European Journal of Endocrinology (1998) 138 59–62

Bone maturation in girls with Turner’s syndrome

Lea Even, Vadim Bronstein and Ze’ev Hochberg1
Department of Pediatrics, Bnai Zion Medical Center, and 1Department of Pediatrics, Rambam Medical Center, Haifa, Israel
(Correspondence should be addressed to Z Hochberg, Faculty of Medicine, PO Box 9697, Haifa 31096, Israel)

Abstract
The mechanism of growth retardation in Turner’s syndrome has not been resolved. It is often referred
to as a bone dysplasia, although endocrine derangement has not been ruled out. The present study was
undertaken to evaluate the maturation of individual bones of the hand and wrist in girls with Turner’s
syndrome and thereby obtain information which may aid in elaborating the possible mechanism of the
growth retardation in girls with Turner’s syndrome. Hand and wrist films of 24 girls with Turner’s
syndrome, 11 normal girls with short stature and 23 normal controls were evaluated, using the
references of Greulich and Pyle. Each bone or epiphysis was given an individual ‘age’. During
childhood the Turner patients showed the greatest delay in bone age of the phalangeal bones while
the least delayed were the radius and ulna (long bones) and metacarpals. The carpal bones showed
intermediate retardation. This pattern and extent of maturational retardation was clearly different
from that of the short stature normal group, who showed uniform retardation of all bones. During
adolescence, the phalangeal bones were further retarded and the carpal bones showed a moderate
retardation. The unique profile of bone maturation in Turner’s syndrome suggests an insult to
chondroplasia, which may be related to estrogen deficiency or to an as yet undetermined endocrine or
paracrine derangement.

European Journal of Endocrinology 138 59–62

Introduction X-ray films of the hand and wrist, however, contain

The mechanism of growth retardation in Turner’s
syndrome has not been resolved. The combination of
short stature with irregular metaphyses, vertebral
anomalies and Madelung’s deformity (1, 2), lead to
the assumption that Turner’s syndrome is a bone
dysplasia, although the radiological findings do not
conform with the definition of any of the known bone-
or chondrodysplasias. The most common bones which
clinicians usually look at in such patients are the hand
and wrist bones, radiographed for the purpose of
assessing a ‘bone age’. This rough measure of bone
maturation gives limited information of growth poten-
tial in girls with Turner’s syndrome, and clinicians
resort to a projection of adult height by an extrapolation
from disease-specific growth curves.
Initial analyses of hand and wrist bones were
reported over two decades ago (3, 4). Metacarpal and
phalangeal bones were reported to be shorter and
narrower in Turner patients, with increasing growth
deficiency from distal to proximal and from lateral to
medial bones. In adult Turner patients the surface areas
of carpal bones were reported to be smaller (5) or not
different (6) from adult controls, and increased with
estrogen replacement. A later report confirmed a delay
in bone maturation in both prepubertal and pubertal
Turner girls (7).
potentially more informative findings on the mecha-
nism of growth retardation. The differential maturation
pattern of the cuboid bones of the wrist from that of the
short bones of the hand or the long bones of the arm,
may signify differential hormonal or growth factors
effects on these various bones.
In the present report we describe the evaluation of
maturation of individual bones of the hand and wrist in
girls with Turner’ syndrome and compare them to those
of a control group of normal sized girls and a control
group of girls with short stature. We show the age-
related changes in these bones and attempt to use the
results to suggest possible mechanisms which may be
affected in the bone maturation of girls with Turner’s
syndrome.
Patients and methods
The age and height characteristics of the subjects used
in this study are presented in Table 1. In all 24 patients
with Turner’s syndrome the diagnosis was confirmed by
chromosomal karyotyping, and included patients with
45,X, 45X,iXq or mosaics. The girls were divided into a
preadolescent group (aged 4.6–10 years, n ¼ 13) and
an adolescent group (aged 10.1–14.7 years, n ¼ 11).
For each patient a single retrospective radiograph was

q 1998 Society of the European Journal of Endocrinology

Downloaded from Bioscientifica.com at 04/25/2021 09:10:36AM

via free access
60 L Even and others EUROPEAN JOURNAL OF ENDOCRINOLOGY (1998) 138

Table 1 Age and height characteristics of girls in the Turner

syndrome group (n ¼ 24), the short normal group (n ¼ 11) and the
control group (n ¼ 23). Results are means 6 S.D.

Turner’s Short Normal

syndrome normal control

Age 10.8 6 0.9 9.7 6 1.0 10.0 6 0.9

Height S.D.s ¹3.0 6 1.2 ¹2.3 6 1.1 0.3 6 1.1

reviewed, and that image was chosen to give a wide

range of ages and to include no patient with sponta-
neous puberty or on any hormonal replacement. The
control group and the short normal group were age-
matched to the Turner patients and retrospectively
selected from the patients of our clinic. Girls in the short
normal group (n ¼ 11) were those referred to us with
height that was smaller than ¹2 S.D.s and who had a
growth velocity of more than ¹1 S.D.s, regardless of their
parents’ height. Retrospectively, in view of their delayed
bone maturation, most of these girls had constitutional
delay of growth and puberty. Normal control girls
(n = 23) were age-matched and of normal height and
growth velocity.
Bone maturation was assessed by a single observer
through evaluation of each individual bone, as shown
in Fig. 1, by the standards of Greulich and Pyle (8). Each
film was evaluated on three separate occasions and the
results did not deviate by more than three months in
any of the films. Whereas the TW-2 method (9) requires
the joint scoring of either the entire plate or the radius,
ulna and short bones (RUS), the Greulich and Pyle
method allows for evaluation of each individual bone in Figure 1 Diagramatic presentation of hand and wrist bones. Bone
terms of ‘years’. For statistical comparison of girls of numbers are explained in Table 2.
different ages, the reading for each bone is given in
terms of ‘years’ of delay against the chronological age, bones’ and the metacarpals, with a mean retardation of
and the standard deviations of values of the two control 0.8–1.1 ‘years’. The carpal bones showed intermediate
groups were evaluated against the results of each group retardation of 1.4–1.5 ‘years’. This pattern and extent
by a Student’s t-test. of maturational retardation was clearly different from
that of the short normal group. In that group all of the
Results bones were retarded to a similar extent, with an average
1.8–2.7 ‘years’ of retardation.
Table 2 presents the individual bone maturation in units The older group (>10 years) of normal controls was
of ‘years’ of retardation relative to the respective similar to the standards. In the older age group, Turner
chronological age. It is divided into girls up to 10 patients showed a further delay of maturation in the
years of age, and girls older than 10 years. This cut-off entire image. The phalangeal bones were further
age was elected as the mean age of the onset of puberty retarded, with a mean retardation of 2.7–3.1 ‘years’.
in normal girls when, under estrogenic activity, bone The ‘long bones’, the metacarpals and the carpal bones
maturation accelerates and on a theoretical basis the showed a moderate retardation of an average 1.5–2.1
gap between Turner’s syndrome and normal control ‘years’. This pattern and extent of maturational
groups would increase. retardation was clearly different from that of the short
The control normal girls showed a slight advance- normal group in whom all of the bones were retarded to
ment of bone maturation of 0.3–0.9 ‘years’ that was a similar extent with an average narrow range of 2.2–
not significantly different from the standards. In the 2.6 ‘years’ of delay.
younger age group Turner patients the phalangeal A third degree regression of specific bones against age
bones were the most retarded with a mean retardation emphasize the unique pattern of bone retardation in
of 1.7–1.8 ‘years’; the least retarded were the ‘long Turner’s syndrome (Fig. 2). Whereas in the two control

Downloaded from Bioscientifica.com at 04/25/2021 09:10:36AM

via free access
EUROPEAN JOURNAL OF ENDOCRINOLOGY (1998) 138 Bone maturation in Turner’s syndrome 61

Table 2 Individual bone maturation assessed by the Greulich and Pyle method. Results are given in ‘years’ of delay behind the
chronological age for thirteen preadolescent Turner syndrome (TS) girls (aged 4.6–10 years) and for eleven adolescent TS girls (aged
10.1–14.7 years). Age-matched normal short (SS) and control groups are also included. Bone numbers (in parentheses) are those shown in
Fig. 1. Results are means 6 S.D.

Preadolescent Adolescent

Bone TS SS Control TS SS Control

Radius (1) ¹1:1 6 1:6 ¹2:0 6 1:4 ¹0:9 6 1:3 ¹1:5 6 1:1a ¹2:4 6 1:4c 0:0 6 1:3
Ulna (2) ¹0:8 6 1:1 ¹1:3 6 1:0 ¹0:9 6 1:5 ¹2:1 6 1:6b ¹2:3 6 1:5c 0:0 6 1:1
Prox. Ph. (3–7) ¹1:8 6 1:4 ¹2:2 6 1:0 ¹0:8 6 2:0 ¹2:7 6 1:5c ¹2:6 6 1:5c ¹0:1 6 1:2
Mid. Ph. (8–11) ¹1:8 6 0:9a ¹2:5 6 1:3b ¹0:4 6 1:6 ¹3:1 6 2:1b ¹2:6 6 1:5c ¹0:1 6 1:2
Dist. Ph. (12–16) ¹1:7 6 1:2 ¹2:4 6 1:1b ¹0:6 6 1:6 ¹2:7 6 1:6b ¹2:6 6 1:5c ¹0:1 6 1:2
Metacarpals (17–21) ¹1:1 6 1:6 ¹2:6 6 1:6b ¹0:6 6 1:5 ¹1:9 6 1:5a ¹2:6 6 1:5c ¹0:1 6 1:1
Capit. & Hamate (22–23) ¹1:5 6 1:6 ¹2:5 6 1:4 ¹0:8 6 1:7 ¹1:9 6 1:6a ¹2:6 6 1:3c 0:0 6 1:1
Pisif. & Triq (24–25) ¹1:4 6 1:6 ¹1:8 6 1:4 ¹0:7 6 1:6 ¹1:1 6 1:7 ¹2:7 6 1:4c ¹0:1 6 1:1
Lunate (26) ¹1:5 6 1:7 ¹3:4 6 2:1c ¹0:6 6 1:2 ¹1:6 6 1:3a ¹2:8 6 2:9c ¹0:1 6 1:1
Scaphoid (27) ¹1:8 6 1:2 ¹2:4 6 1:7a ¹0:9 6 1:8 ¹1:7 6 1:8 ¹2:4 6 1:7c ¹0:1 6 1:0
Trapezoid (28) ¹1:3 6 1:3 ¹2:2 6 1:5a ¹0:6 6 1:6 ¹2:1 6 1:5b ¹2:4 6 1:4c ¹0:1 6 1:1
Trapezium (29) ¹1:1 6 1:4 ¹3:0 6 1:8b ¹0:8 6 1:7 ¹1:6 6 1:4a ¹2:1 6 1:6b ¹0:1 6 1:1
a
P < 0:05, b P < 0:01, c P < 0:005 compared with control group.
Prox. Ph., proximal phalanx; Mid. Ph., middle phalanx; Dist. Ph., distal phalanx; Capit., capitale; Pisif., pisiform; Triq., triquetsal.

groups, ‘long bones’, ‘short bones’ and carpals show

parallel regression with narrow gaps, the pattern in the
Turner patients showed a milder delay in the ‘short
bones’ during early prepuberty, with increased delay
during adolescent age. The carpals and ‘long bones’
were relatively more retarded in prepuberty, and less so
in adolescence.

Figure 2 Third degree regression of the delay (d) in bone maturation
against chronological age in the short normal group (SS), the
control group (Ctr) and the Turner’s syndrome group (TS).
Discussion
As the TW-2 method for evaluation of bone maturation
was developed, Tanner et al. realized that the ‘long bone’
and ‘short bone’ readings (RUS) show a discrepancy
with the carpals in many clinical conditions (9). They
proposed, however, that the mean difference between
carpals and RUS in normal girls is 0 ‘years’ during 2–
11 years of age, with an S.D. of 0.7–1.0 ‘years’. In the
present study we took a similar approach, expanded it
to readings of individual bones and attempted to obtain
an insight from these readings into the mechanisms of
growth retardation in Turner’s syndrome. In order to
deal with bone ‘age’ of individual bones, and to give
the readings in terms of ‘years’ of delay for each
individual bone we had to use the Greulich and Pyle
method (8).
The overall tendency of bone maturation in Turner’s
syndrome is for prepubertal milder retardation of all the
bones compared with short normal girls. Retardation of
the short bones, and to a milder extent that of the long
bones and carpals, increase after the age of ten. This
pattern is consistent with growth curves of Turner
patients (10, 11), where growth retardation during
early childhood is relatively mild, increasing towards
late childhood to become severe during adolescence.
Of the many still obscure aspects of the growth
retardation in Turner’s syndrome, one obvious factor is

Downloaded from Bioscientifica.com at 04/25/2021 09:10:36AM

via free access
62 L Even and others
estrogen deficiency. This deficiency becomes dominant
during adolescence; however it has recently been
appreciated that estrogen deficiency may also have
some impact during childhood (12). It is assumed,
therefore, that the unique pattern of bone maturation
during adolescence in these untreated girls with Turner’s
syndrome is a pattern of estrogen deficiency, which
is characterized by severe retardation of phalangeal
epiphyses, and milder retardation of radius, ulna,
metacarpals and carpals. But this hypothesis needs to
be addressed in a study of bone maturation in girls with
estrogen deficiency from other causes. If this is to become
the maturational profile of estrogen deficiency, it is
obvious that some estrogen deficiency prevails in Turner’s
childhood, as here too the phalangeal epiphyses are the
most retarded.
It is, however, possible that this profile has to do with
another, unknown endocrine derangement of Turner’s
syndrome, leading to the characteristic growth retarda-
tion and its specific ‘bone dysplasia’. By convention, the
appearance on an X-ray film of secondary ossification
centers in the epiphyses of ‘long bones’ and of ‘short
bones’ during early childhood is referred to as the
process of ‘osteogenesis’. It is followed by development
and expansion of the growth plate cleft, a process referred
to as ‘chondroplasia’ (13). Primary ossification of the
cuboid carpal bones develops through ‘osteogenesis’. The
profile of bone maturation in Turner’s syndrome suggests
a major insult to chondroplasia. During the younger age
group, when both cuboid and long bones reflect
osteogenesis, the entire profile of Turner patients was
less delayed, as compared with the short normal group.
Later on, as long bone epiphyses develop to represent
chondroplasia, the maturational profile portrays a
greater retardation of the phalangeal epiphyses. In
descriptive terms, the insult to chondroplasia can be
defined as ‘epiphyseal dysplasia’.
To combine these two approaches, it can be
envisioned that estrogen deficiency results in an insult
to chondroplasia. Yet, chondroplasia is not influenced
only by estrogen. A large volume of literature has been
devoted to the regulation of enchondral ossification by
growth hormone (GH), insulin-like growth factor (IGF)-I
and local growth factors (13). In that respect, serum GH
has been reported to decrease during the adolescence of
girls with Turner’s syndrome (14) and recent indirect
evidence has suggested that the growth retardation in
Turner’s syndrome may be linked to decreased sensitivity
to IGF-I (15).
The enigma of growth retardation in Turner’s
syndrome has yet to be solved. Future research may
be directed to the potential insults to chondroplasia by
EUROPEAN JOURNAL OF ENDOCRINOLOGY (1998) 138
estrogen deficiency, IGF-I decreased sensitivity, or
otherwise.
Acknowledgements
We thank Drs I Sills and H Gutman for letting us review
their patients’ X-rays and Professor M Jaffe for critical
reading of the manuscript.
References
1 Finby N & Archibald RM. Skeletal abnormalities associated with
gonadal dysgenesis. American Journal of Radiology 1965 93 354–
361.
2 Mehta AV, Chidambaram B, Suchedina AA & Garrett AR.
Radiological abnormalities in Turner’s syndrome. Chest 1993
104 1795–1799.
3 Acheson RM & Zampa GA. Skeletal maturation in ovarian
dysgenesis and Turner’s syndrome. Lancet 1961 i 917–919.
4 Park E. Radiological anthropometry of the hand in Turner’s
syndrome. American Journal of Physical Anthropology 1977 46
463–470.
5 Cleveland RH, Done S, Correia JI, Crawford JD, Kushner DC &
Herman TE. Small carpal bone surface area, a characteristic of
Turner’s syndrome. Pediatric Radiology 1985 15L168–15L172.
6 Park E. Cortical bone measurements in Turner’s syndrome.
American Journal of Physical Anthropology 1977 46 455–461.
7 Webber ML, Puck MH, Maresh MM, Goad WB & Robinson A.
Skeletal maturation of children with sex chromosome
abnormalities. Pediatric Research 1982 16 343–346.
8 Greulich WW & Pyle SJ. Radiographic Atlas of Skeletal Development
of the Hand and Wrist, edn 2. Stanford, CA: Stanford University
Press, 1991.
9 Tanner JM, Whitehouse RH, Cameron N, Marshall WA, Healy MJR
& Goldstein H. Assessment of Skeletal Maturity and Prediction of
Adult Height (TW-2 Method), edn 2. London: Academic Press,
1983.
10 Ranke MB, Pfluger H, Rosendahl W, Stubbe P, Enders H, Bierich JR
& Majewski F. Turner’s syndrome: spontaneous growth in 150
cases and review of the literature. European Journal of Pediatrics
1982 141 81–88.
11 Lyon AJ, Preece MA & Grant DB. Growth curve for girls with
Turner’s syndrome. Archives of Diseases in Childhood 1985 60 932.
12 Klein VO, Baron J, Colli MJ, McDonnell DP & Cutler Jr GB. Estrogen
levels in childhood determined by an ultrasensitive recombinant
cell bioassay. Journal of Clinical Investigation 1994 94 2475–2480.
13 Hunziker EB, Wagner J & Zapf J. Differential effects of IGF-I and
hGH on the various developmental stages of growth plate
chondrocytes in vivo. Journal of Clinical Investigation 1994 93
1078–1083.
14 Ross JL, Long LM, Loriaus DL & Cutler GB Jr. Growth hormone
secretory dynamics in Turner’s syndrome. Journal of Pediatrics
1985 106 202–206.
15 Hochberg Z, Aviram M, Rubin D & Pollack S. Decreased sensitivity
to insulin-like growth factor-I in Turner’s syndrome: a study of
monocytes and t-lymphocytes. European Journal of Clinical
Investigation 1997 27 543–547.
Received 25 April 1997
Accepted 29 September 1997
Downloaded from Bioscientifica.com at 04/25/2021 09:10:36AM
via free access