ABSTRACT
For the elaboration of a computer programme for the diagnosis of thyroid diseases the pattern of the results of laboratory tests was studied after dividing 125 hyperthyroid patients into different clinical entities and further also according to the severity of the hyperfunction of the thyroid gland. Comparison of variables was made between different clinical entities and, within these, between different grades of severity. All these groups were compared with the controls. Furthermore, the correlation between the results of laboratory tests was studied. The patients were divided into the following groups: toxic diffuse goitre (TDG; 50 cases), toxic multinodular goitre (TNG; 52 cases) and toxic adenoma (TA; 23 cases), and according to clinical severity of hyperfunction of the thyroid gland, into the following grades: »borderline» (I; 16 cases), mild (II; 35 cases) and moderate to severe (III; 74 cases). The laboratory tests studied were: protein-bound iodine in the serum (PBI), uptake of triiodothyronine by Sephadex (T3U), »free thyroxine index» (T3I), free proportionate and absolute thyroxine (PFT4, AFT4), serum cholesterol, serum and urinary hydroxyproline (HOP), glucose-6-phosphate dehydrogenase activity in red blood cells (G-6-PD), 2.5 h and 24 h thyroid uptake of radioiodine, 24 and 48 h urinary excretion, protein-bound radioactive iodine in the blood (PBRI) and the 24 h conversion ratio (CR). There was a statistically significant difference between TDG and TNG with regard to the thyroid uptake of radioiodine and the free thyroxine, the mean values being lower in the latter group of patients. The TA group differed significantly from the combined TDG-TNG group, the mean values deviating towards the normal range. The TA group did not differ from the controls with regard to PFT4, cholesterol and E0–24. This group evidently represented a separate population.
In the TDG and TNG groups the laboratory data correlated well with the clinical assessment of severity. In most cases the differences were statistically significant. Grade I, however, differed from the controls only with respect to the variables of the hormone concentration in the blood.
The variables of the measurement of hormone concentration in the blood had a strong linear correlation (r = 0.6—0.9) to each other and to the 2.5 h uptake. A significant correlation (r = 0.5—0.6) was also found between the variables mentioned and the urinary HOP and G-6-PD. The correlation between HOP and G-6-PD was of the same order.
The fact that the pattern of PFT4—AFT4 and T3U—T3I differed may indicate variations in the ratios between triiodothyronine and thyroxine in different clinical entities of hyperthyroidism.