Laboratory evaluation in patients with anorexia nervosa: usefulness and limits

Eating Disorders (ED), anorexia nervosa (AN) in particular, are significant causes of morbidity and mortality. The purpose of this study is to evaluate how laboratory studies can help to diagnose AN and to choose the type of care according to the degree of medical compromission, particularly in primary care.

During the years 2002-2009, in our Eating Disorder Centre, we evaluated 298 ED patients diagnosed by criteria Diagnostic and Statistical Manual of Mental Disorders (DMS-IVR) using anthropometric and laboratory tests.

In our 298 ED patients we identified 264 with a BMI value below 17.5 Kg/m(2) (Anorexia Nervosa patients) and 34 subjects with a BMI higher than 17.5 Kg/m(2), but lower than 20 (Eating Disorder Not Otherwise Specified patients). Sixty percent of the subjects showed normal values with respect to the 38 common laboratory tests used in the clinical setting. In the subgroup of AN patients with more severe malnutrition (BMI lower than 14.15 Kg/m(2)), the percentage of abnormal laboratory values was higher, but always below fifty percent; in 19 laboratory test we found a significant correlation between the BMI value and that of the laboratory test.

The laboratory tests may result as normal even in AN subjects with severe malnutrition and their use without a multi-dimensional evaluation may be misleading, and can even delay the care of patients.