Riboflavin Deficiency in Adolescent Girls with Anorexia Nervosa

Reprinted from Eating Disorders Review
July/August 1999 Volume 10, Number 4
©1999 Gürze Books

Riboflavin is essential for energy production, good vision, and healthy skin and mouth tissue. Malnutrition from anorexia nervosa sets in motion a complex chain of events that culminate in riboflavin deficiency.

Riboflavin deficiency has been previously reported in anorexia nervosa patients.1-3 Most recently, C.D. Capo-chichi and colleagues investigated the effects of malnutrition associated with anorexia nervosa and concomitant low thyroid hormone levels on erythrocyte and plasma riboflavin metabolism and urinary excretion of organic acids.4

Riboflavin’s normal pathways

Normally, riboflavin is converted to 2 coenzymes, flavin mononucleotide (FMN) and flavin adenine dinucleotide (FAD) in tissues. Riboflavin kinase and FMN adenyltransferase catalyze these conversions to FMN and FAD. FAD and FMN are stored in cells bound to flavoenzymes, and FAD is the predominant storage molecule. Only a small fraction of riboflavin is stored as free riboflavin in cells. The liver is the major site of riboflavin storage, and contains approximately one-third of total body flavins. The expression of the enzyme riboflavin kinase is regulated by triilodothyronine (T3). Thyroid dysfunction, resulting in hypometabolism, is a well -known metabolic adaptation in anorexia nervosa.

Study design

Thyroid hormones, riboflavin, riboflavin cofactors, and urinary organic acids (as indicators of riboflavin function) were evaluated in 17 adolescent girls with anorexia nervosa (mean age: 16 years; mean BMI 14.8 [10.2-16.3]), and a control group of 17 healthy control adolescent girls (mean age: 13 years; mean BMI 20.5[17.1-25.1]) Refeeding of the anorexia nervosa (AN) group was being accomplished with an ad libitum diet, in which carbohydrate, lipid, and protein provided up to 51%, 40%, and 16% of calories, respectively. Riboflavin intake was 2.5 mg/day throughout the study. Caloric intake on days 0 (day of admission), 15, and 30 was 1625, 2000, and 3196 calories, respectively. Four patients received supplemental nighttime enteral feedings.

Fasting plasma thyroid hormones, erythrocyte and plasma FAD, FMN and riboflavin concentrations and urinary organic acids were assessed from blood and urine samples collected on day 0, and days 15 and 30 of the refeeding period for the AN group and one fasting sample from the controls.

Results: refeeding produced higher riboflavin levels

Compared to controls, the AN refeeding group had higher erythrocyte riboflavin levels (3.5 vs.<0.1nmol/mol hemoglobin; P<0.001), lower plasma FAD (57.8 vs.78.5 nmol/L; P<0.05), and higher urinary ethylmalonic acid levels (7.12 vs. 1.3 micromol/mmol creatinine; P<0.05). At admission, T3 concentrations were low in the AN group and negatively correlated with plasma riboflavin concentrations (r = -0.69; P< 0.01), but became positively correlated (r= 0.59; P<0.05) after 30 days of refeeding.

The authors hypothesize that the insufficient conversion of riboflavin into flavoenzymes could be a result of energy depletion and a decrease in riboflavin kinase activity. Since T3 regulates the expression of riboflavin kinase, a decrease in T3 and resting metabolic rate could be responsible for a decrease in riboflavin kinase biosynthesis. This would decrease the production of FMN and FAD from new riboflavin entering cells, resulting in atypical storage of free riboflavin in the cytoplasm of cells. In addition to the accumulation of riboflavin in the erythrocytes and lower plasma FAD concentrations in the AN group, a rise in the urinary excretion of organic acids was observed. These acids essentially serve as biomarkers of riboflavin-dependent enzyme systems in hepatocytes, and the increased urinary levels that were observed suggested that the riboflavin abnormalities were having an effect on biochemical functions.

The authors conclude that the low T3 concentrations seen in anorexia nervosa could alter riboflavin metabolism, resulting in elevated erythrocyte riboflavin concentrations, low plasma FAD levels, and elevated levels of urinary ethylmalonic acid and isovalerylglycine excretion. These findings have clinical significance because this suggests that an improvement in energy balance, and thus overall nutritional status (rather than simply providing the vitamin), may be necessary to correct the vitamin problem and to restore normal metabolic processes in anorexia nervosa.

— Tami J. Lyon, MPH, RD, CDE


  1. Rock CL, Vasantharajan S. Vitamin status of eating disorder patients: relationship to clinical indices and effect of treatment. Int J Eat Disord 1995;18:257.
  2. Rock CL, Hunt IF, Swenseid ME, Yager J. Nutritional status and bone mineral density in patients with eating disorders. Am J Clin Nutr 1987;46:527.
  3. Philipp E, Pirke KM, Seidl M, et al. Vitamin status in patients with anorexia nervosa and bulimia nervosa. Int J Eat Disord 1998; 8:209.
  4. Capo-chichi CD, Gueant JL, Lefebvre E, et al. Riboflavin and riboflavin-derived cofactors in adolescent girls with anorexia nervosa. Am J Clin Nutr 1999; 69:672.
No Comments Yet

Comments are closed