Changes in Plasma Concentrations of Leptin and Body Fat Composition during Weight Restoration in Anorexia Nervosa

John F. Morgan, Jim Bolton, Philip M. Sedgwick, Sanjeev Patel and J. Hubert Lacey

St. George’s Hospital Medical School London SW17 0RE, United Kingdom Gerard S. Conway Department of Medicine, University College London SW1 8AA, United Kingdom

Recently, Eckert et al. (1) presented longitudinal data confirming previous cross-sectional reports (2, 3) that leptin concentrations in female anorectics were significantly lower than normal weight controls, increased significantly on refeeding, and correlated linearly with body mass index. They also suggested an uncoupling of leptin and adiposity at lowest body weights. However, they acknowledged that "a major limitation of this study is the lack of data on percent body fat," as body mass index does not necessarily correlate with percent body fat. Similarly the cross-sectional reports (2, 3) estimated percent body fat using bioimpedance. The latter assumes uniform hydration, which may not be true in refeeding anorectics (4). To our knowledge, no previous study has longitudinally correlated leptin against total body fat during weight restoration in anorexia nervosa, except by such crude means of body fat estimation.

Like Eckert et al., we carried out a longitudinal study of the relationship between leptin and total body fat. However, we employed dual-energy X-ray absorpitometry (DXA) to measure body fat directly (5), thus avoiding the pitfalls of body mass index or bioimpedance. We monitored plasma leptin concentrations, body mass index, and percent body fat longitudinally in seven female patients with anorexia nervosa (mean age 20.4 yr, SD 4.76, range 16–27 yr) during weight restoration. Patients met DSM-IV criteria for anorexia nervosa. Mean body mass index (BMI) before weight restoration was 15.8 kg/m2 (SD 1.51; range 13.8–17.2 kg/m2) and after weight restoration was 19.8 kg/m2 (SD 0.71; range 19.2–20.9 kg/m2). Plasma was obtained from patients between 0900 and 1000, after a standardized meal containing 250 kcal, in the context of standardized daily intake of between 2100 and 2600 kcal. Serum leptin was measured by RIA using recombinant human leptin as standard with intra- and interassay coefficients of variation of no more than 5%.

The results are shown in TableGo 1. Spearman correlation coefficients, with their associated two-tailed significance levels, were calculated between percentage changes in leptin and other variables. The level of statistical significance was set at 5%. There was a highly significant, positive correlation between percentage changes in leptin and total body fat (r = 0.89; P = 0.007) and also between percentage changes in leptin and BMI (r = 0.86; P = 0.014).


View this table:
[in this window]
[in a new window]
 
Table 1. Measurement of serum leptin and total body fat

 
Using a more precise estimate of percent body fat we have confirmed the core finding of Eckert et al. (1), that changes in leptin with weight gain in anorexia nervosa are consistent with the normal physiological response in nonanorectics. However, we suggest that the apparent uncoupling of leptin from body weight at lowest weights may be explained by the limits of BMI or bioimpedance as estimates of percent body fat, rather than necessarily demonstrating a threshold effect. Further longitudinal studies are required to clarify this and should employ DXA to measure percent body fat.

Footnotes

Received August 4, 1998. Address correspondence to: John F. Morgan, Department of General Psychiatry, St. George’s Hospital Medical School, London SW17 0RE, United Kingdom (

References

  1. Eckert ED, Pomeroy C, Raymond N, Kohler PF, Thuras P, Bowers CY. 1998 Leptin in anorexia nervosa. J Clin Endocrinol Metab. 83:791–795.[Abstract/Free Full Text]
  2. Hebebrand J, Van der Heyden J, Devos R, et al. 1995 Plasma concentrations of obese protein in anorexia nervosa. (Letter to the Editor). Lancet. 346:1624–1625.
  3. Grinspoon S, Gulick T, Askari H, et al. 1996 Serum leptin levels in women with anorexia nervosa. J Clin Endocrinol. Metab. 81:3861–3863.
  4. Lukaski HC. 1987 Methods for the assessment of human body composition: traditional and new. Am J Clin Nutr. 46:537–556.[Abstract]
  5. Hannan WJ, Cowen SJ, Freeman CP, et al. 1993 Evaluation of dual-energy X-ray absorpitometry for the assessment of body composition in anorexic females. In: Ellis KJ, Eastman JD, eds. Human Body Composition. New York: Plenum Press.