Ultrasound assessment of pubertal breast development in girls: intra- and interobserver agreement

Background Clinical assessment of pubertal breast development using Tanner staging is subjective. This has led to the introduction of ultrasound (US), aiming for a more objective analysis. However, information regarding its reliability is lacking. Objective To examine intra- and interobserver agreement of breast maturity staging using US and to examine the precision of direct measurements of the gland. Materials and methods Fifty-seven healthy girls (mean age: 10.9 years, range: 6.1 to 15.9 years) were examined independently by two observers using US of the left breast to score the glandular maturity stage on a 0-5 scale, and to measure the depth and diameter. One observer repeated the examination after 20 to 35 min to assess intra-observer agreement. Cohen's kappa with linear weights was used to examine intra- and interobserver agreement of the US staging, while the measurement precision was analyzed using Bland-Altman plots and 95% limits of agreement. Results The agreement of US staging on a 0-5 scale was very good (kappa 0.84; 95% confidence interval [CI] 0.78-0.91) for intra-observer observation and good (kappa 0.71; 95% CI 0.62-0.80) for interobserver observation. Measurements of glandular depth and diameter were unbiased for a single observer, but the variances were large both within and between observers. Conclusion US using a scale from 0 to 5 is a reliable method to stage the development of glandular breast tissue during puberty in healthy girls. Measurements of glandular depth and diameter were found to be imprecise.