Body Composition Changes in Anorexia Nervosa: A Review

Laurel Mayer, MD, Columbia University and The New York State Psychiatric Institute, New York, NY
Reprinted from Eating Disorders Review
January/February 2003 Volume 13, Number 1
©2002 Gürze Books

One of the cardinal symptoms of anorexia nervosa (AN) is the fear of gaining weight and becoming fat (DSM-IV, criteria B). With near-delusional conviction, patients tell us that if they gain weight, it will be “all fat and no muscle.” Another common complaint during treatment is that weight gain isn’t being evenly distributed, but is collecting “all in my stomach.”

Historically, we have labeled these concerns as persistent distortions, and used them as evidence of the need for continued treatment. Evidence that these concerns may be valid, however, is beginning to surface. This article will provide a brief review of body composition changes in women with AN.

Disturbance of Body Composition

In the underweight state, many studies have documented that body composition is severely disturbed.1,2,3 Although patients may say they “feel huge,” they do not have a disproportionate amount of fat. In fact, total body fat, as well as total body muscle, is severely decreased, consistent with the starved, underweight state of AN.

Surprisingly, however, few studies have systematically examined changes in body composition with weight normalization. There is growing evidence that patients may not gain weight in an evenly distributed pattern. There may be a tendency to deposit a disproportionate amount of fat in the trunk (abdominal) region compared to the extremities (arms and legs).

What Studies Have Shown

Forbes was the first to examine body fat distribution in a cross-section of women with anorexia nervosa.4 Using the waist-to-hip ratio (WHR) as a measure of body fat distribution, he reported no difference in the WHR between women with AN and healthy controls.

Mayo-Smith and colleagues used computerized axial tomography to measure subcutaneous fat (the layer of fat under the skin) and visceral abdominal fat (the fat that surrounds internal abdominal organs, stomach intestines, liver, kidneys, etc.) in three populations. These were patients with AN, patients with Cushing’s syndrome (a state of abnormally high cortisol levels associated with truncal obesity), and normal controls.5 Although limited by the cross-sectional nature of the study, the authors described a five-fold decrease in subcutaneous abdominal fat and only a two-fold decrease in visceral fat in low-weight patients with AN compared to controls.

The authors also reported that the proportion of visceral fat to total fat is significantly larger in patients with AN (0.4) compared to controls (0.14). As visceral fat is concentrated in the abdominal region, the authors suggest that patients with AN show a central concentration of body fat.

Zamboni and colleagues used computerized tomography (CT) to measure changes in the subcutaneous and visceral fat compartments of the abdominal region (more specifically at the L4-L5 level) with weight gain, and extended the findings of Mayo-Smith.6 They replicated the finding of increased visceral compared to subcutaneous fat at low weight in patients with AN, and reported that with a mean 7.3-kg weight gain, subcutaneous fat increased by 212%, and visceral fat increased by 117%. Not surprisingly, they concluded that patients with AN gain abdominal fat. Whether there is a distinct preference to deposit weight centrally cannot be addressed by this study because they did not measure body fat in other regions (i.e., the extremities). Another limitation of this study is that, despite an approximately7-kg weight gain, patients were still significantly underweight (BMI: 17.5+2.0 kg/m2) at the time of retesting. It is possible that changes that occur during the process of refeeding may resolve with normal weight.

Orphanidou et al. used anthropometry (calipers) and dual x-ray absorptiometry (DEXA) to measure changes in body composition and body fat distribution in 26 women undergoing treatment for anorexia nervosa.7 Changes measured by anthropometry supported greater deposition of fat in the central regions than in the extremities. However, the changes in regional fat mass as measured by DEXA did not confirm the anthropometric findings. There is still no satisfactory explanation for this .

Iketani and colleagues used DEXA to assess changes in body fat distribution with weight gain, and extended the Orphanidou study by including a group of normal-weight control women.8 Consistent with previous reports, at low weight, patients had reduced body fat mass. With weight gain, trunk mass increased to levels similar to controls, but extremity fat remained significantly below control values. These authors, too, argue for a disproportionate increase in abdominal fat compared to the extremities during weight gain. However, as in the Zamboni study, patients remained significantly underweight at the second assessment, and thus the distribution of body fat after complete weight recovery cannot be addressed.

An Increase in Truncal Fat

Most recently, Grinspoon et al studied body composition changes in a group of low-weight patients with AN undergoing outpatient treatment and controls.9 (See article on page 4). At baseline, weight, total percent body fat and extremity fat as a percentage of total fat were reduced in patients compared to controls. Truncal fat as a percentage of total fat, however, was not statistically different between the two groups. When patients gained weight, truncal fat as a percentage of total fat increased but percent extremity fat did not. As in previous studies, patients were still underweight at the end of this 9-month trial, leaving uncertain what body fat distribution would be with more complete weight restoration. Regardless, this report of an increase in trunk fat relative to extremity fat contributes to the growing evidence of a central distribution of body fat during weight gain.

What Our Data Revealed

At the recent annual American Psychiatric Association meeting in New Orleans, our group presented body composition data from 20 patients with anorexia nervosa, before and after weight gain, and 20 healthy controls. Patients were all inpatients on the Eating Disorders Unit of the General Clinical Research Unit at the New York State Psychiatric Institute/Columbia University. Data on controls were drawn from the database of healthy subjects studied at the Body Composition Unit at St. Luke’s-Roosevelt Hospital (BCU), New York, NY, where the patients were also studied, and were matched to weight-restored patients for body mass index (BMI).

Body composition was measured by anthropometry and DEXA (Lunar). Patients were assessed within two weeks of admission (prior to the onset of formal weight gain), and again within 2 weeks after reaching a target weight of 90% IBW (1959 Metropolitan Life tables). All subjects were free from medications for a minimum of 2 weeks prior to testing.

As expected, all body compartments were reduced in underweight patients compared to controls. Body weight was significantly lower, as was total percent body fat and lean body tissue. After weight gain, all compartments significantly increased, and those in patients were not significantly different from controls. Patients’ body circumferences (mid-arm, mid-thigh, waist and hip) increased relatively uniformly with weight gain (approximately 20%). However, the waist and hip circumferences of the weight-restored patients were significantly larger than controls, and extremity circumferences were significantly smaller in weight-restored patients compared to controls (Figures 1 and 2).

Attempting to replicate and extend the body fat distribution findings of Grinspoon et al, we examined the changes in trunk and extremity fat with weight normalization, using the regional fat measurements obtained by DEXA.9 At low weight, extremity fat as a percentage of total fat was 52.8%+10, and trunk fat as a percentage of total fat was 40.7%+8.3. Similar to Grinspoon’s findings, percent trunk fat increased with weight normalization. Additionally, whereas Grinspoon’s group reported no change in percent extremity fat, our data revealed a decrease in percent extremity fat.

Body fat is deposited preferentially to the trunk. Although these findings are preliminary, they are consistent with the accumulating evidence that patients with anorexia nervosa may demonstrate an abnormal distribution of body fat (lipodystrophy) that preferentially deposits fat to the trunk and away from the periphery. This study extends the previous study by examining subjects after they regained to 90% of a normal weight.

Future studies would include a larger sample size and a more sophisticated assessment of body fat distribution by total body magnetic resonance imaging. Additional research might also explore whether these changes are predominantly a short-term, post-weight-gain phenomenon or persist with longer-term recovery and maintenance of normal weight.

A link to heart disease

Significant research links increased visceral fat with risk of heart disease.10,11 Traditionally, patients are counseled that the medical complications of starvation resolve with weight restoration.

Additional lines of inquiry might also include the exploration of how these biological changes track with psychological changes. One could easily imagine that patients whose worst fears are realized (“It’s all going to my stomach,” “I look fat,” “I look pregnant”) might have a more difficult time normalizing their weight and/or preventing relapse. This information might help guide therapeutic interventions.


  1. Krahn DD, Rock C, Dechert RE, et al. Changes in resting energy expenditure and body composition in anorexia nervosa patients during refeeding. J Am Diet Assoc 1993;93(4):434.
  2. Probst M, Goris M, Vandereycken W, Van Coppenolle H. Body composition in female anorexia nervosa patients.Br J Nutr 1996;76(5):639.
  3. Dempsey DT, Crosby LO, Lusk E,et al.Total body water and total body potassium in anorexia nervosa. Am J Clin Nutr 1984;40(2):260.
  4. Forbes GB. The abdomen: hip ratio: normative data and observations on selected patients. Int J Obes 1990;14(2):149.
  5. Mayo-Smith W, Hayes CW, Biller BM, et al. Body fat distribution measured with CT: correlations in healthy subjects, patients with anorexia nervosa, and patients with Cushing syndrome. Radiology 1989; 170(2):515.
  6. Zamboni M, Armellini F, Turcato E, et al. Body fat distribution before and after weight gain in anorexia nervosa. Int J Obes Relat Metab Disord 1997;21(1):33.
  7. Orphanidou CI, McCargar LJ, Birmingham CL, et al. Changes in body composition and fat distribution after short-term weight gain in patients with anorexia nervosa. Am J Clin Nutr 1997; 65: 1034.
  8. Iketani T, Kiriike N, Nagata T, et al. Altered body fat distribution after recovery of weight in patients with anorexia nervosa. Int J Eat Disord 1999; 26(3): 275.
  9. Grinspoon S, Thomas L, Miller K, et al. Changes in regional fat distribution and the effects of estrogen during spontaneous weight gain in women with anorexia nervosa. Am J Clin Nutr 2001; 73: 865.
  10. Lapidus L, Bengtsson C, Hallstrom T, et al. Obesity, adipose tissue distribution and health in women—results from a population study in Gothenburg, Sweden. Appetite. 1989;13(1):25.
  11. Bjorntorp P. Abdominal fat distribution and the metabolic syndrome. J Cardiovasc Pharmacol 1992;20 Suppl 8:S26. Review.
No Comments Yet

Comments are closed