Significant improvement in cardiometabolic health in healthy nonobese individuals during caloric restriction-induced weight loss and weight loss maintenance

Abstract

Calorie restriction (CR) triggers benefits for healthspan including decreased risk of cardiometabolic disease (CVD). In an ancillary study to CALERIE 2, a 24-mo 25% CR study, we assessed the cardiometabolic effects of CR in 53 healthy, nonobese (BMI: 22–28 kg/m²) men (n = 17) and women (n = 36). The aim of this study was to investigate whether CR can reduce risk factors for CVD and insulin resistance in nonobese humans and, moreover, to assess whether improvements are exclusive to a period of weight loss or continue during weight maintenance. According to the energy balance method, the 25% CR intervention (n = 34) produced 16.5 ± 1.5% (mean ± SE) and 14.8 ± 1.5% CR after 12 and 24 mo (M12, M24), resulting in significant weight loss (M12 −9 ± 0.5 kg, M24 −9 ± 0.5 kg, P < 0.001). Weight was maintained in the group that continued their habitual diet ad libitum (AL, n = 19). In comparison to AL, 24 mo of CR decreased visceral (−0.5 ± 0.01 kg, P < 0.0001) and subcutaneous abdominal adipose tissue (−1.9 ± 0.2kg, P < 0.001) as well as intramyocellular lipid content (−0.11 ± 0.05%, P = 0.031). Furthermore, CR decreased blood pressure (SBP −8 ± 3 mmHg, P = 0.005; DBP −6 ± 2 mmHg, P < 0.001), total cholesterol (−13.6 ± 5.3 mg/dl, P = 0.001), and LDL-cholesterol (−12.9 ± 4.4 mg/dl, P = 0.005), and the 10-yr risk of CVD-disease was reduced by 30%. Homeostasis model assessment of insulin resistance (HOMA-IR) decreased during weight loss in the CR group (−0.46 ± 0.15, P = 0.003), but this decrease was not maintained during weight maintenance (−0.11 ± 0.15, P = 0.458). In conclusion, sustained CR in healthy, nonobese individuals is beneficial in improving risk factors for cardiovascular and metabolic disease such as visceral adipose tissue mass, ectopic lipid accumulation, blood pressure, and lipid profile, whereas improvements in insulin sensitivity were only transient.

INTRODUCTION

As the average age of the US and worldwide population is increasing, so, too, is the prevalence and incidence of chronic metabolic diseases such as cardiovascular disease (CVD) and type 2 diabetes (47a). Over the past two decades, CVD-related deaths in the US alone increased by more than 30%, which is probably related to an aging population (39). The economic burden of CVD to the society is enormous; one-half of the total health care expenditures in the US ($610 million USD) are devoted to CVD treatment, and current estimates project a threefold increase in these costs by 2030 (15).

With increasing age, physical and mental functionality decline, and the susceptibility to diseases increases (“primary aging”) (18). Vice versa, CVD accelerates the aging process by impairing metabolic health, reducing physical function, and reducing the quality of life (“secondary aging”). The progressive and detrimental interaction between aging and the development of CVD is likely related to a set of factors including increased abdominal adiposity, ectopic lipid accumulation, hypertension, and hyperlipidemia. Treatments to reverse these pathologies may thus attenuate the aging process and the development of CVD.

The largest and longest, controlled studies of calorie restriction (CR) at the National Institute of Aging (n = 121 rhesus monkeys) and at University of Wisconsin at Madison (n = 76 rhesus monkeys) and other studies in nonhuman primates collectively show that CR improves survival and reduces CVD mortality (2, 5, 7, 8, 26). In a review of these studies (25), it was concluded that CR is a promising nutritional intervention to prevent prevalent chronic diseases such as CVD, insulin resistance, and cancer, even in the absence of effects on lifespan. In line with this observation, human retrospective and observational studies have shown that moderate CR has been associated with increased lifespan (50), reduced CVD mortality (17, 43, 50), and improvements in metabolic risk factors (11, 12, 27, 42, 46, 47).

Indeed, numerous randomized clinical trials have supported the potential of CR to improve the cardiometabolic profile of overweight and obese subjects (reviewed in Ref. 14). However, if CR interventions are advocated for lifespan and healthspan extension in all adults, studies are also needed in normal-weight subjects. The most comprehensive assessment of the effects of sustained CR in humans was performed in the initial CALERIE studies followed by a three-site, single protocol across the US in nonobese individuals (9, 16, 33, 37). In an ancillary study to the CALERIE 2 protocol, additional measurements were obtained to evaluate whether CR sustained for 24 mo reduces ectopic fat depots and improves cardiometabolic risk factors (blood pressure and plasma lipid profile). We hypothesized that CR in nonobese subjects significantly improves these outcomes during weight loss (1st year) and that these effects persist during CR-induced weight maintenance throughout the 2-yr intervention.

METHODS

Design

In this multicenter trial (CALERIE 2), participants were randomly allocated to an intervention group aimed to reduce energy intake by 25% (CR) or to a control group with instructions to maintain habitual energy intake on an ad libitum basis (AL) for 24 mo (37). Randomization was in a 2:1 allocation in favor of the CR group and stratified by study site, sex, and BMI dichotomized into normal weight (22.0 ≤ BMI <25.0 kg/m²) or overweight (25.0 ≤ BMI <28.0 kg/m²). The ancillary study is registered at clinicaltrials.gov (NCT02695511) and was monitored by the Institutional Review Board of Pennington Biomedical Research Center. Individuals provided informed consent for the additional visits and procedures.

Subjects

The study was offered to the 80 healthy, nonobese individuals enrolled in CALERIE 2 at Pennington Biomedical Research Center. In addition to the inclusion and exclusion criteria for CALERIE 2 (inclusion: age, 20–50 yr for men, 20–47 yr for women, 22.0 ≤ BMI < 28 kg/m²; exclusion: history or clinical manifestation of CVD and diabetes, abnormal laboratory markers, psychological problems, and regular use of medication except for oral contraceptives), individuals were excluded for contraindications to MRI. To study the true effects of CR, without bias from nonadherent participants, included subjects in the CR group were required to have ≥5% weight loss at month 12 (M12) and/or month 24 (M24), and AL subjects were required to have <5% weight loss. The CONSORT diagram summarizing throughput of individuals in the ancillary study is shown in Fig. 1.

Fig. 1.

CONSORT diagram. Summary of throughput of individuals in the ancillary study.

Study Interventions

From day 1, the CR intervention targeted a sustained 25% restriction of energy intake prescribed according to their energy requirements at baseline. Energy requirements and actual energy intake were determined by the energy-intake balance method, combining measures of daily energy expenditure (doubly labeled water) and changes in body mass, as previously described (31, 34, 36). To facilitate adherence to 25% CR, all meals were provided for the first 27 days of the study, and these included three diets differing in macronutrient composition; standard American (AHA Step 1), Mediterranean, and low fat. Participants also attended regular group and individual meetings with trained interventionists throughout the study (individual meetings weekly in first month, twice monthly with additional biweekly phone contact until M12, and monthly until M24; group meetings monthly from M2 to M24) (36). The intervention was purposely designed to achieve ~15% weight loss during the first 12 mo [commensurate with a 25% CR based on a model of the Phase 1 data (30)] and with sustained CR from baseline in the second year, to promote weight loss maintenance. Participants randomized to the AL group were advised to continue their current diets on a completely ad libitum basis. No specific level of physical activity was required or recommended for either group. All participants received a multivitamin (Nature Made Multi Complete; Pharmavite, Mission Hills, CA) and calcium supplement (1,000 mg/day; Douglas Laboratories, Pittsburgh, PA).

Anthropometrics and Vital Signs

Height was measured at screening using a wall-mounted stadiometer. Weight (Scale Tronix 5200, White Plains, NY) was a metabolic weight measured in the morning after an overnight fast and voiding while wearing a surgical gown, which was subtracted from the total weight. Blood pressure was measured in duplicate while the participant was sitting after a 5-min rest. Mean arterial pressure (MAP) was calculated as follows (2×DBP+SBP)/3.

Body Composition

Body composition [fat mass (FM) and fat-free mass (FFM)] was measured by dual-energy X-ray absorptiometry (DEXA; Hologic QDR 4500A,Bedford, MA). Abdominal adipose tissue (AT) distribution and ectopic lipid accumulation in skeletal muscle and liver were measured by magnetic resonance imaging (MRI) and magnetic resonance spectroscopy (MRS), respectively using a 3.0 T scanner (General Electric, Excite HD System, Milwaukee, WI).

Adipose tissue distribution.

Abdominal tissue (AT) volumes, including total, subcutaneous (SAT), and visceral (VAT) tissues, were quantified between the symphysis pubis and the dome of the diaphragm, using ~8 axial MRI images of 3.4 mm thickness with no interslice gap (38). SliceOmatic 4.2 image analysis software (Tomovision, Montreal, PQ, Canada) was used to analyze images on a PC workstation (Gateway, PIII 500 MHz). All MRI scans were read by the same trained observer.

Muscle and liver lipid content.

Briefly, intramyocellular lipid content (IMCL, soleus muscle) and intrahepatic lipid content (IHL) were measured by the proton magnetic resonance spectroscopy technique (¹H-MRS) on a 3-T whole body imaging and spectroscopy system (General Electric, Excite HD System, Milwaukee, WI) using the PRESS box (Point RESolved Spectroscopy) technique as described previously (20, 22).

Clinical Chemistry

Fasting blood samples were obtained, and the following assays were performed at the CALERIE central biochemistry laboratory at University of Vermont: lipids, using a Beckman-Coulter Synchron CX7 (Brea, CA); total cholesterol, by the cholesterol esterase/oxidase/peroxidase method; triacylglycerols, by the GPO-Trinder method; and HDL-cholesterol (HDL-C), by an assay from Trinity Biotech (Jamestown, NY); LDL-cholesterol (LDL-C), calculated using the Friedwald equation (13); glucose, colorimetric reflectance spectrophotometry (Vitros, Ortho Clinical Diagnostics, Rochester, NY); insulin, chemiluminescent immunoassay (Elecsys 2010; Roche Diagnostics, Indianapolis, IN) concentrations, determined from fasting concentrations; and HOMA-IR, calculated as (insulin (µU/ml) × glucose (mg/dl)/405).

Estimates of 10-Year CVD Risk

Ten-year CVD risk was calculated using the sex-specific equations developed by Anderson et al. (1) that include total and HDL-cholesterol (as ratio), systolic blood pressure, age and sex. Smoking, presence of diabetes, and abnormal ECG were set to 0 in the model, since these variables were not evident in the present study.

Maximal Aerobic Capacity and Physical Activity

V̇o_2peak was measured using the Cornell incremental treadmill test (44), with the speed and/or grade of the treadmill changing every 2 min, as previously described (32). V̇o₂ was measured continuously and calculated at 15-s intervals using a calibrated metabolic cart (TrueOne 2400 or TrueMax 2400; Parvo Medics, Sandy, UT). The single highest 15-s V̇o₂ value during the exercise test was considered the V̇o_2peak, which was expressed per kilogram of body weight. Physical activity was assessed with the 7-day recall questionnaire (40), in which the amount of time spent sleeping and engaging in moderate, hard, and very hard activity was reported; light activity was calculated by difference.

Statistical Methods

This study was powered by the ability to detect a significant adaptation in energy expenditure (a reduction that is greater than expected, based on the reduction in metabolic mass, the primary outcome) from baseline and to detect differences between the two diet groups (AL vs. CR). A random subject effect was included to account for intraindividual correlations over time. Data are presented as least square means (LSM) ± SE that are derived from the linear mixed model. Contrasts of the LSM were used to compare adjusted mean change between intervention groups and to test for group differences in adjusted mean change at M12 and M24. P values represent the significance of the difference of LSM. All analyses were carried out using SAS/STAT software, version 9.4 of the SAS System for Windows (SAS Institute, Cary, NC). All tests were evaluated using significance level α = 0.05.

RESULTS

At baseline, study groups did not significantly differ in age, weight, BMI, or body fat [CR: n = 34 (24 women), AL: n = 19 (12 women)]. Despite randomization, fasting insulin concentrations and HOMA-IR tended to differ between groups at baseline (Table 1).

Table 1.

Subjects’ characteristics

	AL	CR	P
N (men, women)	19 (7, 12)	34 (10, 24)
Age, yr	38.7 ± 1.2	40.0 ± 1.2	0.47
Weight, kg	71.0 ± 1.9	71.9 ± 1.5	0.73
Body mass index, kg/m2	25.5 ± 0.4	25.7 ± 0.3	0.70
Normal weight, n	9	14
Overweight, n	10	20
Fat-free mass, kg	48.3 ± 1.9	47.8 ± 1.5	0.86
Fat mass, kg	23.4 ± 0.9	24.7 ± 0.9	0.36
Body fat, %	32.9 ± 1.3	34.2 ± 1.1	0.47
Abdominal subcutaneous AT, kg	5.0 ± 0.3	5.1 ± 0.3	0.26
Abdominal visceral AT, kg	0.7 ± 0.2	0.7 ± 0.1	0.29
Systolic blood pressure, mmHg	112 ± 2	115 ± 2	0.95
Diastolic blood pressure, mmHg	73 ± 2	76 ± 1	0.83
Glucose, mg/dl	83 ± 1	82 ± 1	0.36
Insulin, mU/ml	6.5 ± 0.5	5.1 ± 0.4	0.06
HOMA-IR, AU	1.33 ± 0.11	1.05 ± 0.09	0.06

Data are presented as LSM ± SE. AT, adipose tissue; HOMA-IR, homeostatic model assessment of insulin resistance. P value indicates statistical significance of difference between the calorie restriction (CR) and ad libitum diet (AL) groups.

Body Weight, Body Composition, and Ectopic Fat

Over the study period of 24 mo, the CR group achieved a mean CR of 14.8 ± 1.5% (16.5 ± 1.5% at M12). Per study design, changes in body weight were different between the groups (P < 0.001). Whereas participants in the AL group slightly increased their body weight over the course of the study (M12: +0.6 ± 0.6 kg, P = 0.30; M24: +1.8 ± 3.0 kg, P = 0.003), CR induced a significant weight loss at M12 (−9.4 ± 0.4 kg, P < 0.001), which was maintained at M24 (CR: −8.7 ± 2.4 kg, P < 0.001). Accordingly, in the CR group, both fat-free mass and fat mass were significantly reduced at both time points (Fig. 2, A and B). Abdominal fat mass in both subcutaneous and visceral depots was significantly decreased by CR at M12 compared with AL and remained decreased at M24 (all, P < 0.001; Fig. 2, C and D). Additionally, ectopic lipid in the liver and skeletal muscle (soleus) was significantly reduced by CR (Fig. 2, E and F). A CR-induced decrease in intrahepatic lipid was significantly different from the change in the AL group at M12, but not at M24 (M12: P = 0.01, M24: P = 0.12; Fig. 2E). Presenting a different pattern, IMCL was reduced at M12, but reached statistically significance only at M24 (M12: P = 0.07, M24: P = 0.03; Fig. 2F).

Fig. 2.

Changes in body composition, abdominal fat, and ectopic fat accumulation during sustained calorie restriction (CR, open symbols, n = 34) or ad libitum diet (AL, filled symbols, n = 19). Data are presented as LSM ± SE. A–B: fat mass (FM, circles) and fat-free mass (FFM, squares). C–D: abdominal subcutaneous (circles) and visceral (squares) adipose tissue (AT, CR: n = 33, AL: n = 18). E: intramyocellular lipid content (IMCL, CR: n = 33, AL: n = 16). F: intrahepatic lipid content (IHL, CR: n = 33, AL: n = 14). *Statistical significance, P < 0.05; #statistical trend, P < 0.10 of the difference of LSM.

Cardiovascular Risk Profile

CR induced a decrease in systolic blood pressure over time compared with AL, which reached statistical significance at M24 (Fig. 3A). Diastolic blood pressure was significantly reduced at M12 and remained reduced during the CR-weight maintenance period compared with AL (Fig. 3B). MAP was therefore significantly reduced from baseline at both M12 and M24 in CR subjects (M12: −5.8 ± 2.1, P = 0.007; M24: −7.6 ± 2.0 mmHg, P < 0.001). Plasma LDL-C concentrations significantly decreased in the CR group at M12 and remained lower at M24 compared with the AL group (P = 0.005 and P = 0.006; Fig. 3C), whereas increases of HDL-C were not significant compared with the AL group (P = 0.12 and P = 0.18; Fig. 3D). As a result, the LDL/HDL ratio decreased over time during CR compared with AL (P = 0.001 and P = 0.005) as well as total cholesterol (P = 0.01, P = 0.01) and triglyceride (M12: −17.6 ± 7.9, P = 0.03; M24: −16.4 ± 7.5 mg/dl, P = 0.03) concentrations. By use of total cholesterol, HDL, systolic blood pressure, and age, the 10-yr risk for CVD was significantly lowered by 30% in the CR group at M12 and maintained throughout the 2-yr intervention (P < 0.001 and P = 0.001 at M12 and M24, respectively; Fig. 3E). Finally, insulin resistance assessed by HOMA-IR was significantly decreased in the CR group at M12 (−0.46 ± 0.15, P = 0.003), but the decrease in HOMA-IR was not maintained at M24 (−0.11 ± 0.15, P = 0.46; Fig. 3F).

Fig. 3.

Cardiovascular risk profile changes during CR (open circles, n = 34) and AL (filled circles, n = 19). Data are presented as LSM ± SE. A: systolic blood pressure (SBP). B: diastolic blood pressure (DBP). C: low-density lipoprotein-cholesterol (LDL). D: high-density lipoprotein-cholesterol (HDL), E: 10-yr CVD risk (1), F: HOMA-IR. *Statistical significance, P < 0.05; #statistical trend, P < 0.10 of the difference of LSM.

Physical Performance

V̇o_2peak, expressed per body mass, was unchanged in the CR group (M24: +0.46 ± 0.75 ml·kg⁻¹·min⁻¹, P = 0.54), whereas it declined significantly in the AL group (M24: −4.40 ± 0.98 ml·kg⁻¹·min⁻¹, P < 0.001). In line with these findings, no difference between the CR and AL groups was observed for daily physical activity by 7-day recall. The hours spent sleeping or in activities of different intensities did not change differently over time between groups; also daily METs were unaffected by the intervention (M12: −0.16 ± 0.29 kcal·kg body wt⁻¹·day⁻¹, P = 0.57; M24: −0.24 ± 0.31 kcal·kg body wt⁻¹·day⁻¹, P = 0.44). Daily energy expenditure, calculated as function of body mass and physical activity ratio assessed by the 7-day recalls, decreased in the CR compared with the AL group (M12: −182 ± 51 kcal/day, P < 0.001; M24: −135 ± 63 kcal/day, P = 0.04).

DISCUSSION

In this novel study of sustained CR in healthy, nonobese participants, we found beneficial effects of CR on body composition, fat distribution, and cardiometabolic risk factors over a study period of 2 yr. We showed that CR-induced weight loss over the first year of the intervention reduced CVD risk factors including ectopic lipid accumulation, blood pressure, and plasma lipids. Importantly, these improvements were sustained between M12 and M24 of the intervention when energy intake was still below baseline levels, and weight loss from year 1 was maintained. The present study was unique in its duration and in its stringent level of control, which allowed us to describe how true CR affects the cardiometabolic, cardiovascular, and cardiorespiratory profile. In addition, we were able to quantify effects of sustained CR during weight loss maintenance and thus independently of weight loss.

In 2015, the Global Burden of Diseases study (12a) estimated that annually 10.7 million deaths are caused by high blood pressure, 5.2 million by hyperglycemia, 4.3 million by high total cholesterol concentrations, and 4.0 million by a high BMI (>25 kg/m²). Amelioration of these chronic health problems is primarily directed at treatments for individuals that present with clinically increased values for these variables. However, an equally favorable approach would be to tackle disease prevention and thereby delay and/or decelerate the development of chronic disease before manifestation. CR has been proposed as a superior nutritional intervention to attenuate the age-associated decline in chronic diseases with the potential to extend lifespan.

There are a myriad of studies in obese subjects showing the benefits of diet-induced weight loss on improvements in cardiometabolic risks (14). We have also previously demonstrated in shorter-term CR studies (6 mo), that overweight individuals experience that weight-loss induced changes ectopic fat, insulin sensitivity, and reduced 10-yr risk for CVD (9, 16, 33, 34). In the current 2-yr study in normal weight and slightly overweight (up to BMI 27.9 kg/m²), the CR intervention significantly reduced the mean BMI (−3.7 ± 0.3 kg/m²) and therefore reduced the prevalence of overweight in the CR group (prevalence of BMI >25; M0: n = 20, M12: n = 0, M24: n = 3 in the CR group; M0: n = 10, M12: n = 11, M24: n = 11 in the AL group). Of the lost body weight, importantly the loss of fat mass was larger than the loss of fat-free mass (70% FM vs. 30% FFM). Neither BMI at enrollment nor sex had a significant effect on weight change or changes in body composition outcomes (10), including the improvements in abdominal fat distribution, ectopic lipid, insulin resistance, and the 10-yr risk for CVD. Because adipose tissue depots differ in morphology, metabolism (24), and association with CVD (3), we quantified abdominal VAT and SAT depots using MRI. VAT, the depot most strongly associated with CVD incidence, decreased by more than 50% in the CR group after 12 mo, and SAT by more than 30%, which is in line with the results of CR imposed for six months (19, 35). The extent of the reductions of VAT and SAT in the present study are, to our knowledge, unprecedented in nonobese individuals and remarkable given the fairly low tissue masses of these depots at baseline (<1 kg and ~5 kg, respectively).

Next, we studied whether CR would induce changes in the plasma profile of cardiometabolic risk factors. We (23) reported previously a reduced CVD risk following 6 mo of CR, and now in this longer study, we confirmed this reduction after 12 and 24 mo (CR: −30%, AL: +15%) but with similar changes in weight. Interestingly, individual factors contributing to the reduced risk for CVD in the 6-mo study, including plasma lipids and blood pressure, were not significantly improved with 6 mo of CR. This finding emphasizes the unique importance of sustained CR, beyond weight loss for benefits to factors related to healthspan.

A central factor in the development of obesity, CVD, and aging is insulin resistance. To investigate possible mechanisms that may relate to reported changes in insulin sensitivity after 6 and 12 mo of CR (16), we measured ectopic lipid accumulation in skeletal muscle (IMCL) and the liver (IHL) using MRS. In contrast to the transient improvement of systemic insulin resistance, there was not a CR-induced improvement in IMCL content after 6 [CALERIE 1 (19)] or 12 mo, but after 24 mo. However the 15% reduction in IMCL that we observed is similar to the effect size observed in obese individuals with comparable weight loss after 4 mo (45), so, although not significant in our study, the reduction in IMCL may be implying some health benefits to the CR subjects. Interestingly, as for insulin resistance, the observed decrease in IHL content after 12 mo disappeared during the weight maintenance period, although percent body fat and abdominal VAT remained reduced. The parallel patterns of hepatic lipid content and systemic insulin resistance support the close relationship between hepatic and systemic metabolism and confirm earlier reports of improved hepatic metabolism following CR (reviewed in Ref. 41). We are not able to fully explain why sustained CR and weight loss was insufficient to maintain an improvement in insulin sensitivity and ectopic lipid accumulation. One might argue that this finding is driven by a reduced compliance with the CR intervention; however, the level of CR achieved from baseline throughout year 2 was estimated at 13.2% CR, only a 2% change from year 1. Alternatively, we acknowledge that the effect size may have been limited by studying a healthy, insulin-sensitive population. Therefore the individuals likely did not have substantially elevated liver fat or insulin resistance (HOMA-IR: 1.05 ± 0.09, liver fat: 1.0 ± 0.2%) as opposed to obese or nonalcoholic fatty liver disease patients who are the subject of most reported studies. Last, HOMA-IR is a crude measure of systemic insulin resistance. Characterization of the CR-induced insulin sensitivity effects on adipose tissue, liver, and skeletal muscle by means of a hyperinsulinemic-euglycemic clamp would provide a more sensitive measurement and thereby allow more definitive conclusions on the effects of CR on insulin sensitivity and its role in reducing cardiometabolic risks.

Finally, we assessed whether the reduction of caloric intake and the loss of body mass affected physical fitness. We found no significant change in maximal oxygen consumption per body weight (V̇o_2peak and V̇o_2max) in the CR-group after 12 and 24 mo. Our findings are in line with previous reports that showed no improvement in aerobic capacity if body (fat) mass is lost through diet, as opposed to weight loss through exercise (21, 45, 48, 49). We conclude that CR-induced improvements on cardiometabolic health appear to be independent of physical fitness.

To our knowledge, this is the first study designed to assess effects of controlled CR during and after weight loss in the same study. In the latter part of the study, CR was maintained from baseline, and the beneficial effects of CR on cardiometabolic risk factors were preserved, but not enhanced. On the basis of this observation, we conclude that reduced and sustained calorie intake is the main driver of the improvements in the cardiometabolic profile. The benefit of long-term CR has been supported by data collected on individuals who self-administer CR. These individuals, who were mostly normal weight at CR onset, have followed a regimen of self-imposed CR for an average of 15 yr and present a remarkably low metabolic risk profile (reviewed in Ref. 29). For example, it has been reported that no member of the CRON society has developed any chronic disease so far, and no use of medication is reported either (4).

In conclusion, we show that the classical CVD risk factors were all improved by sustained CR in nonobese participants. The improved CVD risk profile was associated with a significant reduction in overall adiposity as well as reduced abdominal fat and ectopic lipids, which are highly predictive for the development of chronic metabolic diseases with human aging. We hypothesize that sustained CR would not only prolong average life expectancy by reducing CVD/mortality (secondary aging) but more importantly would improve healthspan, even in individuals who are already “healthy” and not obese. Because sustained CR reduced lipid accumulation in all measured adipose tissue depots did not improve aerobic fitness, we further hypothesize that improvement of cardiometabolic health with CR is mediated through improved body composition and not physical fitness. Arguably the most striking finding of the present study is that the subjects were not obese or at high risk for the development of CVD but were still responsive to the effects of CR on measures of the cardiometabolic profile. Now, it has to be investigated whether such improvements can prevent or delay the onset of metabolic complications in later life of nonobese subjects in a comparable manner, as it has been established for the obese and suggested by data of the CR society members.

GRANTS

This work was supported by the National Institute of Aging [Grants R01 AG-029914, R01 AG-030226, R01 AG-031797, and U01 AG-020478]

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

J.M., L.A.G., H.H., and L.M.R. analyzed data; J.M., L.A.G., E.R., and L.M.R. interpreted results of experiments; J.M. prepared figures; J.M. drafted manuscript; J.M., L.A.G., S.R.S., E.R., and L.M.R. edited and revised manuscript; J.M., L.A.G., S.R.S., H.H., E.R., and L.M.R. approved final version of manuscript; S.R.S., E.R., and L.M.R. conceived and designed research; S.R.S., E.R., and L.M.R. performed experiments.