The purpose of this study was to study how combinations of a regular exercise and diet program can impact anthropometric and health-related outcomes. Additionally and as we previously attempted  we sought to further examine how replacing dietary carbohydrate with protein to varying degrees may impact anthropometric and health outcomes in sedentary, obese women who participated in a regular diet and exercise program. This study represents the second in a series of studies our research group has conducted in an attempt to better understand how various dietary regimens interact with an exercise program and what outcomes can be expected. Currently, the Curves for Women program is followed at over 10,000 fitness franchises across the globe with millions of women worldwide following the program . The study design was based largely upon our previously published study with slight modification in phase I dieting, which was decreased to one week in the present investigation in addition to changes to the weight maintenance period (Phase III). Similar to our previous study, this employed study design has certain strengths such as the overall number of subjects, incorporation of adequate control groups, dietary oversight, and exercise supervision. A potential area of weakness is the unbalanced distribution of participants across all groups which took place as part of study design. The authors acknowledge that increasing (and/or equating) sample size in every group would have addressed this concern and would have been helpful, the primary objective of our study was to determine the impact of different macronutrient distributions and how this alteration impacted the end points measured in the present study. In this respect, however, statistical power analysis of our primary and secondary outcomes revealed power values that ranged from 0.767 - 0.944 and partial eta squared values which ranged from 0.055 - 0.092; providing evidence of appropriate statistical power on our primary and secondary outcomes. We initially hypothesized that following a diet and exercise program would improve all primary and secondary outcomes (waist circumferences and other anthropometric and body composition variables) in comparison to the control groups (no diet + no exercise [CON] and no diet + exercise [ND]). We also hypothesized that those groups ingesting a higher proportion of dietary protein (VLCHP and LCMP) would experience improved responses in our primary and secondary outcomes. The primary findings from this study confirmed our initial hypothesis that following any form of caloric restriction along with the exercise program led to greater improvements in waist circumference (primary) and other secondary outcomes (body mass, DXA fat mass, DXA % fat, resting energy expenditure) in addition to improvements in fitness parameters. This finding also provides continued support that exercise participation alone cannot adequately promote improvements in these end points, a conclusion we  and others have previously made .
A critical component to our study design and investigated questions was the compliance to the recommended dietary regimens prescribed to all groups. As expected, significant changes did occur for energy, carbohydrate and protein intake between diet phases (e.g., Phase I, Phase II, etc.) and dietary groups (e.g., VLCHP and LCMP vs. HCLP). From an overall standpoint, compliance to the dietary regimens was somewhat successful, but some deviations did result in an inability to make all a priori comparison. In this respect, the HED group was unable to achieve the prescribed energy intake of 2,600 kcals•-1; a response which we reported in our initial investigation . Their mean intake of 1,800 kcals•d-1 was approximately 800 kcals•d-1 lower than prescribed, but was still 320 kcals•d-1 greater than the average intake of the other restricted diet groups (e.g., VLCHP, LCMP, HCLP). The higher value still allowed for comparisons between caloric intakes while exercising, which was the main reason for this dietary prescription, even though the associated adaptations we hoped to compare likely did not occur to the extent we planned. The 30% error rate commonly associated with dietary reporting may have contributed to this, and in addition previous reports have suggested that greater errors occur when higher caloric intakes are prescribed. It is also likely that the relatively low fat intake at such a high caloric intake resulted in an overall higher volume of food that was prohibitive versus ingesting a lower volume of higher energy density foods common in the North American diet. Caloric intake, however, was reduced overall by an average of 504 kcals•d-1 in the diet groups (e.g., VLCHP, LCMP, HCLP) from their baseline intakes, which was deemed as a positive response to the prescribed dietary regimens. Lastly and an important point we reported before relates to the relative daily intake of protein for the VLCHP and LCMP groups and inherent safety or risks of this regimen. While both groups ingested a much greater proportion of calories as protein (~50 - 60% daily energy from protein), their relative daily intake of protein (~1.1 gram protein•kg•d-1) was not much greater than RDA guidelines. This result is consistent with our previous work , but is a lower overall protein intake than what other investigators have employed in similar study designs .
Waist circumference was chosen as a primary outcome due to its ability to serve as a predictor of diabetes, cardiovascular disease and other comorbidities . As previously reported, results from this study reveal that regular participation in a resistance-based exercise program, when combined with energy restriction, significantly reduces waist circumference . Independent of macronutrient distribution, reductions in waist circumference were significant over time and greater than CON (Figure 2). Similarly, body mass was also found to significantly decrease in those groups that followed the exercise program and restricted energy intake (Figure 3). No between-group differences were found for those groups that replaced more carbohydrate with protein (VLCHP and LCMP vs. HCLP), although slightly greater decreases in waist circumference and body mass did occur (Table 2); an outcome identical to our previous report . Overall these findings provide support for previous studies in diabetic and non-diabetic populations which also investigated the impact of macronutrient content on changes in weight loss [9, 38, 39]. While none of the studies incorporated an exercise program, our findings refute conclusions made by Layman et al. who suggested that greater total weight loss occurs, independent of exercise participation when increased dietary protein intake occurs . While greater changes did occur in those groups that ingested more protein in the present, the changes were not statistically significant and thus can't be considered real effects. Furthermore, those participants who restricted energy intake and exercise all lost significant amounts of fat mass (Figure 4) and percent fat (Table 2), but again no between-group differences were evident. It is particular interesting to note that only the HCLP group experienced a significant reduction in DXA fat-free mass after completion of the 14-week program. This decrease resulted in a change in DXA fat-free mass that was significantly different than changes in CON after 14 weeks, but the changes were not different than any other groups in the study. At some level, this finding supports our second hypothesis that a greater intake of dietary protein would spare losses of fat-free mass while restricting energy intake. It has been reported that higher intakes of dietary protein during dieting can prevent losses of lean mass, however, these findings are somewhat mixed [9, 22]. Careful interpretation of these data is encouraged as the discussed changes are only within the HCLP and are not a between-group finding. Nonetheless, these findings do suggest that significant amounts of fat can be lost in comparison to changes in fat-free mass while dieting and exercising [5, 21], and to a greater extent than control conditions.
An untoward metabolic response to energy restriction is a decrease in energy expenditure [3, 38, 40]. Our first investigation had participants restrict energy intake to 1,200 kcals•d-1 for two weeks, which resulted in a significant reduction in energy expenditure . To minimize this counteractive response, dieters in the present study restricted calories to 1,200 kcals•d-1 for only week in the present study. While all groups who followed a diet reported a reduction in energy expenditure, only the changes in VLCHP were significant after one week (~84 kcals•d-1). Regular participation in the exercise program for 9 weeks returned these values to baseline and in two groups (VLCHP and HCLP) a significant increase occurred in comparison to week one values (Figure 5). Mixed outcomes have been reported for the influence of an exercise program to stimulate energy expenditure values [22, 41]. Irrespective of these differences, findings from this and our previous investigation demonstrate its successful ability to sustain energy expenditure levels near basal rates, independent of what diet program was being followed (Figure 5).
As we reported previously and as expected, within-group improvements in relative aerobic capacity and relative maximal strength occurred in those groups who followed a diet and exercised [3, 21]. While absolute aerobic capacity (L/min) and strength (1RM) did increase (data not shown) in the no diet + exercise and HED groups, the lack of body mass change precluded statistical significance when represented relative to the body mass changes that occurred (or didn't occur) in these groups. In addition to the changes in fitness status, serum values for total cholesterol, LDL cholesterol and triglycerides decreased non-significantly from the beginning to the end of the protocol (Table 4). Overall the magnitude of changes are less than what has been previously reported [5, 6], but inherent differences in the study designs (presence of exercise program, length of study, protein type and quality, dietary compliance, etc.) in the investigative approaches may explain some of these differences. Overall it remains that regular participation in a diet and exercise and program can help maintain serum and/or promote improvements in serum based measures of cholesterol metabolism and cardiovascular health.
Great interest has developed into understanding the impact of leptin regulation in conjunction with weight loss, energy expenditure and insulin sensitivity [17, 42, 43]. It has been well documented that circulating leptin decreases in response to a decreases in energy availability , however, the influence of exercise and alterations in the macronutrient ratio is still under-researched. As expected, circulating leptin in the present study decreased in all groups who restricted caloric intake, but no differences relative to macronutrient ratio were determined. Volek et al.  reported significant decreases in leptin after an 8-week weight loss program; a finding similar to other studies  and the present study. Using a combined diet and exercise approach, Sartorio et al.  reported acute, significant reduction in leptin which closely mimicked body mass changes; again a finding supported by the present study and our previous reports . Briefly, previous studies have suggested that diets which contain a higher proportion of dietary protein may promote homeostasis of glucose and insulin . In support of these findings, fasting insulin levels in the VLCHP group were the only group to experience significant reductions from baseline (Table 4) after following their prescribed diet and exercise program after 14 weeks. This response also led to a subsequent significant decrease in HOMA-IR, a homeostatic model of insulin resistance. No significant changes in serum levels of ketones and cortisol were found (Table 4).
The overall safety of higher protein diets has been questioned in the literature. Findings from the current study support findings from our previous investigation using a similar investigative approach that no significant changes occurred in any diet or exercise groups for various markers of kidney and liver function as well as various markers of protein breakdown, which provide additional support for higher protein diets in conjunction with a regular exercise program (data not shown). Nonetheless, it remains that this study and multiple previously published investigations provide evidence that a higher relative protein intake do not invoke negative alterations in any of these serum variables in sedentary, obese, but otherwise healthy populations [1, 17, 39, 44].