An Examination of the Relationships Among Cardiorespiratory Fitness Status, Body Mass Index, Appetite and Grelin

Problem Statement: Obesity and its related metabolic disorders rank among the leading causes of illness and mortality worldwide. The prevalence of obesity has led researchers to focus on body weight regulation and the balance between energy intake and energy expenditure. Background: Studies have suggested that ghrelin is involved in both short and long term energy balance. Previous research has examined the relationships among BMI, ghrelin and appetite but few studies have investigated the role that fitness status plays in relation to these. Methods: Data from two studies that were previously conducted in the Energy Metabolism Laboratory were combined for a secondary data analysis. One hundred nine subjects (87 females , 22 males; 19.4 ± 2.1 years; BMI 22.9 ± 3.9 kg/m2) were included in the analysis. Cardiorespiratory fitness (V02max) was estimated by the Queen' s College Step test and self-reported physical activity was determined by the International Physical ~ctivity Questionnaire. Ratings of appetite (by 1 OOmm visual analog scales), plasma total ghrelin (by RIA) and insulin (by ELISA) were analyzed following a 1 Oh fast and 30min after a standardized meal. Results : Ghrelin was significantly correlated with BMI in both the fasting (r=-0.290, p=0.005) and fed (r=-0.381, p=0.001) states. Significant negative correlations were also found between V02max and both fasting ghrelin (r=-0.305, p=0.003) and fed ghrelin (r=-0.336, p=0.005). Fasting ghrelin was also significantly correlated with vigorous physical activity (r=-0.242, p=0.019), waist circumference (r=-0.298, p=0.004) and insulin (r=-0.324, p=0.012). Ghrelin and appetite were not significantly related at single time point measures (t=O, 30 min). BMI was positively correlated with dietary restraint (r=0.258, p=0.007) and disinhibition (r=0.276, p=0.004), whereas V02max was negatively correlated with dietary restraint (r=-0.350, p<0.001). ANOV A indicated that subjects with BMis above the median had significantly lower fasting ghrelin (692.2±237.9 pg/mL vs. 802.2±266.0 pg/mL, p=0.037) and fed ghrelin (575.0±164.4 pg/mL vs. 710.2±226.4 pg/mL, p=0.005) and significantly greater disinhibition (6.5±3.1 vs. 4.8±3.1, p=0.005) compared to those with BMis below the median. Subjects with a higher fitness status reported significantly lower fasting satiety (31.4±22.3 vs. 36.8±22.2, p=0.022), greater desire to eat (53.3±17.5 vs. 45 .5±20.5, p=0.037), and less dietary restraint (7.8±5.2 vs. 11.3±5.4, p=0.001) compared to subjects with a lower fitness status. Significance and Conclusions: These findings confirm previous relationships reported between BMI and ghrelin, subjective appetite and eating behaviors and corroborate lack of associations between single-point time measures of ghrelin and appetite. Additionally, results indicate that cardiorespiratory fitness may be related to circulating ghrelin levels, subjective appetite, and degree of dietary restraint. Further study is needed to determine the nature of these relationships . ACKNOWLEDGEMENTS I would like to thank Dr. Kathleen Melanson, my major advisor, for her long term support. Dr. Melanson's resume enticed me to pursue the URI program and my interactions with her far surpassed any expectations I had regarding the breadth of her knowledge. Dr. Melanson provided me with immeasurable assistance and always made me feel I was a priority despite her numerous other daily demands. Her expertise, patience and attention to detail allowed me to never lose sight of my objectives. It was a privilege to have her as an advisor. Dr. Daniel Kresge played a pivotal role in my thesis development from its early stages through to the fine-tuning process. I am grateful to him for his technical support. Dan is an in-house expert who always provided sound and reliable instruction. I was fortunate to have benefited from his knowledge. I would also like to express my gratitude to my committee members Dr. Greene and Dr. Riebe. Along with Dr. Melanson, they truly formed a "team of experts" and provided me with constructive input throughout the thesis process. Last but certainly not least, Dr. Cathy English, who never failed to see the "bigger picture" regarding my schedule, my thesis and my graduation. Cathy was relentlessly accommodating and flexible with my TA responsibilities. She helped to ensure that my priorities were always on course and made me feel that my individual goals were


ABSTRACT
Studies have suggested that ghrelin is involved in both short and long term energy balance. Previous research has examined the relationships among BMI, ghrelin and appetite but few studies have investigated the role that fitness status plays in relation to these variables. Data from two studies that were previously conducted in our laboratory were combined for a secondary data analysis. One hundred nine subjects (87 females, 22 males; 19.4 ± 2.1 years; BMI 22.9 ± 3.9 kg/m 2 ) were included in the analysis. Cardiorespiratory fitness was estimated by the 3 minute Queen's College Step test and the 51-item Three Factor Eating Questionnaire was used to assess cognitive restraint, disinhibition and perceived hunger. Ratings of appetite (by 100 mm visual analogue scale), plasma total ghrelin (by RIA) and insulin (by ELISA) were analyzed following a 10 h fast and 30 min after a standardized meal. Ghrelin was significantly correlated with BMI in both the fasting (r=-0.290, p=0.005) and fed (r=-0.381, p=0.001) states. Significant negative correlations were also found between estimated cardiorespiratory fitness (V0 2 max) and both fasting ghrelin (r=-0.305, p=0.003) and fed ghrelin (r=-0.336, p=0.005

INTRODUCTION
Obesity and its related metabolic disorders rank among the leading causes of illness and mortality worldwide [1] . The prevalence of obesity has led researchers to focus on body weight regulation and the balance between energy intake and energy expenditure. Ghrelin is a recently discovered hormone that appears to be associated with both short and long term body weight regulation [2]. Ghrelin plays an important role as a short term regulator of appetite through its involvement in meal initiation.
Plasma ghrelin levels rise before meals and decrease after meals [3]. Studies on rodents indicate that ghrelin infusion not only promotes weight gain by increasing food intake but also by decreasing energy expenditure and fat catabolism [4][5][6] .
Intravenous infusion of ghrelin in humans increased food intake in both lean and obese subjects [7]. In addition, higher ghrelin levels have been linked to lower resting metabolic rate and thermic effect of food [8,9]. Collectively, these findings support a role for ghrelin in both energy intake and in energy expenditure.
Ghrelin levels fluctuate with body weight. Research has demonstrated that circulating concentrations are low in obese individuals and increase with weight loss [10,11]. Due to the role that exercise plays in weight loss, recent studies have explored the effects that exercise has on ghrelin [11][12][13][14][15]. Exercise is an established method to increase energy expenditure. Because exercise training improves fitness status, promotes weight loss and improves overall health, there has been recent research devoted to examining the effects of exercise on ghrelin levels. A chronic exercise-induced energy deficit may be associated with higher ghrelin levels. This was demonstrated in long term exercise interventions [11,16]. These increases, however, 3 often occur along with weight loss so it is difficult to determine if the increases in ghrelin are due to the weight loss or to physical activity itself. More research is necessary to determine the mechanism responsible for this increase.
Understanding body weight regulating hormones is also essential in determining the role that fitness plays in regulating appetite. Hunger suppression is observed immediately after exercise [17][18][19] but compensatory effects on food intake occur in the long run to compensate for an average of 30% of energy expended in physical activity [19] . Other research has also explored the relationship between dietary restraint and exercise. It was hypothesized that high dietary restraint would be associated with a tendency for disinhibition and increased post-exercise energy intake [20]. This was not observed in overweight males [20] or in female dieters [21]. This may be attributed to the fact that increases in energy expenditure from physical activity may not be immediately compensated for by increases in energy intake [22]. If fitness status modifies these relationships has yet to be determined.
Few studies have directly examined fitness status and its relation to ghrelin.
The question of whether fit individuals have higher levels of ghrelin has yet to be fully investigated. The primary purpose of this study is to examine relationships among fitness, BMI, and ghrelin and to determine their independent effects on appetite. A better understanding of these relationships will provide insight into the mechanisms that govern body weight regulation and the role that exercise training plays in this process. Determining if ghrelin levels are a function of fitness status may also clarify the relationships between fitness and appetite. This study will also assist researchers in planning human subject studies based on subject characteristics.

Participants
Subjects were normal weight, healthy male and female volunteers recruited from the university community using flyers, class announcements and newspaper advertisements. All participants provided voluntary, written informed consent. The University of Rhode Island Institutional Review Board approved the experimental protocol and procedures.

Study design
Data sets from two studies that were previously conducted in our laboratory were combined for a secondary data analysis including one hundred nine subjects (n=87 females, n=22 males). The Ghrelin and Appetite Response to Feeding (GARF) study consisted of seventy-two females and twenty-two males. Exclusion criteria included: an inability to donate blood due to anxiety or a bleeding disorder, an allergy to any component of the test meal, pregnancy or lactation, and age younger than 18 years. Participants reported to the laboratory after a 12-hour overnight fast. Visual analogue scales, which were used to assess hunger, satiety and desire to eat were completed before and 30 minutes after the consumption of a breakfast shake (Boost,

Waist circumference
Waist circumference was measured in centimeters at the level of the umbilicus after normal expiration using a physician's tape measure.

Appetite
Volunteers were asked to rate their subjective appetite (hunger, satiety, desire to eat) ranging from "not at all" to "extremely" on a 10 cm validated visual analogue scale (VAS) [25]. Appetite assessments were administered in both fasting and fed states.
The 51-item Three Factor Eating Questionnaire (TFEQ) was also administered to measure cognitive dietary restraint, tendency for disinhibition, and perceived general hunger levels [26].

Body composition
Body composition was assessed in the Grain Study after a four hour fast using air displacement plethysmography in a self-contained system using a computer- performed to examine between group differences in fasting and fed ghrelin, in TFEQ restraint, hunger, and disinhibition, and in fasting satiety, desire to eat and hunger.
Waist circumference, blood glucose, and insulin were entered as covariates in ANCOVA to examine their influence on the relationship between BMI and fasting and fed ghrelin. Partial correlations controlling for cardiorespiratory fitness, BMI, waist circumference, blood glucose and insulin were performed to determine their impact on the relationships between ghrelin and appetite in the fasting and fed states. Stepwise multiple regression analysis was then used to determine how well waist circumference, BMI, and insulin predicted fasting and fed ghrelin.

Subject Characteristics
One hundred nine subjects were included in the analysis (87 females, 22 males  Table 1.
Fasting ghrelin was negatively correlated with BMI (r=-.290, p=0.005), waist circumference (r=-0.298, p=0.004) and insulin (r=-0.324, p=0.012). The relationship between fasting ghrelin and BMI was no longer significant (p=0.455) after controlling for waist circumference. Controlling for blood glucose and insulin did not modify the relationship between BMI and fasting ghrelin. In the fed state, controlling for waist circumference, blood glucose, and insulin did not have an effect on the relationship between BMI and ghrelin. In addition, ANCOVA revealed that differences in fasting ghrelin between BMI groups were no longer significant after controlling for waist circumference, insulin, and blood glucose. Significant differences remained, however, in the fed state even after controlling for these variables. Percent body fat (n=l 7) was not significantly related to fasting ghrelin (p=0.378) or fed ghrelin (p=0.279). Also,% body fat did not modify the relationship between ghrelin and appetite in the fasting or fed state.

Predictors of fasting ghrelin
Three models were developed using the variables that were significantly correlated with ghrelin to predict ghrelin levels. The most significant model (p<0.001) 13 was using BMI and V02max to predict fasting ghrelin. Models are presented in Table   4.

DISCUSSION
The results indicate that there is a significant inverse relationship between both BMI and ghrelin and between V02max and ghrelin in the fed and fasting state. The appetite findings suggest that there are also links among eating behaviors, appetite and both BMI and V02max. Subjects with BMis above the median had significantly higher disinhibition scores compared to those with BMis below the median. More fit subjects reported significantly lower fasting satiety, greater desire to eat, and less dietary restraint compared to less fit subjects.
The finding that BMI is negatively associated with ghrelin is in agreement with previous studies [29,30]. Similarly, negative relationships between ghrelin and waist circumference [29] and insulin [30] have previously been reported. The relationship between fasting ghrelin and BMI was no longer significant after statistical control for waist circumference. Waist circumference was slightly more correlated with ghrelin than BMI was. In addition, waist circumference and BMI were highly correlated (r=0.836, p<0.001). This is similarly reported in the literature (r>0.8 in most studies) [31]. This is useful for future studies and suggests that BMI and waist circumference may be similar indexes of body composition [31].
The significant negative association that was found between V0 2 max and ghrelin was inconsistent with our hypotheses. It was counterintuitive that the associations between ghrelin and BMI and ghrelin and V0 2 max were both negative. It was hypothesized that the more fit individuals would have higher ghrelin levels.
Previous research has indicated that ghrelin increases in response to a negative energy balance [ 11,32] . The literature on the effects of short term exercise on ghrelin is inconsistent. Most studies failed to find an effect of exercise on ghrelin levels [12,[33][34][35][36] but two acute exercise studies have reported increases in ghrelin immediately following exercise [ 13,14]. The conflicting results in differences in protocols make it difficult to reach a consensus.
Research on chronic exercise, however, suggests that ghrelin levels rise from long term exercise interventions [11,16]. These findings are complicated by the fact that weight loss also occurred along with the increases in ghrelin, making it difficult to determine if the increases were due to weight loss or to physical activity itself. Weight loss is known to result in increased ghrelin levels [3,32]. These increases are commensurate with the amount of weight lost [11]. There is some evidence that there may not be an independent effect of exercise on ghrelin apart from weight loss. A study that included a weight stable group found that three months of exercise training had no effect on ghrelin levels in the absence of weight loss [10].
The reason for the negative correlation between ghrelin and V0 2 max is unknown. Other studies that have analyzed V0 2 max and ghrelin are inconclusive. An aerobic exercise intervention involving 173 sedentary, overweight postmenopausal women that compared changes in ghrelin levels to changes in fitness levels (changes in V02max) from baseline to 12 months found a non-significant negative association between ghrelin and fitness level [11] . Another study conducted on 65 normal weight females found a non-significant positive relationship between fasting ghrelin and V02max [8]. The authors also found significant negative associations between ghrelin and RMR and TEF independent of fat mass and fat free mass. A more recent study found that ghrelin was negatively associated with RMR in non-obese women independent of body weight [9]. The authors suggested that ghrelin may be regulated by voluntary long-term behavior modification that increases RMR such as aerobic and resistance training [3 7] . This hypothesis would be consistent with the negative association found between fitness status and ghrelin in our study.
Significant differences in appetite and eating behavior were observed between BMI and fitness categories. Subjects with a higher BMI had greater disinhibition compared to subjects with a lower BMI. This finding was consistent with the literature [38]. Disinhibition is positively correlated with both BMI and obesity. It is also predictive of less successful weight loss attempts, weight regain, and low physical activity [38]. BMI has also been previously associated with dietary restraint [39,40].
A significant positive correlation was found between BMI and dietary restraint in our study but there were no significant differences in dietary restraint between BMI groupings.
Subjects with a higher fitness status had significantly lower fasting satiety, greater desire to eat and less dietary restraint. These findings are interesting from a behavioral standpoint. There is a widespread belief that the energy expended from exercise can offset the consumption of more energy dense foods [ 41] . More fit individuals may be more likely to adopt less stringent diets due to the perception that the additional calories consumed will be expended during exercise. This perception could potentially lead to weight gain in individuals engaging in minimal exercise who feel that they do not have to be as conscious of their energy intake. Alternatively, more fit individuals might not need to restrain their energy intake to regulate body weight. Although the total kilocalories and macronutrient composition of the test meals were similar, the perception of satiety may have been altered due to the fact that one test meal was a shake and one was a solid meal.
The strength of the relationships that were found between BMI and ghrelin and between V02max and ghrelin were evidenced by multiple regression analysis. The model including BMI and V0 2 max as the only predictors of fasting ghrelin proved to be the most significant model. This supports the correlation results.
This study contributes to the scientific literature because it is the largest single analysis to our knowledge that examined relationships of both BMI and V0 2 max to both fasting and fed ghrelin. Although many of the variables that were included in this analysis were previously studied in relation to ghrelin, the relationships among BMI, VOzmax, ghrelin and appetite have not been directly compared. In addition, the inclusion of both genders and the use of both fasting and postprandial ghrelin data were unique features of this study. The nature of the secondary data analysis also allowed for a larger sample size (n=109) compared to other studies in this field.

Ghrelin and Energy Balance
Ghrelin is a recently discovered hormone that regulates energy balance [1 , 2].
Since its discovery in 1999, it has piqued much interest in the field of energy metabolism. Studies have indicated that ghrelin plays a pivotal role in both short term and long term energy balance [3] . Ghrelin is a hyperphagic hormone secreted by the stomach that participates in meal initiation [3]. Research also suggests that ghrelin is involved in long term energy balance. Ghrelin is associated with energy storage.
Studies have found that ghrelin levels are elevated in individuals with anorexia nervosa [4] and levels are decreased in obese individuals [5]. This indicates that ghrelin is downregulated in response to excess energy stores. Ghrelin levels return to normal when a healthy body weight is achieved [6]. Hansen's research that measured fasting ghrelin in response to weight loss in obese subjects not only found that fasting ghrelin concentrations increased with weight loss but also found that this increase was positively correlated with the extent of weight loss [5]. In a six month diet program for weight loss, Cummings et al. found a 24 percent increase in the area under the curve for the 24-hour ghrelin profile with a diet induced weight loss of 17 percent [7]. This is consistent with the hypothesis that ghrelin has a role in the long term regulation of body weight [7] . A negative energy balance or weight loss threshold for increasing ghrelin, however, has yet to be determined [8].

Ghrelin and Appetite
Ghrelin stimulates appetite by acting on the arcuate nucleus in the hypothalamus. This region is responsible for controlling food intake [9].
Ghrelin plays an important role as a short term regulator of appetite through its involvement in meal initiation. Studies on rodents indicate that ghrelin infusion promotes weight gain by increasing food intake and decreasing energy expenditure and fat catabolism [10][11][12]. Intravenous infusion of ghrelin in humans increased food intake in both lean and obese subjects [13]. Endogenous ghrelin levels rise before meals to sufficient levels to stimulate appetite [3] . Postprandial ghrelin levels are quickly suppressed by nutrient ingestion [14]. This is evidenced by twenty-four hour plasma profiles that show these preprandial increases in ghrelin and postprandial decreases in ghrelin with every meal [14,15].
A study that analyzed the short term effects of energy restriction on total fasting and postprandial ghrelin reported no significant changes in ghrelin after the four day energy restriction [16]. There were no associations found between total ghrelin and appetite scores. This suggests that ghrelin does not have an immediate effect on appetite in a typical hypocaloric dietary regimen.

Ghrelin Biochemistry
Ghrelin is synthesized predominantly by the stomach [17]. The fundus synthesizes approximately ten times more ghrelin per gram of tissue than other production sites including the duodenum and other parts of the small intestine [18].
Gastrectomy and small bowel resections have indicated that approximately 2/3 of circulating ghrelin is secreted by the stomach with the rest being secreted by the small intestine [19]. Ghrelin is a peptide containing 28 amino acids. It undergoes posttranslational modification that involves the acylation of the serine-3 residue. The serine residue is covalently attached to a medium-chain fatty acid by an ester bond.
This acylation is necessary for ghrelin to bind and activate its receptor, the growth hormone secretagogue receptor (GHS-R) [9] . As a result, most of ghrelin's physiological roles are associated with its acylated form, commonly referred to as active ghrelin. Studies in mice lacking the GHS-R gene have suggested that this receptor is responsible for the orexigenic effects of ghrelin [20]. Ghrelin also exists, however, in another isoform, des-acyl ghrelin, which is suggested to play important roles in glucose and insulin metabolism [21-23] and in adipogenesis [24]. Research is being conducted to identify alternate ghrelin receptors that interact with the desacylated form [2] .

Ghrelin and Exercise
Research has been devoted to examining the effects of exercise on ghrelin due to ghrelin's role in chronic energy balance. Since exercise and ghrelin both affect energy balance, studies have been performed to determine if a relationship exists between the two. Ghrelin levels are increased in states of negative energy balance [5].
This increase may be part of an adaptive response to counteract a negative energy balance [25]. As a result, it has been hypothesized that the energy expenditure resulting from exercise may have an effect on ghrelin. Both short term studies [26][27][28] and long term studies [6,25] have been performed to examine the effects of exercise The literature on the effects of chronic exercise on ghrelin levels, however, is more clear. Previous studies have shown that weight loss resulting from reduced calorie intake increases ghrelin levels [ 5,15]. This finding led researchers to examine whether an exercise-induced calorie deficit would lead to similar results. Multiple studies have examined changes in ghrelin with chronic exercise. A three month diet and exercise intervention produced significant increases in baseline ghrelin, lunch peak, dinner peak, nocturnal rise, and nocturnal peak (31]. A one-year aerobic exercise intervention found that exercise-induced weight loss, in the absence of reduced food intake, was associated with significant increases in ghrelin (25]. These increases in ghrelin were commensurate with the amount of weight lost (25]. In the exercise group, however, ghrelin levels did not increase in those who did not lose weight or lost a mild amount of weight (0.5-3 kg) suggesting that there was no independent effect of exercise.
Another three month exercise intervention reported a significant increase in ghrelin in the weight loss group compared to the controls and the weight stable group [6]. By including a weight stable exercise group, this study was able to demonstrate that exercise training in the absence of weight loss had no effect on ghrelin. These studies add to the body of literature that suggests that ghrelin is a long-term regulator of energy balance. However, since many of these studies include weight loss as a function of the exercise protocols, it is difficult to determine if changes in ghrelin are related to weight loss or to physical activity itself.

Ghrelin, Fitness, and BM!
The results of studies on exercise and ghrelin raise important questions about the relationship between ghrelin, fitness status, and BMI. Ghrelin is associated with key components of energy balance [6]. For instance, ghrelin is negatively correlated with percentage of body fat, fat mass [14]

Ghrelin and Gender
A study examining plasma ghrelin and its relation to body composition reported that there were gender differences in the nature of these associations. Plasma ghrelin levels were inversely associated with fat mass measures in both males and females but there were gender differences in which of these body composition variables were related to ghrelin. For example, in males, plasma ghrelin was inversely correlated with abdominal fat mass and fat distribution index (calculated as the trunkfleg fat mass ratio from dual energy x-ray absorptiometry body composition analysis). In females, inverse correlations were found between plasma ghrelin and BMI, total fat mass, and fat mass/lean mass ratio [35].
Since there are fluctuations in appetite with the menstrual cycle some studies have investigated if ghrelin levels also vary throughout the menstrual cycle. One study aimed specifically at abnormal menstrual cycles found that fasting ghrelin levels were at least 85% higher in women with exercise associated amenorrhea [36]. This supports the role for ghrelin as a metabolic indicator of chronic energy deficient states.
In a normal menstrual cycle, however, a recent study reported that there were no significant changes in ghrelin throughout the cycle [3 7]. This is useful in planning future studies involving female subjects.

Fitness, Appetite, and Energy Intake
The relationship between fitness, appetite and energy intake also warrants further investigation. Appetite plays a central role in the regulation of energy balance [38]. The effects of exercise on appetite and food intake are controversial and depend on the intensity and duration of exercise. Several studies have investigated the effects of acute exercise on appetite and energy intake. There appears to be a weak coupling between energy expenditure and energy intake in the short term [39] .
The effects of exercise on appetite and energy intake are influenced by the intensity of exercise performed. High intensity exercise has been shown to suppress hunger in the short term [39] . A study involving cycling at 35% and 68% ofV0 2 max indicated that hunger ratings were significantly lower in subjects in the highest intensity exercise condition compared to those in the resting condition [ 40]. This hunger suppression did not translate to a reduction in food intake, however. The absence of an effect on food intake has been reported elsewhere (39,(41)(42)(43) . This is not surprising given that appetite is mediated by hormonal, psychological, and environmental factors [ 44]. Low to moderate intensity exercise has not been associated with this suppression (45]. This suggests that the intensity of the exercise has an effect on the extent of appetite and food intake suppression.
Exercise duration also has an effect on appetite and food intake. Erdmann et al. (2007) found that short duration exercise of 30-60 minutes had no effect on hunger/satiety sensations and subsequent food intake (28]. In this same study, a longer duration of exercise of 120 minutes produced an increase in food intake without having an effect on subjective hunger sensations. Most studies dealing with exercise and hunger suppression have focused on aerobic exercise [8]. Little is known about the effects of resistance exercise on appetite. One recent study conducted by Broom et al. (2009)  Although resistance exercise was not found to suppress hunger to the same extent as aerobic exercise did, the suppression was still significant compared to the control. This attenuated response may be attributed to the lower energy expenditure along with the intermittent nature ofresistance exercise compared to aerobic exercise [30].
Some studies have examined the effects of body weight, exercise and food intake. Non-obese women had lower post-exercise energy intakes after strenuous exercise, whereas obese women showed no change in intake across exercise conditions [ 41]. A different study reported no significant differences in post-exercise energy intake between obese and lean individuals [47]. However, obese subjects were found to have a decrease in hunger and appetite after exercise but without an accompanying decrease in energy intake [ 4 7] . Another study found that energy intake was higher in obese women compared to normal weight women after moderate intensity exercise and on the control day [44]. These results demonstrate the complexity of the relationship between exercise, body weight, appetite, and food intake.
While there is evidence of exercise-induced anorexia in the short term [45], a positive relationship exists between physical activity and energy intake in the long term [48]. It has been suggested that physiological and behavioral adaptations occur in the interim [48]. Short term (1-2 days) and medium term (7-16 days) exercise-induced negative energy balances of up to 4 MJ/d occur before compensation in energy intake is observed [ 45]. Beyond this, a compensatory effect takes place and food intake is Increased to compensate for an average of approximately 30% of energy expended in physical activity [ 45]. This compensation is not universally observed. Subjects can be classified as "compensators," those who display increases in energy intake with exercise, or "non-compensators," those who do not show increases in energy intake with exercise [45]. Further research is needed to determine the mechanisms through which this compensation occurs but there is some evidence that body composition plays a role in the extent of this compensation [ 49]. It has been suggested that fat mass acts as an energy buffer and increases in energy intake are not observed until lean body mass is threatened by the energy deficit created from physical activity [ 45] .
A recent review analyzed the effects of chronic exercise on energy intake, energy balance and body weight [50]. While short to medium term exercise is effective at inducing a negative energy balance, long term exercise unless accompanied by calorie restriction, results in only minor weight loss.
The literature highlights many factors that could be responsible for the lack of a substantial effect of exercise on weight loss. These include both physiological and behavioral adaptations to counteract the new energy balance. Weight loss is associated with a compensatory reduction in total BMR [ 48] and an increase in ghrelin levels [5][6][7]. Also, as V0 2 max increases with exercise training, a lower energy expenditure is achieved for the same volume of exercise. Energy expenditure from exercise is also reduced with a lower body weight [ 48].
Behavioral mechanisms such as compensatory increases in energy intake from a less stringent dietary regimen may also counteract the energy deficit created by exercise. This is linked to the popular belief that the energy cost of exercise can offset the consumption of more energy dense foods [51]. Despite these compensations, exercise is essential in preventing weight gain or regain in the long term [50]. The role of exercise in weight maintenance is thought to be a result of an improvement in appetite control and a better coupling between energy intake and energy expenditure [SO]. One possible explanation for this improvement in appetite control may be linked to an increase in satiety hormones including polypeptide YY, glucagon-like peptide-I and pancreatic polypeptide with exercise [52]. However, the extent to which changes in ghrelin may play a role in these relationships remains to be elucidated. The increase in these hormones may be responsible for the previously described phenomenon of exercise-induced anorexia.

Eating Behaviors and Appetite Control
Eating behavior is influenced by a multitude of factors ranging from physiological to cognitive and socio-cultural factors [53]. In order to fully understand eating behavior, it is important to examine both the physiological and nonphysiological factors that affect energy intake. Physiological components of total food intake include hunger, satiation, satiety, and sensory-specific satiety. These factors are regulated by a combination of physical factors including gastric filling, distention, contraction and emptying and metabolic factors including gut peptides, neuropeptides, hormones and the detection of nutrients absorbed into the circulation [3 8].
At the physiological level, appetite and food intake are regulated by the brain and by hormones from the gastrointestinal tract, pancreas, adrenal glands and adipose tissue. The hypothalamus is involved in the central regulation of appetite and continuously receives neural, metabolic, and endocrine signals from the periphery.
Specifically, the arcuate nucleus within the hypothalamus contains many of the receptors for hormones and peptides known to be involved in the regulation of energy intake [50].
Traditional physiologic appetite theories focus on the individual's internal state and explain meal initiation by a "hunger drive" that increases proportionately to individual needs and is decreased by food intake [53]. Eating behavior, however, is much more complex than simple depletion-repletion models can explain and nonphysiological factors must be considered. Appetite is influenced by a combination of cognitive, hedonic, psychological, habitual, cultural and social factors [38].
Physiological mechanisms that control food intake can easily be overridden by social and environmental factors .

Dietary Restraint and Disinhibition
The development of the Three Factor Eating Questionnaire (TFEQ) by Stunkard and Messick in 1985  [58] and fat oxidation [59] after a meal. The lower plasma insulin observed in restrained eaters is thought to be mediated by the reduced secretion of noradrenaline.
Noradrenaline triggers the release of glucose into the bloodstream. Reduced secretion of noradrenaline would therefore result in lower levels of blood glucose in the postprandial state and would require less insulin [55].
A longitudinal study examined cognitive restraint and weight gain in a large sample of normal and overweight adults and adolescents [ 60]. The findings indicated that restraint was positively correlated with BMI in normal weight but not in overweight subjects. Restrained eating was not predictive of weight gain. However, a high initial BMI was associated with a greater increase in cognitive restraint over the two year period. These findings were in contrast to a previous cross-sectional study performed on restrained eating and weight gain that reported that restrained eating is higher with greater BMI [61].
Although most of the literature has focused on restraint, disinhibition also plays an important role in appetite and weight regulation. Disinhibition refers to a tendency towards over-eating. Disinhibition includes over-eating because others are eating, not being able to resist food, eating in response to negative mood states, and over-eating highly palatable foods [62]. Disinhibition is positively correlated with BMI and obesity [ 62]. It is also predictive of less successful weight loss attempts and of weight regain. In addition, disinhibition is associated with lower self-esteem, poor psychological health, and low physical activity [62].
Relatively little research has directly examined the relationship between disinhibition and physical activity but a few studies have been conducted. One study suggested that women who exercised and had high disinhibition scores increased their energy intakes leading to a positive energy balance [59]. On the other hand, acute exercise was shown to have a beneficial effect on women who had high disinhibition scores by decreasing their motivation to eat and increasing their preference for low-fat foods [54].

Effects of Dietary Restraint and Exercise on Appetite Control
Exercise has been shown to be more effective than dieting in inducing a negative energy balance in the short term [ 51]. A study that compared the effects of a diet induced negative energy balance to an exercise induced negative energy balance found that hunger ratings and food cravings were significantly higher in the diet condition [ 51]. Another study that examined the impact of dietary restraint and a single bout of moderate-intensity exercise on post-exercise energy intake in sedentary males found that there was not a significant effect of exercise on energy intake either immediately after exercise or over a twelve hour time period [63]. The authors expected to find the overweight males with high restraint to become disinhibited by exercise and have significantly higher energy intakes but this was not observed. This could be attributed to the fact that increased energy expenditure from physical activity is not immediately compensated for by increases in energy expenditure [48]. A study on exercise, dietary restraint and energy intake in females who had recently lost weight found that exercise did not result in a significant increase in energy intake. This was not surprising since the majority of the subjects were classified as highly restrained. The subjects were also categorized into overeaters or undereaters depending on whether or not their self-selected energy intake levels exceeded their maintenance requirements. Compared to the undereaters, the overeaters had higher disinhibition scores [64].

Queen's College Step Test
Maximum oxygen uptake (V0 2 max) is a widely accepted measure of cardiorespiratory fitness. The experimental protocol used to determine V0 2 max, however, is difficult, exhausting, and requires a well-equipped laboratory [65] .

TRUE FALSE
18. While on a diet, i f I eat food that is not allowed, I consciously eat less for a period of time to make up for it.

19.
Being with someone who is eating often makes me hungry enough to eat also. 20. When I feel blue, I of ten overeat .

.
I enjoy eating too much to spoil it b y c ounting calories or watching my weight .
22 . When I see a real delicacy , I often get so hung ry that I hav e to eat it right away.
23 . I often stop eating when I am not really fUll as a conscious means of limiting the amount that I eat. 24. I get so hungry that my stomach of ten feels like a bottomless pit.

25.
My weight has hardl y changed at all in the last ten years. 26 . I am al ways hungry so it is hard for me to stop eating before I finish the food on my plate .

.
When I feel lonely , I console myself by eating .
28 . I consciously hold back at meals in order not to gain weight.
29 . I sometimes get v ery hungry late in the ev ening or at night.
30. I eat any thing I want, any time I want.

31.
Without even thinking about it, I take a long time to eat.
32 . I count calories as a conscious means of controlling my weight.
33. I do not eat some foods because they make me fat .
34. I am always hungry enough to eat at any time .
35 . I pay a great deal of attention to changes in my figure . 36 . While on a diet, i f I eat a food that is not allowed, I of ten then splurge and eat other high c alorie foods.
Please answer the questions in this section by circling the number above the response that is appropriate for you.