Many patients with TS may not be the shortest child in kindergarten but will have a significant decrease in linear growth rate by third or fourth grade. Some present only when the normal pubertal growth spurt fails to occur.[20
Our patients had a mean age at onset of diagnosis of 8.8 ± 3.0 years and mean CA at the onset of therapy of 12.7 ± 2.4 years, with a mean delay of treatment 3.9 ± 2.4 years. It is easy to misinterpret the absence of puberty and small size of these patients as due to constitutional delay.[20
Turner patients showed some degree of disproportion where their US/LS ratio SDS was 1.8 ± 1.8 , which may give an evidence for an abnormality of the skeletal and/or growth plate as a cause of short stature. Being a type of skeletal dysplasia was previously suggested by Bridges and Brook;[21
] however, X-rays lacked convincing evidences for skeletal dysplasia.[22
GH is approved for use in children with TS in most industrialized countries and is recommended in the recently issued guidelines.[23
] Numerous studies have reported that recombinant human growth hormone (rhGH), with or without anabolic steroids, can accelerate growth and lead to a height greater than that predicted in girls with TS.[24
] In one cohort, the mean height of girls who completed a mean of 7.6 years of rhGH therapy (n
= 17) was 150.4 cm, a gain of 8.4 cm over the expected average height.
The Canadian randomized control trial was published recently. They randomly assigned 154 children to treatment with GH or no treatment and both the groups were observed until the adult height was reached. They were able to follow-up 61 of 76 treated patients and 43 of 78 controls to adult height and observed a 7.2 cm difference between the two groups, with a 95% confidence interval of 6.0–8.4 cm. The mean duration of treatment was 5.7 years, and puberty was induced pharmacologically at a chronological age of 13 years.[28
Growth response of our cases to GH showed that height SDS improved from –4.3 to –3.3 using Tanner standard and from –1 to 0.2 using Turner standard .
The patient’s height became much closer to the target height as itchanged from 32.8 to 17 cm (3.5 to 2.6). Estimated mature height in TS improved from 144.6 to 150.7 cm .
The growth velocity in TS decreased from 6.1 cm (2 and 2.1 SDS using Tanner and Turner standards, respectively) to 4.9 cm (0.8 and 1.9 SDS using Tanner and Turner standards, respectively) .
Delta changes of 20 patients with TS, followed up for 2 years, showed a significant difference for target height – patient’s height in centimeters (P = 0.002) .
In the first year, the most important predictive factor was BA delay (negative correlation), which explained 61% of the variability in response with an error SD of 1.9 cm. For the second year, growth velocity of the first year was the strongest predictive factor (positive correlation), followed by target height SDS (negative correlation) and IGF-1 SDS (positive correlation). For the third year, growth velocity of the preceding years was the most important predictive factor (positive correlation). Predictive factors in the second and third years of GH therapy explained all of the variability in the response. So, it is apparent that during the period of catch up growth in girls with TS, BA delay was the most important predictor; the more the delay, the better is the growth response, giving a potential chance for better growth. During the period of stable growth, the growth velocity in the preceding year is the most important predictor; the higher the GV in the preceding year, the better is the response.
The growth responses of 686 girls with TS were determined during the first4 years of GH therapy. For the first year, GH dose was the most important predictor (positive correlation). Age, weight SDS, additional treatment with oxandrolone, the difference between the height SDS and mid-parental height SDS and frequency of injections were other predictors. These variables explained 46% of the variability of the response with an error SD of 1.26 cm. Height velocity in the preceding year was the most important predictor in the second to fourth year, with GH dose, age, weight SDS and addition of oxandrolone to treatment as the other predictors.[29
] It was further reported that although the accuracy of prediction in all4 years was high, indicated by the low error SDS, the predictive power was relatively low.[30
] In contrast to this result, we reported a high predictive power in the second and third years where the predictive factors explained all the variability in response to GH treatment in TS. In disagreement with this study, GH dose (dose/week), frequency of GH injections and birth weight were not included as variables in the present work. Also, none of our patients received oxanodrolone therapy. Otherwise, predictive variables were similar. In the present study, we had only one girl with TS who reached FAH; she had a height of 143.5 cm (3.1 SDS). A final adult height that varies between 147.7 ± 5.6 and 152.3 ± 5.3 has been reported in girls with TS. Monitoring of IGF-1 and IGF-BP3 levels in girls with TS during the initial 3 years of GH therapy showed a significant rise of IGF-1 values in the first 2 years, whereas the IGF-BP3 values showed a significant rise in the first year followed by a significant drop in the second year. The significant rise of growth factors IGF-1 and its binding protein-3 during early period of GH treatment, which is followed by their decline, may be related to the initial catch up growth which is followed by a steady pattern on the continuation of GH treatment.
In conclusion, this study seems to show that GH therapy provides a satisfactory auxological result. Bone age delay is to be considered as a predictive factor that may negatively influence the effect of GH therapy on final height. The growth velocity in the preceding year is the most important predictor of GH therapy response.
These observations help to guide rhGH prescription to reduce the risks and costs.