Triiodothyronine (T3) and metabolic rate in adolescents with eating disorders: Is there a correlation?

AIM To examine the correlation between T3 and resting energy expenditure (REE) in adolescent patients with eating disorders (ED) to assess whether T3 can be used to predict metabolic rate suppression and recovery. METHODS A retrospective chart review was performed on patients with ED (Anorexia Nervosa [AN], Bulimia Nervosa [BN], and Eating Disorder NOS [EDNOS]), aged 11-22 years, who had T3 and REE measured within 1 month (N=38 AN, 32 BN/EDNOS). REE was measured by indirect calorimetry (IC) and represented as the percentage of expected REE (%EREE) predicted by the Harris-Benedict equation. Pearson correlation coefficients were calculated to examine the relationship between T3 and %EREE and how each correlates with anthropometric data, laboratory values, and diagnosis. RESULTS T3 was significantly correlated with %EREE in the AN group but not in the total population or BN/EDNOS group. In the total study population, T3 alone correlated significantly with weight, Body Mass Index (BMI), BMI percentile, %Ideal Body Weight (%IBW), %Maximum Weight Lost (%MWL), LH, and estradiol. In the AN group, T3 and %EREE both correlated with BMI, BMI percentile, LH, and estradiol; however, only T3 correlated with %IBW and %MWL. In the BN/EDNOS group, T3 correlated with BMI, BMI percentile, %IBW, and estradiol while %EREE correlated with none. CONCLUSION In patients with AN, T3 correlated significantly with markers of malnutrition and %EREE and may serve as a surrogate measure when IC is unavailable. Following T3 during treatment of AN may assist clinicians in assessing metabolic suppression and recovery and help guide caloric prescriptions and goal weights.