR group than in the WD group (Table 3).
Serum CRP, TNF-α, and TGF-β1concentrations were lower in the CR group than in the WD group (p < 0.05) (Table 1).
Nothing is known regarding the effects of long-term CR with optimal nutrition on cardiac function in non-obese humans. In this study, we assessed heart function in subjects practicing CR for 3 to 15 years. Our data show that long-term CR with optimal nutrition improves DF relative to healthy, age- and gender-matched control subjects eating a standard WD.
It has been shown in mice that indexes of DF are beneficially affected by CR (13). Interestingly, the DF changes observed in our CR subjects were very similar to those found by Taffet et al. (13) in lifelong CR mice. Systolic function assessed by fractional shortening changes relatively little with normal aging (1–2). Both groups in our study had normal SF; however, DF was significantly better in the CR than in the WD group and similar to that of younger individuals (Fig. 2).The CR group had lower Apeakvelocities, higher early filling fraction, and higher E/A, demonstrating better DF. Consequently, lower Apeakvalues could mitigate against left atrial hypertrophy. Moreover, TDI-derived E′, which normally decreases with aging, was higher in the CR subjects, suggesting improved longitudinal function compared with the WD group. Importantly, E′ has also been shown to be negatively associated with both diastolic blood pressure and BMI (14).
Further evidence of better DF in the CR subjects is provided by the MBIP-derived LV chamber stiffness parameter k, which was significantly lower in the CR subjects compared with the WD group. The viscoelastic chamber constant cwas also lower in the CR group, demonstrating a decreased impediment to filling. Because Epeakwas not different between groups, the WD group must expend more energy to achieve the same peak velocity during early filling.
Data from rat studies show that collagen content increases in the aging heart and correlates with increasing myocardial stiffness, whereas CR results in less accumulation of cardiac connective tissue (15). Transforming growth factor-β1, an anti-inflammatory cytokine, plays a major role in regulating myocardial extracellular matrix protein deposition and degradation (16). A sustained increase of TGF-β1facilitates the development of fibrosis (16). Brooks et al. (17) demonstrated that TGF-β1-deficient mice survived longer and exhibited less myocardial fibrosis and lower myocardial stiffness compared with age-matched control subjects. Furthermore, overexpression of TNF-α leads to upregulated expression and release of TGF-β1, which in turn increases myocardial collagen content (18). Our findings demonstrate that systemic inflammation was markedly reduced in CR compared with WD subjects, as reflected by low serum concentrations of CRP and TNF-α. These findings taken in concert suggest that CR has a salutary effect on DF by lowering systolic blood pressure and decreasing systemic inflammation and myocardial fibrosis.
Because our study was not randomized, our cross-sectional study design does not allow us to assign causation, although the observed significant differences on the basis of cross-sectional data are an important first-step in elucidating the effects of long-term CR in humans. Another minor limitation is that some echocardiographic parameters were not obtained in all subjects for technical reasons.
We conclude that CR has cardiac-specific effects that ameliorate aging-associated changes in DF. These beneficial effects on cardiac function might be mediated by the effect of CR on blood pressure, systemic inflammation, and myocardial fibrosis.
Research supported by National Institutes of Health grants HL54179, HL04023 (Dr. Kovács), RR00036 (General Clinical Research Center), DK20579 (Diabetes Research and Training Center), and DK56351 (Clinical Nutrition Research Unit), the Whitaker Foundation (Dr. Kovács), American Heart Association (Dr. Kovács), and the Alan A. and Edith L. Wolff Charitable Trust (Dr. Kovács). Dr. Meyer was supported by Institutional National Research Service Award AG-00078.
- Abbreviations and Acronyms
- body mass index
- caloric restriction
- C-reactive protein
- diastolic function
- left ventricle/ventricular
- model-based image processing
- systolic function
- tissue Doppler imaging
- transforming growth factor-beta1
- tumor necrosis factor-alpha
- Western diet
- Received April 7, 2005.
- Revision received July 26, 2005.
- Accepted August 25, 2005.
- American College of Cardiology Foundation
- Kitzman D.W.,
- Sheikh K.H.,
- Beere P.A.,
- Philips J.L.,
- Higginbotham M.B.
- Fontana L.,
- Meyer T.E.,
- Klein S.,
- Holloszy J.O.
- Schakel S.F.,
- Sievert Y.A.,
- Buzzard I.M.
- Nagueh S.F.,
- Mikati I.,
- Kopelen H.A.,
- Middleton K.J.,
- Quinones M.A.,
- Zoghbi W.A.
- Appleton C.P.,
- Firstenberg M.S.,
- Garcia M.J.,
- Thomas J.D.
- Kovács S.J. Jr.,
- Barzilai B.,
- Pérez J.E.
- Lisauskas J.B.,
- Singh J.,
- Bowman A.W.,
- Kovács S.J.
- Taffet G.E.,
- Pham T.T.,
- Hartley C.J.
- Thomas D.P.,
- McCormick R.J.,
- Zimmerman S.D.,
- Vadlamudi R.K.,
- Gosselin L.E.
- Border W.A.,
- Ruoslahti E.
- Sivasubramanian N.,
- Coker M.L.,
- Kurrelmeyer K.M.,
- et al.