Comparative Evaluation of the Efficacy of Hand-Wrist and Cervical Vertebrae Radiography for the Determination of Skeletal Age

BACKGROUND Prediction of skeletal growth is necessary for growth modification and surgical orthodontic treatments and is usually done by assessing skeletal maturity indicators in hand-wrist radiographs. The use of growth stages of cervical vertebrae in lateral cephalograms has been suggested to avoid overexposure. OBJECTIVES This study seeks to assess the degree of agreement between hand-wrist and cervical vertebrae maturation stages for skeletal age determination and prediction of the peak growth spurt (PGS). PATIENTS AND METHODS This cross-sectional study was conducted with 67 boys and 66 girls between 8 and 18 years of age, divided into 11 age groups; 266 hand-wrist radiographs and lateral cephalograms were obtained and analyzed. Hand-wrist maturation stages were evaluated according to the Grave and Brown, Bjork system (stages 1 - 9). The cervical vertebral maturation stage (CVMS) was determined on lateral cephalograms based on a system described by Baccetti et al. (CVMS 1-5). To apply the Cohen's kappa index, the stages of growth were reduced to 5 intervals (A - E) to relate the 5 CVMS to the 9 stages of Bjork hand-wrist analysis. RESULTS In all age groups, the skeletal maturity stages of the hand and wrist bones and the cervical vertebrae of the girls were ahead of the boys. Cohen's kappa test revealed a low level of agreement between the two methods [Kappa (95% CI) = 0.312 (0.290 - 0.377)]; concordance was slightly higher in males (K = 0.33 for males versus 0.27 for females). Evaluation of concordance coefficients between the stages determined by the two methods indicated the highest concordance in 8- and 9-year-olds and the lowest in 12- and 14-year-olds. The level of agreement between the two methods was only acceptable in 8- and 9-year-olds of both genders and 10-year-old boys. The level of agreement between the two methods in other age groups was not acceptable. CONCLUSION The level of agreement between the two methods was low; thus, they cannot be used alternatively to estimate patients' skeletal age or to predict the PGS. This may be due to the effect of different maturation levels (influenced by the environment, ethnicity, and gender) on the agreement between methods for skeletal age determination.