Q & A: Constitutionally Thin’

Q: A patient of mine who is quite thin (BMI: 16.3) insists that she doesn’t have an eating disorder, but rather that she is “constitutionally thin.” What is constitutional thinness, and how can it be distinguished from anorexia nervosa (AN) when a patient is so thin? (E.J., WV).

A: Constitutional thinness (CT) has long been recognized as a clinical entity, separate from AN. It is usually characterized by a low and stable body mass index (BMI, or kg/m2) without any hormonal abnormality. Some research studies comparing individuals with CT and AN have limited samples to individuals whose BMIs were under 16.5, so you can appreciate that this is truly a thin population.

In one recent study, CT individuals differed from those with AN in several ways. For example, compared to patients with AN, CT individuals had normal levels of free T3, IGF-I, and leptin, confirming the absence of under-nutrition in CT. Their psychological profiles revealed a desire to gain weight. Their food intake (7,565 +/- 908 kJ/day) was similar to that of controls (7,961 +/- 1,452 kJ/day) and higher than in those with AN (4,894 +/- 703 kJ/day). However, an increased energy expenditure to fat-free mass ratio differentiated CT from controls and could account for the resistance to weight gain observed in CT (Am J Physiol Endocrinol Metab 2007; 292:E132-7. Epub 2006 Aug 15).

In other studies, levels and ratios of ghrelin and obestatin (a pro-grehlin hormone that inhibits grehlin’s appetite stimulation) differed between individuals with CT and AN (Psychoneuroendocrinology 2009; 34:413. Epub 2008 Nov 7), and yet other research also confirmed opposite concentrations of orexigenic and antiorexigenic hormones (such as peptide YY, glucagon-like peptide 1, ghrelin, and leptin) that could be used to differentiate CT and AN.

That said, other research clearly shows that being constitutionally thin is not without risk. One recent study showed that bone mineral density in the lumbar spine and in total were similar in CT and AN individuals, in both cases less than in a control group, leading the authors to hypothesize that anthropometric parameters could contribute more significantly than estrogen deficiency in the achievement of peak bone mass in AN patients (Br J Nutr 2009; Mar 23:1-6. [Epub ahead of print]).

The second study confirmed lower bone density and decreased “breaking strength” in the bones of women with CT and AN compared to controls. However, the same study showed that markers of bone turnover were similar in women with CT and controls, greater than those in patients with AN (J Clin Endocrinol Metab 2008; 93:110). 

The important point for clinicians is to skillfully distinguish individuals who claim to be constitutionally thin but who are in fact masking an eating disorder from those who truly have constitutional thinness. A hormonal checkup should routinely be part of that differential diagnostic workup.

–J.Y.