Comparison of bone age readings by pediatric endocrinologists and pediatric radiologists using two bone age atlases

Abstract

Background: Pediatric endocrinologists at a children's hospital were concerned that bone age readings by radiologists frequently differed from their readings by >1 year.

Objective: To compare consistency of bone age readings by endocrinologists vs. radiologists, and the Greulich and Pyle (GP) atlas vs. the recently published Gilsanz and Ratib (GR) digital atlas.

Materials and methods: Sixteen radiographs were read by seven radiologists and five endocrinologists, using both the GP and GR atlases. Time to complete all 16 readings was recorded. An intraclass correlation coefficient (ICC) was used to assess the level of agreement between the two disciplines and the use of different atlases.

Results: The ICCs for the seven radiologists and five endocrinologists were both 0.95; the ICC for readings done by GP vs. GR for all raters was 0.96. The number of outlier readings was similar for radiologists and endocrinologists but was higher for GR than for GP. Radiologists spent less time (1.15 min) reading each bone age than endocrinologists (2.4 min).

Conclusion: Under controlled conditions, pediatric endocrinologists and radiologists displayed nearly identical consistencies in reading bone ages. The GR atlas performed similarly to the GP atlas, except for an increased number of outliers, and might be suitable as a replacement for GP.