RESUMO
OBJECTIVE: There are few published data on thyroid size and body size variables in children. We aimed to develop an index for normal thyroid size in schoolchildren using ultrasonography and to evaluate the accuracy of clinical examination in the diagnosis of goitre. DESIGN: A study to screen and diagnose goitre. SUBJECTS: A population based sample of 1094 schoolchildren aged 6-14 years (556 boys and 538 girls) were submitted to clinical examination of the thyroid. One hundred and nineteen were considered to have goitre. Ultrasonography was performed in 85 who agreed to participate in the study and in 62 schoolchildren selected randomly from those thought to have no goitre. MEASUREMENTS: Clinical examination and definition of goitre were performed according to WHO criteria. Thyroid volume was measured by ultrasonography (7-5 MHz transducer). RESULTS: The gland volume of 62 schoolchildren without goitre on clinical examination showed a positive correlation with body surface area, age and skinfold thickness. When these factors were entered in multiple regression analysis, only body surface area remained associated with thyroid volume. The upper limit (95th percentile) of the ratio of thyroid volume to body surface area (Ecobody index) was 6.2 ml/m2. Taking this index as the standard criterion, the performance parameters of clinical examination for the defection of goitre in the 1094 schoolchildren were: sensitivity 41%, specificity 91%, positive predictive value 27% and negative predictive value 95%. CONCLUSIONS: Clinical examination of the thyroid is unreliable in establishing the presence of goitre in schoolchildren because it lends to overestimate thyroid size. Nevertheless, it can be used as a screening test. Ultrasonography should be performed when the thyroid gland is thought to be enlarged.
