Discordance between knee and hand-wrist bone age in adolescents: implications for skeletal maturity assessment and its association with thyroid-stimulating hormone levels.

BACKGROUND

Hand-wrist radiographs may not fully reflect skeletal bone age (BA) during mid-to-late puberty. Knee magnetic resonance imaging (MRI), a non-radiative imaging technique, may provide detailed evaluation of the lower limbs.

OBJECTIVE

To validate the San Diego MRI-based knee BA atlas in a Chinese adolescent cohort presenting with growth deceleration, evaluate the concordance between knee and hand-wrist BA, and explore factors influencing potential discrepancies.

MATERIALS AND METHODS

This study involved 227 adolescents. Hand-wrist BA was assessed using the Greulich and Pyle atlas with hand-wrist radiographs, and knee BA using the San Diego atlas with knee MRI. Correlations between knee BA, hand-wrist BA, and chronological age (CA) were analyzed. Discrepancies between the two sites were further evaluated for associated clinical factors.

RESULTS

Knee BA correlated better with CA (r = 0.816, P < 0.001) than with hand-wrist BA (r = 0.763, P < 0.001). The mean difference between hand-wrist BA and CA was 1.08 ± 1.19 years, and between knee BA and CA was 0.71 ± 1.06 years (P < 0.001). 14.1% of adolescents showed more advanced knee maturity than hand-wrist bones. Thyroid-stimulating hormone (TSH) levels were positively associated with the difference between hand-wrist and knee BA (r = -0.198, P = 0.011). Subclinical hypothyroidism significantly increased the risk of knee overmaturity (OR = 6.022, P < 0.001).

CONCLUSION

Knee MRI is a reliable imaging modality for assessing skeletal maturity in adolescents with growth retardation. A moderate correlation was observed between hand-wrist and knee BA. In adolescents with elevated TSH levels, knee MRI may be preferable to hand-wrist radiographs for predicting final height, evaluating growth potential, and guiding growth hormone therapy decisions.