Validity of the use of a few hand-wrist bones for assessing bone age

Bone age is widely used as an osseous maturation method to assess biological development in clinical and auxological studies. Numerical methods for calculating bone age have better replicability; however, they require a wide data manipulation. The aim of this study was to evaluate the accuracy of bone age estimation by using just a few ossification centers. In 205 hand and wrist radiographs of children and adolescents, aged 0.9-17.4 years old (111 males and 94 females), bone age was determined by two trained observers employing the five-bone (B5) and the TW2 methods. To compare the results of the two methods, the mean differences by age and sex were tested by the Mann-Whitney test. The relationship of the bone age distribution estimated by B5 and TW2 was calculated and the mean of the bone age determined by the two methods was plotted by age and by differences between bone age calculated by the two methods +/- 2 SD. The bone age determined by B5 was between 1.2 and 16.8, and by TW2 was between 1.2 and 18.0 years. The mean differences between B5 and TW2 (-0.06 +/- 0.6) were not significant (p > 0.05). The distribution of mean differences by age and method demonstrated that all deviations were encompassed into +/- 2 SD with no particular bias. In general terms, a good agreement was obtained between these two methods.