Growth hormone significantly increases the adult height of children with idiopathic short stature: comparison of subgroups and benefit

Of the 123 children with idiopathic short stature treated with rhGH (0.32 ± 0.03 mg/kg/week – 6 days per week), from the late 80’s to 2012, 88 attained adult or near adult height, 68 males and 20 females. Twenty-seven of the children were lost to follow-up (22 males out of 98, 22%, and 5 females out of 25, 20%). Eight males with NFSS and delayed puberty, who were treated with depo-testosterone with 50 mg a month for 6 months or 100 mg for 3 months to induce pubertal changes, were not included in the results reported, to avoid questions in the interpretation of results, even though it is known that testosterone increases the growth velocity but does not improve adult height [57, 58], if any it may decrease the adult height if treatment is prolonged or with higher doses, by advancing bone age (123 children -27-8 = 88).

Of the 68 males, forty-seven were classified as NFSS and 21 (30%) as FSS; 16 females were classified as NFSS and 4 as FSS. The groups were further divided into those with normal puberty and delayed puberty. The duration of treatment was 5.2 ± 2.7 years for the 68 males and 3.5 ± 0.9 for the 20 females, with a range of 2 to 8 years.

All the children met the criteria for ISS with the height below -2.0 SD (except for 1 female who was -1.81), normal stimulated growth hormone levels, normal birth weight and length, and absence of comorbid conditions.

The age at start of treatment ranged between 4.7 years and 16 years, 11.9 ± 3.3 years for the 68 males and 12 ± 1.9 years for the 20 females. The children were seen every 3 months and the rhGH dose was adjusted at each visit. The beginning dose of GH in the late 80’s was 0.22 mg/kg/week 3 times a week, but the dose was increased to 0.3 mg/kg/week or more, because the response was not as good as expected. With the normal weight gain the dose would decrease to 0.28 mg/kg/week and adjustments were made. The dose ranged from 0.28 to 0.4 mg/kg/week with adjustments, for a calculated average dose, throughout the years, of 0.32 ± 0.03 mg/kg/week. The dose was not increased for puberty.

A number of determinations were obtained before the onset of treatment; bone age, CBC, sedimentation rate, chemistry panel, anti-endomysial antibodies, tissue transglutaminase antibodies, glucose, hemoglobin A_1C, TSH, free T₄, IGF-I, IGFBP-3 and cortisol; and most of them annually. An MRI of the brain was obtained at the time of the growth hormone stimulation testing on many of them in the past. This is no longer felt needed. Karyotypes were obtained in many. Bone age was determined by Greulich and Pyle standards and adult height predicted by Bayley-Pinneau tables.

Heights were measured by a wall-mounted stadiometer. The heights at the onset of treatment were expressed as SDS for chronologic age. The height SDS was calculated in the usual manner: height of the subject in cm minus the mean height for age or adults divided by the SD in cm from the mean. Adult heights were obtained at 19 ± 2.45 years, range 15.7 to 27.53, for males and at 18.3 ± 2.34 years, range 15 to 24 years, for females. Sixty-seven of the adult heights were obtained by us (76%) and 9 in a doctor’s office (10%). Twelve (14%) were obtained at home, following detailed instructions and were consistent with predicted heights. The last bone age available in our records was 16.0 ± 1.0 years for males and 14.1 ± 0.46 for females, but some adult heights were obtained later. To obtain final adult heights, the potential remaining growth of some children whose heights were obtained before closure of the epiphyses (near adult height) was calculated. These numbers for calculated final adult heights were not included in the numbers reported and it will be addressed later (Additional file 1: Table S1).

Pubertal development was assessed by the method of Tanner. Males with testicular volume of less than 4.0 ml by the age of 14.0 years and females with no breast development by the age of 13 years were classified as delayed onset of puberty. Children whose heights were below the midparental height SDS target range of -1.6 SD were classified as NFSS and those within the target range as FSS. Children whose fathers or mothers were below -2.0 SDS were also classified as FSS. Of the 25 subjects (males and females) with FSS, 9 (36%) had a father or a mother with a height below -2 SD (-2.25 to -2.87 SD). The classification of delayed onset of puberty includes the previously used definition of constitutional delay of growth and puberty (CDGP) in accordance with the international consensus [45].

Target height (MPH) was calculated from the self-reported parental heights by the method of Tanner: (height of the father + height of the mother + 13)/2 for the males and (height of the father + height of the mother – 13)/2 for females and expressed in cm [47]. No addition for secular trend was needed, since there was no or minimal increase in secular trend in the United States National Health Statistics between 1977 and 2000. Midparental height SDS was calculated by the following equation: (father’s height SDS + mother’s height SDS)/√2(1 + r (M,F); r is the correlation between the parent’s height which is 0.3 [50]. The SDS was based on the US National Center Health Statistics of 1977. The values for MPHs SDSs were not different than those calculated by target height, by the method of Tanner, when corrected for assortative mating (correlation between the parent’s heights). The characteristics of NFSS and FSS children treated to adult height are illustrated in Table 1.

Twenty-two males (22% of the total of 98) were lost to follow-up: 6 were treated for less than 6 months and 16 were lost at 10 to 14.4 years of age. Five females (20% of the total of 25) were lost to follow-up: 2 were treated for less than 6 months and 3 were lost at 9.4 to 12.5 years of age. All had a good response to rhGH treatment. An analysis of intent to treat to adult height was conducted.

We definitely observed the variability in response. The AH gain of all the 88 children (males and females) was +1.9 SDS (0.76 SD) with a range for +0.29 to +4.13 SDS, quite a range, (95% CI 1.71 to 2.06). AH was -0.71 SDS (0.74 SD), that would give a range, ± 2 SD, of -2.19 SD to -0.77 SD (page 12). Range of Ahs could be seen in Additional file 3: Figure S2a, b, c, and Additional file 4: Figure S3a and b.

For the 68 males (non-familial and familial short stature), the AH gain was 1.86 SDS (0.69 SD) with a range of +0.29 to 3.92 SDS (95% CI, 1.69 to 2.03). The AH was -0.72 SDS (0.72 SD) giving a range of ± 2 SD of -2.16 to +0.72 SDS.

The AH gain for the 47 males with NFSS was 2.04 SDS (0.64 SD) with a range of 0.94 to 3.92 SDS. The AH gain for the 21 males with FSS was 1.45 SDS (0.62 SD) with a range of 0.29 to 2.40 SDS (p = 0.001).

The AH was -0.54 SDS (0.64 SD), for a range ± 2 SD of -1.82 to +0.74 SDS for the 47 males with NFSS. For the 21 males with FSS the AH was -1.14 SDS (0.72 SD) for a range of ± 2 SD of -2.58 to +0.3 SDS (p = 0.003).

The comparison of the values for different measures obtained for all the children with normal puberty and delayed puberty, males 68 and 20 females, in all the subgroups, NFSS and FSS, males and females are shown in Table 2. There was no difference in any of the values for the baseline height, MPH, father’s height, mother’s height, adult height, and gain in height, for any of the subgroups, for normal and delayed puberty cohorts; all the p values were more than 0.05. There was a difference, however, between NFSS and FSS for normal and delayed puberty.

In the study of Rekers-Mombarg, et al. [48] comprising 132 children with ISS, the children declined gradually in growth from a length of -0.8 SDS in boys and -1.3 SDS in girls at birth, to a height of -2.7 SDS at 16 years in boys and 13 years in girls, and increasing to a mean of -1.5 SDS in boys and -1.6 SDS in girls. The gain in SDS from childhood height to adult height (adult height gain) varied from 0.1 SDS (1.22 SD) at 3 years to 0.1 SDS (0.60 SD) at 14 years, and to 1.2 for boys at 16 years and 1.1 for girls at 13 years.

It became of interest to know if the changes that we observed in adult height and adult height gain of our treated subjects were somewhat related to age and not to treatment, even though we did not see differences in our children with normal puberty (usually younger) and those with delayed puberty (usually older). The effect of age on adult height and adult height gain in males with NFSS and FSS (for whom we had adequate numbers) is shown in Table 5. There was a significant difference in the age of children with NFSS with normal puberty (8.6 (2.7 SD) years) and delayed puberty (14.4 (1.0 SD) years) (p < 0.001), but there was no difference in the adult height or adult height gain (p >0.6 and >0.3, respectively).

Similarly, there was a significant difference in the age of children with FSS with normal puberty (10.3 (2.47 SD) years) and delayed puberty (14.1 (1.31 SD) years) (p < 0.001), but no difference in the adult height and adult height gain (p, 0.729 and 0.949, respectively).

In the 13 young females with NFSS, the adult height and adult height gain was the same as in young males with NFSS.

We correlated AH gain and age at the start of treatment and AH gain and duration of treatment in our 47 NFSS males and 21 FSS males, and there was no significant correlation (correlations shown in supplemental graphs in Additional file 5: Figure S4.

The gain in height is very variable for individuals and averages are not accurate or useful to predict the benefit that an individual will obtain. Based on the averages, Table 5, a male with delayed puberty and NFSS with treatment for 3.5 years could gain 2.11 SDS (16 cm in the USA). The gain in height depends on the pubertal growth, benefit of growth hormone, and bone age. In the aforementioned case, Table 5, depending on the age, bone age, progress of bone age, pubertal growth, and growth hormone response, the gain in height could range from 1.3 SDS to 3.9 SDS (10 to 29 cm). The only way that we could inform the subjects of the benefit that he or she could obtain, is the way that we do it now. Based on the height and the bone age, the predicted adult height is obtained. The subject can be informed that based on our experience, growth hormone should be of benefit to him or her, to improve the adult height. The benefit could be 5 cm, 7.5, 10 cm or more but cannot be predicted exactly, because it depends on the response to treatment, progress of bone age, and years of treatment. Then the provider and the subject decide whether treatment is continued to attain the maximum height or ended when he or she is satisfied with the height.

Children were observed for a period of time prior to treatment, so that their growth rates could be assessed. In a retrospective review we could know the children who were lost to follow up and, consequently, analyze their growth rates prior to and during treatment.

Growth rates determined for less than 6 months were not included. The growth rates in centimeters per year and the change in the height SDS, prior to treatment, on treatment prior to puberty, and on treatment during the first two years of puberty were not different for treated children lost to follow-up and for those treated to adult height, (p >0.05) (Figure 3 and Table 6). Thus, there was no bias on the reported effect of GH on children treated to adult height.

Complaints of myalgia or arthralgia of the legs early in the treatment, which promptly subsided without adjustment of the GH dose, were very rare. Two of the 123 children had a low serum TSH and free T₄ (subclinical hypothyroidism) and were treated with levothyroxine, which was discontinued after cessation of treatment.

There were no other side effects.

Concerns have been raised on the possibility of side effects in the future from high levels of IGF-1 because of its mitogenic effect [16, 59]. Therefore, it became of interest to know the levels of IGF-1 with treatment. The levels of IGF-1 prior to treatment, during treatment for different stages of puberty, and after treatment are illustrated in Figure 4 and Table 7. Except for a few, all the values were within the range expected for the pubertal stage. Only 8.4% of the values (20 out of 237) (19 out of 198, 9.6% for the males and 2.5%, 1 out of 39, for the females) exceeded the normal range at one time during the treatment. These values are less than those reported in patients with growth hormone deficiency treated with 0.24 mg/kg/week. Seventeen percent of the values were higher than 2 SD after 2 years of treatment [60].

One of the limitations of our study is the lack of controls. Fortunately, there are, presently, a number of randomized and nonrandomized controlled studies that would permit assessment of the benefit, by comparisons of the adult height attained and AH gain in our treated children with those of historical untreated controls.

The adult height and adult height gain (AH minus Baseline height) are measurements that the investigators obtain and should be more accurate than comparisons based on attainments of PAH or MPH (see later). MPH was not used to calculate benefit of treatment.

The adult height gain corrects for baseline differences in the different studies in treated subjects and controls, provides information on the benefit of treatment, and permits comparison of groups not matched for baseline heights.

The comparison of adult heights provides also information on the benefit, and when the baselines heights are not different or the AHs are corrected for baseline heights differences, yields the same results as the adult height gain.

The reports on SDS permit comparisons of different populations and calculation of the benefit in centimeters based on the centimeters for SD of adults in a particular population. In this presentation we used 6.75 cm for males and 6.14 cm for females for SD of adult height, the numbers from the US National Health Statistics of 1977. The benefit in centimeters would be different in different populations depending on the centimeters for SD of adults.

In Figure 6 are similar comparisons for the adult height gain (mean SDS and 95% CI) for the controls and for our treated subjects. The numbers for the different studies for the baseline, adult height and adult height gain of controls and our treated subjects are in Table 8.A glance at the Figures 5 and 6 clearly shows that there is a significant benefit from treatment.

The benefit from treatment for the adult height for our children (-0.71 SDS) versus the AH (-2.16 SDS) of the controls of the 9 studies, Table 8, was + 1.45 SDS with a range of +1.17 SDS to +1.69 SDS (an average of 9.8 cm for males with a range of 7.9 to 11.4 cm and an average of 8.9 cm for females with a range of 7.2 to 10.3 cm).

The analysis of the adult height gain, Table 8, showed somewhat similar results. The gain of controls in the 9 studies was, on the average, +0.49 SDS (with a range of 0.18 to 0.8 SDS). In our treated children the gain was on the average +1.90 SDS for a difference with controls of +1.41 SDS, range +1.1 to +1.72 SDS (9.5 cm for males with a range of 7.4 to 11.6 cm and 8.6 cm for females, with a range of 6.7 to 10.5 cm).

The average baseline height of the 9 published studies, Table 8, was -2.66 SDS, not different than in our study -2.61 (±0.62) SDS. The baseline of 5 of the 9 studies was not different than ours p >0.1 or >0.5; in two it was higher and in two lower.

In the study of Rekers-Mombarg et al. [48] on the height outcomes of untreated children with ISS, the adult height of 48 NFSS boys of -1.74 SDS (0.91 SD) was 8.3 cm (95% CI, 7.1 to 9.5 cm), less than their target height. The adult height of our 47 treated NFSS boys of -0.54 SDS (0.64 SD) was equal to the target height, suggesting that our treated boys gained 8.3 cm. Actually, the adult height of our 47 NFSS boys of -0.54 SDS (0.64 SD) (95% CI, -0.35 to -0.73) was 1.2 SDS, 8.1 cm (6.8 to 9.3 cm) higher than the adult height of the untreated boys, indicating the benefit obtained by treatment. This benefit of 8.1 cm (6.8 to 9.3 cm) is similar to the benefit found by comparison with other studies and adds confidence in the results of our comparisons.

Table 9 shows the benefit of different doses of growth hormone on the adult height. In published studies the benefit obtained in adult height was usually less than 4 cm over controls with doses of less than 0.3 mg/kg/week. And it was more than 5 cm (6.8, 7.4, 7.5) with doses of 0.3 or more mg/kg/week.

In our study, the benefit with a dose of 0.32 (±0.03 SD) mg/kg/week was 9.8 cm for males and 8.9 cm for females, about 5 cm to 6 cm more than the 4 cm in published studies treated with less than 0.3 mg/kg/week, but, in the range, only 1.5 or 1.35 cm more than in those treated with 0.3 or more mg/kg/week.

There have been only 2 studies comparing NFSS and FSS, the one by Wit et al. [9] and by Albertsson-Wikland et al. [3], Table 10.

The benefit in adult height gain (that corrects for differences in baseline heights) of our treated children were higher than the adult height gain of the controls in the published studies by 1.26 SDS and 1.56 SDS for the NFSS (8.5 and 10.5 cm for males and 7.7 and 9.5 cm for females) and 1.21 SDS and 1.6 SDS for the FSS (8.1 to 10.8 cm for males and 7.4 to 9.8 cm for females); an equal benefit for children with NFSS and FSS.

Even though the benefit with treatment was the same for children with NFSS and FSS, our 25 children with FSS were significantly lower in adult height and adult height gain than our 63 children with NFSS, Table 11; adult height of -1.08 SDS in FSS versus -0.56 SDS in NFSS (p <0.01) and adult height gain 1.51 in FSS versus 2.06 in NFSS (p < 0.001).

Of interest is that the results in the only two studies that have been published are similar, Table 11. The adult height gain of the untreated controls and of the treated subjects were significantly less in the children with FSS than in NFSS (p <0.05). In other words, the adult height gain of treated children with NFSS was higher than in FSS, but also higher was the adult height gain of the NFSS than the FSS controls, so the benefit was the same. This was also the conclusion of Dahlgren J [12] in his analysis of these 2 studies.

This observation seems to be consistent in the three different studies and has implications for interpretation of results with treatment, in studies where the number of children with FSS in the controls and treated subjects is not known, and in the selection of controls for non-randomized or randomized studies.

The reported impression, by review of published studies, is that the benefit of treatment of children with ISS is less than in other conditions for which GH has been licensed [20, 21, 61].

Our results compare well with those obtained in GH treated children in other conditions for which GH has been approved, growth hormone deficiency (GHD), small for gestational age (SGA), or Turner syndrome.

The adult height of our 88 children with ISS of -0.71 SDS (0.74 SD) is equal to the adult height of -0.7 SDS (1.2 SD) achieved in 121 children (males and females) with GHD treated with 0.3 mg/kg/week, 3 or 6 times a week, reported by Blethem et al. [62].

In a review of four randomized controlled trials (RCTs) comprising 391 children with SGA treated with growth hormone (range of 0.23 to 0.47 mg/kg/week) [63], the adult height exceeded controls by 0.9 SDS. The adult height gain was 1.25 SDS (1.5 SDS in treated minus 0.25 SDS in untreated subjects). There was no difference between the 2 dose regimens. The adult height gain in our study exceeded controls by 1.41 SDS (1.90 SDS minus 0.49 SDS in controls), Table 8.

Our results of an increase of 7.1 to 10.3 cm of adult height in treated females over controls (Table 8) compares favorably with those obtained in Turner syndrome. In 61 patients with Turner syndrome [64], treated with 0.3 to 0.375 mg/kg/week of growth hormone, the final height was approximately seven cm higher (95% CI, of 6 to 8 cm) than in 43 untreated control patients. Despite this increase, the adult height of treated patient with Turner syndrome was still outside the normal range.

Our observations in 88 children clearly show a significant benefit of GH treatment (0.32 ± 0.03 mg/kg/week) when compared to 305 untreated controls in 9 different studies, a benefit based on adult height gain (that corrects for baseline differences) of 9.5 cm (7.4 to 11.6) for males and 8.6 cm (6.7 to 10.5) for females (Table 8).

The benefit in the adult height of our treated subjects over the adult height of published historical controls was 9.8 cm for males and 8.9 cm for females (Table 9). This benefit is higher by 5 or 6 cm than the benefit obtained in subjects treated with 0.16 or 0.16 to 0.28 mg/kg/week of GH, usually of less than 4 cm. The benefit was only 1.35 or 1.5 cm higher than that obtained in other studies using 0.3 mg/kg or more per week.

This dose dependent effect has been reported previously in studies using 0.3 mg/kg or more of GH a week [65]. A dose effect for children with ISS was shown by Wit et al. [66] with a mean adult height gain of 1.85 SDS (similar to ours) and a benefit of 7.2 cm with a dose of 0.37 mg/kg/week. Also in a randomized controlled study by McCaughey et al. [2], 8 girls (6 of them with FSS) treated with GH 0.35 mg/kg/week (range 0.4 to 0.31), were 7.5 cm taller than the controls; and by others: Buchlis et al. [8] 6.8 cm for females, Albertsson-Wikland et al. [3] 8 cm based on adult height gain, Hintz R et al. [12] 9.2 cm higher for boys and 5.7 cm for girls, and Kemp et al. [67] (Genentech registry) 1.7 to 2.0 SDS in AH gain in children treated for 7 years, results similar to ours.

In view of the aforementioned, it is possible that the stated modest benefit of 4 cm with the treatment of ISS is related to the low dose of GH used and to averaging the results without taking in consideration the difference in the benefit from different doses.

In the meta-analysis conducted by Finkelstein et al. [10], there were only four controlled studies reporting adult height, -1.51 SDS. The reported benefit of treatment over controls, on the average was 5.7 cm, range 2.3 to 8.7 cm, a wide range. Two of the studies used dosages of GH of 0.25 and 0.26 mg/kg/week and the other two doses of 0.3 and 0.4 mg/kg/week, which may account for the 8.7 cm benefit.

In the review of Guyda H et al. [68], comprising 413 treated children in 11 studies the average adult height was -1.7 SDS (range -1.1 to -2.4 SDS). Seven of the studies used doses of GH ranging from 0.21 to 0.26 mg/kg/week for an average of 0.24 mg/kg/week. The difference of the adult height range from -1.1 to -2.4 SDS is 1.3 SDS, a large difference, that could be owing to a difference in the GH dose used. In three of the studies using 0.3 mg/kg/week or more the adult height was -1.1, -1.3, and -1.4 SDS. So the difference of adult height from -1.4 to -2.4 SDS was owing mainly to studies using the lower dose of GH.

As a consequence of the averaging, the improved benefit obtained, in SDS or cm, with the higher doses of GH is not reflected in the reports, as is illustrated in Table 9, averaging results of published studies with no benefit (-0.1 cm) with others with a benefit of 7.5 cm.

By our results, the adult height and adult height gain were significantly higher by 0.6 SDS (4 cm) in the 47 treated males with NFSS than in the 21 with FSS (Table 4), and in our 63 (males and females) with NFSS (0.55 SDS) higher than in the 25 (males and females) with FSS (Tables 4 and 11).

Similar results were obtained in the only 2 studies reporting the results of NFSS and FSS. The adult height gain of the untreated controls and of treated subjects was significantly higher in the children with NFSS than in FSS (Table 11).

This seems to be a consistent finding in different studies and has implications for the interpretation of the results observed in published studies. The inclusion of a high number of children with FSS in the treated cohort will yield lower results. And also, it has implications for the selection of controls in randomized or non-randomized controlled studies, since the number of children with FSS in the treated or control groups may affect the results.

It is difficult to assess how this has affected the results of previous studies, since most or practically all of the published studies included all the children as ISS and did not mention whether they were NFSS or FSS.

It would seem advisable to identify the two groups NFSS and FSS in future trials.

The assessment of the benefit from treatment with GH is sometimes difficult, even in controlled studies, because the proportion of children with FSS and NFSS, normal and delayed puberty, and at times IUGR, may not be the same in the treated and control groups.

As previously mentioned, to assess the benefit of treatment, comparison of the AH gain (adult height minus the baseline height) and the AH attained of the treated and control groups would seem to be the best method, because they are based on measurements by the investigator.

Also useful would be the comparison between the groups on the attainment of target height. Target heights, however, are often based on self-reported parental heights. In addition, children with FSS may attain heights near or equal to the MPH without treatment, so the inclusion of a different number of children with FSS in the group may influence the attainment of MPH and render the interpretation as benefit of treatment inaccurate. Furthermore, there are a number of reported methods to calculate MPH [47], which may give different results for the benefit when comparing AHs and MPHs. It would also be difficult to compare the benefit of different studies using different methods for calculating MPH.

Calculation of the benefit to children with ISS, treated with GH, based on the difference of AH attained and PAH at the baseline does not appear accurate. The imprecision and inaccuracy of PAH for both sexes have been indicated in a number of reports and only 33% of the variability in achieved AH is predictable. Commonly used methods tend to over-predict AH, especially in young males and in children with delayed bone age, and under-predict if the bone age is not delayed [69]. Also, the individual variability of the tempo of progression of the bone age, either faster or slower than usually expected, as illustrated in Figure 1, affects the accuracy of the PAH. In a randomized controlled trial, girls with ISS treated with GH were 3.5 cm taller than the originally PAH, but 7.5 cm taller than controls when calculated on the basis of AH and AH gain in SDS, a meaningful difference [2].

Recent critical reviews [20, 21, 61] concluded that to date no study has fulfilled the criteria for high quality and strong recommendations, in part owing to the small number of children in the studies, and felt that additional high quality trials up to the achievement of adult height would be necessary to determine the efficacy, ideal dosage, long term safety of growth hormone therapy and to address health related quality of life and cost issues.

There is agreement with their recommendations. Even though rhGH has proven to be a remarkably safe medication for 27 years at the doses recommended, long surveillance studies have been suggested by many, because of the mitogenic effect of IGF-1.

Regarding the dose, there has been many and probably enough trials with a dose of less than 0.3 mg/kg/week of GH, to know that it is not an adequate dose to induce a meaningful or satisfactory gain. Doses higher than 0.3, whether it be 0.32, 0.35, 0.37 mg/kg/week, or starting with 0.32 with adjustments may give a satisfactory gain, but the studies have been few; so additional studies would be helpful. It may be preferable to calculate the dose on mg/kg/week; calculations based on m² of surface area, would provide higher doses per kg at 6 years of age than later because of the nonlinear relationship of m² of surface area and weight.

Additional genetic studies may give useful information to explain the variability of response.

In the past 27 years, since rhGH became available in 1985, there have been 3 or 4 randomized studies, up to the achievement of adult height, and, apparently, they did not provide the answers. It may take 8 or more years to get results from more randomized trials. It is hoped that we do not wait for the answer; many children could benefit while we wait.

One of the main problems that has been often addressed in the past is the significant cost, which limits the availability to children who may need it, raises questions about the use of health resources, and a number of ethical considerations [25]. One may question whether studies would seem applicable to solve the problem of cost; participation from the pharmaceutical industry would be required. The high cost of biopharmaceuticals is a problem concerning public health services around the world, is in the public domain^a, and to reduce cost and increase affordable access to treatment may not be as simple as one may think [70‐72].

The wholesale acquisition cost (the list price paid by the distributing pharmacies from the supplier), generally, the price put by the manufacturer of the GH was $50,000 per gram in 2007 [73] and apparently $76,000 in 2010 [74]. The cost to the patients (or insurances) for the purchase of GH from the distributing pharmacies, presently, may be as much as $88,000 to $100,000 per gram. This is based on the pharmacy bills given to the family for the purchase of GH^b.

The estimated cost of GH therapy compared with no therapy, in 2006, was $52,634 per inch (2.54 cm) or $99,959 per child reflecting an incremental growth of 1.9 in (4.8 cm) (Lee JM et al.) [22] and by others, in 2011, at $113,000 per inch or more (Durand-Zaleski I et al.) [23], depending on unit cost and height gain. This cost is applicable to any child treated with GH, whether it be growth hormone deficiency, Turner syndrome, intrauterine growth retardation or ISS.

If the price was reasonable, many of the objections to treatment, concerns for use of heath resources, and ethical considerations would subside. Cost influences pediatric endocrinologists in their decision to treat [75], and third party payers (private insurances or health agencies, state or national) in their decision to support treatment [74]. Also, if the price was reasonable, it, probably, would be the right thing to do to help the children to attain an adult height within the range judged to be normal by National Health Standards and by society. It would not harm anybody.