Statistical analysis

The statistical analysis of data was performed with Statistical Package for Social Science software, version 21 (SPSS Inc., Chicago, Ill, USA). Qualitative data were described with counts and percentages. Associations between categorical variables was tested with the chisquared test. For parametric data, continuous variables were presented as means and standard deviations (mean ±SD). The Kolmogorov-Smirnov test was used to test the normality of quantitative data. The ADC values were normally distributed, whereas the Tc-99m uptake levels and TRAb, TSH, FT3, and FT4 levels were non-normal. For parametric data, Student’s t tests and one-way analysis of variance (ANOVA) were used. Mann-Whitney U tests and Kruskal-Wallis tests were used to compare non-parametric variables. ROC (Receiver Operating Characteristic) curves were drawn to detect optimal cut-off ADC values for differentiation between patients with thyrotoxicosis and controls. This ADC value, Tc-99m uptake levels, and TRAb levels discriminated between Graves’ disease and painless thyroiditis with the best accuracy. P value of £0.05 was considered to be statistically significant. The Spearman’s coefficient was used to evaluate the degree of correlation between the ADC value of the thyroid gland, Tc-99m uptake levels, and laboratory tests (TRAb, TSH, FT4, and FT3).