Exercise Decreases the Risk of Metabolic Syndrome in Elderly Females


Medicine & Science in Sports & Exercise: February 2009 - Volume 41 - Issue 2 - pp 297-305
doi: 10.1249/MSS.0b013e31818844b7
Clinical Sciences

Purpose: To determine the effect of exercise based on a multiple purpose, high-intensity philosophy on parameters of metabolic syndrome (MetS) in older females with the MetS.

Methods: Thirty-two women (68.7 ± 3.4 yr) with the MetS according to the International Diabetes Federation (IDF) took part in the exercise training (EG), and 33 females (69.5 ± 4.3 yr) also with MetS served as control group (CG). Beside the diagnostic tools of the MetS (waist circumference, triglycerides, HDL-C, blood pressure, and glucose), corresponding anthropometric parameters, total cholesterol, and high-sensitivity C-reactive protein (hsCRP) were determined.

Results: After 12 months of exercise, significant effects were observed for total body fat [EG: −1287 g vs CG: +484 g; P = 0.001; confidence interval (CI) = −2787 to −744 g], trunk fat (−1070 vs −203 g; P = 0.005; CI = −1460 to −274 g), hip circumference (−2.1 vs +1.0 cm; P = 0.001; CI = −4.8 to −1.3 cm), triglycerides: (−24.1 vs −0.3 mg·dL−1; P = 0.021; CI = −43.9 to −3.7 mg·dL−1), total cholesterol: (−19.8 vs +0.2 mg·dL−1; P = 0.008; CI = −34.6 to −5.5 mg·dL−1), and HDL-C: (+3.8 vs −0.2 mg·dL−1; P = 0.036; CI = 0.3 to 7.6 mg·dL−1). Waist circumference slightly decreased in both groups (EG: −0.7% vs CG: −0.6%; P = 0.94). Resting systolic (−7.1 vs −6.4 mm Hg; P = 0.84) and diastolic (−8.0 vs −+9.8 mm Hg; P = 0.39) blood pressure significantly decreased in both groups. Changes of glucose and hsCRP did not significantly vary between groups (glucose: +0.7 vs +1.0 mg·dL−1, P = 0.88; hsCRP: −0.68 vs −0.50 mg·L−1, P = 0.36). The number of criteria of the MetS significantly decreased in the EG (4.13-3.66) and slightly decreased in the CG (4.12-3.97); however, significant time group interactions were not observed (P = 0.15; CI = −0.74 to 0.12).

Conclusion: Our multipurpose exercise program significantly affects most parameters of the MetS in elderly women.

Institute of Medical Physics, University of Erlangen, GERMANY

Address for correspondence: Wolfgang Kemmler, Ph.D., Institute of Medical Physics, University of Erlangen, Henkestrasse 91, 91052 Erlangen, Germany; E-mail: wolfgang.kemmler@imp.uni-erlangen.de.

Submitted for publication June 2008.

Accepted for publication July 2008.

Article Outline

Age-related diseases and multimorbidity are increasingly straining health care systems in many countries around the globe. In Germany, over 50% of the population 70 yr and older has two to four chronic diseases and more than 25% have five or more chronic diseases (27). Although physical exercise in general plays a vital role in disease prevention and treatment (20), a large variety of studies in elderly females have shown positive effects of exercise training on bone mineral density, osteoarthritis, low back pain (31), risk of falls (8), insulin resistance/glucose intolerance (24), blood lipids (10), body composition (28), and abdominal adiposity (13).

However, interventions targeting these risk factors and diseases vary widely, ranging from high-intensity resistance exercise and high-intensity aerobics or brisk walking to Tai Chi and balance training. In parallel, there are vast differences in exercise components. For example, exercise interventions that promote bone strength generally use intense mechanical strains with high-strain intensities and/or strain rates but low to moderate repetition numbers and durations (18). In contrast, the opposite strategy is used to improve metabolic or cardiovascular parameters: low to moderate intensities and moderate to high numbers of repetitions and durations (9). These discrepancies may explain the failure of studies (22,29) that focus on "contradictory" end points (i.e., weight loss and/or maintenance of bone mineral density) with isolated training strategies.

Nevertheless, the implementation of effective multi purpose exercise programs is highly relevant for our mostly sedentary and ageing society because due to time constraints or simple reluctance, even subjects with multiple risk factors will refuse to simultaneously participate in several specified exercise programs (19).

Thus, the primary goal of the Senior Fitness and Prevention Study (SEFIP) was to validate a general purpose exercise program with an acceptable training volume of two sessions per week and low requirements for technical training facilities and materials that could easily be transferred from a university environment to community-based rehabilitation and prevention programs. In this contribution, we focused on the effectiveness of our exercise program that was based on a high-intensity philosophy on the metabolic syndrome (MetS) as a cluster of cardiovascular and metabolic risk factors. We used a subgroup of the SEFIP cohort with MetS as defined by the international diabetes foundation (IDF) (1).

Back to Top | Article Outline


The Senior Fitness and Prevention Study (SEFIP) was a randomized, single-blinded intervention study with 246 females 65 yr and older. The study was approved by the ethics committee of the University of Erlangen (Ethik Antrag 3354) and the Bundesamt für Strahlenschutz (Z5-22462/2-2005-026). All study participants gave written informed consent. The study was registered under www.clinicaltrials.gov.

Back to Top | Article Outline

End points

The primary end points of the SEFIP-study were fracture parameters (bone mineral density and fall frequency). Secondary end points were CHD-risk, determined by MetS criteria. As mentioned above, in this subgroup analysis, we focused on subjects with the MetS according to IDF.

Back to Top | Article Outline


Table 1 outlines the patient flow of the SEFIP study. Briefly, 5855 postmenopausal women aged 65 yr and older, living independently in the community of Erlangen-Nürnberg, Germany, were contacted by mail between May 2005 and January 2006. Mailings lists were obtained from the database of the Siemens Health Insurance Company (Siemens Betriebs Krankenkasse). Six hundred and fifty-nine (659) women responded and were contacted from research assistants by telephone. Two hundred and eighty-three (283) subjects were not admitted to the study due to 1) medication or diseases affecting bone metabolism during the last 2 yr (n = 243); 2) participation in exercise studies during the last 2 yr (n = 10); 3) acute medication affecting fall risk (n = 11); 4) history of profound coronary heart diseases (stroke, cardiac events; n = 13) and inflammable diseases (n = 2); 5) known secondary osteoporosis (n = 2); and 6) very low physical capacity (<50 W at cycle ergometry, n = 2).

The remaining 376 subjects were invited to meetings with detailed information concerning the study protocol. From these, 77 subjects disliked study protocol mainly due to the randomization procedure. Two hundred and ninety-nine subjects were screened. However, immediately before the randomization procedure, three more subjects refused to participate because they could not choose their preferred study arm. Thus, using computer-generated block randomization stratified for age, 296 subjects were finally assigned to three interventions: 1) exercise (EG: n = 123), 2) wellness control (CG: n = 123), and 3) exercise and whole-body vibration (WBV; n = 50). Subjects receiving WBV were part of the Erlanger Longitudinal Vibration Study, a substudy with extended eligibility criteria (30).

Ninety-five subjects from the SEFIP cohort of 246 subjects with MetS as defined by IDF (1) were included in the present analysis. Of those, 45 subjects participated in the EG and 50 in the CG group (Table 1).

Back to Top | Article Outline

Baseline and 12 Month Measurements

In the baseline procedure and at the follow-up visits, measurements were taken at the same time during the day (±1 h) and by the same research assistant.

Back to Top | Article Outline

MetS as Defined by the IDF

According to the new definition of IDF (1), the MetS consists of abdominal adiposity (waist circumference ≥80 cm for European females) as a necessary criteria and two out of the following criteria: 1) raised triglyceride levels (≥150 mg·dL−1); 2) reduced HDL-C (<50 mg·dL−1 for females or specific treatment for previously detected hypertriglyceridemia/reduced HDL-C); 3) raised blood pressure (systolic blood pressure ≥130 mm Hg, diastolic blood pressure ≥85 mm Hg, or specific treatment); and 4) raised fasting plasma glucose (≥100 mg·dL−1).

Back to Top | Article Outline

Anthropometric Data

We measured height, weight, waist and hip circumferences, and total and regional body composition. Waist circumference was measured as the minimum circumference between the distal end of the rib cage and the top of the iliac crest along the midaxillary line. Body composition was determined by dual-energy x-ray absorptiometry (Discovery; Hologic, Bedford, MA).

Back to Top | Article Outline

Blood Pressure

Blood pressure was determined in a sitting position 5 min after rest with an automatic oscillometric device (Bosco; Bosch, Jungingen, Germany). All measurements were taken in the nonfasting condition. Subjects were refrained from coffee or tea for at least 2 h before testing and more than 12 h after the last relevant physical exertion or exercise session.

Back to Top | Article Outline

Blood Lipids

After an overnight fast, blood was sampled in the morning (7:00-9:00 a.m.) in a sitting position from an antecubital vein, which was subsequently analyzed for fasting glucose, total and HDL-C, and triglycerides. Serum samples were centrifuged at 3000 RPM for 20 min and frozen at −70°C. Glucose, total cholesterol, HDL-C, triglycerides (Olympus Diagnostica GmbH, Hamburg, Germany), and hsCRP (Roche Diagnostics, Mannheim, Germany) were determined.

Back to Top | Article Outline


Detailed baseline questionnaires completed by all participants combined several parts: 1) social status, demographic parameters, and living conditions of the participants; 2) health status and medical conditions; 3) general health risk factors and specific osteoporotic risk factors including falls and fall history; 4) prestudy physical activity and exercise levels (17); and 5) frequency and volume of medical care structured by type of physician.

To determine changes for confounding parameters during the intervention period, we performed structured interviews with the participants at the end of the study each by the same physician as during the baseline measurements.

Back to Top | Article Outline

Nutritional Analysis

The individual dietary intake was assessed by a 4-d protocol completed by each study participant at baseline and after 18 months. Subjects were carefully instructed on correct weighting and reporting by research assistants. The complete food intake was recorded. For precise weighting of the consumed food, the participants were equipped with identical digital household food scales. The analysis of the protocols was performed using Prodi-4.5/03 Expert software (Wissenschaftlicher Verlag, Freiburg, Germany), which extracts a total of 1500 different basic nutritional ingredients. On the basis of the results from this analysis, all study participants were individually supplemented with calcium (Ca) and cholecalciferol (vitamin D) to ensure a total daily intake of 1500 mg of Ca and of 500 IU of vitamin D. The nutritional analysis was performed in close collaboration with the Department of Sports Medicine, University of Bayreuth.

Back to Top | Article Outline


Two interventions were compared in this study: a complex high-intensity aerobic and resistance exercise program that primarily focussed on bone strength versus a "wellness" program with low training frequency that focussed on well-being. Both interventions was supervised by certified instructors. Participants training were rearranged by expert supervisors every 6-12 wk. The attendance protocols and the individual training logs taken by the participants were analyzed every 2 months (protocols) and every 6 months (training logs) to monitor attendance and compliance.

All subjects were individually supplemented with Ca and cholecalciferol (vitamin D) as described above and, apart from the added intervention, were asked to maintain their habitual lifestyle.

Participants were blinded to the underlying hypotheses because benefits of exercise and the wellness program were discussed before the study. Exercise and wellness groups were trained at different locations by different supervisors to prevent contact between the intervention groups. During the measurements, outcome assessors and research assistants were not allowed to ask subjects about their allocated intervention.

Back to Top | Article Outline

Exercise Group

The exercise program consisted of four sessions per week, divided into two supervised group sessions with 8-13 participants lasting 60 min each and two nonsupervised individual home training sessions of 20 min each. In the following sections, we summarize the joint exercise program.

Back to Top | Article Outline

Group Exercise Session

The joint exercise session consisted of four sequences: 1) warm-up and endurance; 2) coordination; 3) isometric strength training, functional gymnastics, and stretching; and 4) dynamic strength training.

Back to Top | Article Outline

Warm-up/endurance sequence.

Twenty minutes of aerobic dance exercises with a progressively increasing amount of high-impact exercises was performed. HR ranged between 70% and 85% HRmax during the sequence.

Back to Top | Article Outline

Coordination sequence.

Besides the aerobic dance exercises that focussed on general coordinative skills, a brief (5 min) specific coordination sequence focussed on static and dynamic balance. Three to five exercises under progressively increased postural instability conditions were performed in a standing position.

Back to Top | Article Outline

Resistance training sequence.

Resistance training was subdivided into two sequences: 1) functional gymnastics, stretching, and isometric exercises and 2) dynamic exercises.

Back to Top | Article Outline

Isometric strength training, functional gymnastics, and stretching.

The 10-15 isometric floor exercises with one to two sets each with special emphasis on trunk extensors/flexors, hip extensors/flexors, and leg abductors/adductors were performed with 6-8 s of maximum intensity and a 20- to 30-s active rest period between the sets/exercises. Intensity of this section was progressively increased by introducing more strenuous isometric exercises every 8-12 wk. Between the isometric exercises, stretching exercises (20-30 s of continuous stretching) for the corresponding muscle group were performed.

Back to Top | Article Outline

Dynamic strength training.

Three exercises with two sets each, 12 to 15 repetitions, and a 2-s concentric/1-s static/2-s eccentric time under load protocol were additionally carried out with special regard to the upper trunk (low and high belt rowing and belt shoulder raises) using elastic belts (Thera-Band, Hadamer, Germany). Intensity of the belt exercises was increased by using belts with different tension grades (3.5, 4.5, and 6 kg per 100% extension) and progressive belt shortening. Subjects were instructed to perform at a maximum minus two-repetition exertion level. Again, continuous stretching exercises for the corresponding muscle groups were established during the rest periods of the dynamic exercises (≍30-40 s).

Three dynamic leg exercises (initially: leg heel raises, front lounge, and leg abduction) with two sets each were conducted on a 25-cm platform in a circuit mode, with 1 min of exercise and 1 min of active rest with stretching exercise for the corresponding muscle groups. Time under tension for each resistance exercise was given 2-0-2 s; thus, eight to nine repetitions per leg could be performed during 60 s. Subjects were asked to perform at a maximum minus a two-repetition exertion level. Intensity was progressively increased by enlarging the amplitude of the movement, changing the velocity of concentric execution, and introducing more strenuous exercises. Thus, finally a deep front lounge with toe raise, a deep front lounge with an explosive concentric movement, and a leg abduction exercise with toe raises were executed.

Back to Top | Article Outline

Home Training Session

Home-based training focussed on strength training consisting of one to two sets of six to eight isometric exercises, two to three belt exercises with two sets of 10-15 repetitions, and four to six intermitted stretching exercises. Participants were provided with instruction booklets describing each exercise and elastic belts. To ensure proper technique, home exercises were first discussed and trained in the joint training sessions. The home training exercises were replaced every 6 months by different and more intense ones.

Back to Top | Article Outline

Wellness Control Group

The central aim of the wellness program was to briefly introduce and to try out training topics of health-related physical activity and to increase well-being with little impact on physical parameters. Subjects of the CG performed a 60-min low-intensity physical activity and relaxation program once a week for blocks of 10 wk with intermitted 10-wk intervals without training. After 5-10 min of moderate walking at 50-60% HRmax, each of the 10 sessions of a block focussed on another objective: 1) relaxation, 2) games interaction, 3) general coordination, 4) endurance, 5) balance, 6) dances, 7) body sensitivity, 8) muscle strength, 9) breathing, or 10) flexibility.

Each session was terminated by a 10-min progressive muscular relaxation sequence. During the sessions with endurance or resistance exercises, subjects were asked to perform with low to moderate effort. This procedure was repeated during each of the following 10-wk blocks without increasing exercise volume or intensity

Back to Top | Article Outline

Statistical Analysis

All baseline values are reported either as means with SD or medians with interquartile ranges. The Kolgomorov-Smirnov test was used to ascertain for normal distribution. Homogeneity of variance was investigated using Levine's F-test. The difference in the reduction of the prevalence of the MetS at 12 months was checked using the χ2 test. For normally distributed variables, differences within groups were assessed with paired t-tests; otherwise, the Wilcoxon test was used. Twelve-month data were reported as absolute changes compared with baseline.

According to the data, nonparametric tests, t-tests, or repeated-measures ANOVA was performed to check time-group interactions. Between-group differences were given as absolute difference along with 95% confidence interval (CI). All tests were two tailed, and a 5% probability level was determined statistical significant. Further effect sizes (ES) based on the absolute difference between baseline and 12-month control in both groups were calculated using Cohen's d (6). We used SPSS 14.0 (SPSS Inc., Chicago, IL) for all statistical procedures.

Back to Top | Article Outline


Complete 12-month data were available for 86 subjects with MetS. As presented in Table 1, in both groups three subjects quit the study. Three of these cited study-related reasons in respect to either the exercise protocol or the Ca/vitamin D supplementation. The other three subjects were unable to visit the 12-month assessment.

Twenty-one women were excluded from the statistical analysis by protocol: 14 subjects changed their medication affecting blood pressure and blood lipids, and 7 subjects lost more that 5% of their body mass by weight reduction programs (Table 1). Thus, the results given here are from 33 subjects of the EG and 32 subjects of the CG.

Attendance rate of the exercise group averaged 74% ± 7% for the joint session and 41% ± 9% for the home training session resulting in an average attendance of 2.3 sessions per week per year.

Tables 2-5 show the baseline values for anthropometric parameters and risk factors in both groups. There were no significant differences between exercise and wellness group for known confounding anthropometric, metabolic, or physical fitness parameters. Further, no changes of possibly confounding parameters were observed during the intervention period.

Back to Top | Article Outline

Metabolic syndrome.

Table 3 shows numerical changes of the parameters defining MetS during the study. Waist circumference was not significantly changed in either group (EG: −0.7 ± 3.1 cm, P = 0.18 vs CG: −0.6 ± 3.3 cm, P = 0.26), although hip circumference significantly decreased in the EG (−2.1 ± 2.6 cm, P < 0.001) resulting in a significant difference between both groups (Table 4). Triglycerides (P = 0.007) and HDL-C (P = 0.014) favorably changed in the exercise group and slightly deteriorated in the CG (P > 0.90), resulting in significant between-group differences for both parameters.

At baseline, systolic and diastolic blood pressure differed between both groups; thus, the analysis was adjusted for baseline data. Systolic and diastolic blood pressure both significantly decreased in the exercise (diastolic: −8.0 ± 6.6 mm Hg; systolic: −7.1 ± 11.3 mm Hg) and the control group (diastolic: −9.8 ± 9.7 mm Hg; systolic: −6.4 ± 14.9 mm Hg). However, after 12 months, between-group differences were not significant. Resting fasting plasma glucose slightly increased in both groups (EG: +0.7 ± 8.8 mg·dL−1, P = 0.67 vs CG: +1.0 ± 9.9 mg·dL−1, P = 0.57) but did not result in significant between-group changes.

Figure 1 demonstrates that in the exercise group the number of criteria (i.e., values above the threshold given by the IDF definition) of the MetS significantly (P = 0.002) decreased from 4.13 ± 1.18 to 3.66 ± 1.23. In the control group favorable but not significant (P = 0.34) changes were determined (4.12 ± 0.79 to 3.97 ± 0.79). However, significant between-group differences (P = 0.15; CI = −0.74 to 0.12 scores; effect size (ES) = 0.36) were not observed. Prevalence of MetS decreased by 30.3% in the EG compared with 15.6% in the CG (P = 0.15).

Back to Top | Article Outline

Body fat and body composition.

Total body fat (EG: −1287 ± 1513 g, P < 0.001 vs CG: +483 ± 2463 g, P = 0.28) and trunk fat (EG: −1070 ± 990 g, P < 0.001 vs CG: −203 ± 1342 g, P = 0.41) significantly decreased in the EG but did not change in the CG. Between-group differences were significant for both parameters (Table 4). Total lean body mass slightly increased in both groups (EG: +394 ± 1123 g, P = 0.056 vs CG: +137 ± 1213 g, P = 0.53). With the exception of lean body mass (LBM) within-group changes significantly differed between EG and CG for all parameters listed in Table 3.

Back to Top | Article Outline

Total cholesterol and high-sensitivity C-reactive protein (hsCRP).

Results for total cholesterol and hsCRP are given in Table 5. Total cholesterol (P = 0.001) significantly improved in the exercise group and was stable in the CG (P > 0.90). Significant differences between both groups were determined for this parameter (Table 5). High-sensitivity C-reactive protein decreased in both groups (EG: 2.38 mg·L−1 to 1.71 mg·L−1, P = 0.042 vs CG: 3.37 mg·L−1 to 2.86 mg·L−1, P = 0.086); however, the reduction was significant in the EG only. No significant group/time effect was determined.

Back to Top | Article Outline


In this study, we investigated the effect of a multipurpose exercise program on the MetS defined as a cluster of cardiovascular and metabolic risk factors. In summary, we demonstrated that a high-intensity strength and endurance exercise training impacted a variety of MetS criteria. It is of further interest that even the rather moderate intervention of our (wellness) control group also substantially benefited blood pressure. Our results extend the existing data by the important message that general purpose exercise programs primarily designed to reduce bone fracture risk (16) are also effective to reduce CHD-risk factors in elderly females at risk.

Our study has several strengths. 1) We specifically targeted a homogenous community living group of females 65 yr and older with MetS according to IDF. 2) Covariates such as diseases, medication, nutrition, or lifestyle changes were strictly controlled throughout the study. Significant changes resulted in an exclusion of the corresponding subject from the analysis. 3) Our control group carried out a physical activity program focussing on well-being without impacting on physiological parameters. As exercise studies cannot be blinded, it is more acceptable to the participants of the "placebo group" if they also carry out some activity, which, however, should not affect the end points. 4) The study duration of 1 yr was long enough to detect relevant changes of body composition. 5) The intensity of the exercise regimen was progressively increased (more strenuous exercises, increased ROM, increased tension grade of the belts) during the interventional period, and the joint sessions were strictly supervised by certified trainers. 6) The requirements for auxiliary technical training materials (i.e., Thera Band) were low; thus, the program could be easily adopted by others. 7) In addition to MetS criteria, we determined changes of adjuvant and extended risk factors for CHD.

One limitation of the study was the subgroup analysis strategy that retrospectively reduces the power of the randomized protocol. However, as the evaluation of exercise effects on MetS was a secondary end point of the overall study design, we believe that this approach was acceptable. Also for all parameters listed in Table 1, there were no differences between the randomized cohort and the MetS subjects. Further, despite our effort to exercise with very low intensity and magnitude, it cannot be excluded that the wellness intervention affects parameters with low-strain thresholds (4).

Contrary to other exercise studies (2,12), we prefer to determine the effect of our exercise program on the number of risk factors of the MetS because it can be quantified on finer scale compared with just analyzing the prevalence of the MetS. This is particular true in studies investigating cohorts with 100% prevalence because negative changes that may be induced by the exercise protocol could not be determined. However, for a better comparison, we also calculated percentage changes of the prevalence for our study groups.

Due to the different definitions of MetS, it was difficult to directly compare our results with those from other studies. As reported by Sandhofer et al. (23), the prevalence of MetS in healthy, elderly European females (50-70 yr) is highest for the IDF definition (19.5%) compared with 16.2% for the World Health Organization or 17.0% for the National Cholesterol Education Program (NCEP) definition. Therefore, in the following discussion, it must be considered that concerning changes, the exercise effect on the specific criteria selected by these definitions varies; thus, changes of the prevalence may differ despite the same absolute changes.

At least three studies demonstrated a reduction of MetS prevalence by exercise interventions (2,12,25). Anderssen et al. (2) evaluated the effect of a supervised endurance program (3 × 60 min·wk−1 at 60-80 HRmax) with or without diet on IDF-defined MetS prevalence in middle-aged men [45 ± 3 yr, body mass index (BMI): 29 ± 3 hg·m2. Similar to our study, the males (n = 137) with MetS were recruited from a larger trial, the "Oslo Diet and Exercise Study." In the exercise group, the MetS prevalence was reduced by −24%; however, contrary to the combined diet and exercise group (−67%), the MetS reduction in exercise group was not significantly different from the control group (−12%). Although we did not focus on the reduction of MetS prevalence in our cohort, our data were comparable (EG: −30% vs CG: −16%; P = 0.15).

In a subgroup of the HERITAGE Family Study, Katzmarzyk et al. (12) used the NCEP Adult Treatment Panel (ATP) III definition and determined MetS prevalence reduction in black and white males and females (17-65 yr). Initially, the prevalence was 16.9% for the entire, apparently healthy cohort (n = 621), with a lower prevalence in white females (12.6%). After 20 wk of progressive endurance exercise (3 × 30-50 min·wk−1 at 55-75% V˙O2max), 28% of the females (n = 50) no longer had MetS. However, 4% had developed MetS during the exercise trial. Comparable to our study, Katzmarzyk et al. (12) showed a reduction of the number of MetS criteria poststudy. As a limitation, there was no control group in the study of Katzmarzyk et al. (12).

Comparable to our study, Steward et al. (25) investigated the effect of a mixed strength and endurance exercise protocol (three times a week, seven exercises with two sets of 10-15 repetitions at 50% one repetition maximum; 45 min of endurance stationary exercise at 60-90 HRmax) on MetS prevalence in elderly subjects (n = 104, age = 55-75 y). Initially, 42% of their cohort with elevated blood pressure showed MetS according to the NCEP ATP III definition. After 6 months, nine subjects of the exercise (18%) and eight of the control group (15%) no longer had MetS; however, four participants (8%) of the control group developed MetS (overall between-group difference: P = 0.06). In parallel, the number of MetS criteria decreased by −0.65 (CI = −0.89 to −0.40) in the exercise group and by −0.30 (CI = −0.56 to −0.04) in the control group (between-group difference: P = 0.06).

To evaluate which criteria were primarily affected by the exercise program and were thus responsible for the reduction of our primary end point, we determined changes of MetS criteria according to IDF as well as some adjuvant CHD-risk factors. Abdominal obesity as measured by waist circumference was not affected. In contrast, related obesity parameters like hip circumference and total body and trunk fat showed significant exercise effects. However, although waist circumference was confirmed as a predictor of abdominal body fat (11), longitudinal changes of abdominal fat may be not adequately detectable by this technique. Likewise, Kay and Fiatarone (13) reported exercise-induced reductions of abdominal visceral fat up to 48% as determined by CT or MRI without corresponding changes of waist circumference.

In our study, MetS status was primarily affected by favorable changes of triglycerides (−15%) and HDL-C (+6%) in the EG with minor negative changes in the CG. In their meta-analysis of randomized controlled trials, Kelley et al. (15) reported average reductions of 2-5% for total cholesterol, LDL-C, and triglycerides and increases of 3% for HDL-C after aerobic exercise in females.

In their review of exercise effects on blood lipids in postmenopausal women, Asikainen et al. (3) reported positive changes in overweight and dyslipidemic subjects but minor or no effects in healthy postmenopausal females. Although most randomized controlled trials focus on aerobic training, favorable changes of triglycerides and HDL were also detected in postmenopausal women after resistance training (5) or mixed aerobic/resistance training programs (25). Comparing the effect of vigorous (>60% V˙O2max) versus moderate exercise intensity in their review, Swain and Franklin (26) reported no intensity effect on blood lipids.

Interestingly, Swain et al. (26) determined greater improvements of diastolic blood pressure after vigorous exercise programs. In contrast to these results, we detected comparable changes of diastolic and systolic blood pressure in both groups. As mentioned above, it was difficult to address the significant improvements of blood pressure in the control group. Although there is no evidence, we speculate that the underlying pathway for these improvements is not caused by exercise-induced physiological changes.

We also failed to detect longitudinal differences for fasting plasma glucose among the groups. Although this parameter is quite variable from day to day, we believe that our sample size is high enough to detect relevant changes. It was difficult to compare these data with the corresponding literature because most studies in this area focussed on other cohorts and end points. Kelley and Goodpaster (14), who summarized the effect of physical exercise on insulin resistance and glucose tolerance in a meta-analysis, reported favorable changes of insulin resistance in obese subjects and in subjects with diabetes mellitus. Improvements of glucose tolerance were less consistent and seem to be related with baseline values of glucose intolerance and exercise intensity. Swain et al. (26) confirmed the latter finding with greater improvements after vigorous exercise.

High-sensitivity C-reactive protein decreased in both groups (−29% vs −15%); however, changes were significant in the exercise group only. Thus, at least changes of the exercise group were within the range of 16-41% hsCRP reductions reported for dedicated longitudinal exercise interventions (21). Interestingly, changes of hsCRP seemed to be independent of baseline levels, body composition, and weight loss. Although further studies are required to determine the dose-response effect, Pleasance and Grandjean (21) recommended in their review three to four sessions per week with 40-80 min·d−1 duration at an intensity of 70-80% V˙O2max to lower hsCRP.

In summary, our general purpose exercise program based on a low-volume, high-intensity philosophy affected direct and indirect MetS risk factors and significantly lowered the severity of MetS in females with MetS. Our high-intensity strategy was confirmed by the review of Ford and Li (7), who noted that moderate or high intensity was more strongly inversely related with the MetS compared with low-intensity exercise. The data of our exercise group confirm this statement, although we observed a trend that low-volume and low-intensity exercise as performed in our (wellness) control group also positively affected the MetS prevalence. Thus, dose-response studies are required to exactly determine the minimum effective exercise dose.

The authors gratefully acknowledge the support of Opfermann GmbH und Co. (Whiel, Germany) who supplied calcium and vitamin D for all study participants. Further, the authors warmly thank the Siemens Betriebs Krankenkasse Erlangen (SBK, Erlangen, Germany) for their general support. The authors also acknowledge the support of the Thera-Band GmbH, Mtd-Systems (Neuburg v. Wald, Germany) and the Viasys (Conshohuocken, PA, USA). None of the authors has any conflict of interest. Further, the results of the study do not constitute endorsement by ACSM.

No funding was received for this work from NIH, Wellcome Trust, HHMI.

Back to Top | Article Outline


1. Alberti KG, Zimmet P, Shaw J. Metabolic syndrome-a new world-wide definition. A consensus statement from the International Diabetes Federation. Diabet Med. 2006;23(5):469-80.
2. Anderssen SA, Carroll S, Urdal P, Holme I. Combined diet and exercise intervention reverses the metabolic syndrome in middle-aged males: results from the Oslo Diet and Exercise Study. Scand J Med Sci Sports. 2007;17(6):687-95.
3. Asikainen TM, Kukkonen-Harjula K, Miilunpalo S. Exercise for health for early postmenopausal women. Sports Med. 2004;34(11):753-78.
4. Asikainen TM, Miilunpalo S, Oja P, et al. Randomised, controlled walking trials in postmenopausal women: the minimum dose to improve aerobic fitness? Br J Sports Med. 2002;36(3):189-94.
5. Castaneda C, Layne JE, Munoz-Orians L, et al. A randomized controlled trial of resistance exercise training to improve glycemic control in older adults with type 2 diabetes. Diabetes Care. 2002;25(12):2335-41.
6. Cohen J. Statistical Power Analysis for the Behavioral Sciences. 2nd ed. Hillsdale (NJ): Lawrence Earlbaum Associate; 1988. pp. 20-5.
7. Ford ES, Chaoynag L. Physical activity or fitness and the metabolic syndrome. Expert Rev Cardiovasc Ther. 2006;4(6):897-915.
8. Gillespie LD, Gillespie WJ, Robertson MC, Lamb SE, Cumming RG, Rowe BH. Interventions for preventing falls in elderly people. Cochrane Database Syst Rev. 2003;1-112.
9. Haennel RG, Lemire F. Physical activity to prevent cardiovascular disease. How much is enough? Can Fam Physician. 2002;48:65-71.
10. Houde S, Melillo KD. Cardiovascular health and physical activity in older adults: an integrative review of research methodology and results. J Adv Nurs. 2002;38(3):219-34.
11. Kamel EG, McNeill G, Han TS, et al. Measurement of abdominal fat by magnetic resonance imaging, dual-energy x-ray absorptiometry and anthropometry in non-obese men and women. Int J Obes Relat Metab Disord. 1999;23(7):686-92.
12. Katzmarzyk PT, Leon AS, Wilmore JH, et al. Targeting the metabolic syndrome with exercise: evidence from the HERITAGE Family Study. Med Sci Sports Exerc. 2003;35(10):1703-9.
13. Kay SJ, Fiatarone Singh MA. The influence of physical activity on abdominal fat: a systematic review of the literature. Obes Rev. 2006;7(2):183-200.
14. Kelley DE, Goodpaster BH. Effects of physical activity on insulin action and glucose tolerance in obesity. Med Sci Sports Exerc. 1999;31(11 suppl):S619-23.
15. Kelley GA, Kelley KS, Tran ZV. Aerobic exercise and lipids and lipoproteins in women: a meta analysis of randomized controlled trials. J Womens Health. 2004;13(10):1148-64.
16. Kemmler W, von Stengel S, Kalender WA, Engelke K. Exercise prevents falls and maintain bone mineral density in elderly postmenopausal women. Preliminary data of the Erlangen Senior Fitness and Prevention Study. J Back Musculoskelet Rehabil. 2007;22(suppl 1):S336.
17. Kemmler W, Weineck J, Kalender WA, Engelke K. The effect of habitual physical activity, non-athletic exercise, muscle strength, and V˙O2max on bone parameters is rather low in early osteopenic postmenopausal women. J Musculoskelet Neuronal Interact. 2004;4(3):325-34.
18. Kemmler W, Weineck J, Hensen J, Lauber D, Kalender WA, Engelke K. Exercise recommendations for an increase of bone strength based on animal models and studies with athletes. (Empfehlungen für ein körperliches Training zur Verbesserung der Knochenfestigkeit: Schlussfolgerungen aus Tiermodellen und Untersuchungen an Leistungssportlern). Dtsch Z Sportmed. 2003;54(11):306-16.
19. Marcus R. Exercise: moving in the right direction. J Bone Miner Res. 1998;13(12):1793-6.
20. Pedersen BK, Saltin B. Evidence for prescribing exercise as a therapy in chronic disease. Scand J Med Sci Sports. 2006;16(suppl 1):3-63.
21. Plaisance EP, Grandjean PW. Physical activity and high-sensitivity C-reactive protein. Sports Med. 2006;36(5):443-58.
22. Salamone LM, Cauley JA, Black DM, et al. Effect of a lifestyle intervention on bone mineral density in premenopausal women: a randomized trial. Am J Clin Nutr. 1999;70(1):97-103.
23. Sandhofer A, Iglseder B, Paulweber B, Ebenbichler CF, Patsch JR. Comparison of different definitions of the metabolic syndrome. Eur J Clin Invest. 2007;37(2):109-16.
24. Sigal RJ, Kenny GP, Wasserman DH, Castaneda-Sceppa C. Physical activity/exercise and type 2 diabetes. Diabetes Care. 2004;27(10):2518-39.
25. Stewart KJ, Bacher AC, Turner K, et al. Exercise and risk factors associated with metabolic syndrome in older adults. Am J Prev Med. 2005;28(1):9-18.
26. Swain DP, Franklin BA. Comparison of cardioprotective benefits of vigorous versus moderate intensity aerobic exercise. Am J Cardiol. 2006;97(1):141-7.
27. Tesch-Römer C. Sustainable retirement benefits-national and international perspectives. In Situation of Elderly Subjects in Germany. (Zukunftsfähige Altersvorsorge- deutsche und internationale Perspektiven. In "Lebenslagen älterer Menschen in Deutschland.") Berlin; 2004. p. 19-21.
28. Toth MJ, Beckett T, Poehlman ET. Physical activity and the progressive change in body composition with aging: current evidence and research issues. Med Sci Sports Exerc. 1999;31(11 suppl):S590-6.
29. Uusi-Rasi K, Sievanen H, Pasanen M, Oja P, Vuori I. Maintenance of body weight, physical activity and calcium intake helps preserve bone mass in elderly women. Osteoporos Int. 2001;12(5):373-9.
30. von Stengel S, Kemmler W, Engelke K, Kalender W. First results of the Erlangen Longitudinal Vibration study (ELVIS). (Erste Ergebnisse der Erlanger Longitudinalen Vibrations-Studie (ELVIS).) Osteologie. 2007;16(suppl 1):20.
31. Vuori IM. Dose-response of physical activity and low back pain, osteoarthritis, and osteoporosis. Med Sci Sports Exerc. 2001;33(6 suppl):S551-86.


©2009The American College of Sports Medicine