Combined effects of resistance training and carbohydrate-restrictive or conventional diets on weight loss , blood variables and endothelium function

Objective To compare the effects of either a carbohydrate-restrictive diets or a conventional hypoenergetic diet combined with resistance training. Methods Twenty-one overweight and obese adults participated in an eight-week program consisting of progressive resistance training combined with carbohydrate-restrictive diets (initially set at <30 g carbohydrate; n=12) or conventional hypoenergetic diet (30% energetic restriction; carbohydrate/protein/lipid: 51/18/31% of total energy consumption; n=9). It was hypothesized that the carbohydrate-restrictive diets would induce greater weight loss but that both diets would elicit similar effects on selected health markers. Body mass, and body composition, blood variables and flow-mediated brachial artery dilation (flow-mediated brachial artery dilation; by ultrasound) were used to assess changes due to the interventions. Results Significant within-group reductions in body mass (-5.4±3.5%; p=0.001 versus -3.7±3.0%; p=0.015) and body fat (body fat; -10.2±7.0%; p=0.005 versus -9.6±8.8%; p=0.017) were identified for carbohydrate-restrictive 544 | CM MEIRELLES & PSC GOMES Rev. Nutr., Campinas, 29(4):543-554, jul./ago., 2016 Revista de Nutrição http://dx.doi.org/10.1590/1678-98652016000400009 diets and conventional hypoenergetic diet, respectively, but there were no significant differences between groups as the result of the interventions. Fat free mass, blood variables and flow-mediated brachial artery dilation did not significantly change, except for the total cholesterol/high-density lipoprotein ratio, which was reduced 10.4±16.9% in carbohydrate-restrictive diets (p=0.037) and 0.5±11.3% in conventional hypoenergetic diet (p=0.398). Conclusion Carbohydrate-restrictive diets associated with resistance training was as effective as conventional hypoenergetic diet in decreasing body mass and body fat, as well as maintaining fat free mass, blood variables and flow-mediated brachial artery dilation, however it was more effective at lowering the total cholesterol/low density lipoprotein ratio.


I N T R O D U C T I O N
Traditionally, Conventional diets (hypoenergetic and hypolipidic) have been recommended to promote losses in body mass and fat 1 .However, interest in the effect of restricting dietary carbohydrates on body mass is increasing in the academic community.A Carbohydrate-Restrictive Diet (CRD) is defined as a carbohydrate intake of less than 150 g daily 2 .
According to a meta-analysis that included 13 randomized and controlled trials conducted for six or twelve months 3 , a carbohydrate-restrictive diets accompanied by a reduction in energy intake http://dx.doi.org/10.1590/1678-98652016000400009led to greater losses in body mass and lowered triacylglycerol levels as well as increases in High Density Lipoprotein-cholesterol (HDL-c) compared to the conventional diet.However, the overall effect on the blood lipid profile warrants further attention because significant elevations in total and Low Density Lipoprotein-cholesterol (LDL-c) were observed at the end of both six and twelve months of CRD.
Evidence for the coronary risks attributed to CRD as evaluated by endothelium function is scarce and controversial.Endothelium function can be accurately assessed by Flow-Mediated Brachial Artery Dilation (FMD), which is highly sensitive for identifying coronary artery disease 4 .Very few studies have examined the effects of CRD on FMD.Some of them have observed significant reductions after very short intervention periods 5,6 , while others observed no significant changes after short-or long-term adherence to a CRD 7,8 .
High-sensitivity C-reactive protein, an acute-phase plasma protein, is an inflammatory biomarker strongly associated with coronary artery disease 9 .Studies investigating the effects of CRD on this marker have reported controversial results showing either no changes 7 or decreases in this marker 8,10 .
There is a lack of knowledge about the influence of exercise on the overall metabolic response to hypoenergetic diets containing different contents of carbohydrate because most of the previous studies conducted with CRD were performed in sedentary individuals.In addition, resistance training has been recognized as useful in reducing health risks during periods of energy restriction 11 .However, the combined effects of CRD and resistance training on markers of health remain to be determined.
The aim of the present study was to compare the effects of CRD and conventional diet hypoenergetic diets in combination with progressive resistance training on body mass loss, body composition, blood lipids, high-sensitivity C-reactive protein and FMD.It was hypothesized that a CRD would induce greater weight loss, but that both diets would elicit similar effects on the selected health markers.

M E T H O D S
Apparently healthy subjects were selected in response to announcements on four fitness clubs in the city of Rio de Janeiro, Brazil.A total of 38 were eligible for testing, and 32 met study inclusion/exclusion criteria.Seventeen subjects were assigned to the CRD and 15 to the conventional hypoenergetic diet group.In the CRD, twelve (71%) completed the eight weeks program (32±10 years; 1.69±0.09m; 8 women), and in the conventional diet group, nine (60%) completed the study (45±10 years; 1.69±0.13;5 women).All subjects expressed the desire to lose body mass and had previously been enrolled in a resistance training program for at least three months, as well as in aerobic physical activities.Exclusion criteria included upper or lower limb injury, current pregnancy, diabetes, Body Mass Index (BMI) lower than 25 kg•m -2 , use of ergogenics, stimulants or drugs, and a history of dyslipidemia or arterial hypertension within the previous six months.For participants who were subjected to the CRD, additional exclusion criteria were as follows: urea >50 mg•dL -1 , creatinine >1.5 mg•dL -1 and uric acid >7 mg•dL -1 in men or >5.7 mg•dL -1 in women.
Subjects were permitted to choose the diet, considering dietary habits and food preferences.
All volunteers provided written informed consent.The study was approved by the Universidade Antonio Carlos Institutional Ethical Review Board (#097/06), and was registered at the Clinical Trials registration site under NCT01096836.
Prior to the beginning of the study, all subjects completed a self-report dietary and physical activity history questionnaire.They also underwent initial body composition assessment, blood sampling and flow-mediated brachial artery dilation measurements.Thereafter they underwent strength tests to determine total strength (the sum of the total load multiplied by total repetitions performed during tests consisting of 8-10 maximum repetitions for three exercises, i.e., leg press, triceps pushdown, and biceps pulldown).
The intervention consisted of eight weeks of diet and progressive resistance exercise training (2 sets of 10 repetitions of 11 exercises for lower and upper body muscle groups with 2 minutes intervals between sets performed 3 times per week).Subjects were permitted to choose the dietary intervention based on their dietary habits and food preferences.All volunteers met with a nutritionist weekly for the first four weeks and every two weeks after the first month.At these meetings, the subjects underwent nutritional counseling and body mass measurement with subjects wearing light clothing and without shoes.To ensure adherence to the diet, they completed a 24-hour food recall at each meeting.
At the end of the training program, all laboratory and ultrasound tests were repeated.Body composition assessments and strength tests were also repeated after the first four weeks.
Subjects in the carbohydrate-restrictive diets group received handouts listing the amounts of carbohydrate in frequently consumed foods, and they were instructed to reduce carbohydrate intake to less than 30 g per day during the initial four weeks and to add 10 g of carbohydrate each week thereafter until the end of the study.
The conventional diet was designed to provide approximately 70% of each individual's energy requirements 12 with 55% of the energy from carbohydrates, 15% from proteins, and 30% from fats, according to the recommendations of the Institute of Medicine (IOM) 13 .The diet consisted of a structured plan based on the nutritional habits of each subject.Dietary data were analyzed using the software Avanutri (Avanutri Informática, Rio de Janeiro, Brazil).Body mass, height, waist girth and skinfolds were always measured by the same experienced evaluator.Body density was estimated using the Jackson & Pollock 14 and Jackson et al. 15 equations for men and women, respectively.Siri's equation 16 was used to predict the percentage of body fat.
Fasting blood samples were taken and analyzed for glucose, triacylglycerol, total cholesterol, HDL-c urea, creatinine, uric acid, Alanine Aminotransferase (ALT), and high-sensitivity C-reactive protein.LDL-c was determined by the Friedewald's equation 17 .
All blood analyses were performed using a Roche/Hitachi 917 system (Hoffmann-La Roche AG., Switzerland) and standard kits.The inter-day coefficients of variation for each analysis were 1.8% for glucose and triacylglycerol, 1.7% for total cholesterol, 1.3% for high-density lipoprotein, 2.3% for urea and creatinine, 1.7% for uric acid, 4.4% for ALT, and 3.0% for high-sensitivity C-reactive protein.
Flow-mediated brachial artery dilation was performed using a two-dimensional color spectral Doppler ultrasound equipped with a 14-MHz linear transducer (Toshiba Nemio, Japan).Measures were obtained with subject in supine position.The transducer was placed on the anteromedial face of the right arm perpendicular to the centerline of the arm, 5-10 cm above the antecubital fossa and over the artery.Basal and post-occlusion diameters were measured between the intima-lumen interfaces at the end of diastole.The occlusion was maintained for 5 minutes using a cuff on the arm to apply pressure slightly above the systolic artery pressure, which was confirmed by the lack of a pulse on the Doppler screen.The post-occlusion was measured at the same site 60s after the blood flow was released.
The day-to-day intra-class correlation coefficient (R) and the absolute Typical Error of Measure (TEM) were R=0.948 and TEM=0.05mm for basal, R=0.948 and TEM=0.07 mm for post occlusion, and R=0.842 and TEM=2.7% for flow-mediated brachial artery dilation, as reported in a previous study 18 . http://dx.doi.org/10.1590/1678-98652016000400009 The Shapiro-Wilk test showed that the data were not normally distributed.A Wilcoxon and Mann-Whitney tests were used to identify within-group and between-group differences, respectively.When multiple comparisons within a subject were necessary (for body mass, body composition, dietary and strength variables), a Bonferroni adjustment was applied; i.e., the p-value was divided by the number of comparisons (0.05/2: p<0.025).For all other analyses, the

R E S U L T S
Subjects on the CRD significantly reduced their energy consumption by 48% (p=0.046),their carbohydrate consumption by 72% (p=0.028), and their fiber intake by 57% (p=0.028).Subjects on the conventional hypoenergetic diet significantly reduced their energy intake by 30% (p=0.028) and their total fat intake by 39% (p=0.028),but they did not significantly change their carbohydrate and fiber intakes (Table 1).
Individual analyses showed that half of the subjects in the carbohydrate-restrictive diets group (6/12) and one third of (3/9) subjects in the conventional diet group lost more than 5% of their initial body mass.Changes in anthropometric variables are shown in Table 2.
Subjects in both groups increased their total strength as a result of resistance training.The carbohydrate-restrictive diets subjects increased their strength by 14% (p=0.005), and Glucose, triacylglycerol, total cholesterol, HDL-c, and LDL-c did not change significantly within or between groups.A significant reduction (12.8%; p=0.037) was observed in the total cholesterol/HDL ratio in the CRD subjects (Table 3).Changes in high sensitivity C-reactive protein, urea, creatinine, uric acid and ALT were not significant in either group (Table 3).Individual analyses showed that one subject in the CRD group had increased high sensitivity C-reactive protein levels, whereas another from the same group who had begun with high ALT levels reached normal values at the end of study.
Flow-mediated dilation was not significantly changed after eight weeks of intervention in either group (Figure 1).hypoenergetic CRD nor a conventional diet led to significant different changes in body and fat masses, maintenance of fat-free mass, FMD, or other metabolic variables.
Both groups significantly reduced their energy intake, although the dietary methods differed.The mean carbohydrate intake among the CRD group was 83±43 g at week 8, and all subjects ingested amounts below the upper limit that characterizes a CRD (<150 g daily) 2 .In the CRD group, the ingestion of carbohydrate was below the recommendations of IOM 13 , while protein and fat intakes were above the upper limits.In conventional hypoenergetic diet, the three macronutrients were ingested within the range recommended by IOM 13 .In CRD group, voluntary energy restriction might be due either to the moderate protein ingestion and/or to the low glycemic index inherent of this type of diet.
Paddon-Jones et al. 19 reported that protocols involving moderate protein diets and resistance exercises may help maintain fat-free body mass due to their appetite-suppressant properties as well as the increased total energy expenditure caused by the thermic effect of protein.Ludwig 20 showed that low glycemic index diets result in higher satiety, decreased appetite and voluntary energy ingestion.
The lack of a significant difference in loss of body mass between the CRD group and the conventional diet group is in agreement with other studies of six to ten weeks duration 21,22 , although it does not agree with a previous systematic review designed to investigate the effects of CRD on body mass loss 3 .
The magnitude of body mass loss in the current study was lower than that reported in the published studies, which found decreases of about 7% during the intervention period, however, the current study has combined progressive resistance training with CRD and conventional diets.Johnstone et al. 23 reported that four weeks of non-ketogenic CRD (approximately 35% of total energy from carbohydrate) led to a significantly lower body mass loss (-4.4 kg versus -6.3 kg) than Analyses of individual cases showed that 2/7 subjects in the CRD group and 3/7 in the conventional diet group had low baseline FMD values.At the end of intervention, one subject in each group had improved FMD values.Two subjects in the conventional hypoenergetic diet group with normal initial FMD values showed reductions to less than desirable values.None of participants in the CRD group reduced their initial values to below the desirable FMD value.

D I S C U S S I O N
The main findings of the present study are that during an eight-week intervention program (resistance training and diet), neither a http://dx.doi.org/10.1590/1678-98652016000400009an isoenergetic ketogenic CRD (approximately 4% of total energy from carbohydrate), suggesting that carbohydrate restriction may maximize losses in body mass.Previous studies comparing the effects of CRD and conventional diets without resistance training reported significant decreases in fat-free mass in both interventions (from -1.4 to -3.2 kg, representing up to 32% of total body mass reduction) 2,6,10,24,25 .However, despite the short duration of the present intervention, CRD associated to resistance training seemed to prevent fat-free mass loss.
A relevant point to be discussed is the need for gluconeogenesis from amino acid carbon skeletons during a low-carbohydrate diet.This gluconeogenic role of dietary proteins has been hypothesized as being involved in their greater satiating power when compared with other macronutrients.In addition, protein exhibits a high thermic effect, as a result of the energy cost of urea synthesis and gluconeogenesis from amino acids.Both effects.-.satiating power and thermogenesis -are referred to help during body mass loss.It is expected that the decreased supply of carbohydrate would lead to low blood levels of insulin and reduce the lean body mass preservation.However, in the present study, the association between CRD and resistance training could possibly translate to a healthier body composition status, since even low quantities of amino acids are useful to promote muscle protein synthesis when stimulated by the resistance training 26 .
It should be pointed out, however, that subjects in CRD ingested hypoenergetic diets with protein content of 29-32%, what is only apparently high in protein compared to a conventional diet group with a protein level of 22-26%, although the total grams of protein relative to body mass did not differ between the two types of diet.
Waist girth significantly decreased in both groups after the intervention, but no significant differences were observed between CRD and conventional diets, similar to other reports 7,8 , and discrepant from Volek et al. 24 .
Other than the total cholesterol/HDL ratio in the CRD group, no significant changes were observed either within or between groups in the present study.These results were similar to other reporting data from four weeks of intervention 23 .However, it is typically observed that a CRD promotes higher reductions in triacylglycerol and increases in HDL-c levels compared to conventional diets 22,25 .Wood et al. 27 also reported that improvements in HDL-c were better in the CRD group versus the low-fat diet group after 12 weeks of hypoenergetic diets with progressive resistance training.
According to a cohort study carried out in more than 8,000 subjects 24 , the total cholesterol/ HDL ratio is a superior measure of risk for coronary heart disease compared to either total cholesterol or LDL-c cholesterol levels.Therefore, the greater reduction in total cholesterol/HDL-c ratio noted in the CRD group in the present study might indicate some additional benefit of carbohydrate restriction beyond calorie restriction.
Subjects in the present study did not present significant reductions in high sensitivity C-reactive protein levels either within or between groups, although significant reductions were observed in body mass.On the contrary, Seshadri et al. 10 observed that reductions in high sensitivity C-reactive protein were directly associated with body mass loss.Results similar to the present study were reported by Phillips et al. 5 and Keogh et al. 7 , who found that significant body mass reductions were independent of modifications in high sensitivity C-reactive protein levels in subjects submitted to six to twelve weeks of a CRD or conventional diet without exercise.
Pre-to post-intervention changes in FMD were also not statistically significant, as previously reported after six weeks to two years of eight weeks of CRD [6][7][8]28,29 . Convesely, Phillips et al. 5 observed significant reductions in FMD in CRD dieters (-1.4±0.6%) and significant increases in that of conventional dieters (+1.9±0.8%)along with an approximately 5 kg body mass reduction in both groups after six weeks of intervention. T available evidence on the effects of resistance training on endothelial function is scarce, but it appears that changes are manifested in subjects with some endothelial dysfunction, as observed in young prehypertensive subjects 30 .Among subjects with abnormal FMD, similar to one subject in the CRD and one in the conventional diet group, values increased to above the cut-off point.The lack of significant effects of the intervention on FMD might be explained by the small changes in body mass, blood lipids and high sensitivity C-reactive protein levels.It could be speculated that greater changes in those variables would be necessary to provoke significant alterations of FMD.
No significant changes were observed in renal or liver functions as assessed by urea, creatinine, uric acid and ALT levels.These results were probably due to the maintenance of total protein intake.Although it is well accepted that a high-protein diet may be detrimental to individuals with existing kidney dysfunction, there is little evidence that high protein intake is dangerous for healthy individuals 31 .Based on the current evidence, CRD interventions do not seem to impair organic functions in obese subjects, even over a two-year period 32 .But further research is needed to confirm their safety over a longer term and with broader study populations.
A potential limitation of the present study is that subjects could choose their own diet intervention, although this approach probably strengthened the ecological validity of the study.It is recognized that dieters tend to drop out of restrictive dieting programs if they find it too difficult to adhere to some of the food choices either because they are not familiar with them or because they dislike the choices.
In conclusion, the present study showed similar effects of hypoenergetic diets and resistance training on body mass loss and health markers in overweight and obese individuals, regardless of their carbohydrate intake.Therefore, a CRD may be an alternative intervention for treating individuals in this population provided there is no specific contraindication.Further research is needed in other groups and for longer periods to determine the effects of CRD associated to resistance training in supporting weight maintenance or weight regain.

R E F E R E N C E S
p<0.05 compared to baseline values (Wilcoxon test); † p<0.05 between CRD and CONV diet (Mann-Whitney diet).M: Mean; SD: Standard Deviation; CRD: Carbohydrate-Restrictive Diet; CONV: Conventional Diet.

Figure 1 .
Figure 1.Changes in flow-mediated brachial artery dilation of subjects on carbohydrate restrictive diet (n=7) and conventional diet (n=7) associated to a resistance training program during eight weeks.Data are mean and standard deviation.Rio de Janeiro (RJ), Brazil, 2008.

C
O N T R I B U T O R S CM MEIRELLES and PSC GOMES jointly developed the study's concept and design, collected and analyzed the data, and wrote the manuscript.

Table 1 .
Changes in the dietary composition of subjects on a Carbohydrate-Restrictive Diet or a Conventional Diet enrolled in an eight-week resistance training program.Rio de Janeiro (RJ), Brazil, 2008.

Table 2 .
Changes in anthropometric characteristics of subjects on a Carbohydrate-Restrictive Diet or a Conventional Diet enrolled in an eight-week resistance training program.Rio de Janeiro (RJ), Brazil, 2008 * Note: * p<0.025 compared to baseline (Wilcoxon test); † p<0.05 between CRD and CONV (Mann-Whitney test); Four skinfolds were triceps, abdominal, supra-ilium and anterior thigh.M: Mean; SD: Standard Deviation; CRD: Carbohydrate-Restrictive Diet; CONV: Conventional Diet.

Table 3 .
Changes in blood variables of subjects on a carbohydrate-restrictive diet or a conventional diet enrolled in an eight-week resistance training program.Rio de Janeiro (RJ),Brazil, 2008.