Abstract
Purpose
Data on the clinical characteristics of adolescent males with eating disorders are limited. The purpose of this study was to describe the demographic characteristics, presenting vital signs, laboratory results, and relevant risk factors for eating disorders among males presenting to an outpatient adolescent and young adult medicine practice.
Methods
Retrospective chart review of male eating disorder patients aged of 11–25 years presenting to the University of California, San Francisco Adolescent and Young Adult Eating Disorder Program between June 1, 2011, and November 1, 2014. Charts were reviewed for demographic and clinical characteristics and risk factors for eating disorders.
Results
Thirty-three patients were included; mean age was 16 years. Patients presented with mean heart rate was 58.7 bpm, and orthostatic heart rate change was 22 bpm, with 51.5% meeting Society for Adolescent Health and Medicine hospital admission criteria. Mean percent of median body mass index was 88%. Of patients with available laboratory data, 33.3% were anemic, 23.8% leukopenic, 19.0% thrombocytopenic, and 10.0% neutropenic. Half had a history of a psychiatric disorder; 41.5% had a history of overweight or obesity, and 12.1% had a family history of an eating disorder. The DSM-IV-TR and DSM-5 diagnostic criteria were retrospectively applied to patients, with an increase in diagnosis of anorexia nervosa from 36.4% to 48.5%. Diagnoses of Eating Disorder Not Otherwise Specified, now Other Specified Feeding or Eating Disorder in DSM-5, decreased from 62.6% to 45.5%.
Conclusions
Male patients with eating disorders presented with significant abnormalities; patients were bradycardic and orthostatic; and more than half met Society for Adolescent Health and Medicine admission criteria. Patients with available laboratory data demonstrated significant abnormalities consistent with malnutrition. Given that eating disorders are less likely to be detected in males, it is important to recognize early signs of malnutrition, particularly in those who present within the normal body mass index range for age.
Keywords
Implications and Contribution
This study describes presenting vital signs and laboratory findings in male adolescents and young adults presenting to an outpatient clinic with restrictive eating disorders. Males were found to have low heart rates, abnormal changes in heart rate from lying to standing, and changes in laboratory values consistent with malnutrition.
See Related Editorial p. 371
Males are estimated to make up 5%–10% of cases of anorexia and bulimia nervosa [
[1]
]. Prevalence estimates have been reported to be higher in younger males; one study of 5- to 13-year olds found that one fourth of patients meeting criteria for eating disorders were male [[2]
]. However, this is likely to be a drastic underreporting of the actual prevalence. Reasons for the underreporting of eating disorders in males may be related to the female-centric diagnostic criteria. Specifically, before 2013, diagnostic criteria focused on typically female presenting symptoms, including drive for thinness and amenorrhea. After the implementation of the new DSM-5 diagnostic criteria, there has been a reported 28.9% increase in lifetime eating disorders in men [3
, 4
].Despite the increasing recognition of disordered eating behaviors in males, there are limited data on clinical presentation and risk factors for eating disorders among adolescent and young adult males. One study of 10 hospitalized adult males reported moderate to severe malnutrition on presentation, with patients weighing <80% of ideal body weight on admission [
[5]
]. However, despite being significantly underweight, vital signs in these patients were notable for a normal average heart rate (68.3 beats per minute). Another study of 135 boys and adult males (ages 6–60 years) noted a high mean premorbid percent estimated body weight of 133% [[6]
]. In fact, premorbid obesity has been reported to be a major risk factor among boys with restrictive eating disorders [6
, 7
]. Other characteristics common in males with restrictive eating disorders are less well understood.Unfortunately, the relative lack of data on presenting signs and symptoms may contribute to delays in recognizing males with disordered eating and lead to treatment delays and medical complications, including loss of growth potential and cardiovascular sequelae [
8
, 9
]. In an effort to more clearly illustrate the presenting characteristics of males with restrictive eating disorders, this study will describe presenting characteristics, including age at presentation, as well as their presenting vital signs, including body mass index (BMI), percent of median BMI, temperature, heart rate, and orthostatic blood pressure measurements in 33 adolescent males with disordered eating who presented to an outpatient adolescent and young adult medicine practice between 2011 and 2014. We will also describe relevant historical information, including sports participation, past medical and psychiatric history, and relevant family history, including history of eating disorders.Methods
Eligible patients were all males between the ages of 11 and 25 years presenting to the University of California, San Francisco (UCSF) Adolescent and Young Adult Eating Disorder outpatient clinic for a first visit for restrictive eating between June 1, 2011, and November 1, 2014. These dates coincide with the use of a new electronic medical record for UCSF outpatient clinics. Patients were identified by visit type: New Eating Disorder, and by diagnostic code: “eating disorder” or “malnutrition,” as our institution does not use specific diagnoses for billing, such as anorexia nervosa, bulimia nervosa, and so forth. Patients were excluded if they were determined to have weight loss or malnutrition due to a medical condition other than disordered eating, and as we were interested in restrictive eating disorders, patients were excluded if they met criteria for bulimia nervosa or binge eating disorder.
Charts were reviewed for patients' (1) demographic characteristics (including age and race/ethnicity); (2) clinical characteristics (BMI, percent of median BMI, vital signs at initial visit, including temperature, heart rate, and blood pressure, laboratory results within 30 days of intake visit, and severity of illness as evidenced by need for hospitalization per 2003 Society for Adolescent Health and Medicine admission criteria [
[10]
]); and (3) presence of risk factors for eating disorders (including family history of eating disorders or other mental health disorders, and history of exercise or sports participation [1
, 6
]). Laboratory results were coded as normal or abnormal based on the normal range reported for the individual laboratory, which was included in all laboratory result reports; results were not adjusted based on age. Median BMI for age (in months) and sex was determined by using the 50th percentile BMI per the Center for Disease Control BMI tables; percent of median BMI was calculated as follows: the patient's BMI on presentation/50th percentile BMI for age and sex. Descriptive statistics, including mean, standard, deviation, and range, were used to analyze the data.The authors also retrospectively applied DSM-IV-TR and DSM-5 criteria for eating disorders, given the recent change in diagnostic criteria; charts were reviewed for presence of eating disordered behaviors, body dysmorphia, and mode of weight loss (restricting, purging, and so forth). For the low-weight criterion for DSM-IV-TR diagnosis of anorexia nervosa, we defined low weight as <85% of median BMI. For DSM-5, given the low-weight criterion was necessarily liberalized to reflect patients' overall health status in determining diagnosis, we used a cutoff of <87% of median BMI [
11
, 12
, 13
].Results
Thirty-three patients were included in the analysis; the mean age of patients was 16.0 years (standard deviation [SD] = 2.7, range = 11–23 years); 51.5% of patients were Caucasian, 24.2% Hispanic/Latino, 9.1% Asian, 9.1% black; race/ethnicity was not reported for 6.1% of patients.
The significant clinical and laboratory findings are summarized in Tables 1 and 2. The mean percent of median BMI was 88.8% on presentation; mean percent weight loss was 20.0% from premorbid weight. Patients' mean heart rate while supine was 58.7 bpm (SD = 17.5); mean heart rate change from supine to standing was 22 bpm (SD = 21.6). The mean change in systolic blood pressure from supine to standing was 5.5 mm Hg (SD = 2.1), and mean change in diastolic blood pressure was 7.0 mm Hg (SD = 7.1). The mean temperature was 36.5°C (SD = .5). Of all patients included in this study, 51.5% presented with vital sign abnormalities meeting criteria for hospital admission (bradycardia, orthostatic heart rate change, orthostatic blood pressure change, and/or hypothermia); 39.4% of patients were bradycardic, and 12.1% had orthostatic heart rate changes. The mean duration of illness was 5.8 months. Among patients who presented with heart rates <50 bpm, the mean duration of illness was 6.8 months, compared with 5.2 months among patients whose heart rates on presentation were >50bpm. There were no differences in percent of weight lost between the two groups (19.9% among patients with heart rate <50 bpm vs. 20.0% among patients with heart rate >50 bpm).
Table 1Laboratory findings
| Mean | Standard deviation | Range | % Abnormal | |
|---|---|---|---|---|
| Complete blood count (n = 21) | ||||
| White blood cell count (×109/L) | 5.9 | 1.3 | 4.1–8.5 | 23.8 |
| Hemoglobin (g/dL) | 12.6 | 1.3 | 10.2–15.6 | 23.8 |
| Hematocrit (%) | 39.8 | 4.1 | 31.3–45.5 | 33.3 |
| Platelet count (×109/L) | 206.3 | 70.9 | 115–443 | 19.0 |
| Absolute neutrophil count (×109/L) | 3.1 | 1.1 | 1.6–5.0 | 10.0 |
| Metabolic panel (n = 20) | ||||
| Sodium (mmol/L) | 138.7 | 2.3 | 135–142 | .0 |
| Potassium (mmol/L) | 4.0 | .4 | 3.5–4.8 | 25.0 |
| Chloride (mmol/L) | 104 | 2.2 | 101–108 | .0 |
| Creatinine (mg/dL) | .7 | .2 | .41–1.15 | 20.0 |
| Magnesium (mg/dL) | 2.1 | .2 | 1.8–2.3 | .0 |
| Phosphorus (mg/dL) | 3.9 | .4 | 1.8–4.8 | 5.3 |
| Calcium (mg/dL) | 9.4 | .7 | 8.7–10.2 | 10.0 |
| Thyroid function tests (n = 19) | ||||
| TSH (mIU/L) | 2.1 | 1.4 | .32–6.72 | 10.5 |
| Free T4 (pmol/L) | 11.1 | 1.4 | 2.5–16.0 | 5.6 |
| Transaminases (n = 20) | ||||
| AST (U/L) | 34.1 | 30.8 | 19–158 | 5.3 |
| ALT (U/L) | 35.1 | 39.5 | 7–189 | 40.0 |
| Cholesterol panel (n = 15) | ||||
| Total cholesterol (mg/dL) | 163.5 | 44.9 | 89–231 | 40.0 |
| Triglycerides (mg/dL) | 56.3 | 15.4 | 33–92 | .0 |
ALT = alanine transaminase; AST = aspartate transaminase; Free T4 = free thyroxine; TSH = thyroid stimulating hormone.
Table 2Clinical characteristics
| Vital signs (N = 33 patients) | Mean | Standard deviation | Range |
|---|---|---|---|
| Heart rate (beats per minute) | 58.7 | 17.5 | 33–102 |
| Orthostatic heart rate change (beats per minute) | 22.0 | 21.6 | 0–110 |
| Systolic blood pressure, supine (mm Hg) | 110.0 | 11.4 | 91–127 |
| Systolic blood pressure, standing (mm Hg) | 110.9 | 13.4 | 91–127 |
| Orthostatic systolic blood pressure change (mm Hg) | 5.5 | 2.1 | 1–23 |
| Diastolic blood pressure, supine (mm Hg) | 59.9 | 7.1 | 44–71 |
| Diastolic blood pressure, standing (mm Hg) | 65.9 | 12.3 | 47–93 |
| Orthostatic diastolic blood pressure change (mm Hg) | 7.0 | 7.1 | 0–26 |
| Temperature (C) | 36.5 | .5 | 34.8–37.4 |
| % Median body mass index | 88.8 | 14.1 | 62.3–124.0 |
Of patients with available laboratory data within 30 days of intake, 33% of patients had abnormal hematocrits, 24% had abnormal white blood cell counts, and 10% had abnormal absolute neutrophil counts. Among patients with available metabolic panels, 25% patients had abnormal potassium levels on presentation, and 5% had abnormal phosphorus levels. Aspartate transaminase and alanine transaminase were abnormal in 5.3% and 40% of patients, respectively. Forty percent had elevated total cholesterol levels, whereas none had abnormal triglycerides. Ten percent had abnormal thyroid-stimulating hormone levels, and 5.6% had abnormal thyroxine values.
DSM-IV-TR and DSM-5 diagnostic criteria were retrospectively applied to patients, with an increase in diagnosis of anorexia nervosa from 36.4% to 48.5%. Diagnoses of Eating Disorder Not Otherwise Specified, now Other Specified Feeding or Eating Disorder in DSM-5, decreased from 62.6% to 45.5%. Criteria for Avoidant/Restrictive Food Intake Disorder, a new diagnosis under DSM-5, were met by 6.1% of patients.
Regarding relevant risk factors for disordered eating, 42.4% of patients had a history of overweight or obesity. Fifty percent had a comorbid psychiatric disorder, the most common being depression (27.3%) or anxiety (15.2%). Family history of an eating disorder in a first degree relative was present in 12.1% of patients. While 36.4% had a history of any sports participation, only 12.1% were currently active in sports at the time of intake.
Discussion
The results of this descriptive study suggest that male patients with eating disorders present with clinical laboratory and vital sign derangements from normal despite being only moderately underweight. On average, patients were mildly bradycardic and had orthostatic heart rate changes, and more than half met 2003 Society for Adolescent Health and Medicine admission criteria at intake. Importantly, despite the fact that the average percent of median BMI was relatively high, 88.8%, the mean percent body weight lost was 20.0%, consistent with severe malnutrition. Three quarters of patients meeting admission criteria qualified for hospitalization based on bradycardia alone. The patients with available laboratory data described in this study also had laboratory abnormalities consistent with malnutrition, including anemia, leukopenia, thrombocytopenia, and hypokalemia.
Cardiac complications of anorexia nervosa and malnutrition are well described and include loss of left ventricular muscle mass, impaired contractility, and decreased compliance commonly leading to bradycardia and orthostasis [
14
, 15
]. Similar to previous studies in adult and adolescent males, patients in this study had notable cardiac sequalae related to weight loss, most notably bradycardia and orthostasis. The mean heart rate of our patients was lower than what was previously reported (58.7 vs. 68.3 bpm) [[5]
]. Causes of this difference are unclear and need to be investigated further with a larger sample. With regards to orthostatic changes in vital signs, our patients had a mean change in heart rate of 22 bpm but were notably normotensive and did not have blood pressure changes with change in position, consistent with previous studies reporting positive orthostatic heart rate changes in the absence of blood pressure changes in males [[5]
]. Of note, among patients who presented with heart rates <50 bpm, the mean duration of illness was longer (6.8 months) than patients whose heart rates on presentation were >50 bpm (5.2 months); however, the two groups did not appear to differ in percent of body weight lost.Disorders of hematopoiesis are well described in malnutrition; in severe malnutrition, gelatinous transformation of the bone marrow has been recognized and has also been demonstrated to be reversible with refeeding [
[16]
]. Consistent with findings from studies including adults, our patients with available laboratory data also demonstrated laboratory abnormalities consistent with malnutrition, including anemia, leukopenia, thrombocytopenia, and hypokalemia. Our patients also had lower rates of leukopenia and anemia than other male studies [9
, 17
] and higher rates of thrombocytopenia [[17]
]; the differences between the findings of this study and those of other studies may be related to our small sample size and our inclusion of both adolescent and young adult males. Of note, laboratory results were categorized as normal or abnormal based on laboratory-reported normal ranges.Electrolyte and liver function abnormalities are also well recognized in eating disorders, with findings consistent in our sample. Of note, our patients (25%) have a significantly higher rate of hypokalemia than patients from other studies (4.6%–7%). While vomiting is a known risk factor for hypokalemia, only three patients in our study reported vomiting at the time of presentation. Thus, the hypokalemia in our patients is likely related to their malnutrition. None of our patients had derangements in magnesium on presentation, and only 5% had abnormal phosphorus levels. Notably, 40% of our patients had elevated alanine transaminase levels. This rate is similar to another study including hospitalized males (37%) [
[9]
], though it is difficult to compare the outpatients in this study to hospitalized patients, who likely had more severe illness given their need for hospital admission.Nearly half of the patients in this study had a history of being overweight or obese. More importantly, the mean percent of median BMI of the patients in this study was 88.8%; despite this, patients in this study lost an average of 21.5% of premorbid body weight, consistent with severe malnutrition. Consistent with severe weight loss, these patients were clinically malnourished as evidenced by vital sign and laboratory abnormalities. This is consistent with other studies, which note that higher premorbid weights are more common in males with disordered eating than females [
6
, 18
], which likely contributes to delay in diagnoses.Our findings also reinforce the importance of the recent changes in DSM-5 diagnostic criteria, given the relatively high mean BMI despite significant medical complications of malnutrition in our patients. Consistent with previous studies, using the diagnostic criteria from DSM-5, fewer male patients qualified for a diagnosis of Eating Disorder Not Otherwise Specified (62.6% using DSM-IV-TR to 48.5% using DSM-5), with more meeting criteria for anorexia nervosa (32.4% using DSM-IV-TR to 45.5% using DSM-5). This is largely due to the change in “low-weight” criterion, with the removal of a numerical weight cutoff to qualify for “significantly low weight.” These findings demonstrate the improved ability of DSM-5 criteria in capturing males within specified diagnoses, consistent with the overall goals of DSM-5.
Regarding risk factors for eating disorders, 50% of patients reported a history of psychiatric disorder, with the majority having a history of anxiety or depression; previous studies have demonstrated higher rates of psychiatric comorbidities in patients with eating disorders than the general population. Our sample was also consistent with previously described risk factors for males, with 12.1% of patients in our sample with a family history of eating disorder and 27.3% with a family history of any psychiatric disorder [
[6]
]. Notably, other studies have described sports participation as a risk factor for males; however, only 36.4% of our patients had a history of sports participation, though 75% described that they participated in other forms of physical activity at the time of presentation, usually in the form of school physical education. Sports participation appears to be less of a prevailing risk factor for our patients.Clearly, our study is limited by retrospective design and small sample size, as well as only partial availability of laboratory data. It should be noted that this study is comprised of a clinical, treatment-seeking sample of patients presenting to an outpatient eating disorders program and therefore may not be generalizable to other populations. This study only included outpatients presenting to clinic and as such may not have been as ill as patients admitted or referred directly from an emergency department. Therefore, it is possible that there are some patients who present to care with greater abnormalities than we have found in our sample. Further studies are needed to examine these findings in larger samples. Furthermore, future studies are needed to compare the presenting clinical status of male and female adolescents to further examine similarities and differences between genders.
The results of this study are consistent with previous studies of males with restrictive eating disorders and demonstrate that despite having relatively high percent median body weight, males with restrictive eating disorders can present with abnormalities in vital signs and laboratory values, making early detection and intervention all the more critical.
Acknowledgments
This work was performed at the University of California, San Francisco.
Funding Sources
This study was supported by the Leadership Education in Adolescent Health Training grant T71MC00003 from the Maternal and Child Health Bureau, Health Resources and Services Administration, U.S. Department of Health and Human Services and the Norman Schlossberger Research Fund, Division of Adolescent and Young Adult Medicine, University of California, San Francisco.
Supplementary Data
- Appendix A
References
- Eating disorders in men: An update.J Mens Health Gend. 2005; 2: 186-193
- Burden of eating disorders in 5-13-year-old children in Australia.Med J Aust. 2009; 190: 410-414
- Prevalence and severity of DSM-5 eating disorders in a community cohort of adolescents.Int J Eat Disord. 2014; 47: 610-619
- American Psychiatric Association Diagnostic and statistical manual of mental disorders DSM-5. American Psychiatric Association, Washington, DC2013
- Medical complications in male adolescents with anorexia nervosa.J Adolesc Health. 1995; 16: 448-453
- Eating disorders in males: A report on 135 patients.Am J Psychiatry. 1997; 154: 1127-1132
- Eating disorders in adolescent boys: A review of the adolescent and young adult literature.J Adolesc Health. 2003; 33: 427-435
- Poor catch-up growth in late adolescent boys with eating disorders, weight loss and stunting of growth.Eur Eat Disord Rev. 2013; 21: 395-398
- Medical findings in outpatients with anorexia nervosa.Arch Intern Med. 2005; 165: 561-566
- Eating disorders in adolescents: Position paper of the society for adolescent medicine.J Adolesc Health. 2003; 33: 496-503
- Body mass index cut offs to define thinness in children and adolescents: International survey.BMJ. 2007; 335: 194
- DSM-IV-defined anorexia nervosa versus subthreshold anorexia nervosa (EDNOS-AN).Eur Eat Disord Rev. 2013; 21: 1-7
- Strict versus lenient weight criterion in anorexia nervosa.Eur Eat Disord Rev. 2004; 12: 51-60
- Medical complications of eating disorders in adolescents.Pediatrics. 1988; 8: 613-623
- Cardiac function in adolescents and young adults with anorexia nervosa.J Adolesc Health. 1983; 4: 9-14
- Gelatinous transformation of bone marrow in a patient with severe anorexia nervosa.Int J Hematol. 2013; 97: 157-158
- Medical complications of anorexia nervosa and bulimia.Am J Med. 2015; 129: 30-37
- A review of eating disorders in males.Cur Opin Psychiatry. 2014; 27: 426-430
Article info
Publication history
Published online: June 07, 2016
Accepted:
April 6,
2016
Received:
January 13,
2016
Identification
Copyright
© 2016 Society for Adolescent Health and Medicine. All rights reserved.
