|Home | About | Journals | Submit | Contact Us | Français|
In their recent article, Rappold et al1 investigated the presence of SHOX defects in a large cohort of 1608 children of short stature. Of the total number of SHOX mutations/deletions identified, 58% were found in 55 children with Leri–Weill dyschondrosteosis (LWD) and 2.2% in 1534 cases considered to have idiopathic short stature. The authors created an evidence‐based scoring system based on the clinical features of 68 patients with SHOX defects to identify the most appropriate children for SHOX gene testing. The following criteria were used: arm span:height ratio <96.5%, sitting height:height ratio >55.5%, body mass index >50th centile and the presence of cubitus valgus, short forearm, bowing of the forearm, appearance of muscular hypertrophy and/or dislocation of the ulna. This scoring system had some limitations, such as a low positive predictive value (11%) when using the lower cutoff (score of 4) and a lower sensitivity (61%) when using the upper score (score of 7, out of a maximum of 24).
To select those likely to have mutations in the SHOX gene from a population of children with short stature, previous studies have already suggested a limbs:trunk ratio, ((calculated subischial leg length + arm span)/sitting height)2 and sitting height:height ratio (SH:H), expressed as standard deviation score (SDS) for age and sex.3 Rappold et al.1 analyzed the SH:H ratio as absolute values, even though their cohort represented a wide age range, and age is known to strongly influence this ratio.4
It would be useful if Rappold et al. were to report the limbs:trunk ratio proposed by Binder et al.2 and SH:H ratio expressed as SDS4 in this large cohort of patients with SHOX mutations. These parameters could also improve the proposed score system.
Competing interests: None declared.