Variability in developmental timings of the knee in young American children as assessed through Pyle and Hoerr's radiographic atlas.

This study examines the accuracy of the Pyle and Hoerr radiographic atlas technique in an effort to document the extent of normal variation associated with developmental timings in the knee for purposes of age estimation. The atlas has been previously tested; however, accuracy rates were produced from a dataset, which spread in age from mostly 7-16 years. This study took a closer look at the younger age groups, examining radiographs from 297 children (147 female and 150 male) from birth to 6 years. Standard deviations representing the difference between the skeletal and chronological age were calculated according to two groupings. Each group represents episodes, or time periods, of differential developmental rates as expressed through the number of plates within the atlas dedicated to documenting each year of life. The beginning year of life is characterized by the most rapid of development as represented by the numerous image plates used to depict this time period. Individuals assigned to plates with a skeletal age between birth and 1 year were grouped collectively to document the variation associated with such rapidly changing morphology (SD = 2.5 months in female children; 2.3 months in male children). Years 1-3.8 years (female) and 1-4.5 years (male) were represented by two or three images within the atlas, and therefore, individuals assigned to plates with a skeletal age falling within this range were placed within a second grouping (SD = 5.2 months in female children; 7.0 months in male children). As expected, variation was observed to decrease as developmental processes accelerated in the younger children. The newly calculated standard deviations offer tighter predictions for estimating age in young children while at the same time maintaining an acceptable width that accounts for normal variation in developmental timings.