Journal of the American College of Nutrition, Vol. 24, No. 4, 244-249 (2005)
Published by the American College of Nutrition
Effects of Binge Eating Behavior on Fuel Oxidation and Body Composition
Donatella Gniuli, MD, PhD,
Esmeralda Capristo, MD, PhD,
Elisabetta Liverani, MD, PhD,
Aldo V. Greco, MD,
Menotti Calvani, MD and
Geltrude Mingrone, MD, PhD
Department of Metabolic Disease, Catholic University, School of Medicine, Rome, ITALY
Address reprint requests to: Donatella Gniuli MD, Istituto di Medicina Interna e Geriatria, Sezione di Malattie del Ricambio Università Cattolica del Sacro Cuore, Largo Gemelli 8, 00168 Rome, Italy. E-mail:dgniuli{at}hotmail.com
 |
ABSTRACT
|
|---|
Objective: To investigate energy expenditure and glucose metabolism after a standard oral glucose load (75 g) in 8 normal weight bulimic women and 8 normal weight control women and to evaluate the relative endocrine implication.
Design: Serum glucose and insulin were measured both in basal conditions and after the glucose load; a basal endocrine assessment and body composition was evaluated and glucose induced thermogenesis (GIT) was calculated during 300 min following the glucose load.
Results: Serum glucose levels were significantly lower in bulimics both in fasting and in post-prandial state. Insulin levels were similar in bulimic and control women before and after the glucose load. FSH, leptin and free urinary cortisol (FUC) were all within the normal ranges, but significantly lower in bulimic patients compared with controls (p < 0.001). Fat mass (FM) and Fat-free mass (FFM) were reduced in bulimic patients, even if they normalized after correction per body weight. Resting energy expenditure (REE) was similar in the two groups even after FFM normalization, while GIT was lower in bulimic patients and it was strongly related to free urinary chortisol. Glucose oxidation was higher in fasting state and post glucose load, while lipid oxidation was strongly reduced.
Conclusion: An energy preservation mechanism seems to be the key element for normal-weight bulimic patients’ metabolism, consisting in leptin levels and GIT reduction, and lipid oxidation inhibition.
Key words: bulimia nervosa, glucose-induced thermogenesis, leptin, fat-free mass, fat mass, energy expenditure
|
INTRODUCTION
|
|---|
Bulimia nervosa is a syndrome characterized by recurrent episodes of binge eating associated with purging mechanisms such as self-induced vomiting, laxative or diuretic abuse, or extenuate physical activity [1]. These compensatory mechanisms are linked to the altered body image and fear of weight gain of these patients. While the emotional and hormonal pattern typical of this pathological status has been extensively evaluated, very little evidence has been offered about the consequences of binge eating habits on fuel metabolism. Energy expenditure (EE) in bulimia nervosa has been studied principally in the fasting state, showing a clear discordance between predicted Resting Energy Expenditure (REE), computed by Harris Benedict equation, and measured REE by indirect calorimetry evaluation [2,3]. Bulimic pattern is characterized by various type of purging strategies, so that there are no unanimous findings in REE values, having been reported both low [4,5] normal [6–8] and higher [9] even after normalization by lean body mass. These discrepancies may be due to differences in methodology and in clinical status of the subjects, severity of binge eating and purging, impact of diet, smoking and physical activity. Another important limitation is that REE is affected by activity during the hours preceding its determination, and may not accurately reflect 24-hour energy utilization.
Total energy expenditure (TEE) does not differ between patients and controls [7,10], suggesting that the energy conserving metabolic adaptations characteristic of semi-starvation do not occur in patients with bulimia nervosa.
To our knowledge, no other studies concerning metabolic changes, in terms of energy expenditure and substrate oxidation induced by meal have been evaluated in bulimic patients. Aim of this study was to investigate biological and metabolic features and glucose induced thermogenesis (GIT) following a standard oral glucose load in a cohort of normal weight bulimic patients without apparent pathological modifications.
|
SUBJECTS AND METHODS
|
|---|
The sample consisted of 8 normal weight women with bulimia nervosa and eight control women that were similar in age, weight and body mass. Their anthropometric and metabolic characteristics are shown in Tables 1 and2. Biological parameters were all within the normal range, but their hormonal and biochemical metabolites concentration significantly differed from those of control subjects. Diagnosis of bulimia nervosa was based on a 1.5 h structured psychiatric interview [11,12] on the established criteria of DSM IV [1]; our patients did not have history of anorexia nervosa, and self-induced vomiting was the only purging mechanism accepted for being included in the study. The patients were clinically controlled for 10 days before they underwent the study at the Division of Metabolic Diseases of the Catholic University. They were evaluated by an alimentary diary for their free-caloric intake and for the vomiting frequency. The healthy volunteers were recruited from the medical staff of the Department, and they were subjected to the same procedures as the bulimic patients. In all subjects, body weight oscillations did not exceed 1 kg during the 10 days of enrollment. Both bulimic patients and controls were asked to exercise no more than 30 minutes per day during this period. Each patient gave his informed consent according to the Institutional Review Board guidelines, and all clinical investigation described in the paper was conducted in accordance with the guidelines in The Declaration of Helsinki.
Body Composition DXA
Body composition was evaluated by Dual-energy X-ray absorptiometry (DXA) (LUNAR DPX, Adult Whole Body, Single-beam, Analysis version 3.65). DXA measurement was performed with a whole-body scanner operated in the single-beam configuration. Three sets of body composition values were obtained: grams of fat tissue (FM), gram of lean mass (FFM), and the percent fat body mass (PFM) compared to the total soft tissue mass. The areal bone mineral density (BMD), defined as the ratio to bone area (number of pixels in the total body or regional scan classified as containing bone) was also calculated. As a patient is analyzed, the program calculates R-value during each iteration. R consists of the ratio of 38 keV attenuation to 70 keV attenuation in soft tissue and bone. The program distinguishes between soft tissue points R-values and bone points R-value. Percentage of body fat is directly obtained by the R-Value. The body-composition distribution was obtained by manually separating the whole-body scan image into specific regions of interest (ROI).
Oral Glucose Load for GIT Determination
The study was performed after an overnight fast. After voiding, patients were placed on beds and a venous catheter was inserted into an antecubital vein for blood sampling; the line was kept patent with saline. Respiratory gas exchange measurement was started 40 minutes prior to the glucose load, in order to measure the REE, and continued for 300 min after glucose intake in order to calculate the glucose induced thermogenesis (GIT). At the end of the REE-measurement period, the glucose load was taken up in less than 3 min. Blood samples were collected immediately before the glucose challenge (time 0), and at 10, 20, 30, 40, 50, 60, 80, 100, 120, 140, 160, 180, 210, 240, 270 and 300 min post-load for serum glucose and insulin determination. At the end of the experiment, urine was collected for urinary nitrogen and glucose loss assay.
Respiratory gas exchange was measured by an open-circuit ventilated hood system (Deltatrac, Datex Instrumentarium Corp, Helsinki, Finland). Energy expenditure (EE), npRQ, and substrate oxidation were calculated from oxygen consumption, carbon dioxide production, and nitrogen urinary excretion according to standard formulas [13].
Biochemical Analysis
Serum glucose was measured by the glucose oxidase method (Beckman Glucose analyzer 2, Galway, Ireland), and serum insulin by the Micro-Particle Enzyme Immunoassay (MEIA, Abbot, IL60064, U.S.A.). Hormonal assays (FSH, LH, TSH, fT3, fT4, leptin, free urinary chortisol) were performed at the central lab of our University. Urinary nitrogen production was calculated from daily urinary urea elimination.
GIT Calculation
GIT was calculated as the total increase of EE above REE during 300 min following the oral glucose challenge, using the trapezoidal rule for the calculation of the area under the curve (AUC). GIT was also expressed as percent EE for glucose uptake according to the following equation:
where 15.67 kJ is the energy equivalent for 1 g of glucose [14].
Statistics
All values are expressed as mean ± SD. Independent sample t-test for the unequality of variance was used in order to assess the significance of differences between means of the examined variables. A linear stepwise regression analysis was performed in order to evaluate the influence of both FM and FFM on GIT.
|
RESULTS
|
|---|
Daily Energy Intake
Self-induced vomiting was the only purging mechanism repeatedly utilized by all patients, and was reported to occur almost 4 times a day. Daily purged energy intake, resulting from the dietary diary recall, was 9764.99 ± 3798.32 kJ/day, while the non-purged caloric intake was supposed to be 4515.73 ± 532.69 kJ/day. The daily caloric intake in normal weighing controls was 5623.18 ± 531.36 kJ/day.
Bulimic patients’ diary recall analysis revealed that binges were especially based on carbohydrates (58.22 ± 3.82%) and saturated fatty acids (30.48 ± 1.24%), with a small amount of protein (11.31 ± 5.05). Their energy intake out of the binges was lower than controls one, with proteins and fibers preference (72.31 ± 2.54%), and a smaller amount of carbohydrates (16.3 ± 1.68%) and lipids (11.49 ± 4.1%). Binges occurred preferentially in the late evening and during the night. Controls diary recall was compatible with a Mediterranean diet with a higher amount of carbohydrates (51.23 ± 0.9%), a medium content of proteins and fibers (31.43 ± 1.03%), and a medium-low intake of lipids (17.34 ± 1.93%).
Body Composition
Body mass index (BMI) was lower in bulimic patients (p < 0.001), and both FFM and FM were reduced in bulimics (FFM p < 0.05; FM p < 0.001). After correction of FFM and FM for total body weight, FFM and FM percentage did not differ in both groups of subjects(Table 1).
Biochemical Findings
Glucose and insulin concentrations time course after the oral glucose load are reported inFig. 1(A and B). Both fasting serum glucose and glucose time-course after the glucose load were significantly lower in bulimic patients than in controls (p < 0.01).
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 1. Plasma glucose and insulin concentration before (time 0) and during the 300 minutes after the 75 g oral glucose load in bulimic patients and control subjects. Solid line represents control subjects time-course, while dashed line represents bulimic patients time-course.
|
|
No differences were found in fasting serum insulin levels between bulimic patients and controls(Table 1). Insulin time-course showed a slightly reduced response to the glucose load in all the bulimic patients evaluated, even if a high inter-individual variability was observed in these series (p = n.s.), with higher levels of insulinemia maintained after 150 minutes since the load ingestion (Fig. 1). All the evaluated endocrine parameters were within the normal range, but FSH, Leptin and CLU were significantly lower in bulimic patients compared with controls (p < 0.001).
REE and GIT
REE was significantly lower in our series of bulimic patients (p < 0.05), but it become similar to that of controls after FFM normalization. REE and EE changes after the oral glucose load are shown inFig. 2. GIT, expressed as a percentage of energy content of glucose uptake, was lower in bulimic patients (p < 0.01). Using a linear stepwise regression, in a model including FM, FFM, FSH, Leptin, LH, TSH and FUC, the GIT was positively correlated only with FUC (R = 0.61; p < 0.05).
View larger version (14K):
[in this window]
[in a new window]
|
Fig. 2. Energy expenditure before (time 0) and 300 minutes after the glucose load (75 g). Time 0 is the average of resting and fasting EE 40 minutes before the glucose intake. Solid line represents control subjects time-course, while dashed line represents bulimic patients time-course.
|
|
Glucose and lipid oxidation were remarkably unstable within the bulimic series, with a high intra and inter-individual variability (Fig. 3). Basal and post-prandial glucose oxidation were significantly higher in bulimic than in control subjects (namely p < 0.001 and p < 0.01). In bulimic patients, lipid oxidation became negative at 50 min remaining around negative values until 220 minutes after the load; in control subjects lipid oxidation was positive during the whole study period. Using a linear stepwise regression, in a model including FM, FFM, FSH, Leptin, LH, TSH and FUC, the lipid oxidation was positively correlated with leptin (R = 0.45; p < 0.05).
View larger version (25K):
[in this window]
[in a new window]
|
Fig. 3. Total glucose and lipid oxidation rate basely (time 0) and 300 minutes after the oral glucose load (75 g). Time 0 is the average of indirect calorimetric measurements 40 minutes before the glucose intake. Solid line represents control subjects time-course, while dashed line represents bulimic patients time-course.
|
|
|
DISCUSSION
|
|---|
In spite of its relatively high prevalence [15] in young subjects, only a few studies on energy metabolism have been published on bulimia nervosa [2–10]. In the present study glucose and lipid oxidation, GIT, glucose and insulin response to an oral glucose load were investigated in 8 bulimic patients and in 8 normal weight control subjects.
High basal glucose oxidation is a common feature in very lean subjects with malabsorption [16] or anorexia nervosa [17,18], who are underweight depending on both FM and FFM depletion. In anorectic patients, this phenomenon seems to be linked to the extremely low body weight which determines a relative excess of glucose per body weight compared to control subjects. So far, normal weight subjects receiving the same amount per body weight of glucose show the same glucose oxidation trend as the anorectic patients [17]. Our series of bulimic patients, on the contrary, maintained a satisfactory body composition preservation, with a FFM and FM percentage similar to that of controls, so that it is reasonable to consider the increase of glucose oxidation after the oral glucose load as a real phenomenon. Fasting npRQ changes according to diet composition so that a high carbohydrates diet makes the npRQ increasing progressively over 0.90 [19]. During binges, bulimic patients currently utilize carbohydrates, especially sucrose, as primary choice of food items [20]; as secondary choice, they consume saturated fatty acids, monounsaturated fatty acids, and proteins [21] in decreasing preference. Alimentary daily recall analysis revealed that our patients preferentially eat carbohydrates and saturated fatty acids during binges, while they maintain a very low energy intake out of the binges, with a favorite consumption of proteins and fibers. As a consequence, high fasting glucose oxidation could be induced by the high carbohydrates content of the frequent binges (almost 4 per day).
Binge eating disorder dramatically affects patient’s quality of life both psychologically and physically. Body composition is seldom affected in these patients, for purging behaviors, even when efficacious, have a higher impact on gastro-intestinal tract than on calories balance. Nevertheless a continues feast-famine diet cycle determines a maximum metabolic efficiency, with increased hepatic gluconeogenesis and lipogenesis [22], so that alimentary habits of bulimic subjects might enhance the ability of the body to retain energy during feast period implementing energy stores for famine period, according to the thrifty genotype hypothesis [23]. The negative lipid oxidation observed in our series contributes to explain the preservation of fat mass in our patients. Evidence has accumulated that in both acutely ill and recovered patients with either anorexia or bulimia nervosa circulating leptin levels are lower than in controls matched for BMI. It is unknown if these lower leptin levels represent a state or trait marker [24]. A positive, significant correlation was observed between leptin levels and lipid oxidation in our bulimic patients. Leptin is primarily regulated by insulin-induced changes of adipocyte metabolism, and dietary fat and fructose, which do not increase insulin secretion, lead to reduced leptin production, suggesting a mechanism for high-fat/high-sugar diets to increase energy intake and weight gain. As a matter of fact, leptin production is strongly influenced by nutritional status. Leptin main biological role appears to be adaptation to reduced energy availability, regulating adipose tissue energy expenditure together with other proteins and hormones. In this complicated physiological pattern, it is possible to hypothesize that leptin may contribute to energy expenditure reduction acting through lipid oxidation inhibition [25].
Anorectic patients show low lipid oxidation, increased glucose oxidation and low GIT, as well [17]. The low GIT can be explained in the same view of energy preservation mechanism, as demonstrated by its correlation with FFM. Binges occurred preferentially during the night period, strongly interfering with the circadian bio-rhythm and the adrenal function. Free Urinary chortisol levels were within the normal range, but they were significantly lower than those of controls. A direct relationship was observed, in our bulimic patients, between GIT and FUC. We are observing, therefore, a normal phenotype of a pathological situation where the balance between energy intake and body composition-hormonal pattern is becoming inadequate. As a consequence, we are identifying a process which tend to preserve body composition by avoiding energy waist through thermogenesis.
Glucose time-course after the glucose load is similar to that of anorectic patients, while insulin secretion after the oral load is normal. The glucose time-course is certainly related to the delayed gastric emptying commonly described in bulimia nervosa [26–29], and it can be one of the biological stimuli for the auto-maintenance of binges in these patients. The ancillary result of a blunted glycemia even after a glucose load is a slightly reduced, but time-maintained insulin response which can be co-responsible for the FM preservation in these patients.
In conclusion, our observations on glucose and energy metabolism in active, normal-weight, bulimic patients can be relevant for the comprehension of the patho-physiology of this disease and can provide useful information for the medical, dietetic and psychological therapy of binge eating disorders.
Received October 9, 2004.
Accepted March 13, 2005.
|
REFERENCES
|
|---|
- American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorder, (DSM IV), 4th ed. Washington, DC: American Psychiatric Association,1987
.
- Sedlet KL, Ireton Jones CS: Energy expenditure and the abnormal pattern of a bulimic: a case report. J Am Diet Assoc89
:74
–77,1989
.[Medline]
- Scebendach J, Golden NH, Jacobson MS, Arden M, Pettei M, Hardoff D, Bauman N, Reichert P, Copperman N, Hertz S: Indirect calorimetry in the nutritional management of eating disorders. Int J Eat Disord17
:59
–66,1995
.[Medline]
- Bennett SM, Williamson DA, Powers SK: Bulimia nervosa and resting metabolic rate.Int J Eat Disord8
:417
–424,1998
.
- Obarzanek E, Lesem MD, Goldstein DS, Jimerson DC: Reduced resting metabolic rate in patients with anorexia nervosa. Arch Gen Psychiat48
:456
–462,1991
.[Abstract/Free Full Text]
- Devlin MJ, Walsh BT, Kral JG: Metabolic abnormalities in bulimia nervosa.Arch Gen Psychiat47
:144
–148,1990
.[Abstract/Free Full Text]
- Pirke KM, Trimborn P, Fichter M: Average total energy expenditure in anorexia nervosa, bulimia nervosa, and healthy young women.Biol Psychiat30
:711
–718,1991
.[Medline]
- Detzer MJ, Leitenberg H, Poehlman ET, Rosen JC, Silberg NT: Resting metabolic rate in women with bulimia nervosa: a cross sectional study.Am J Clin Nutr60
:327
–332,1994
.[Abstract/Free Full Text]
- Leonard T, Foulon C, Lajeunesse BS: High resting energy expenditure in normal weight bulimics and its normalization with control of eating behaviour.Appetite27
:223
–233,1996
.[Medline]
- Kotler LA, Devlin MJ, Mattehws DE, Walsh BT: Total Energy Expenditure as measured by doubly-labeled water in outpatients with bulimia nervosa.Int J Eat Disord29
:470
–476,2001
.[Medline]
- Cooper Z, Fairburn C: The eating disorders examination: a semistructured interview for the assessment of specific psychopathology of eating disorders.Int J Eat Disord6
:1
–8,1987
.
- Cooper Z, Cooper PJ, Fairburn CG: The validity of eating disorders examination and its subscales. Br J Psychiat154
:807
–812,1986
.
- Ferrannini E: The theoretical bases of indirect calorimetry: a review.Metabolism37
:287
–301,1988
.[Medline]
- Tataranni PA, Mingrone G, Greco AV, Caradonna P, Capristo E, Raguso CA, De Gaetano A, Tacchino RM, Castagneto M: Glucose-induced thermogenesis in postobese women who have undergone biliopancreatic diversion.Am J Clin Nutr60
:320
–326,1994
.[Abstract/Free Full Text]
- Striegel-Moore RH, Franko DL: Epidemiology of binge eating disorder.Int J Eat Disord34
:S19
–S29,2003
.
- Capristo E, Mingrone G, Addolorato G, Greco AV, Corazza GR, Gasbarrini G: Differences in metabolic variables between adult coeliac patients at diagnosis and patients on a gluten-free diet. Scand J Gastroenterol32
:1222
–1229,1997
.[Medline]
- Gniuli D, Liverani E, Capristo E, Greco AV, Mingrone G: Blunted glucose metabolism in anorexia nervosa.Metabolism50
:876
–881,2001
.[Medline]
- Vaisman N, Rossi MF, Goldberg E, Dibden LJ, Wykes LJ, Penxharz PB: Energy expenditure and body composition in patients with anorexia nervosa.J Pediatr113
:919
–924,1988
.[Medline]
- Acheson KJ, Schutz Y, Bessard T, Ravussin E, Jequier E: Nutritional influences on lipogenesis and thermogenesis after a carbohydrate meal.Am J Physiol246
:E62
–E70,1984
.
- Van Der Ster Wallin G, Norring C, Holmgren S: Binge eating versus nonpurged eating in bulimics: is there a carbohydrate craving after all?Acta Psychiatr Scand89
:376
–381,1994
.[Medline]
- Gendall KA, Sullivan PE, Joyce PR, Carter FA, Bulik CM: The nutrient intake of women with bulimia nervosa. Int J Eat Disord21
:115
–127,1997
.[Medline]
- Zimmet PZ: Hyperinsulinemia—How innocent a bystander? Diabetes Care16
:56
–70,1993
.
- Neel JV: Diabetes mellitus: a thrifty genotype rendered detrimental by progress? Am J Hum Genet14
:353
–362,1962
.[Medline]
- Frey J, Neuhauser-Berthold M, Elis SA, Duncker S, Rose F, Blum WF, Remschmidt H, Geller F, Hebebrand J: Lower serum leptin levels in female students of the nutritional sciences with eating disorders. Eur J Nutr42
:142
–148,2003
.[Medline]
- Havel PJ: Update on adipocyte hormones: regulation of energy balance and carbohydrate/lipid metabolism. Diabetes53
(Suppl 1):S143
–151,2004
.[Abstract/Free Full Text]
- Robinson PH, Clarke M, Barrett J: Determinants of delayed gastric emptying in anorexia nervosa and bulimia nervosa. Gut29
:458
–464,1988
.[Abstract/Free Full Text]
- Cuellar RE, Kaye WH, Hsu LK, Van Thiel DH: Upper gastrointestinal tract dysfunction in bulimia. Dig Dis Sci33
:1549
–1553,1988
.[Medline]
- Hutson WR, Wald A: Gastric emptying in patients with bulimia nervosa and anorexia nervosa.Am J Gastroenterol85
:41
–46,1990
.[Medline]
- Kamal N, Chami T, Andersen A, Rosell FA, Schuster MM, Whitehead WE: Delayed gastrointestinal transit times in anorexia nervosa and bulimia nervosa.Gastroenterology101
:1320
–1324,1991
.[Medline]
TOP
ABSTRACT
INTRODUCTION
