Psychologic and physiologic effects of dieting in adolescents. (Review Article).

ABSTRACT

Obesity in adolescents has increased by 75% in the past three decades. Cross-sectional and prospective surveys have shown that a large percentage of adolescents, particularly females and even those of normal weight, diet at some time. While moderate changes in diet and exercise have been shown to be safe, significant psychologic and physiologic consequences may occur with extreme or unhealthy dieting practices. Moderate dieting has been shown to be associated with negative self-esteem in some adolescents. The very act of starting any diet increases the risk of eating disorders in adolescent girls. Extreme methods of weight loss can have adverse physiologic effects if not closely monitored,. Electrolyte disturbances, cardiac dysrhythmias, and even sudden cardiac death can result from unhealthy or extreme dieting practices. Such practices are associated with other problem behavior in adolescents. We review current information on dieting in teenagers and discuss psychologic and physiologic effects of t hese practices.

**********

ADOLESCENTS, especially females, are bombarded with messages from the media about thinness, images of so-called beauty, and ways to achieve a lower body-weight. These images, combined with a society that places a high value on physical beauty, send mixed messages to teenagers and may result in unhealthy, frequently unnecessary, attempts to lose weight. (1)

Dieting has become increasingly widespread, not only to achieve current ideals of thinness, but also because of the increased prevalence of adolescent obesity (an increase of 75% in the past three decades). Approximately 16% of adolescents are mildly overweight with body mass index (BMI) in the 85th to 95th percentile, while 9.9% of adolescents are severely overweight (BMI [greater than or equal to]95th percentile). (2,3) Being overweight as an adolescent (BIVU > 75th percentile) is a significant predictor of increased risk for later atherosclerotic coronary artery disease, colorectal cancer, gout, and arthritis. (4) Being overweight as an adolescent may be a greater risk factor for these conditions than being overweight as an adult. (4) Additionally, obese adolescents have a 70% chance of being obese adults, and obesity in adulthood is associated with several disease states, including obstructive sleep apnea, hypertension, adverse lipoprotein profiles, diabetes mellitus, coronary heart disease, stroke, color ectal cancer, and death from all causes. Aside from its association with various medical conditions and disease states, obesity is also associated with fewer years of education, increased poverty, and a lower marriage rate. (2,5) Given its prevalence, there is an obvious need for attention to obesity, preferably starting at an early age, in an effort to prevent such problems. The answer does not lie in the current trend of attempts to attain quick fixes using unhealthy dieting practices, including those that promise rapid weight loss over short periods of time.

Because dieting is commonplace among adolescents, health care practitioners should be aware of its potential adverse effects, both psychologic and physiologic. Concerns about dieting include its possible association with cycles of weight loss and re-gain that increase the likelihood of developing eating disorders and obesity; decreased self-esteem and other psychologic issues; and potential increases in cardiovascular risk factors and mortality, both long-term and acute. The potentially devastating outcome of dieting practices is tragically illustrated by the case report elsewhere in this issue, outlining the sudden death of a 16-year-old girl who was attempting to follow a high-fat, high-protein, low-carbohydrate diet regimen. (6) In this paper, we assess the current dieting practices among adolescents and review the potential adverse physiologic and psychosocial consequences associated with dieting during adolescence.

THE PREVALENCE OF DIETING

The Centers for Disease Control and Prevention (CDC) monitors 6 categories of health risks in adolescents, including unhealthy dietary behavior, with the Youth Risk Behavior Surveillance (YRBS) system. (3) Several studies have used the YRBS or a similar tool to assess current dieting practices among various populations of adolescents. Current data show that significantly more female adolescents practice weight-control behavior than their male counterparts. Storz and Greene (7) reported that 169 adolescent girls (83%) in their study wanted to lose weight, despite the fact that 104 (62%) of them were within the normal range for body weight. An overwhelming majority of the girls reflected a suboptimal body concept, used more negative than positive adjectives to describe their appearance, and chose ideal figure outlines that were smaller than their actual figures. In a study of Taiwanese female college students, only 16.2% were actually overweight or obese, yet 51.4% perceived themselves as such. (8) A high perce ntage of the subjects practiced dieting and/or exercise to lose weight, regardless of their actual body weight. The prevalence estimates from several studies of weight-loss and/or weight-control efforts by adolescents are summarized in Table 1.

Dieting can include a wide range of behavior; therefore, it is important to determine not only the prevalence of dieting among adolescents, but also to specify the types of weight-loss methods that are used (Table 2). In many cases, dieting is inappropriate, in that normal-weight adolescents are frequently the ones that are using dieting and other weight-loss methods. A minority (10% to 27%) report the use of unhealthy, and possibly dangerous, weight-loss methods such as skipping meals, use of very-low-calorie diets, eating only 1 food per day, or use of powdered diet drinks (Table 2). A more alarming figure was the percentage (1% to 8.9%) using dangerous weight-loss methods, including fasting, self-induced vomiting, laxatives, or diet pills (Table 2).

Different interpretations of the, term dieting may cause variation in the prevalence estimates of weight-loss behavior among adolescents. Some define dieting as simply watching what they eat, while those at the other extreme define dieting as using drastic measures such as fasting, vomiting, or diet pills. This distinction often is not accounted for in research. In a study evaluating the gradations of adolescent weight-control behavior, 44% of adolescent females reported trying to lose weight on the previous day, yet only 8.6% of the food records reflected dieting. (18) This discrepancy was accounted for by adolescents' misinterpretation of dieting to mean just watching what they ate. This variation among adolescents in interpretation of dieting must be considered when reviewing the literature on adolescent weight-loss behavior.

POSITIVE CONSEQUENCES OF DIETING

Despite potential dire outcomes with improper use, positive health consequences are associated with moderate dieting behavior in adolescents. Moderate dieters, those using all dieting methods except vomiting, diet pills, and extreme calorie-restriction, reported more health-promoting behavior than their nondieting and extreme-dieting (those vomiting or using diet pills) counterparts. (9,10) Moderate dieters had increased fruit and vegetable intake, a lowered fat intake, and increased physical activity when compared with the other 2 groups. Those trying to lose weight are more likely to decrease servings of high-fat meats and dessert foods and increase days of hard exercise, compared with nondieters. (11)

Dieting and weight loss in obese teens have several potentially positive health outcomes. Insulin resistance, hyperinsulinemia, and high cholesterol levels decrease with weight reduction in obese adolescents. (5,9) Escobar et al (19) found that altered hepatic insulin clearance in obese children was reversible with weight loss, resulting in near-normalization of serum insulin levels. In addition to correction of abnormal results of serum chemistry tests and insulin resistance, weight loss in morbidly obese subjects decreases upper-airway obstructive symptoms and improves sleep function. (5)

Although energy restriction with moderate dieting has been associated with decreased linear growth, the decreased linear growth velocity is temporary and not associated with permanent stunting. (20) A multidisciplinary weight-reduction (calorie restriction with moderate exercise) program for obese adolescents showed that the majority of subjects were able to sustain normal growth velocity while effectively decreasing body weight. (21)

NEGATIVE CONSEQUENCES OF DIETING

Physical Consequences

Negative physical consequences of excessive dieting can include delayed linear growth and delayed pubarche. (22) In a study evaluating children with short stature and/or delayed puberty, approximately 7% of the subjects fit a pattern of growth-failure due to malnutrition resulting from self-imposed calorie restriction, arising from a fear of obesity. (23) Normal weight gain decreased in these adolescents at 12 years of age, followed by a decline in linear growth velocity. Puberty was also delayed in half of the subjects, and bone age was delayed 0.5 to 5.5 years.

Adolescent weight-loss behavior has been associated with an increased risk of long-term weight gain. A prospective cohort study of high school females found that those who restricted their dietary intake, exercised for weight control, and labeled themselves as dieting were at increased risk for obesity later in life. (2) Appetite-suppressant and laxative use, vomiting, and binge eating also predicted greater weight gain over the 3-year study period. After controlling for initial body mass, adolescents who reported dieting and other weight-loss efforts, either healthy or unhealthy, were more likely than those who did not report these efforts to gain weight.

Dieting in obese adolescents is associated with a decreased resting energy expenditure. A low-energy diet, comprised of calorie restriction only, causes a significant reduction in resting energy expenditure. (24-26) The reduction in resting energy expenditure exceeded that accounted for by changes in lean body mass by 15% in one study. (25) At a certain point in weight-loss efforts, obese patients have to either further decrease caloric intake or preferably increase physical activity to continue to lose weight. The high rate of failure to lose weight in obese children and adolescents may be attributed to this decreased resting energy expenditure.

Psychosocial Consequences

Dieting can also have a significant impact on psychologic health in adolescents, during a time of rapid physical, psychologic, and social development. Several studies have correlated adolescent problem behavior with dieting status.(1,21,22) The most negative patterns of psychosocial and health behavior were found among frequent dieters and purgers, while the most positive patterns were found among the group of never-dieting nonpurgers. (27)

Independent of age, BMI, and known smoking predictors, contemplation of cigarette use has been shown to be positively related to weight concerns, and experimentation with cigarettes has been shown to be positively related to engaging in weight-control behavior. (28) Each day in the United States, more than 3,000 adolescents become habitual smokers and more than 6,000 try their first cigarette. It is unknown what percentage of these new smokers start, at least in part, as a means of controlling weight; however, nearly 75% of them will become addicted, and the physical consequences of smoking are well documented in the literature. Unhealthy weight-loss methods have been shown to be frequently associated with other problem behavior, such as alcohol or marijuana use, school delinquency, engaging in unprotected sex, and suicide attempts. (27) Increased frequency of dieting and purging behavior have been shown to be independently associated with increased health-risk behavior in adolescent females and to a lesser e xtent in males. (3,28) Compared with nonpurging adolescents, purgers (and frequent dieters, to a lesser extent) were more likely to use alcohol, tobacco, and drugs. Additionally, they performed more poorly in school, were more likely to be sexually active, and more frequently reported physical abuse or sexual abuse than their nonpurging counterparts. They were almost three times as likely as their healthy-dieting peers to score high on measures assessing suicide risk.3 Both purging status and dieting frequency were correlated with negative body image, emotional and family stress, and lower "connectedness" with family and others. (1)

Analysis of self-esteem changes among 60 obese adolescents enrolled in a 12-week multidisciplinary weight-loss program found a significant decrease in self-esteem scores, most notably in the physical attributes and appearance subscales. (2) Despite attempts at dieting using various methods, they had no significant decrease in weight when compared with control subjects. It has been suggested that recruiting obese adolescents into weight-loss programs diminishes their coping mechanisms for maintaining self-esteem. (29)

The biggest concern about dieting practices of adolescents is the relationship between dieting and the subsequent development of eating disorders, including anorexia nervosa, bulimia nervosa, and binge eating. Approximately 1% of teenage girls have anorexia nervosa, 2% to 3% of young women have bulimia, and 3% to 5% more have variants of disordered eating. (30) Several obvious health complications are associated with eating disorders. Complications of anorexia nervosa are usually due to starvation and include amenorrhea, estrogen deficiency, osteopenia, hypophosphatemia, renal dysfunction, anemia, leukopenia, and cardiac abnormalities. (31) Complications of bulimia include hypokalemia and other electrolyte disturbances, with the potential for cardiac arrhythmias, menstrual irregularities, gastric dilation and rupture, esophageal rupture, Mallory-Weiss tears with upper gastrointestinal bleeding, parotid gland enlargement, tooth erosion, and esophagitis. (30)

Binge eating has been associated with frequent dieting and purging in adolescents. Girls and boys who reported dieting frequently (more than 5 times in the past year) were almost twice as likely to have fears of being unable to stop eating and to binge eat. (1) Purging adolescents were approximately 3 times as likely to binge eat or fear being unable to stop eating. (1) A study of adolescent females found that binge eating was prevalent among dieters using unhealthy or dangerous weight-control methods and those with a high-risk psychologic profile. (15) The researchers suggested that dieting is probably not appropriate for those with poor self-perception because of the subsequent increased risk for eating disorders.

In a population-based cohort study for predictors of eating-disorder onset in adolescents, fewer than 1-in-500 nondieting female subjects developed an eating disorder over 12 months. (17) Adolescent females dieting at a moderate level were 5 times more likely to develop a new eating disorder within 6 months and had a 1-in-40 chance of developing a new eating disorder over 12 months. At highest risk were female severe dieters. They were 18 times more likely to develop a new eating disorder within 6 months than those who did not diet and had almost a 1in-5 chance of developing a new eating disorder over 12 months.

Seventy-four percent of patients with bulimia attributed the development of their eating disorder to the inability to maintain a low-carbohydrate diet, leading to carbohydrate craving and subsequent cycles of binging and purging. (30) Restrained eating and dieting practices, independent of body weight, was found to be a better predictor of future eating problems than obesity. These findings are similar to those of Keys et al. (32) Male military volunteers were placed on a very-low-calorie diet. All of them had symptoms of eating disorders, including preoccupation with food, food hoarding, and emotional instability.

Several studies have also suggested an association between psychiatric morbidity or disorders and the development of eating disorders; however, the question still remains whether dieting behavior and eating disorders result in the development of the psychiatric disorder or vice versa. Buddeberg-Fisher et al (33) found that a substantial number of adolescent females had more than one psychiatric disorder, and that psychiatric morbidity had the strongest correlation with negative body image. This study raises the concern that psychiatric disorders may be related to dieting behavior among adolescents.

Biochemical and Physiologic Consequences

The primary acute effect of most concern regarding various types of diets and dieting behavior are the anecdotal reports of death in otherwise healthy patients. Several potential etiologies exist for such problems, including arrhythmias or hypertension from stimulant medications (diet pills), acute electrolyte disturbances, or primary effects of the diet on myocardial contractility. These problems may be related purely to the diet or dieting practices or may result from the exacerbation of a previously undiagnosed condition (long QT syndrome).

Both acute and chronic adverse effects have been reported from over-the-counter and prescription diet pills. (34) Many of these medications contain sympathomimetic agents, which act to control hunger as well as to increase the basal metabolic rate. Similar physiologic effects are seen with thyroid supplements, which have also been used to increase the basal metabolic rate in an attempt to facilitate weight loss. While these pharmacologic agents may be effective in these respects, significant adverse effects related to their effects on the cardiovascular system (arrhythmias, hypertension, myocardial ischemia) and the central nervous system (tremulousness, insomnia, psychosis, and seizures) may occur. (30,31) Dieter and Vorperian (35) reported the case of a 59-year-old woman who was using an over-the-counter diet medication containing the sympathomimetic ephedrine, who had cardiac arrest and polymorphic ventricular tachycardia. The patient was resuscitated successfully and subsequent electrophysiologic workup r evealed no etiology other than the diet pills.

In addition to the potential arrhythmogenic effects of diet pills containing stimulants and sympathomimetics, their effects on heart rate and systemic vascular resistance can increase myocardial oxygen consumption to the point where an imbalance occurs between the supply-and-demand ratio, resulting in myocardial ischemia. Gibbs (36) reported the case of a 37-year-old woman who had chest pain, arm pain, and subsequent ventricular fibrillation while using thyroid supplements, a sympathomimetic agent, and thiazide diuretics to facilitate weight loss. He concluded, "The common use of inappropriate prescription drugs for the treatment of obesity, even in patients without underlying organic heart disease, can be hazardous and may result in sudden catastrophic events.

In addition to acute effects described, recent history reminds us of the potential for chronic effects from diet pills. Fenfluramine and dexfenfluramine were introduced into clinical practice as replacements for the sympathomimetic agents. Since they lacked the sympathomimetic effects of previously-used agents, their potential for morbidity and mortality was thought to be limited. They were subsequently withdrawn from clinical use when a significant proportion of patients had cardiac valvular lesions and pulmonary hypertension after long-term use of these drugs. (37)

Aside from the direct effects of diet pills on cardiac function, with their potential for arrhythmias or myocardial ischemia, past clinical experience shows additional morbidity of dieting and dieting practices. The potential for morbidity and mortality related to dieting is highlighted by the experience from the 1960s and 1970s, with reported deaths in patients using liquid protein products. (38,39) One potential etiology for these deaths was presented by Isner et al, (40) who reported on the clinical and morphologic findings in 17 patients who died suddenly while using the liquid protein diet. Of the 17 patients, 16 were women and all had lost a significant amount of weight (mean loss of 35% of their body weight over a 5-month period). Eight of the patients had had one or more episodes of syncope. Multiple-lead electrocardiograms (ECGs) from 10 patients were available and showed ventricular tachycardia and prolongation of the Q-T interval. The authors concluded that semi-starvation, using a very-low-calorie diet is a cause of acquired Q-T prolongation. Other potential etiologies of the Q-T prolongation, including drug-related causes and alterations in potassium, magnesium, and calcium, were ruled out. Autopsy findings showed myocardial fiber atrophy and increased lipofuscin pigment.

In many of the reports, the exact cause of death could not be definitely determined and may have been related to a direct toxic effect of the diet, starvation-induced myocardial atrophy or degeneration, electrolyte imbalance, arrhythmogenic effects of diet pills, preexisting repolarization defects (long QT syndrome), or myocardial ischemia. A full review of the investigations and potential etiologies for these deaths is outlined by the review of Surawicz and Waller. (41)

Subsequently, additional anecdotal reports surfaced regarding morbidity and mortality during dieting using a very-low-calorie diet. (12) These reports noted cardiac complications, including decreased QRS voltage, prolonged Q-T interval, ventricular arrhythmias, and cardiac arrest related to a very-low-calorie diet. Patients with a preexisting prolonged Q-T interval corrected for heart rate (QTc) and those who experienced excessive weight loss were thought to be at higher risk. Thwaites and Bose (42) described an adolescent on a physician-prescribed very-low-calorie diet who lost 24 pounds in 2 weeks. She subsequently had a syncopal episode, and an ECG showed classic ventricular tachycardia with numerous episodes of torsades de pointes. Later, the patient was found to have a preexisting prolonged QTc.

Schmidinger et al (43) compared the course of 8 children (aged 7.7 to 15 years) who were treated with a commercially prepared very-low-calorie diet with that of 1 child who ate a homemade dietary regimen. Caloric intake was similar between the 2 diets; however, the homemade regimen was deficient in sodium and calcium. On day 14, the patient receiving the homemade regimen had arrhythmias, including couplets and nonsustained ventricular tachycardia, thereby potentially implicating electrolyte and mineral imbalance in the etiology of dieting-related mortality.

Additional evidence for the potential role of electrolyte, trace element, and mineral disturbances as the culprits in the morbidity and mortality from very-low-calorie diets and liquid protein diets is provided by Amatruda et al. (44,45) In a prospective evaluation of 6 moderately obese patients who were treated with a 300 kcal liquid protein diet, they noted arrhythmias on 24-hour ambulatory Holter electrocardiograms. (44) There was no association with any single metabolic imbalance or combination of metabolic imbalances. Evidence for the potential role of electrolyte, mineral, or trace element deficiency as an etiologic factor was provided by the follow-up study, which showed no evidence of arrhythmias in patients following the same diet supplemented with minerals, electrolytes, and trace metals. (45)

Another factor that may account for morbidity and mortality related to dieting is the onset of ketosis related to decreased caloric intake, with the mobilization of tissue-stored lipids via lipolysis and release of fatty acids into the blood stream. This ketosis may be augmented by the inclusion of low carbohydrate intake, and, therefore, low endogenous insulin production. Low-carbohydrate/high-protein/high-fat diets have fallen in and out of popularity since the 1960s, resurfacing in various forms, including the Stillman Diet, the Scarsdale Diet, the Zone Diet, the Carbohydrate Addicts' Diet, and the Atkins Diet. These plans are attractive to many dieters because they set no limit on the amount of certain types of foods that can be eaten, propose to exclude hunger from the dieting experience, reduce appetite, and, at times, produce steady weight loss, even after dramatic failures or weight gain on other diets. Few data from prospective, controlled clinical studies have evaluated the long-term physiologic con sequences of such diets, however. These diets are based on several suppositions. Their promoters suggest that obese patients make less efficient use of insulin and overproduce insulin in response to a carbohydrate load, which favors conversion of carbohydrate to fat rather than to energy, even when there is an ongoing energy requirement. It is suggested that the chronically high insulin level in obese patients prevents the liberation of fat stores during times when normal weight loss would occur, such as short fasts and overnight sleep. When a greater quantity of carbohydrate is ingested than can be used immediately or stored as glycogen, the excess is converted into triglycerides and transported to adipose tissue to be stored as fat. A secondary premise is that triglycerides do not stimulate the release of insulin and that amino acids do so only in the presence of a simultaneous glucose load. Therefore, on a low-carbohydrate diet, a state of dietary ketosis is achieved, regardless of the amount of fat-rich a nd protein-rich food consumed. A catabolic state is maintained, allowing weight loss with no intake limitation. With a high-protein/high-fat intake, there is the release of cholecystokinin, a known appetite suppressant. The lack of glucose in the diet also inhibits the 1-hour post-prandial insulin surge associated with the stimulation of neurogenic feeding centers. Additionally, the increased reliance on fat to meet metabolic demands results in the production of ketone bodies, which may also suppress the appetite and curb feelings of hunger.

Low-carbohydrate/high-protein diets are not without potential for adverse biochemical and physiologic consequences, however. Ketone bodies compete with uric acid for renal tubular excretion; ketogenic diets increase serum uric acid concentrations, which can exacerbate disease in patients with gouty diathesis. Bloom and Azar (46) have reported that subjects on low-carbohydrate diets can have postural hypotension. Using an isolated, perfused rat-heart model, Russell and Taegtmeyer (47) showed that the beating heart loses contractile function (more than 50% loss of function in 60 minutes) when oxidizing acetoacetate alone. Elevated levels of free kitty acids may promote both vascular thrombosis and cardiac arrhythmias. (48,49) Plasma lipolysis and infusion of albumin-bound free fatty acids have been shown to lead to the development of arrhythmias in animal models, possibly through a detergent effect on mitochondrial and cell membranes. (48) These diets have also been associated with hypothyroidism, characterized by decreased levels of thyroid-stimulating hormone and triiodothyronine. (50) Stockholm et al (51) have shown that the decrease of stroke volume and cardiac index that occur with a very-low-calorie diet can be prevented by the concomitant administration of thyroid supplementation using triiodothyronine.

The potential role of these physiologic factors in the morbidity and mortality of dieting remains to be determined; however, given previous reports showing alterations in the histology of the myocardium, (40) alterations in myocardial structure and function may play some etiologic role.

Restriction of caloric intake to less than the basal needs results in catabolism of protein and glycogen stores, which have a high water content, leading to a diuretic effect and the loss of free water and electrolytes. Additionally, mobilization of peripheral fat stores and a high fat intake leads to a ketotic state with the production of various ketone bodies, including acetoacetate and [beta]-hydroxybutyrate. These ketone bodies undergo urinary excretion with a cation to maintain electrical neutrality, resulting in the additional loss of cations, including calcium, magnesium, and potassium. (52) When compounded by inadequate intake, deficiencies of these cations may ensue. Mao et al (53) found altered cardiac-conduction function in rats fed a high-fat, marginal-copper diet, suggesting another trace mineral that may have some role in dieting-related cardiac problems.

In addition to its use in weight loss, non-calorie-restricted ketogenic diets have been used for several years for the treatment of seizure disorders in children that are resistant to standard therapy. (54) Best et al (55) evaluated 20 patients receiving a ketogenic diet and noted a prolonged Q-T interval in 3 patients. There was a correlation between Q-T prolongation, decrease in serum bicarbonate level, and increase in serum [beta]-hydroxybutyrate level. Three additional patients had echocardiographic evidence of cardiac chamber enlargement, while 1 patient had severe dilated cardiomyopathy with pulmonary edema, which resolved after discontinuance of the diet.

The final potential factor that may play a role in the etiology of dieting-related morbidity and mortality is direct myocardial effects related solely to inadequate caloric intake. Starvation, regardless of its cause, results in atrophy of the heart, bradycardia, right shift of the QRS axis, decreased QRS complex and T wave amplitude, and QTc interval prolongation. (56) Autopsy studies have shown histologic alterations of the myocardium, including atrophy, ganglionitis, and neuritis. (40,57)

SUMMARY

Approximately 60% of adolescent females and 25% of adolescent males have dieted at some point in their lives (Table 1). Among these adolescent dieters, a significant percentage report unhealthy or dangerous weight-loss methods, including use of diet pills, fasting, skipping meals, or using very-low-calorie diets (Table 2). Dieting can be associated with both positive and negative consequences. Dieting adolescents report more health-promoting behavior, such as increasing fruit and vegetable intake, decreasing fat intake, and increasing exercise; however, dieting in adolescence is also associated with increased health-risk behavior, such as alcohol and tobacco use, delinquency, and suicide attempts. Additionally, dieting is often undertaken by adolescents who are not above the ideal or normal body weight for their age and physical stature.

Dieting appears to have more negative than positive consequences on the psychologic health of adolescents. Adolescent weight-loss behavior is associated with poor body image, negative psychosocial risk factors, and possibly decreased coping mechanisms, leading to decreased self-esteem. A strong correlation between dieting and the onset of eating disorders exists. One of the most alarming issues regarding dieting is the report of associated mortality. While many of the reports are anecdotal and the majority of patients diet without sequelae, scientific data support potential adverse physiologic consequences of dieting and weight loss, including loss of electrolytes, minerals and trace metals, histologic changes of the myocardium, as well as the potential adverse effects of free fatty acids on myocardial function. Alterations of electrolytes (sodium, potassium, magnesium, and calcium), minerals, and trace metals may be related to decreased intake (anorexia, deficient diets) or increased losses from the gastroin testinal tract (bulimia or laxative use) and the kidneys (ketosis).

Given all of the potential consequences of dieting and its associated behaviors, obesity prevention should be the primary intervention point, with improved education of parents, adolescents, and children regarding appropriate eating patterns and behaviors. When weight loss is necessary, the most appropriate method remains modest caloric restriction incorporating a balanced intake of macronutrients and micronutrients, along with increased physical activity. Considering the previous reports of serious cardiac sequelae, very-low-calorie diets or ketogenic diets should be undertaken only under a physician's direct supervision, with a pre-diet ECG and frequent cardiac monitoring during the course of the diet. The common theme in many of the reports of morbidity and/or mortality related to dieting practices is the use of diets that induce ketosis (very-low-calorie diets or low-carbohydrate/high-fat diets). With very-low-calorie diets, ketosis is induced by inadequate caloric intake and fatty-acid mobilization, whil e with low-carbohydrate diets, ketosis is induced by decreased insulin-release. When such diets are followed without supplementation, alterations of electrolytes, minerals, and trace elements may result in Q-T prolongation and the potential for cardiac dysrhythmias.

TABLE 1.

Prevalence of Weight-Loss or Weight-Control Attempts in Adolescents


         Study Question               Sample

Trying to lose weight in the          9th-l2th grade students
  past '30 days (3)
Actively attempting to lose           9th-l2th grade students
  or control weight in
  the past 7 days (9)
Trying to lose weight in              9th-12th grade females
  the past year (10)
Trying to lose weight (14)            9th-12th grade students
Trying to lose weight (12)            6th-8th grade sttidents
Trying to maintain weight (13)
Wanted to lose weight (14)            6th grade females
Ever dieted (14)                      6th grade females
Currently trying to lose weight (15)  10th grade females
Ever been on a diet (16)              5th-12th grade students
Subgroups:
  5th-8th grade students
  9th-l2th grade students

                                                         Yes (%)
         Study Question               No. of Subjects   Female

Trying to lose weight in the              15,349         59.4
  past '30 days (3)
Actively attempting to lose               16,125         57.6
  or control weight in
  the past 7 days (9)
Trying to lose weight in                   1,015         41.4
  the past year (10)
Trying to lose weight (14)                 3,055         61.6
Trying to lose weight (12)                 2,331         50.6
Trying to maintain weight (13)                           22.2
Wanted to lose weight (14)                   206         58.0
Ever dieted (14)                             206         53.0
Currently trying to lose weight (15)         341         53.7
Ever been on a diet (16)                   6,728         45.4
Subgroups:
  5th-8th grade students                                 35.6
  9th-l2th grade students                                56.3

                                            Yes (%)
         Study Question               Male      Total

Trying to lose weight in the          26.1      42.7
  past '30 days (3)
Actively attempting to lose           24.9      41.5
  or control weight in
  the past 7 days (9)
Trying to lose weight in               -          -
  the past year (10)
Trying to lose weight (14)            21.5      41.1
Trying to lose weight (12)            30.5      40.5
Trying to maintain weight (13)        23.1      22.7
Wanted to lose weight (14)             -          -
Ever dieted (14)                       -          -
Currently trying to lose weight (15)   -          -
Ever been on a diet (16)              20.2      32.5
Subgroups:
  5th-8th grade students              18.2
  9th-l2th grade students             22.5
TABLE 2.

Prevalence of Specific Weight-loss Methods


Study                                       Sample

CDC, (3) 2000                       9th-12th grade students
  Decreased caloric intake
  Exercise
  Fasting
  Diet pills, powders, or liquids
  Vomiting and/or laxatives
Story et al, (9) 1998               9th-12th grade students
  Moderate weight-control
    methods (all methods except
    vomiting or diet pills)
  Extreme weight-control
    methods (vomiting and/or
    diet pill use)
French et al, (1) 1995              9th-12th grade females
  Healthy weight-loss methods *
  Both healthy and
    unhealthy methods +
  Vomiting
  Diet pills
  Appetite suppressants
  Laxatives
  Diuretics
Krowchuk et al, (13) 1998           6th-8th grade students
  Exercise
  Dieting
  Vomiting/Laxatives
  Diet pills
Neumark-Szatainer et al, (16) 1995    10th grade females
  Moderate weight-loss
    methods **
  Unhealthy methods ++
  Dangerous methods ss
Patton, (17) 1999                   9th-12th grade students
  Moderate dieters
  Severe dieters
Neumark-Szatainer et al, (16) 2000  5th-12th grade students
  Ever binge-purge cycling
  Binge-purge behavior
    at least once per day

                                                            Yes (%)
Study                               No. of Subjects  Females    Males

CDC, (3) 2000                           15,349
  Decreased caloric intake                            56.1      25.0
  Exercise                                            67.4      49.5
  Fasting                                             18.8       6.4
  Diet pills, powders, or liquids                     10.9       4.4
  Vomiting and/or laxatives                            7.5       2.2
Story et al, (9) 1998                   16,125
  Moderate weight-control                             52.1      23.4
    methods (all methods except
    vomiting or diet pills)
  Extreme weight-control                               5.5       1.5
    methods (vomiting and/or
    diet pill use)
French et al, (1) 1995                   1,015
  Healthy weight-loss methods *                       21.5       -
  Both healthy and                                    15.2       -
    unhealthy methods +
  Vomiting                                             4.4       -
  Diet pills                                           5.4       -
  Appetite suppressants                                3.3       -
  Laxatives                                            1.6       -
  Diuretics                                            1.0       -
Krowchuk et al, (13) 1998                2,324
  Exercise                                            71.0      48.0
  Dieting                                             44.9      24.4
  Vomiting/Laxatives                                   9.7       4.0
  Diet pills                                           8.8       4.7
Neumark-Szatainer et al, (16) 1995         341
  Moderate weight-loss                                17.9       -
    methods **
  Unhealthy methods ++                                27.9       -
  Dangerous methods ss                                 7.9       -
Patton, (17) 1999                        1,685
  Moderate dieters                                    60.0      29.0
  Severe dieters                                       8.0       0.5
Neumark-Szatainer et al, (16) 2000       6,728
  Ever binge-purge cycling                            13.4       7.1
  Binge-purge behavior                                 8.9       4.1
    at least once per day

                                     Yes (%)
Study                               Total

CDC, (3) 2000
  Decreased caloric intake          40.4
  Exercise                          58.4
  Fasting                           12.6
  Diet pills, powders, or liquids    7.6
  Vomiting and/or laxatives          4.8
Story et al, (9) 1998
  Moderate weight-control            -
    methods (all methods except
    vomiting or diet pills)
  Extreme weight-control             -
    methods (vomiting and/or
    diet pill use)
French et al, (1) 1995
  Healthy weight-loss methods *      -
  Both healthy and                   -
    unhealthy methods +
  Vomiting                           -
  Diet pills                         -
  Appetite suppressants              -
  Laxatives                          -
  Diuretics                          -
Krowchuk et al, (13) 1998
  Exercise                          59.4
  Dieting                           34.5
  Vomiting/Laxatives                 6.8
  Diet pills                         6.7
Neumark-Szatainer et al, (16) 1995
  Moderate weight-loss               -
    methods **
  Unhealthy methods ++               -
  Dangerous methods ss               -
Patton, (17) 1999
  Moderate dieters                  45.0
  Severe dieters                     4.0
Neumark-Szatainer et al, (16) 2000
  Ever binge-purge cycling          10.2
  Binge-purge behavior
    at least once per day

* Healthy weight loss methods were defined as calorie reduction,
increased exercise, increased fruit and vegetable intake, eliminating
snacks, decreasing fat intake, eliminating sweets, reducing the amount
of food consumed, changing the type of food eaten, eating less meat,
eating less high carbohydrate food, and eating low-calorie foods.

+ Unhealthy weight loss methods were defined as fasting, skipping meals,
smoking, lexative use, diuretic use, appetite suppressant use, diet pill
use, and vomiting.

** Moderate weight loss methods were defined as exercise, balanced diet
inclusive of food groups, behavior modification, eating less fats and
sugar, consultation with a health care professional, and participation
in a weight loss group.

++ Unhealthy methods were defined as skipping meals, use of very-
low-calorie diet, eating one food a day, and use of powdered diet
drinks.

(ss.)Dangerous methods were defined as fasting, vomiting, use of
laxatives, and use of diet pills.

References

(1.) French SA, Story M, Downes B, et al: Frequent dieting among adolescents: psychological and health behavior correlates. Am J Public Health 1995; 85:695-701

(2.) Stice E, Cameron RP, Killen JD, et al: Naturalistic weight-reduction efforts prospectively predict growth in relative weight and onset of obesity among female adolescents. J Consult Clin Psychol 1999; 67:967-974

(3.) Centers for Disease Control and Prevention: Youth risk behavior surveillance--United States, 1999. MMWR CDC Surveill Summ 2000; 49(SS05):1-96

(4.) Must A, Jacques PF, Dallal GE, et al: Long-term morbidity and mortality of overweight adolescents: a follow-up of the Harvard Growth Study. N Engl J Med 1992; 327:1350-1355

(5.) Willi S, Oexmann MJ, Wright N, et al: The effects of a high-protein, low-fat, ketogenic diet on adolescents with morbid obesity: body composition, blood chemistries, and sleep abnormalities. Pediatrics 1998; 101:61-67

(6.) Stevens A, Robinson DP, Turpin J, et al: Sudden cardiac death during dieting in an adolescent. South Med J 2002; (in press)

(7.) Storz N, Greene W: Body weight, body image, and perception of fad diets in adolescent girls. J Nutrit Educ 1983; 15:15-18

(8.) Wong Y, Huang YC: Obesity concerns, weight satisfaction and characteristics of female dieters: a study on female Taiwanese college students. J Am Coll Nutr 1999; 18:194-200

(9.) Story M, Neumark-Sztainer D, Sherwood N, et al: Dieting status and its relationship to eating and physical activity behavior in a representative sample of US adolescents. J Am Diet Assoc 1998; 98:1127-1135

(10.) French SA, Perry CL, Leon GR, et al: Dieting behavior and weight change history in female adolescents. Health Psychol 1995; 14:548-555

(11.) Middleman AB, Vazquez I, DuRant RH: Eating patterns, physical activity, and attempts to change weight among adolescents. J Adolesc Health 1998; 22:37-42

(12.) Ahmed W, Flynn MA, Alpert MA: Cardiovascular complications of weight reduction diets. Am J Med Sci 2001; 321:280284

(13.) Krowchuk DP, Kreiter SR, Woods CR, et al: Problem dieting behavior among young adolescents. Arch Pediatr Adolesc Med 1998; 152:884-888

(14.) Koff E, Rierdan J: Perceptions of weight and attitudes toward eating in early adolescent girls. J Adolesc Health 1991; 12:307-312

(15.) Neumark-Sztainer D, Butler R, Palti H: Dieting and binge eating: which dieters are at risk? J Am Diet Assoc 1995; 95:586-589

(16.) Neumark-Sztainer D, Hannan PJ: Weight-related behavior among adolescent girls and boys: results from a national survey. Arch Pediatr Adolesc Med 2000; 154:569-577

(17.) Patton GC: Onset of adolescent eating disorders: population based cohort study over 3 years. BMJ 1999; 318:765-768

(18.) Nichter M, Ritenbaugh C, Nichter M, et al: Dieting and "watching" behavior among adolescent females: report of a multimethod study. J Adolesc Health 1995; 17:153-162

(19.) Escobar O, Mizuma H, Sothern M, et al: Hepatic insulin clearance increases after weight loss in obese children and adolescents. Am J Med Sci 1999; 13:282-286

(20.) Dietz W, Hartung R: Changes in height velocity of obese preadolescents during weight reduction. Am J Dis Child 1985; 139:705-707

(21.) Sothern MS, Udall JN, Suskind EM, et al: Weight loss and growth velocity in obese children after very low calorie diet, exercise, and behavior modification. Acta Paediatr 2000; 89:1036-1043

(22.) Pugliese M, Lifshitz F, Grad G, et al: Fear of obesity: a cause of short stature and delayed puberty. N Engl J Med 1983; 309:513-518

(23.) Davis DR, Apley J, Fill G, et al: Diet and retarded growth. BMJ 1978; 1:539-542

(24.) Sothern MS, Loftin M., Suskind R, et al: The impact of significant weight loss on resting energy expenditure in obese youth. J Invest Med 1999; 47:222-226

(25.) Menozzi R, Bondi M, Baldini A, et al: Resting metabolic rate, fat-free mass and catecholamine excretion during weight loss in female obese patients. Br J Nutr 2000; 84:515-520

(26.) Tounian P, Frelut M-L, Parlier G, et al: Weight loss and changes in energy metabolism in massively obese adolescents. Int J Obes Relat Metab Disord 1999; 23:830-837

(27.) Neumark-Sztainer D, Story M, French SA: Covariations of unhealthy weight loss behavior and other high-risk behavior among adolescents. Arch Pediatr Adolesc Med 1996; 150:304-308

(28.) Tomeo CA, Field AE, Berkey CS, et al: Weight concerns, weight control behavior, and smoking initiation. Pediatrics 1999; 104:918-924

(29.) Cameron JW: Self-esteem changes in children enrolled in weight management programs. Issues Comp Pediatr Nurs 1999; 22:75-85

(30.) Shisslak CM, Crago M, Neal ME, et al: Primary prevention of eating disorders. J Consult Clin Psychiatry 1987; 55:660-667

(31.) Haglin L: Hypophosphataemia in anorexia nervosa. Postgrad Med J 200l; 77:305-311

(32.) Keys A, Brozek J, Henschel A, et al: The Biology of Human Starvation. Minneapolis, University of Minnesota Press, Vol II, 1950, pp 819-853

(33.) Buddeberg-Fischer B, Klaghofer R, Reed V: Associations between body weight, psychiatric disorders, and body image in female adolescents. Psychother Psychosom 1999; 68:325-332

(34.) Zahn KA, Li EL, Purssell RA: Cardiovascular toxicity after ingestion of "herbal ectasy." J Emerg Med 1999; 17:289-291

(35.) Dieter ES, Vorperian VR: Sudden cardiac death in a patient taking an over-the-counter diet pill. Wis Med J 2000; 99:20-22

(36.) Gibbs HR: Diet pills and sudden death (Letter). N Engl J Med 1988; 318:1127

(37.) Teramae CY, Connolly HM, Grogan M, et al: Diet drug-related cardiac valve disease: the Mayo Clinic echocardiographic laboratory experience. Mayo Clin Proc 2000; 75:456-461

(38.) Lantigua RA, Amatruda JM, Biddle TL, et al: Cardiac arrhythmias associated with a liquid protein diet for the treatment of obesity. N Engl J Med 1980; 303:735-738

(39.) Michiel RR, Sneider JS, Dickstein RA, et al: Sudden death in a patient on a liquid protein diet. N Engl J Med 1978; 298:1005-1007

(40.) Isner JM, Sours HE, Paris AL, et al: Sudden, unexpected death in avid dieters using the liquid-protein modified fast diet. Circulation 1979; 60:1401-1412

(41.) Surawicz B, Waller BF: The enigma of sudden cardiac death related to diet. Can] Cardiol 1995; 11:228-231

(42.) Thwaites BC, Bose M: Very low calorie diets and pre-fasting prolonged QT interval, a hidden potential danger. West Ind Med J 1992; 42:169-171

(43.) Schmidinger H, Weber H, Zwiaer K et al: Potential life-threatening cardiac arrhythmias associated with a conventional hypocaloric diet. Int J Cardiol 1987; 14:55-63

(44.) Lantigua RA, Amatruda JM, Biddle TL, et al: Cardiac arrhythmias associated with a liquid protein diet for the treatment of obesity. N Engl J Med 1980; 303:735-738.

(45.) Amatruda JM, Biddle TL, Patton ML, et al: Vigorous supplementation of a hypocaloric diet prevents cardiac arrhythmias and mineral depletion. Am J Med 1983; 74:1016-1022

(46.) Bloom WL, Azar GJ: Similarities of carbohydrate deficiency and fasting: weight loss, electrolyte excretion, and fatigue. Arch Intern Med 1963; 112:333-337

(47.) Russell RR III, Taegtmeyer H: Changes in the citric acid cycle flux and anaplerosis antedate the functional decline in isolated rat hearts. J Clin Invest 1991; 87:384-390

(48.) Oliver MF, Yates PA: Induction of ventricular arrhythmias by elevation of arterial free fatty acids in experimental myocardial infarction. Cardiology 1972; 56:359-364

(49.) Peyreigne C, Bouix D, Aissa Benhaddad A, et al: Hemorheologic effects of a short-term ketogenic diet. Clin Hemorheol 1999; 21:147-153

(50.) Ullrich IH, Peters PJ, Albrink MJ: Effect of low-carbohydrate diets high in either fat or protein on thyroid function, plasma insulin, glucose and triglycerides in healthy young adults. J Am doll Nutr 1985; 4:451-459

(51.) Stockholm KH, Astrup A, Breum L, et al: Cardiac stroke work output during a short term very low calorie diet, noradrenaline-induced thermogenesis and T3 supplementation. Int J Obes Relat Metab Disord 1989; 13(suppl 2) :151-153

(52.) Zeman FJ, Hansen RJ: Diabetes mellitus, hypoglycemia and other endocrine disorders. Clinical Nutrition and Dietetics. Seman FJ (ed). New York, Macmillan Publishing Co, 2nd Ed, 1991, pp 409-410

(53.) Mao S, Medeiros DM, Hamlin RL: Marginal copper and high fat diet produce alterations in electrocardiograms and cardiac ultrastructure in male rats. Nutrition 1999; 15:890-898

(54.) Hemingway C, Freeman JM, Pillas DJ, et al: The ketogenic diet: a 3 to 6 year follow-up of 150 children enrolled prospectively. Pediatrics 2001; 108:898-905

(55.) Best TH, Franz DN, Gilbert DL, et al: Cardiac complications in pediatric patients on the ketogenic diet. Neurology 2000; 54:2328-2330

(56.) Swenne I, Larsson PT: Heart risk associated with weight loss in anorexia nervosa and eating disorders: risk factors for QTc interval prolongation and dispersion. Acta Paediatr 1999; 88:304-309

(57.) Siegel SJ, Cabeen WR Jr, Roberts WC: Prolonged QT interval-ventricular tachycardia syndrome from massive rapid weight loss utilizing protein modified fast diet: sudden death with sinus node ganglionitis and neuritis. Am Heart J 1981; 102:121-122

RELATED ARTICLE: KEY POINTS

* Approximately 16% of adolescents are mildly overweight, while 9.9% are severely overweight.

* Of dieting adolescents, 1% to 8.9% report using potentially dangerous methods of weight loss.

* Potential causes for morbidity and mortality related to dieting practices include a direct toxic effect of the diet, starvation-induced myocardial atrophy, imbalances of electrolytes, and preexisting repolarization defects.

* Given the potential consequences of dieting, obesity prevention should be the primary intervention point.

* When weight loss is necessary, the most appropriate method remains modest caloric restriction, consisting of a balanced macronutrient and micronutrient intake combined with increased physical activity.

From the Departments of Child Health and Anesthesiology, University of Missouri, Columbia.

Reprint requests to Joseph D. Tobias, MD, University of Missouri, Department of Anesthesiology, 3W40H, One Hospital Dr, Columbia, MO 65212.