Body impedance and thyroid function
associated with its prediction from RI (22%) and bw
(24%). The TSH-R relationship is given in Figure 1.
After the contribution of R to TSH was taken into
account, anthropometric variables were not able to
explain any additional part of TSH variance (p=NS,
stepwise regression; data not shown).
The limitations of this study should be kept in mind.
It was performed in a convenience sample that is
not representative of the general population. For
instance, the average BMI of our subjects was 21
kg/m2, which is substantially lower than the aver-
age national BMI. Since there is some evidence that
the relationship between anthropometry and thy-
roid size may be different in obese subjects (3), fu-
ture studies should investigate this relationship al-
so in overweight (and underweight) subjects. It is
also possible that sample-related characteristics
may be responsible for the great association be-
tween BIA and TSH in our study. However, this
seems unlikely in view of the fact that body com-
position relationships in homogenous samples such
as that employed in our study, are commonly less
strong than in heterogeneous samples and this is
especially true for BIA (6, 16).
In conclusion, this study of healthy subjects of both
sexes shows that bioelectrical resistance may be a
better indicator of thyroid function than anthro-
pometry, possibly because of its more direct rela-
tionship with fat-free tissues. Further studies are
needed to test whether this relationship held in un-
der- and over-weight subjects. The question whe-
ther FFM is a better predictor of TSH than bw
awaits a final answer from the employment of ref-
erence models for the assessment of FFM.
DISCUSSION
Thyroid size is known to be related to anthropo-
metric dimensions and this relationship may partly
explain the differences in thyroid volume observed
between sexes (1). However, this correlation is on-
ly low to moderate and even if it is useful to work
out reference values for thyroid size (1, 15), there
is a need for more accurate predictors of thyroid
volume and, most importantly, function. As this lat-
ter is concerned, it is to be noted that the correla-
tion existing between thyroid size and TSH in
healthy subjects is too low [r=-0.26, p=0.001,
n=296 (1)] for thyroid volume to be considered a
good proxy of thyroid function. FFM may be a bet-
ter “denominator” for thyroid function than weight
because it comprises the most metabolically active
tissues of the body (2). R offers a qualitative as-
sessment of fat-free tissues because of its inverse
relationship with TBW, which is the major con-
stituent of FFM (5, 12). This simple use of BIA
avoids the problems connected with the popula-
tion-specificity of quantitative algorithms and the
fact that many of them include bw among the pre-
dictors (3).
In this study of healthy subjects, we found that R
was a better predictor of TSH than anthropometry.
We studied healthy subjects because we were in-
terested to know whether a physiological relation-
ship existed between TSH and R. Moreover, since
hyperthyroid subjects have often suppressed lev-
els of TSH, the study of the TSH-R relationship as a
continuous one would have been difficult in con-
ditions of thyroid disease [besides the fact that
physiological relationships between thyroid func-
tion and fat-free tissues may be lost during thyroid
disease (7)]. The relationship between TSH and R
was an inverse one (Fig. 1) and since R is inversely
proportional to FFM, this relationship can be in-
terpreted as showing a direct relationship between
TSH and FFM, which is biologically sound. It is of
interest however that R was a better predictor of
TSH than RI. Since RI is generally a better predictor
of FFM than R (5, 16), this suggests that factors oth-
er than the relationship of R with body composi-
tion may explain the greater power of R in esti-
mating thyroid function as compared to RI and an-
thropometry.
REFERENCES
1. Gomez J.M., Maravall F.J., Gomez N., Guma A., Soler J.
Determinants of thyroid volume as measured by ultra-
sonography in healthy adults randomly selected. Clin.
Endocrinol. 2000, 53: 629-634.
2. Severi S., Malavolti M., Battistini N., Bedogni G. Some ap-
plications of indirect calorimetry to Sports Medicine. Acta
Diabetol. 2001, 38: 23-26.
3. Wesche M.F., Wiersinga W.M., Smits N.J. Lean body mass
as a determinant of thyroid size. Clin. Endocrinol. 1998,
48: 701-706.
4. Heymsfield S.B., Wang Z.M., Withers R.T. Multicomponent
molecular level models of body composition analysis. In:
Roche A.F., Heymsfield S.B., Lohman T.G., Human body
composition. Human Kinetics, Springfield, 1996, p. 129-
147.
5. Deurenberg P. International consensus conference on
impedance in body composition. Age Nutr. 1994; 5: 142-
145.
6. Guo S.S., Chumlea W.C., Cockram D.B. Use of statistical
methods to estimate body composition. Am. J. Clin. Nutr.
1996, 64 (Suppl.): 428S-435S.
7. Edmonds C.J., Smith T. Total body potassium in relation to
thyroid hormones and hyperthyroidism. Clin. Sci 1981, 60:
311-318.
622