0021-972X/99/$03.00/0
The Journal of Clinical Endocrinology & Metabolism
Copyright © 1999 by The Endocrine Society
Vol. 84, No. 12
Printed in U.S.A.
Final Height of Patients with Turner’s Syndrome
Treated with Growth Hormone (GH): Indications for GH
Therapy Alone at High Doses and Late Estrogen Therapy
EMANUELE CACCIARI, LAURA MAZZANTI, AND
THE ITALIAN STUDY GROUP FOR TURNER SYNDROME*
Department of Pediatrics, University of Bologna, 40138 Bologna, Italy
ABSTRACT
We report final height data of patients with Turner’s syndrome
collected by the Italian Study Group for Turner’s Syndrome. One
hundred and thirty-five patients reached their final height during GH
therapy with different therapeutic regimens (dose and combination).
They were divided into 3 groups: group A, 74 patients with high doses
of GH (1 IU/kg/week) for at least 2 yr; group A1, GH alone and
estrogen therapy added not before 14 yr of chronological age (47
patients, of whom 30 were treated for .4 yr and 10 for .6 yr); group
A2, GH plus ethinyl estradiol (17 patients) or GH plus oxandrolone
(10 patients); group B, 51 patients with low doses of GH (0.5 IU/
kgzweek) and high doses of GH for less than 2 yr; and group C, 10
patients with high doses of GH with spontaneous menarche. In contrast to the patients of groups B and C, the patients of group A showed
a significantly higher final height (mean, 147.5 6 6.5 cm) than their
projected height (mean, 142.9 6 6.4 cm). They showed also a signif-
T
HE MOST common clinical feature of Turner’s syndrome (TS) is short stature (1). Ranke et al. (2) and Lyon
et al. (3) confirmed this finding and characterized it by underlining the lack of pubertal spurt. Short stature remains
one of the most serious defects of this condition. Rosenfeld
(4) obtained short term results using high doses of GH and
low doses of oxandrolone. This opened the way to achieving
a satisfactory auxological result. Recently published final
results, although encouraging, differ (5–12) as the therapeutic regimens are different in duration, age at start of therapy,
Received December 1, 1998. Revision received June 28, 1999. Accepted
July 7, 1999.
Address requests for reprints to: E. Cacciari, M.D., Department of
Pediatrics, University of Bologna, Via Massarenti 11, 40138 Bologna,
Italy. E-mail: [email protected].
* Participating investigators: The Italian Study Group for Turner Syndrome: Rosalba Bergamaschi, M.D., Annamaria Perri, M.D., and Emanuela Scarano, M.D. (Bologna); Giuseppe Chiumello, M.D., and Maria Pia
Guarnieri, M.D. (Milan); Franco Rigon, M.D., and Anna Licursi, M.D.
(Padova); Anna Maria Pasquino, M.D., and Ida Pucarelli, M.D. (Rome);
Salvatore Di Maio, M.D., and Mariacarolina Salerno, M.D. (Naples);
Francesca Severi, M.D., and Daniela Larizza, M.D. (Pavia); Francesco
Prisco, M.D. (Naples); Sergio Bernasconi, M.D. (Parma); Fabio Buzi,
M.D. (Brescia); Patrizia Matarazzo, M.D. (Torino); Luciano Cavallo,
M.D. (Bari); Giuseppe Saggese, M.D. (Pisa); Giorgio Tonini, M.D. (Trieste); Maria Carlotta Ragusa, M.D. (Catania); Maddalena Sposito, M.D.
(Perugia); Orazio Gabrielli, M.D. (Ancona); Fabrizio De Matteis,
M.D. (L’Aquila); Lodovico Benso, M.D. (Torino); Brunetto Boscherini,
M.D. (Rome); Giorgio Radetti, M.D. (Bolzano); Carmelo La Cauza, M.D.
(Firenze); Filippo De Luca, M.D. (Messina); Patrizia Borrelli,
M.D. (Rome); Francesco Morabito, M.D. (Piancavallo); and Gianni Bona,
M.D. (Novara).
icantly higher final height compared to the subjects of groups B
(mean, 145.6 6 5.7 cm) and C (mean, 143.0 6 5.3). Among the patients
of group A, the best results were obtained in the patients of group A1
treated with GH alone at high doses and for a longer period (4 yr,
149.3 6 6.4 cm; 6 yr, 153.8 6 4.0 cm). Karyotype, GH secretion, and
birth weight did not influence the efficacy of GH therapy. A low target
height and a high prevalence of a spontaneous ovarian activity or
menarche may negatively influence the effect of GH therapy. Estrogens did not improve final height when added to GH therapy. The use
of small doses of oxandrolone was not effective in our experience. GH
therapy provides a satisfactory auxological result, especially with
high doses of GH alone, given for a long period of time. Optimization
of the treatment would seem to require the identification of the ideal
age for starting therapy, and this is only possible with a specially
designed multicenter study. (J Clin Endocrinol Metab 84: 4510 –
4515, 1999)
GH doses, and therapeutic combinations. We believe, therefore, that the results of other studies can help to optimize
treatment. Here we report the final height data collected by
the Italian Study Group for TS.
Subjects and Methods
The Italian Study group for TS collected 135 patients, who reached
their final height during GH therapy following different protocols. In
Italy, GH treatment for short stature in patients with TS is given free of
charge by the National Health Service. We defined final height in our
patients as when they showed a growth of 0 cm or below 0.5 cm for at
least 1 yr after the discontinuation of GH therapy. All patients were
prepubertal at the start of treatment, euthyroid, and without any relevant cardiac or renal abnormalities. Karyotypes were as follows: 45% of
patients were 45,X, 41% had X-structural abnormalities, and 14% had
X-mosaicism. Real-time pelvic ultrasonography was performed in 102
patients. In 114 of the 135 patients, parental height was available to
calculate the target height (TH). Birth weight was available for 107 of the
135 patients.
Therapy regimens
Patients with TS were divided into three groups (Table 1).
Group A. Seventy-four patients were treated with high doses of recombinant human GH (1 IU/kgzweek) for at least 2 yr. These patients were
divided into 2 subgroups. Group A1 consisted of 47 patients treated with
GH alone. High doses of GH were given for 3.6 6 1.5 yr (range, 2– 6);
23 of them had previously been treated with low doses of GH (0.5
IU/kgzweek) for 1.1 6 0.6 yr (range, 0.3–3.1). All of these patients did
not begin estrogen therapy before 14 yr of chronological age.
Group A2 consisted of 27 patients, who, before the treatment with
high doses of GH, were treated with a combination of GH at low doses
for 1.2 6 0.9 yr (range, 0.5– 4.7) and ethinyl estradiol (100 ng/kgzday for
4510
FINAL HEIGHT IN TURNER’S SYNDROME WITH GH THERAPY
0.9 6 0.4 yr; range, 0.3–1.8) in 17 patients or oxandrolone (0.05 mg/
kgzday for 1.8 6 0.7 yr; range, 0.8 –3.0) in 10 patients.
The karyotypes of the patients of group A were 45,X (55%), X mosaicism (10%), and X structural abnormalities (35%; Table 2). Forty-five
4511
patients of this group had an evaluation of endogenous GH secretion
and underwent 2 pharmacological tests (arginine and l-dopa), 13 of
them were GH deficient (GH peak ,8 mg/L in both tests), as previously
reported (13, 14), whereas 32 were not. Our patients were evaluated for
TABLE 1. Auxological data at start of GH therapy and final height in 135 patients with Turner syndrome subdivided according to
therapy groups
Therapy groups
Pretherapy
CA (yr)
BA (yr)
Ht (cm)
Ht
SD
score
Projected ht (cm)
Final ht
CA (yr)
BA (yr)
Ht (cm)
Ht
Ht
SD
SD
score
score increment
Group A
(n 5 74)
Group A1
(n 5 47)
Group A2
(n 5 27)
Group B
(n 5 51)
Group C
(n 5 10)
11.7 6 2.6
(5.3–17.3)
9.6 6 2.1
(4.2–13)
125.0 6 10.3
(100.4 –144.6)
0.03 6 0.92
(22.1 to 2.9)
142.9 6 6.4
(127.5–162.8)
12.6 6 2.3
(8 –17.3)
10.5 6 1.7
(7–13)
128.4 6 9.8
(104.7–144.6)
0.05 6 0.81
(21.6 to 1.1)
142.9 6 5.7
(131.6 –150.3)
11.0 6 2.6
(5.3–15.9)
9.0 6 2.2
(4.2–12.3)
122.5 6 9.9
(100.4 –140.4)
0.04 6 1.05
(22.1 to 2.9)
142.8 6 7.3
(127.5–162.8)
12.4 6 2.5
(6.7–16.8)
10.8 6 2.1
(6.3–13.7)
129 6 11.4
(105.3–146)
0.24 6 0.97
(21.7 to 2.1)
144.2 6 6.8
(130.5–157.3)
10.8 6 1.8
(8.7–12.9)
8.7 6 2.4
(6.5–11.3)
121.8 6 9.1
(111.8 –130.4)
0.13 6 0.49
(20.9 to 0.3)
141.6 6 3.4
(136.1–144.5)
18.6 6 0.2
(14.0 –24.7)
16.3 6 0.9
(15–18)
147.5 6 6.5a
(132.0 –158.8)
0.72 6 0.93
(21.50 to 2.33)
0.57 6 0.76
(21.65 to 2.85)
18.6 6 0.2
(15.6 –24.7)
16.2 6 1.01
(15–18)
148.5 6 7.1a,b
(132.0 –158.8)
0.85 6 1.02c
(21.50 to 2.33)
0.65 6 0.69d
(20.62 to 2.85)
18.5 6 0.4
(14 –23)
16.4 6 0.7
(15–18)
146.1 6 5.1b
(136.5–158.0)
0.51 6 0.73c
(20.86 to 2.21)
0.41 6 0.87d
(21.65 to 1.73)
18.7 6 0.3
(14.6 –26.1)
16.4 6 1.0
(15–18)
145.6 6 5.7b
(131.4 –159.0)
0.45 6 0.81c
(21.59 to 2.36)
0.19 6 0.63d
(21.91 to 1.43)
17.1 6 0.8
(13.1–21.1)
16.6 6 1.04
(15–18)
143.0 6 5.3b
(131.0 –147.0)
0.07 6 0.75c
(21.64 to 0.64)
0.08 6 1.06d
(22.15 to 1.07)
CA, chronological age; BA, bone age. Values are the mean 6 SD, with the range in parentheses.
a
By ANOVA (groups A, A1, final height vs. projected height): P 5 0.0001.
b
By ANOVA (groups A1, A2, B, C): F 5 3.19; P 5 0.026.
c
By ANOVA (groups A1, A2, B, C): F 5 3.08; P 5 0.03.
d
By ANOVA (groups A1, A2, B, C): F 5 3.78; P 5 0.01.
TABLE 2. Auxological data at start of therapy and final height in patients with Turner syndrome treated for at least 2 yr with high
doses of GH (group A) and in patients with Turner syndrome with spontaneous menarche treated for at least 2 yr with high doses of GH
(group C)
Pretherapy
Group A
Karyotype
45,X (n 5 40)
X-SA (n 5 26)
X-mosaicism (n 5 8)
GH secretion
2 tests ,8 mg/L (n 5 13)
1 or 2 tests .8 mg/L (n 5 32)
Birth wt
Normal weight (n 5 40)
SGA (n 5 7)
LGA (n 5 6)
TH
TH #160 cm (n 5 43)
TH .160 cm (n 5 20)
Ovarian activity
Absent (n 5 24)
Present (n 5 11)
Spontaneous menarche (n 5 10; group C)
Projected ht
(cm)
Final ht (cm)
0.15 6 1.02
20.04 6 0.78
20.48 6 0.56
143.5 6 7.1
142.2 6 5.4
139.1 6 3.9
148.6 6 6.0
145.9 6 7.1
145.7 6 4.9
126.1 6 14.0
123.1 6 9.8
0.72 6 1.22a
20.19 6 0.83
147.5 6 8.5a
141.2 6 5.8
149.5 6 5.8b
145.4 6 5.5
8.4 6 1.9
9.5 6 1.6
11.0
122.4 6 10.1
122.3 6 10.6
137.7 6 5.2
0.21 6 1.01c
20.66 6 0.75
0.01 6 0.74
143.9 6 7.1c
137.9 6 5.3
142.6 6 5.1
147.4 6 5.2c
142.2 6 7.1
146.7 6 7.6
11.4 6 2.8
12.6 6 2.6
9.6 6 2.2
10.3 6 2.1
122.8 6 9.9
132.3 6 9.3
20.13 6 0.73
0.73 6 0.79d
141.6 6 5.1
147.6 6 5.5d
145.9 6 6.9
152.0 6 4.0d
11.4 6 3.0
12.2 6 2.9
10.8 6 1.8
8.8 6 2.3
9.2 6 1.8
8.7 6 2.4
125.5 6 12.0
127.2 6 9.5
121.8 6 9.1
0.32 6 1.08
0.09 6 1.18
20.13 6 0.49
144.8 6 7.5
143.1 6 8.3
141.6 6 3.4
148.9 6 5.5e
145.3 6 6.5
143.0 6 5.3
CA (yr)
BA (yr)
Ht (cm)
11.3 6 2.7
12.0 6 2.5
13.5 6 1.7
9.2 6 2.1
10.4 6 1.9
11.0
124.0 6 11.5
125.8 6 9.1
129.7 6 6.1
10.9 6 2.7
11.5 6 2.6
8.8 6 2.3
8.8 6 1.7
10.7 6 2.3
11.6 6 2.0
15.7 6 0.4
CA, Chronological age; BA, bone age.
a
P 5 0.006, 2 tests ,8 mg/L vs. 1 or 2 tests .8 mg/L.
b
P 5 0.03, 2 tests ,8 mg/L vs. 1 or 2 tests .8 mg/L.
c
P 5 0.03, normal weight subjects vs. SGA.
d
P 5 0.0001, TH .160 cm subjects vs. TH #160 cm subjects.
e
P 5 0.007, spontaneous menarche vs. absent ovarian activity.
Ht
SD
score
4512
GH secretion without estrogen priming (14). Thirty-five patients were
followed longitudinally for ovarian data with real-time pelvic ultrasonography and were evaluated for the spontaneous appearance of
breast stage 2 or greater; 24 patients had no gonadal activity and 11
patients showed transient ovarian function (detectable ovaries, spontaneous breast development, but not spontaneous menarche). Four of
these 11 patients were not given sex steroid replacement therapy, because they demonstrated low gonadotropin levels, at least until they
reached final height. In 63 patients parental height was available to
calculate TH. For 53 patients of group A birth weight was recorded, and
they were divided into 3 groups: normal weight subjects (10th-90th
percentile; n 5 40), small for gestational age (SGA; ,10th percentile; n 5
7), and large for gestational age (LGA; .90th percentile; n 5 6) subjects.
Furthermore, we considered separately the patients that had completed
at least 4 (n 5 30) and 6 (n 5 10) yr of therapy (Table 3). All of these
patients began estrogen therapy after 14 yr of chronological age.
Group B. Fifty-one patients were treated with low doses of GH for 4
months to 6 yr and with high doses of GH for less than 2 yr (Tables 1
and 3). Nineteen of them received a combination of GH and estrogen for
a mean period of 1.81 6 1.32 yr (range, 0.3–5.2 yr), 15 received GH plus
oxandrolone for 1.8 6 0.9 yr (range, 0.5– 4.2 yr), and 17 received GH
alone. Karyotypes were 45,X (36%), X-mosaicism (16%), and X-structural
abnormalities (48%).
Group C (Tables 1 and 2). Ten patients were treated with high doses of GH
for more than 2 yr and experienced spontaneous menarche. Five patients
had previously received low doses of GH for 1.1 6 0.3 yr (range, 0.8 –1.5
yr), and 1 patient had received oxandrolone for 2.5 yr. Karyotypes were
45,X (20%), X-mosaicism (30%), and X-structural abnormalities (50%). At
pelvic ultrasonography all of them showed detectable ovaries.
In patients of all groups GH therapy was given with six or seven
weekly injections sc. In the patients without spontaneous menarche,
puberty was induced at a bone age above 13 yr, taking into account the
psychological situation of each patient, giving at the beginning only
ethinyl estradiol at a dose of 100 ng/kgzday, then 200 ng/kgzday. Once
an acceptable uterine maturation was reached, medroxyprogesterone
was added to ethinyl estradiol from the 11th to the 21st day of the cycle.
At the discontinuation of every cycle of treatment menses occurred
regularly.
Growth and bone age were evaluated each year. The mean final
height of untreated Italian girls with TS is 142.5 cm (15). Height sd score
was calculated from these Turner-specific standards.
Projected adult height was calculated by the method of Lyon et al. (3),
using the sd score for chronological age (standards for Italian girls with
TS). Height was measured using a Harpenden stadiometer. Bone ages
were analyzed according to the method of Greulich and Pyle (16).
Statistical analysis
For statistical analysis, the computer program Statistical Package for
Social Science (SPSS, Inc., Chicago, IL) was used. Mean and sd values
were calculated for each baseline parameter and after each year of
therapy. Statistical significance was assessed using Student’s t test and
ANOVA. All results nominally significant at P , 0.05 were indicated.
TABLE 3. Auxological data at start of therapy and final height of
patients with Turner syndrome treated with low doses of GH
(group B) and patients in group A treated for more than 4 yr (n 5
30) and 6 yr (n 5 10) with high doses of GH
Therapy groups
Pretherapy
CA (yr)
BA (yr)
Ht (cm)
Ht SD score
Projected ht (cm)
Final ht (cm)a
JCE & M • 1999
Vol 84 • No 12
CACCIARI ET AL.
Group B
(n 5 51)
Group A
$4 yr
(n 5 30)
Group A
$6 yr
(n 5 10)
12.4 6 2.5
10.8 6 2.1
129.0 6 11.4
0.24 6 0.97
144.2 6 6.8
145.6 6 5.7
10.5 6 2.6
8.9 6 2.1
120.6 6 10.7
0.07 6 0.86
143.0 6 6.0
149.3 6 6.4
9.7 6 2.7
8.4 6 2.8
118.3 6 10.9
0.21 6 0.66
143.9 6 4.6
153.8 6 4.0
CA, Chronological age; BA, bone age.
a
By ANOVA: F 5 10.03; P 5 0.0001.
Results
In Table 1, the data for patients of group A (74 patients)
with the 2 subgroups (A1 and A2), and of groups B and C are
summarized. In contrast to the patients of groups B and C,
the patients of group A showed a significantly higher final
height than their projected height. They also showed a significantly higher final height (centimeters andsd score) and
a greater height sd score increment (final height sd score 2
pretherapy height sd score) compared to the subjects of
groups B and C. Pretreatment projected height was not significantly different in the 3 groups. Among the patients of
group A, the best results were obtained by the patients of
group A1 treated with GH alone.
Bone ages at final height were determined for only some
patients, and no difference was found between the therapy
groups (group A, n 5 49; group A1, n 5 30; group A2, n 5
19; group B, n 5 20; group C, n 5 8).
The height sd score increment correlated positively with
the years of duration of high dose GH treatment (P 5 0.0001)
and negatively with pretreatment bone age (P 5 0.006).
Table 2 summarizes auxological data before therapy and
final height results in the 74 patients of group A subdivided
by karyotype, GH secretion, birth weight, ovarian activity,
and TH less than or more than 160 cm and in 10 patients with
spontaneous menarche (group C).
Final height, projected height, and pretreatment height sd
score were significantly higher in the patients with GH deficiency than in the patients without GH deficiency, in normal weight subjects than in SGA, and in the patients with TH
above 160 cm than in the patients with TH below or equal to
160 cm. The patients with spontaneous menarche had a significantly lower final height than the patients without ovarian activity.
Table 3 shows auxological data before therapy and the
final height results in 51 patients of group B compared with
patients from group A, who were treated with high doses of
GH, 30 of them for at least 4 yr and 10 for 6 yr. At ANOVA
final height was significantly different in the 3 groups. In
particular, the subjects treated for 6 yr showed a significantly
higher final height than the patients treated for 4 yr and the
patients of group B. Furthermore, the patients treated for 4
yr had a significantly higher final height than the patients of
group B. No differences were found in pretreatment height
sd score and projected height in the groups considered, even
though the group B patients were older both in chronological
and bone age than group A patients at the start of therapy.
In Table 4 the pretreatment values for chronological age,
bone age, height, and projected height are reported with the
final height and the TH in 30 patients treated with high doses
of GH therapy for at least 4 yr, subdivided by chronological
age less than or equal to or more than 11 yr and by bone age
less than or equal to or more than 9.5 yr. The patients treated
at younger chronological and bone ages showed a higher
final height, although no statistically significant differences
were found between the 2 groups.
Some patients (Fig. 1) showed a final height below 142.5
cm (50th percentile for untreated Italian TS girls) (15). There
was no difference in the percentage of these patients in the
FINAL HEIGHT IN TURNER’S SYNDROME WITH GH THERAPY
4513
TABLE 4. Patients with Turner syndrome treated for at least 4 yr with high doses of GH, subdivided according to chronological age (CA)
and bone age (BA) at start of therapy
Pretherapy CA
Pretherapy
CA (yr)
BA (yr)
Ht (cm)
Ht SD score
Projected ht (cm)
TH (cm)
Final ht (cm)
Pretherapy BA
#11 yr
(n 5 7)
.11 yr
(n 5 23)
#9.5 yr
(n 5 6)
$9.5 yr
(n 5 21)
7.7 6 1.5
6.9 6 1.7
109.1 6 6.9
0.23 6 0.70
144.1 6 4.9
156.7 6 8.7
148.4 6 7.7
11.3 6 2.3
9.6 6 1.8
123.9 6 9.3
0.03 6 0.91
142.7 6 6.4
158.6 6 5.2
149.6 6 6.2
9.5 6 1.7
8.4 6 1.1
118.4 6 7.4
0.38 6 0.27
145.1 6 1.9
160.4 6 7.9
152.0 6 5.6
10.8 6 2.9
9.0 6 2.4
121.2 6 12.0
20.07 6 0.96
142.0 6 6.7
158.3 6 5.7
149.2 6 6.7
FIG. 1. Final height (centimeters) of each patient according to the
different GH therapy groups. The line of 142.5 cm corresponds to the
50th percentile for untreated Italian TS girls.
various therapy groups. The number of patients with Xmosaicism was no higher in the group with final height
below 142.5 cm than in the group with final height above
142.5 cm, and no differences were found in the number of
patients who required sex steroid replacement therapy.
However, with the multiple regression analysis it was possible to see that a poorer outcome was mainly influenced by
the low TH (t 5 3.82; P 5 0.0004) and by a higher prevalence
of spontaneous ovarian activity (t 5 2.1; P 5 0.04) and not by
karyotype.
Discussion
Even though our patients did not follow a common therapeutic regimen, our final results offer some significant indications. The mean final height of untreated Italian girls
with TS is 142.5 cm (15), and therefore, therapy added an
average of 5 cm in height to the patients of group A, 6 cm to
those of group A1 treated with GH alone (first with low and
then with high doses), and 3.6 cm to those of group A2
treated with combinations of GH plus estrogens and GH plus
oxandrolone (using steroids at classic low levels). As the
mean projected heights are similar (142.9 and 142.8 cm) in the
two subgroups (A1 and A2) and virtually identical to the
final height of the untreated TS subjects (142.5 cm) (15), it is
reasonable to suppose that the results obtained are encouraging. They show a greater efficacy in treatment with GH
alone, especially at high doses, even though we have to
underline that our high dose of GH was still lower than that
recommended in the United States. This is confirmed both by
the final height of the patients in group B (mean, 145.6 cm)
and by that of the 30 patients treated for more than 4 yr
(mean, 149.3 cm) and 6 yr (mean, 153.8 cm) with high doses.
The results for the only 10 patients treated for more than 6
yr appear excellent; they achieved 10 cm more than projected
height and 11.3 cm more than the mean final height of untreated patients. On the other hand, it should be underlined
that the older untreated patients may be slightly penalized
by the secular trend. In other studies (4 – 6, 12) the combination of GH and small doses of oxandrolone gave the best
results, although some doubts remain (12); in our series the
same combination could not be used for an accurate comparison. In fact, the mean final height in group B, which
included patients treated with low doses of GH combined for
various time periods with oxandrolone or estrogens showed
that the result of therapy is strongly influenced by the adequacy of the therapy, particularly using high doses of GH.
Our experience seems to rule out any kind of auxological
advantage in the early use of estrogens (6, 17, 18), as both
provoked (group A2) and natural (group C) estrogenization
gave unsatisfactory results.
In each therapy group some patients demonstrated an
unsatisfactory result. In fact, some of them reached a final
height below 142.5 cm (15). The percentage of these patients
was not different in the various therapy groups considered.
Among the predictive factors that influenced this result negatively, the low TH and the high prevalence of spontaneous
ovarian activity are the only ones with statistical relevance.
In any case a progressively better height sd score increment
was reached in those patients treated with high dose GH, and
a high positive correlation was found with the duration of the
therapy and the younger pretreatment age.
The reasons for short stature in TS subjects have not yet
been completely clarified. However, both hormonal status
and an intrinsic bone factor linked to the loss of material on
the X chromosome have been suggested. The study of growth
factors in subjects with TS suggest a reduced secretion of GH,
although at differing ages (19, 20). If this pattern of GH
behavior has an effect on growth, the results of GH therapy
and pretreatment auxological status should be different in
patients with a GH deficit from those in patients with normal
GH secretion. This seems to be confirmed by our study; the
32 patients with normal GH reached a mean final height of
145.4 cm, whereas that of the 13 patients with pathological
4514
JCE & M • 1999
Vol 84 • No 12
CACCIARI ET AL.
GH secretion was 149.5 cm (P 5 0.03). However, considering
the initial height sd score and the projected height, we found
that both of these values were significantly higher (P 5 0.006)
in the group with GH deficit than in the group with normal
GH. This seems to suggest that GH secretion did not significantly affect the growth of the patients examined before or
during therapy, in agreement with the work of Schmitt et al.
(21).
In our patients the factors that seem to influence the improvement in final height are TH and the absence of spontaneous ovarian activity or menarche. This is shown by the
well known positive correlation between TH and final
height, by the different final heights of our patients, subdivided according to TH more than 160 cm or 160 cm or less.
This is confirmed by the finding that the patients in each
therapy group who had a final height below 142.5 cm showed
a lower TH and a higher prevalence of spontaneous ovarian
activity than the patients with a final height above 142.5 cm.
Other important genetic factors influence the growth of
these patients. Weight for gestational age clearly affects final
height. This is not due to an unsuccessful response to therapy, but to the fact that SGA are probably already conditioned at birth. In fact, the difference between projected and
final heights is the same in normal subjects and SGA, but the
final height of the latter is significantly lower (P 5 0.03) than
that of normal subjects.
In our series, considering the 30 patients with at least 4
yr of high dose GH therapy subdivided by chronological
age less than or equal to or more than 11 yr and bone age
less than or equal to or more than 9.5 yr at the start of
treatment, no differences between the final heights of the
2 groups were found, as reported by Haeusler et al. (22).
Unfortunately, the small number of cases and the many
variables characterizing the 2 groups (therapy duration,
karyotype pattern, height sd score by chronological age at
the start of treatment, TH) make it impossible to give a
statistical significance to these results. Nevertheless, there
is a positive correlation between height sd score increment
and the duration of high dose GH treatment, particularly
at a younger bone age. Perhaps only a multicenter study
will be able to state the importance of the optimal duration
of GH therapy and the age for starting treatment. In fact,
the different studies published are discordant (6, 23–28)
with regard to the correlation between age at start of
therapy and statural gain, with different effects of the
therapy on bone maturation at different ages (4, 18, 28 –31).
As observed by Rochiccioli et al. (7), our series did not
show any apparent differences in the final heights reached
by patients subdivided by karyotype. Therefore, the numerous variables that affect auxological outcome, in the
three different karyotype groups make the results less
significant.
In conclusion, this study seems to show that GH therapy
provides a satisfactory auxological result, especially with
high doses and long periods of GH alone, even though we
used GH doses lower than those recommended in the U.S.
We suggest postponing the addition of estrogens as long
as possible, taking into consideration the psychological situation. The use of small doses of oxandrolone is not effective
in our experience. GH secretion did not influence final height
in our patients, and we may conclude that pretreatment GH
provocative tests are not useful and need not be performed.
A low TH and a high prevalence of a spontaneous ovarian
activity or menarche are to be considered as predictive factors that may negatively influence the effect of GH therapy
on final height. Optimization of the treatment requires identification of the ideal age for starting therapy, and this is only
possible with a specially designed multicenter study.
Acknowledgment
We are grateful to Dr. Stefano Gualandi, Ph.D. (Department of Pediatrics, University of Bologna), for his helpful assistance with statistical
analysis.
References
1. Ferguson-Smith MA. 1965 Karyotype-phenotype correlation in gonadal dysgenesis and their hearing on the pathogenesis of malformation. J Med Genet.
2:142–155.
2. Ranke MB, Pfluger H, Rosendahl W, et al. 1983 Turner syndrome: spontaneous growth in 150 cases and review of the literature. Eur J Pediatr. 141:81– 88.
3. Lyon AJ, Preece MA, Grant DB. 1985 Growth curve for girls with Turner
syndrome. Arch Dis Child. 60:932–935.
4. Rosenfeld RG, Hintz RL, Johanson AJ, et al. 1988 Three-years results of a
randomozed prospective trial of methionyl human growth hormone and oxandrolone in Turner syndrome. J Pediatr. 113:393– 400.
5. Rosenfeld RG, on behalf of the Genentech National Cooperative Study
Group. 1993 Long term results of GH therapy in Turner syndrome. In: Hibi
I, Takano K, eds. Basic and clinical approach to Turner syndrome. Amsterdam:
Elsevier; 339 –344.
6. Nilsson KO, Albertsson-Wikland K, Alm J, et al. 1996 Improved final height
in girls with Turner’s syndrome treated with growth hormone and oxandrolone. J Clin Endocrinol Metab. 81:635– 640.
7. Rochiccioli P, Battin J, Bertrand AM, et al. 1995 Final height in Turner
syndrome patients treated with growth hormone. Horm Res. 44:172–176.
8. Massa G, Otten BJ, de Muinck Keizer-Schrama SMPF, et al. 1995 Treatment
with two growth hormone regimens in girls with Turner syndrome: final
height results. Horm Res. 43:144 –146.
9. Attanasio A, James D, Reinhardt R, Rekers-Monbarg L. 1995 Final height and
long term outcome after growth hormone therapy in Turner syndrome: results
of a German multicentre trial. Horm Res. 43:147–149.
10. Takano K, Shizume K, Hibi I, et al. 1995 Long-term effects of growth hormone
treatment on height in Turner syndrome: results of a 6-year multicentre study
in Japan. Committee for the treatment of Turner’s syndrome. Horm Res.
43:141–143.
11. Bareille P, Massarano AA, Stanhope R. 1997 Final height outcome in girls with
Turner syndrome treated with a combination of low dose oestrogen and
oxandrolone. Eur J Pediatr. 156:358 –362.
12. Rosenfeld RG, Attie KM, Frane J, et al. 1998 Growth hormone theraphy of
Turner’s syndrome: beneficial effect on adult height. J Pediatr. 132:319 –324.
13. Tassoni P, Cacciari E, Cau M, et al. 1990 Variability of growth hormone
response to pharmacological and sleep tests performed twice in short children.
J Clin Endocrinol Metab. 71:230 –234.
14. Cacciari E, Tassoni P, Parisi G. 1992 Pitfalls in diagnosing impaired growth
hormone (GH) secretion: retesting after replacement therapy of 63 patients
defined as GH deficient. J Clin Endocrinol Metab. 74:1284 –1289.
15. Bernasconi S, Larizza D, Benso L, Volta C, Vannelli S, Milani S, on behalf
of the Italian Study Group for Turner Syndrome. 1994 Turner’s syndrome
in Italy: familial characteristics, neonatal data, standards for birth weight
and for height and weight from infancy to adulthood. Acta Paediatr.
83:292–298.
16. Greulich WW, Pyle SI. 1959 Radiographic atlas of skeletal development of the
hand and wrist, 2nd Ed. Stanford: Stanford University Press.
17. Page LA. 1993 Final heights in 45,X Turner’s syndrome with spontaneous
sexual development. Review of European and American reports. J Pediatr
Endocrinol. 6:153–158.
18. Massa G, Maes M, Heinrichs C, Vandeweghe M, Craen M, VanderschuerenLodeweyckx M. 1993 Influence of spontaneous or induced puberty on the
growth promoting effect of treatment with growth hormone in girls with
Turner’s syndrome. Clin Endocrinol (Oxf). 38:253–260.
19. Ross JL, Long LM, Loriaux DL, Cutler Jr GB. 1985 Growth hormone secretory
dynamics in Turner syndrome. J Pediatr. 106:202–206.
20. Albertsson-Wikland K, Rosberg S. 1991 Pattern of spontaneous growth hormone secretion in Turner syndrome. In: Ranke MB, Rosenfeld RG, eds. Turner
syndrome growth promoting therapies. Amsterdam: Excerpta Medica; 23–28.
21. Schmitt K, Haeusler G, Blümel P, Plöchl E, Frisch H. 1997 Short and long-term
(final height) growth responses to growth hormone (GH) therapy in patients
FINAL HEIGHT IN TURNER’S SYNDROME WITH GH THERAPY
22.
23.
24.
25.
26.
27.
with Turner syndrome: correlation of growth response to stimulated GH
levels, spontaneous GH secretion, and karyotype. Horm Res. 47:67–72.
Haeusler G, Schmitt K, Blümel P, Plöch E, Waldhor T, Frisch H. 1996 Growth
hormone in combination with anabolic steroids in patients with Turner syndrome: effect on bone maturation and final height. Acta Paediatr.
85:1408 –1414.
Ranke MB. 1995 Growth hormone therapy in Turner syndrome. Analysis of
long term results. Horm Res. 44:35– 41.
Pasquino AM, Passeri F, Municchi G, et al. 1996 Final height in Turner
syndrome patients treated with growth hormone. Horm Res. 46:269 –272.
Van den Broeck J, Massa GG, Attanasio A, et al. 1995 Final height after
long-term growth hormone treatment in Turner syndrome. J Pediatr.
127:729 –735.
Chu CE, Paterson WF, Kelnar CJ, Smail PJ, Greene SA, Donaldson MD. 1997
Variable effect of growth hormone on growth and and final adult height in
Scottish patients with Turner’s syndrome. Acta Paediatr. 86:160 –164.
Takano K, Ogawa M, Tanaka T, Tachibana K, Fujita K, Hizuka N. 1997
28.
29.
30.
31.
4515
Clinical trials of GH treatment in patients with Turner’s syndrome in Japan:
a consideration on final height. The committee for the treatment of Turner’s
syndrome. Eur J Endocrinol. 137:138 –145.
Rongen-Westerlaken C, Nienhuis HE, Wit JM, et al. 1993 The effect of four
years of growth hormone therapy on body stature and proportions in patients
with Turner syndrome. In: Hibi I, Takano K, eds. Basic and clinical approach
to Turner syndrome. Amsterdam: Elsevier; 293–302.
Van Teunenbroek A, De Muinck Keizer-Schrama SMPF, Stijnen T, et al.
1996 Yearly stepwise increments of the growth hormone dose results in a better
growth response after four years in girls with Turner syndrome. J Clin Endocrinol Metab. 81:4013– 4021.
Joss E, Zuppinger K. 1984 Oxandrolone in girls with Turner’s syndrome. A
pair-matched controlled study up to final height. Acta Pediatr Scand.
73:674 – 679.
Takano K, Shizume K, Hibi I. 1992 Treatment of 46 patients with Turner’s
syndrome with recombinant human growth hormone for three years: a multicenter study. Acta Endocrinol (Copenh). 126:296 –302.