This randomized controlled trial showed that, in this group of elderly men and women, a 12-month PA intervention and an SA intervention reduced the prevalence of MetS, but the PA intervention did not reduce the prevalence of MetS, the average number of MetS components, or the z-MetS more than the SA intervention. However, in the subgroup of participants not taking any medication for treating MetS components, those in the PA intervention had lower odds of having MetS compared with the SA intervention. This suggests that the effect of PA in this group of elderly men and women may be blunted by the use of medications to treat MetS components.
The findings of no significant differential effects of the 12-month PA intervention compared with SA intervention were somewhat unanticipated considering the beneficial effects of PA shown in previous studies (17
). However, in a recent randomized controlled study in 67- to 76-year-old men and women (24
), 12 weeks of aerobic exercise training did not significantly change a composite metabolic risk score (systolic and diastolic BP, 2-hour postprandial glucose, fasting insulin, HDL-C, triglycerides, and waist circumference) or its constituents (except waist circumference). Our results were in line with their findings but with a much longer intervention. However, our results do not exclude metabolic effects of the PA intervention because in the entire sample, the prevalence of MetS was reduced and triglycerides and systolic BP were also reduced after 12 months.
In comparison with other studies in older adults (17
), our participants were older, and a large proportion (78.1%) were taking medications for treating at least one of the MetS components. Because using medication for treating MetS components was considered having met the criterion, it is not sensitive to changes in medication dose or the number of medications taken for the same MetS component. It is also unlikely that a participant’s physician would discontinue the medication during the PA intervention. In fact, we observed greater beneficial effects of PA compared with SA in those not taking any medication for treating MetS. Also, the exercise intensity and volume in our study were lower than in the few studies showing beneficial effects of exercise training alone on metabolic risks and MetS in older adults (17
Moreover, in the above-mentioned studies showing metabolic effects of exercise training, two had a small degree of exercise-induced weight loss (17
), and one showed approximately 4% fat mass loss (22
). Another study in middle-aged and older adults showed that 16-week moderate-intensity continuous aerobic exercise and high-intensity interval training both improved MetS, and both groups had small but significant weight loss (3%–4%) (15
). In the study that did not show exercise improved metabolic composite, the decreases in body weight (from 77.2 to 77.0 kg) and waist circumference (from 98.6 to 97.8 cm) in the exercise group were almost negligible (24
). In our study, the PA intervention did not affect body weight, fat mass, or waist circumference overall, and the odds of having MetS were lower in those who lost 5% or more weight than those who did not, regardless of intervention assignment. Additionally, a recent study in obese postmenopausal women showed that those who lost more fat were more likely to resolve MetS (19
). Thus, the metabolic benefits of exercise training may be less clear in the absence of weight loss or fat loss.
Importantly, the SA intervention in our study was not a no-intervention control. Although the PA intervention did not result in greater changes in MetS than the SA and both interventions decreased the prevalence of MetS, it should not be interpreted as the PA intervention had no effect. Perhaps in this group of elderly men and women, the health education SA intervention enabled some unmeasured behavioral changes in those participants that led to beneficial health effects.
At baseline, higher HOMA score and greater amount of fat mass were associated with greater odds of having MetS. Fat mass change and HOMA score change, however, were not associated with the presence of MetS during the intervention after adjusting for baseline MetS and covariates. It is widely accepted that insulin resistance and adipose tissue play important roles in the development of MetS (1
). However, HOMA score is an index, not a direct measure for insulin resistance, and may not be adequate to detect changes in insulin resistance. Similarly, more specific measures of adiposity, such as visceral adipose tissue, may be more sensitive to show changes due to intervention.
In summary, this study showed that in a group of men and women aged 70–89 years old, a PA intervention did not result in lower prevalence of MetS, above beyond that of an SA health education intervention. Our findings suggest that medication use may override any beneficial metabolic effects of PA in this age group. It should be noted that the SA group also demonstrated slightly increased PA calories during follow-up visits, which may have contributed to the reduced MetS prevalence at 6 months in this group. Further studies are needed to determine whether a greater PA volume and/or intensity that elicits weight loss will result in greater beneficial metabolic effects in this age group.