Original Report
Return of Menses in Previously Overweight Patients with Eating Disorders

https://doi.org/10.1016/j.jpag.2019.11.002Get rights and content

Abstract

Study Objective

Adolescents with eating disorders and a history of overweight present with higher weights, longer duration of disease, but equally severe symptomatology compared with previously normal weight patients. To better delineate treatment goals for this understudied population, we compared weight at menses resumption with premorbid maximum weight among previously overweight and normal weight patients.

Setting

Outpatient adolescent medicine clinic at an eating disorder referral center.

Participants

Postmenarchal patients meeting criteria for Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition eating disorders. History of overweight was defined as a body mass index (BMI) greater than or equal to the 85th percentile or 25 or more.

Interventions and Main Outcome Measures

Weight characteristics at presentation and menses resumption (BMI, BMI z-score, change from maximum weight to presentation weight)

Results

Previously overweight patients presented with greater mean weight, longer duration of disease, and higher BMI than previously normal weight patients. No difference was found in rates of amenorrhea at presentation or menses resumption. Previously overweight patients resumed menses at a younger age and higher BMI z-scores. The difference between weight at menses resumption and premorbid maximum weight was greater for previously overweight patients.

Conclusion

Previously overweight patients with eating disorders present differently than their normal-weight peers, so reliance on weight status as a screening criterion might result in underdiagnosis. Although BMI z-scores associated with menses resumption are higher for previously overweight patients, there is no difference in weight gain between presentation and menses resumption and time to menses resumption compared with previously normal-weight patients. Moreover, menses resumption occurred at weights significantly lower than premorbid maximum weight for previously overweight patients, so restoration to highest premorbid weight is not necessary.

Introduction

The rising prevalence of adolescent overweight and obesity has resulted in an emphasis on weight loss by clinicians and health education curricula. Likely reflecting this messaging, overweight adolescents report greater body dissatisfaction and concern about weight than normal-weight adolescents.1 These risk factors result in a higher likelihood of eating disorder thoughts and behaviors such as bingeing, purging, diet pill, and laxative use among these patients.1, 2, 3, 4 The reported prevalence of a history of overweight or obesity among pediatric patients with eating disorders ranges from 11% to 36%.5, 6, 7 Previously overweight adolescents with eating disorders present differently than their previously normal-weight peers. Specifically, they present with greater weight suppression (difference between premorbid highest weight and presentation weight), longer duration of disease, and higher body mass index (BMI).6,8 However, despite no difference in symptomatology, including comparable rates of amenorrhea and with similar scores on psychological measures, previously overweight adolescents with eating disorders are less likely to be hospitalized.6,8

Patients with eating disorders, including anorexia nervosa, atypical anorexia nervosa, avoidant/restrictive food intake disorder (ARFID), and bulimia nervosa, can present with primary and secondary amenorrhea.9 Amenorrhea is associated with comorbidities such as lower peak bone mass and impaired neuropsychological functioning; resumption of menses is a well established marker of physiologic recovery among patients with eating disorders.10,11 In a previous study, previously overweight adolescents with secondary amenorrhea resumed menses at a significantly higher weight than their previously normal-weight peers, suggesting that goal recovery weights might differ on the basis of premorbid overweight status.7

There remains a paucity of longitudinal data on cohorts of previously overweight patients. Because previous weight status might affect treatment goal weights, a comparison between premorbid maximum weight and weight at menses resumption is necessary. Further, there is lack of literature on incidence of history of overweight as well as amenorrhea and subsequent return of menses in patients with bulimia nervosa and ARFID, because previous studies have focused their investigation exclusively on patients with anorexia nervosa and atypical anorexia nervosa.

To address these gaps in knowledge, we aimed to investigate the differences in presenting characteristics, including: (1) eating disorder diagnosis; (2) weight status; (3) weight suppression; and (4) duration of disease between patients with eating disorders with and without a history of overweight. We also compared menstrual characteristics, including: (1) rates of secondary amenorrhea; and (2) weight status at menses resumption between the groups. In keeping with the previous literature, we hypothesized that previously overweight patients would present at higher weights but with greater weight suppression and longer duration of illness with no differences in rates of amenorrhea. We further hypothesized that previously overweight patients would resume menses at higher weights than previously normal-weight patients.

Section snippets

Study Population

A retrospective longitudinal cohort study was conducted on female patients hospitalized and followed in an outpatient clinic for nutritional insufficiency at the Cleveland Clinic, an academic medical center, between January 2008 and January 2017 (n = 420). Data were collected from the electronic medical record for up to 5 years or until the last documented clinic visit from the time of hospitalization. Inclusion criteria included female sex, presentation for evaluation of nutritional

Presentation Characteristics

Of the 420 female patients who presented for evaluation of nutritional insufficiency between 2008 and 2017, 114 did not qualify: 5 had diagnoses other than an eating disorder, 41 were premenarchal, 7 were underweight at their maximum weight, and 61 had missing presentation data. A total of 306 patients were included in the study. Of these patients, 97/306 (31%) of patients had a history of overweight.

Patient demographic characteristics and diagnoses are presented in Table 1. A greater

Discussion

We found that 31% of the patients who presented for treatment of eating disorders had a history of overweight. They presented at higher weight status, but with greater weight suppression, and longer duration of disease than their previously normal-weight peers. They were more likely to be diagnosed with atypical anorexia or bulimia nervosa than their previously normal-weight peers. Although the rates of amenorrhea at presentation and menses resumption did not differ according to group,

References (35)

  • M. Leitner et al.

    Prophylactic phosphate supplementation for the inpatient treatment of restrictive eating disorders

    J Adolesc Health

    (2016)
  • K.K. Ridout et al.

    Daily laboratory monitoring is of poor health care value in adolescents acutely hospitalized for eating disorders

    J Adolesc Health

    (2016)
  • N.D. Brener et al.

    Reliability and validity of self-reported height and weight among high school students

    J Adolesc Health

    (2003)
  • F.J. Elgar et al.

    Validity of self-reported height and weight and predictors of bias in adolescents

    J Adolesc Health

    (2005)
  • D. Neumark-Sztainer et al.

    Five-year longitudinal predictive factors for disordered eating in a population-based sample of overweight adolescents: implications for prevention and treatment

    Int J Eat Disord

    (2009)
  • M. Whitelaw et al.

    Restrictive eating disorders among adolescent inpatients

    Pediatrics

    (2014)
  • A. Poyastro Pinheiro et al.

    Patterns of menstrual disturbance in eating disorders

    Int J Eat Disord

    (2007)
  • Cited by (0)

    The authors indicate no conflicts of interest.

    View full text