SHOX haploinsufficiency and Leri-Weill dyschondrosteosis: prevalence and growth failure in relation to mutation, sex, and degree of wrist deformity.

SHOX mutations causing haploinsufficiency were reported in Leri-Weill dyschondrosteosis (LWD), which is characterized by mesomelic short stature and Madelung deformity of the wrists. The aim of this study was to determine the prevalence of SHOX mutations in LWD and to investigate the degree of growth failure in relation to mutation, sex, age of menarche, and wrist deformity. We studied 20 families with 24 affected children (18 females) and nine affected parents (seven females). All patients presented with bilateral Madelung deformity and shortening of the limbs. Height, sitting height, parental height, birth length, age of menarche, and presence of minor abnormalities were recorded. The degree of Madelung deformity was estimated by analysis of left hand radiographs. Microsatellite typing of the SHOX locus was used for detection of SHOX deletions and PCR direct sequencing for the detection of SHOX point mutations. In 14 of 20 families (70%), SHOX mutations were detected, with seven deletions (four de novo) and seven point mutations (one de novo). The latter included five missense mutations of the SHOX homeodomain, one nonsense mutation (E102X) truncating the whole homeodomain, and one point mutation (X293R) causing a C-terminal elongation of SHOX. Median age of the affected children was 13.4 yr (range, 6.1-18.3), mean height sd score (SDS) (sd in parentheses) was -2.85 (1.04), and mean sitting height/height ratio SDS was +3.06 (1.09). Mean birth length SDS was -0.59 (1.26). Growth failure occurred before school age. Height change during a median follow-up of 7.4 yr (range, 2.3-11.3) was insignificant with a mean change in height SDS of -0.10 (0.52). Mean height SDS of affected parents was -2.70 (0.85) vs. -0.91 (1.10) in unaffected parents. Height loss due to LWD was estimated calculating delta height defined by actual height SDS minus target height SDS of the unaffected parent(s). In the children, mean delta height SDS was -2.16 (1.06), the loss being greater in girls at -2.30 (1.02) than in boys at -1.72 (1.09) (P = 0.32). In patients with SHOX deletions, it was -2.14 (1.15) vs. -1.67 (0.73) for the SHOX point mutation group (P = 0.38). Mean delta height SDS was -2.26 (0.68) for the girls with early menarche (<12 yr) vs. -2.08 (0.91) for the other postmenarcheal girls (P = 0.72). Height loss in patients with radiologically severe wrist deformities in comparison with those having milder radiological signs was -2.81 (1.01) vs. -1.70 (1.04) (P = 0.03). GH treatment in five children during a median duration of 3.4 yr (range, 1.5-9.8 yr) with a median dosage of 0.23 mg/kg.wk (range, 0.14-0.25) resulted in a mean height SDS gain of +0.82 (0.34). In conclusion, SHOX defects were the main cause of LWD. Growth failure occurred during the first years of life with a mean height loss of 2.16 SDS whereas pubertal growth may only be mildly or not affected. Children with a severe degree of wrist deformity were significantly shorter than those with mild deformities. No statistically significant effects of type of mutation, age of menarche, or sex on height were observed. The effect of GH therapy varied between individuals and needs to be examined in controlled studies.