[Accuracy of the methods for height prediction in patients treated for congenital primary hypothyroidism]

The objective of this study was to determine how accurately three methods of height prediction (the Bayley-Pinneau, Tanner-Whitehouse Mark 1 and Roche-Wainer-Thissen methods) estimate adult height in a group of 17 girls and 7 boys treated for congenital primary hypothyroidism.

The patients were diagnosed at a mean chronological age of 1.2 years. Their thyroxine treatment dose ranged between 3.1 and 8.6 micrograms/kg/day. Height predictions at various chronological ages were retrospectively compared to final adult height. The first prediction was made at a chronological age of 7.3(1.1) years, corresponding to the sixth year of treatment. Heights were expressed in centimeters and standard deviation scores (SDS) and related to Tanner standardized curves.

The final heights reached were significantly higher than the target heights (-0.5 (0.7) SDS vs -1.1 (0.9) SDS, p < 0.01). The three methods accurately estimated adult height. Tanner and Bayley-Pinneau height predictions were similar. The Roche method over-predicted height at all chronological ages with a mean error which ranged between +2 (3.2) centimeters and +2.8 (2.1) centimeters.

1) The patients who were diagnosed and treated early for congenital primary hypothyroidism reach an adult height in the normal range, which is probably related to the total recovery of their retarded bone age. 2) In this pathological condition, the Bayley-Pinneau and Tanner-Whitehouse Mark 1 are the most reliable methods of height prediction.