The major finding of the present study was that a small volume of home-based exercise in addition to a slight caloric restriction promoted a higher reduction in body weight during the first six months in comparison with a caloric restriction only. However, this change was not sustainable during one year of follow up. In addition, the exercise program did correlate with a substantial increase in serum HDL cholesterol levels, even at 12 months, indicating continued compliance and resulting in important health benefits. Also, weight loss was greater among those with higher compliance.
The role of exercise on body weight maintenance is considered important (Foster, 2006
; Jequier, 2002
), but there is no recommendation based on well designed randomized controlled trial. Redman et al. (2007)
compared the effects of a 25% energy deficit by diet alone or diet plus exercise in 35 overweight subjects during 6 months. The authors found no additional effects of physical exercise on body weight, fat mass, and fat-free mass in comparison with diet alone, and concluded that exercise plays a role equivalent to caloric restriction in terms of energy balance.
On the other hand, in a randomized controlled trial, McTierman et al.(2007) studied the effects of 300 minutes (high volume) of moderately vigorous exercise without dietary intervention on prevention of weight gain during 12 months of follow up. Women lost 1.4+1.8 kg compared to an increase of 0.7+0.9 kg in controls and men lost 1.8 +1.9 kg compared to an increase of 0.1+0.1 in controls. It is important to highlight that our study had a small caloric restriction in both groups allowing examining the effects of a smaller amount of exercise that could be recommended in clinical practice.
A meta-analysis by Curioni and Lourenço (2005)
compared six randomized clinical trials (n = 265) with follow-up ranging from 10 to 52 weeks, and found a 20% greater weight loss in diet-plus-exercise groups, as well as a 20% greater sustained weight loss after 1 yr, when compared with the diet-only groups. Another recent meta-analysis (Shaw, et al., 2006
) found a small yet significant decrease of 1kg in body weight in the diet-plus-exercise group when compared to the diet-only group. In addition, Bond Brill et al. (2002)
studied the effects of walking in combination with a low-fat diet on weight loss and other health-related variables in 88 overweight women and, in accordance with our finding, a relatively small amount of exercise (30 min/d) during eight months of follow-up reversed the weight gain experienced in the control group.
The effects of different amounts of exercise training on body weight was evaluated by Slentz et al. (2004)
in 120 overweight dyslipidemic subjects during eight months, with no changes in diet. They found that a modest amount of exercise (30 min/d) during the follow-up was positively associated with weight maintenance and improved the adherence to exercise programs. Poor adherence to a given exercise protocol may be one of the main reasons why randomized controlled trials often fail to find an association between physical activity and weight loss, especially in obese individuals. The prescribed amount of exercise may have paramount importance on exercise adherence, with more demanding exercise programs associated with poorer adherence (Fogelholm and Kukkonen-Harjula, 2000
). The compliance to the exercise protocol in the present study was high at all times, and the HB exercise arm had almost the same percentage of loss to follow-up compared to the non-exercise group.
As reported in many other studies, maintenance of the weight loss after 6 months is difficult to sustain, and a substantial proportion of the patients will eventually revert to their original body weight (Jeffery, et al., 2000
; Wing and Hill, 2001
). Although the results regarding the role of PA on prevention of weight regain are still considered controversial, its application has become a part of the strategies for weight maintenance following weight loss programs (Haskell, et al., 2007
; Saris, et al., 2003
; Wu, et al., 2009
), and officially endorsed by the American College of Sports Medicine (Donnelly, et al., 2009
). This institution states the need for a greater amount of physical activity, reaching 250 to 300 min per week of moderate-intensity PA, in order to help prevent weight regain. Our study suggest that a small volume of PA, about 120 min per week, may be a good start for increasing in PA and it also suggest that after 6 months this volume should be increased to maintain long-term weight losses.
We did not find any major differences between the groups for measures of central obesity, except at the third month, when the HB group showed a greater reduction in waist circumference. Mourier et al. (1997)
studied 24 overweight and obese subjects and showed that, although the exercise group showed major reduction of abdominal fat as evaluated by magnetic resonance, no differences were found for waist circumference and waist-to-hip ratio in comparison with non-exercisers. Thus, more accurate methods to determine small changes in abdominal fat would be necessary. In addition, no differences between the groups were found for fat-mass percentage during the follow-up and bio-impedance is not also a golden measure of fatness (Dehghan and Merchant, 2008
; Pateyjohns et al., 2006
Numerous studies have documented that PA has favorable effects on metabolic parameters such as lipid profile and insulin sensitivity (Kraus, et al., 2002
; Tuomilehto, et al., 2001
; Warburton, et al., 2006
). For the metabolic parameters analyzed in our study, the most striking difference between the groups was for HDL-cholesterol, with a two-fold elevation for the HB in comparison with controls after 1 year of follow-up. These results have been extensively described (Donnelly, et al., 2000
; Durstine, et al., 2001
; Leon and Sanchez, 2001
) among obese and hypercholesterolemic subjects, and our findings confirmed this association for middle-aged non-obese and non-hypercholesterolemic women, an important finding since high HDL cholesterol is one of the few protective factors for cardiovascular disease (Alwaili, 2010
; Singh et al., 2007
Although many studies have shown beneficial effects of PA on insulin sensitivity, (Holloszy, 2005
; Matthaei, et al., 2000
; Tuomilehto, et al., 2001
) others which evaluated the combined effects of PA and diet in comparison with diet-only showed no differences in the results (Larson-Meyer, et al., 2006
; Weinstock, et al., 1998
). Our results suggest no significant additional effects of PA on insulin sensitivity evaluated by HOMA-IR. However, most women evaluated in the present study were not insulin resistant at baseline, according to the Brazilian criteria for insulin resistance (Geloneze, et al., 2006
), which may explain the lack of a difference between the groups.
Limitations of the present studies include: 1) We have not used a golden standard measure of PA but self reported diary shows good correlation with other methods such as accelerometers and double labeled water (Besson, et al., 2010
; Meriwether, et al., 2006
). In addition, we did not assess PA in control group during the intervention, but a lack of major changes for metabolic parameters in this group, mainly for HDL cholesterol, may confirm no changes in PA status; 2) adiposity was measured by BIA, which is a measurement with many limitations (Dehghan and Merchant, 2008
); 3) subjects were only females and not obese, therefore we may not extrapolate the results to obese and males individuals. These measurements error appears to be non differential and would change findings towards the null hypothesis.
In conclusion, a home-based exercise program was feasible and promoted greater weight reduction in the first 6 months, after which no further weight reduction was observed. Favorable changes in HDL-cholesterol after 1 year suggest that home-based exercise is a good strategy to initiate and increase physical activity and promote health benefits. Whether modifying training modalities after 6 months would increase these clinical benefits could be an important topic for future research.