Anorexia nervosa: A quantitative approach to follow-up

Prospective Longitudinal Assessment of Linear Growth and Adult Height in Female Adolescents With Anorexia Nervosa

CONTEXT

Growth retardation is an established complication of anorexia nervosa (AN); however, findings concerning the adult height of AN patients are inconsistent.

OBJECTIVE

The objective of this work was to assess linear growth and adult height in female adolescents with AN.

DESIGN AND SETTING

A prospective observational study was conducted in a tertiary university hospital.

PARTICIPANTS

Participants included all 255 female adolescent AN patients hospitalized in the pediatric psychosomatic department between January 1, 2000 and May 31, 2015.

INTERVENTIONS

Height and weight were assessed at admission and during hospitalization. Patients were subsequently invited for measurement of adult height. Additional data collected included premorbid height data, menstrual history, skeletal age, pertinent laboratory studies, and parental heights.

MAIN OUTCOME MEASURE

The main outcome measure of this study was adult height.

RESULTS

Mean age at admission was 15.4 ± 1.75 years, mean body mass index (BMI) was 15.7 ± 1.8 kg/m2 (BMI SDS = -2.3 ± 1.45 kg/m2). Premorbid height SD scores (SDS) were not significantly different from those expected in normal adolescents (0.005 ± 0.96). However, height SDS at admission (-0.36 ± 0.99), discharge (-0.34 ± 0.96), and at adult height (-0.29 ± 0.95), were significantly (P < .001) lower than expected. Furthermore, adult height was significantly (P = .006) shorter compared to the midparental target height. Stepwise forward linear regression analysis identified age (r = 0.32, P = .002) and bone age (r = -0.29, P = .006) on admission, linear growth during hospitalization (r = 0.47, P < .001), and change in luteinizing hormone during hospitalization (r = -0.265, P = .021) as independent predictors of improvement in height SDS from the time of admission to adult height.

CONCLUSIONS

Whereas the premorbid height of female adolescent AN patients is normal, linear growth retardation is a prominent feature of their illness. Weight restoration is associated with catch-up growth, but complete catch-up is often not achieved.

Medical assessment of adolescent girls with eating disorders: an evaluation of symptoms and signs of starvation

AIM

To investigate the prevalence of symptoms and signs related to starvation at the initial examination of adolescent girls with eating disorders (ED).

METHODS

Two hundred and eleven girls with eating disorders recruited for a multicentre research and evaluation programme of six specialist eating disorder services in Sweden have been studied. The presence or absence of 12 symptoms, reported by the patients, and 16 signs, observed by the examiners, were registered and related to body weight.

RESULTS

Eleven observed signs--loss of subcutaneous fat, loss of muscle mass, loss of muscular force, dry and scaly skin, brittle nails, dry and brittle hair, lanugo hair, resting pulse <60, systolic blood pressure <110, peripheral hypothermia, and peripheral cyanosis--were related to body weight expressed in standard deviation scores (SDS). When the number of observed signs for each patient was calculated, there was a strong relationship with weight. The odds ratio for having more than two signs was 4.35 (95% CI 2.67-7.04; p = 2.8 x 10(-9) for every one-unit change in weight SDS. Of the symptoms reported by the patients, only three were related to weight. When the number of reported symptoms for each patient was calculated, a relationship with weight was not observed.

CONCLUSION

In adolescent girls with ED, physical signs observed at medical examination can be related to weight. However, reported symptoms are poorly related to weight and may be influenced by other factors. The finding emphasizes the importance of medical assessment at presentation of patients with ED.

Struggling with recovery: participant perspectives on battling an eating disorder

Recovery from an eating disorder encompasses a complex matrix of mental, physical, and social factors. Although methodological differences affect rates of reported recovery (from 24 to 76%), many women are not experiencing full recoveries from eating disorders. This trend is apparent in persons who discontinue severely destructive behaviors but continue to exhibit psychiatric problems and impairment in social and occupational roles. The purpose of this grounded theory study was to provide an explanatory schema about recovery. Data collection included 17 interviews, a focus group, and participant observation. Findings included repeating wavelike patterns of disease and recovery that seldom returned to a state of normal eating but, instead, remained at a level of atypical eating.

The changing profile of anorexia nervosa at the Toronto Programme for Eating Disorders

The present study is a retrospective comparison of anorexic patients of the restricting subtype (AN) with those of the bulimic subtype (AN+BN) over three consecutive 5-year time periods. Subjects consisted of 806 female out-patients diagnosed as either AN or AN+BN between 1978 and 1994 during an assessment of their eating problem at the University of Toronto Programme for Eating Disorders in Toronto, Canada. During each time period an increasing proportion of subjects was diagnosed with AN+BN than with AN. Subjects with AN+BN reported a longer duration of illness prior to being assessed. Over time, subjects in both groups appeared to weigh more, and were less likely to report amenorrhea. Subjects from both groups reported higher frequencies of purging behavior, impulsive behaviors, and associated affective symptoms in the later time periods. Over time, the differences between the two subgroups of patients became less distinct. The results indicate changes in the clinical profile of patients with anorexia nervosa seen at the Toronto Programme for Eating Disorders over the past 16 years. The implications of these changes are explored.

Long-term outcome of children and adolescents with anorexia nervosa: study of comorbidity

Eighty-seven children and adolescents with anorexia nervosa, admitted to the Gaslini Department of Child Neurology and Psychiatry between 1976 and 1990, were followed up after a mean of 9.6 years. Outcome measures included the Morgan Russell Outcome Schedule as modified by Jeammet. Outcome was good in 43 (53%) cases, intermediate in 27 (34%) cases, and negative in 11 (14%) cases. No deaths occurred. Based on the Jeammet assessment schedule, the most significant items predicting outcome were insight; sexual, familial, and social relationships; and mental state. Gender of patients and early disease onset did not seem to be predictive measures. Poor outcome was associated with a severe initial clinical picture and length of in-patient treatment. In regard to comorbidity, mood and personality disorders seemed to be negative prognostic indicators, whereas anxiety disorders did not show prognostic value.

Stature, drive for thinness and body dissatisfaction: a study of males and females from a non clinical sample

A study of 174 subjects selected from a Canadian shopping mall revealed an inverse correlation among the males but not among females between body stature and body image parameters related to dieting, body weight and shape (for example, drive for thinness, and body dissatisfaction). These correlations were independent of body weight. Among the females, there was a direct correlation between the Body Mass Index and these body image parameters. Height is generally associated with desirable looks and academic success among men and most likely has a positive effect upon body image. Therefore, males who are dissatisfied with their stature, a feature that is generally not within their control, may attempt to alter an aspect of their body that they can control (for example, weight). This finding may have important implications among males with eating disorders.

The process of recovering from anorexia nervosa

The authors describe the process of recovering from anorexia nervosa as it is revealed by 13 women who have recovered from the illness. Emphasizing the patient's perspective, the paper reviews the perceived causes of the disorder, helpful and harmful therapy-related and life experiences, features hardest to change, the defensive function of anorexia nervosa and residual traits. Generally, the movement toward health entails forming a therapeutic relationship in which the anorexic can identify and express feelings, experience the empathic, nonjudgmental understanding of another person, separate from a pathological family system, resolve hostile dependent attachment to parents, assuage primitive guilt, and engage in the trials of adolescent psychosexual development to enter adulthood with the beginnings of a firm, cohesive sense of self.

Recovery of reproductive function in patients with anorexia nervosa: a 10-year follow-up study

The recoveries of reproductive function and body weight of 21 patients with anorexia nervosa were studied in a 10-yr period from the time of their first presentation. The recovery rates of both menstruation and eating behavior were 81.0%, although at the end of the study period the rate of amenorrhea (19.0%) was still higher than that (6%) in women in the general population. The change of mean body weight in patients in whom regular menstruation was restored was not significantly different from that of patients in whom amenorrhea persisted. Of the patients, 16 married after treatment and 14 of these (87.5%) became pregnant after spontaneous (11 patients) or induced (3 patients) ovulation, and 12 delivered a baby. The present study suggests that resumption of regular menstruation may not depend only upon body weight gain, and that most patients may be able to have a child after appropriate treatments.

The adolescent ballet dancer. Nutritional practices and characteristics associated with anorexia nervosa

A questionnaire was administered to 45 female ballet dancers aged 12-21 years and 44 nonathletic female subjects aged 14-16 years. Dietary practices and nutrition beliefs of the dancers were assessed, and a comparison was made of the prevalence of anorexia nervosa characteristics among the two groups. Ballet dancers reported characteristics of anorexia nervosa significantly more often than did controls. Characteristics used to differentiate between the groups were underweight (p less than 0.05), distorted body image (p less than 0.005), amenorrhea (p less than 0.02), and binge eating (p less than 0.005). Nutrition practices and beliefs among adolescent ballet dancers included frequent use of weight reduction strategies such as fasting, binging, and selective food restriction. Supplements were used by 60% of the dancers, primarily a multivitamin or vitamin C supplement. Fluids or carbohydrates wer not viewed as an important aid to performance by the majority of dancers. This study indicates the need for improved education for ballet dancers, coaches, trainers, and parents, including information on nutrition needs and the management of behaviors associated with anorexia nervosa.

Evaluation of growth in anorexia nervosa from serial measurements

Measured longitudinal growth data in a sample of 66 patients with anorexia nervosa (AN) (58 females and 8 males) is compared to expected growth from normative standards. The data was subject to cross-sectional as well as longitudinal analysis. Females had a normal growth rate and pattern, in particular a normal sequence and timing of pubertal events. Distribution of weight for age indicated a slight (non-significant) overweight before onset of anorexia nervosa. Weight by height was significantly above the expected before onset of AN. Mean age at Peak Height Velocity (PHV) was 11.5 yr. Mean size of PHV 8.03 cm/yr. A subsample of females (those with onset of AN after menarche) had a mean age of menarche of 12.88 yr. Mean distance in time from age at PHV to age at menarche was 1.34 yr. These results are normal for Scandinavian girls. Male anorexics were tall for age, heavy for age and, where a PHV could be seen (5 out of 8 cases), had signs of early puberty. Mean age at PHV was 11.8 yr, a highly significant difference from the normal mean age at PHV which is 14.06 yr for Scandinavian boys. Weight by height was distributed as in the reference population, that is, with no indication of pre-morbid obesity for male probands.