Abstract 494

Background: Often in patients with growth disorders, arm span is used to assess long-bone growth versus mixed spine and long-bone growth in height. This is normally considered a 1:1 ratio. We examine this relationship in boys with constitutional delay and boys and girls with idiopathic short stature

Methods: Patients were seen in an endocrinology clinic in a large children's hospital between 1992 and 1998 and diagnosed with either constitutional delay or familial or idiopathic short stature. Inclusion criteria were all encounters during which a simultaneous arm-span and height were measured. Outliers were removed in an automated manner. Cross-sectional data of height versus arm-span deviation from height was plotted. A linear model was fit to the cross-sectional distribution using the least-squares method, and slope was compared for significance against zero slope, using a T statistic.

Results: Records were reviewed from 155 boys diagnosed with constitutional delay in 226 encounters, and 208 boys and 122 girls diagnosed with only familial or idiopathic short stature in 286 and 169 encounters respectively. The regression line for boys with constitutional delay was y = 0.075 x - 9.16, where x was height in cm and y was arm-span minus height in cm, and is shown in the figure. Arm-span equalled height when height was 122.4 cm; below this, arm-span was wider than height. The regression line had a slope statistically significantly different from zero (p < 0.001). There was no significant difference in the model when the data was partitioned by pubertal status. Boys and girls with familial or idiopathic short stature also had regression lines with slopes statistically significantly different than zero (y = 0.036 x - 3.9 and y = 0.071 x - 8.45, respectively).

Fig 1
figure 1

NO caption available

Conclusion: In boys with constitutional delay and boys and girls with idiopathic short stature, arm-span is wider than height at shorter heights, and more narrow than height at taller heights.