Subjects: 23 Obese (44 +/- 4% BF), Healthy, Premenopausal Women (Range 21-47, late 30's on average)
Compared: Two diets -- one low fat LF , one low carb LC, and Two exercise states -- one aerobic EX, one no exercise NX -- Four groups were compared LF/NX, LF/EX, LC/NX, LC/EX.
Study Length: 12 Weeks of weight loss, preceded by 5 weeks maintenance (MI), and followed by 6 weeks maintenance (MII).
Diets - Carb/Protein/Fat: Maintenance - 45/20/35 ; LC - 25/25/50 ; LF - 60/25/15 It is worth noting that the LC vs. LF comparison kept protein constant. The caloric content of the reducing diets was individualized to each participant to be ~75% of measured RMR.
Exercise: 3X/week, 45 minutes = 15 min on each of three different cardio machines, 60-65% VO2max
Measured: Changes in body composition, Resting Metabolic Rate (RMR), thermic effect of a meal (TEM aka TEF), and total daily energy expenditure (TDEE).
- LC lost a bit more weight than LF: 10.6 kg ± 2.0 vs 8.1 ± 3.0 kg (P 0.037)
- Diet composition did not significantly influence body composition or energy expenditure changes
- RMR was reduced similarly for both diets: 0.54 MJ/d (~130 calories)
- TEM/TEF did not differ between the two diets
- Exercise resulted in greater loss of fat mass 8.8 ± 2.1 kg vs. 6.1 ± 2.3 kg (P = 0.008)
- Exercise maintained TDEE better +0.07 ± 1.23 vs. -1.46 ± 1.04 (P 0.004) due directly to exercise, and +0.75 ± 1.06 for the exercisers vs -0.61 ± 1.03 MJ/d (P = 0.006) for the non-exercisers not solely attributed to the exercise sessions. I'm not sure these are additive, but +0.82 MJ vs -2.07 MJ between the two. This is a swing of almost 3 MJ or ~ 700 calories!!
Some of the tabulated hormone & blood glucose findings: (You can click to enlarge)
Interestingly, the LF/NX group had the same fasting insulin as the LC/EX group during the reducing phase., the LC/NX group had the highest fasting insulin. I don't think these were statistically significant (especially because it looks like some samples were lost), but this counters the general belief that low carb diets necessarily impact fasting insulin levels. One can see that fasting insulin decreased in all groups probably attributed to weight loss in general.
From the Discussion:
More important than total weight losses, however, are the relative changes in FFM and FM. Exercise training was a major determinant of the changes in body composition, with FM comprising 89.4% of the weight loss in the Ex group, compared with only 71.3% in the Nx group.
Exercise led to more FAT loss irrespective of diet.
The increased fat loss in the Ex group indicates a greater energy deficit, but this cannot be attributed to the energy costs of the exercise sessions alone because the diet prescriptions for the Ex subjects were increased to compensate for the energy expended during the exercise sessions. The rationale for this compensation was to avoid a greater energy deficit in the Ex subjects relative to the Nx group, because changes in FFM (6) and RMR (30, 35) have been shown to depend on the degree of energy restriction.So they held the calorie deficit constant for the caloric expenditure associated with the exercise session itself. But interestingly:
The increased fat loss in the Ex group cannot be attributed to maintenance of RMR, because in this study aerobic exercise failed to have a protective effect on RMR. ... In the present study, RMR, which comprised 58 ± 6% of the subjects’ daily energy expenditure, demonstrated the commonly observed decrease during the reducing diet in all treatment groups.
RMR was also measured during MII after the subjects consumed an isoenergetic diet for 4 wk. RMR values increased relative to energy restriction levels, but remained significantly depressed during MII.That second quote is depressing and probably is what causes relapses and regain. I contend that a goodly portion of the adult obesity epidemic (especially in women) is probably due to dieting in the first place. Repeated bouts of caloric restriction depress RMR (and thus TDEE), and there's a "hangover" effect that is not attributable just to reduced caloric needs to maintain a lower mass.
A unique feature of this study is that we measured free-living TDEE. Aerobic exercise played an important role in maintaining TDEE during weight reduction. The lack of a significant change in TDEE in the Ex group (+ 1.2 ± 12.2%) is in sharp contrast with the decrease of 12.3 ± 8.8% observed in the Nx group during low-energy feeding. The major effect of Ex on TDEE during weight reduction was its effect on physical activity. Because the change in TDEE is equal to the sum of the changes in RMR, TEM/TEF, and physical activity, and because the changes in RMR and TEM/TEF were comparable in the Ex and Nx groups, we were able to deduce that the significant TDEE difference between groups was attributable to a difference in physical activity (Figure 3). Although compensatory reductions in spontaneous physical activity are classic responses to undernutrition and must also be considered as part of the energy balance picture during energy restriction (43), the addition of aerobic exercise during the reducing diet proved to be an effective method for preventing this decrease. In contrast with the Nx group, the Ex group increased their physical activity by 0.75 ± 1.06 MJ/d, offsetting the decreases in RMR and TEM/TEF, yielding virtually no change in TDEE. Note: 0.1 MJ (mega joule) is roughly equal to 25 food calories.What this says to me is that ELMM IS the most effective method to lose weight after all. I gotta say, my personal journey included rapid weight loss without formal exercise, but I'm a more active person by nature. As weight was lost, I did become more physically active (taking stairs, parking further away from destination, etc.) so I think I probably staved off some of the TDEE reductions. But this can be highly individual as the large variation in the mean (>140%) meant that for some of the exercisers, TDEE was not maintained and/or was still reduced. Still, your chances seem better when including moderate aerobic exercise into the mix.