COPYRIGHT 2005 BY THE JOURNAL OF BONE AND JOINT SURGERY, INCORPORATED

Io8,
Accuracy of the Sauvegrain
Method in Determining
Skeletal Age During Puberty
BY ALAIN DIMGLIO, MD, YANN PHILIPPE CHARLES, MD,
JEAN-PIERRE DAURES, PHD, MD, VINCENZO DE ROSA, MD, AND BONIFACE KABOR, MD
Investigation performed at the Service dOrthopdie Pdiatrique, Hpital Lapeyronie, Montpellier,
and the Institut Universitaire de Recherche Clinique, Montpellier, France
Background: The method of Sauvegrain et al. for the assessment of skeletal age from radiographs of the elbow is
useful during the two years of the pubertal growth spurt. The purpose of this study was to determine the accuracy of
the method and its value in pediatric orthopaedics.
Methods: The Sauvegrain method uses four anatomical landmarks of the elbow: the lateral condyle, trochlea, olecra-
non apophysis, and proximal radial epiphysis. It is based on a 27-point scoring system. The scores for these struc-
tures are summed, and a total score is determined. A graph is then used to determine the skeletal age. The method
was evaluated by three independent observers who used it to assess skeletal age on anteroposterior and lateral ra-
diographs of the left elbow of sixty boys and sixty girls and compared the results with assessments made with use of
the Greulich and Pyle atlas on posteroanterior radiographs of the left hand and wrist. Skeletal age determinations
were performed twice by each observer at a four-week interval.
Results: The skeletal age determination from radiographs of the elbow was more precise because a clear semian-
nual age determination was possible. On the basis of the rating by the observers, the Sauvegrain method presented
excellent interobserver correlation (r = 0.93) and excellent reproducibility (r = 0.96). The correlation between the
methods of Sauvegrain et al. and Greulich and Pyle was good (r = 0.85). Nevertheless, certain elbow growth centers
showed an intermediate developmental morphology, which failed to correspond to the score described by Sauvegrain
et al. This led to errors in the interpretation of data. We suggest an intermediate score for these cases, and we mod-
ified the original graph to make it more accurate.
Conclusions: The modified method of Sauvegrain et al. is simple, reliable, and reproducible, and it complements the
Greulich and Pyle atlas. In clinical practice, maturity can best be evaluated by associating skeletal age, annual growth
rate, and secondary sexual characteristics. Therefore, this method is useful when major decisions such as the timing
of epiphysiodesis or spinal arthrodesis are necessary during puberty.
Level of Evidence: Diagnostic Level II. See Instructions to Authors for a complete description of levels of evidence.
he determination of skeletal age is a controversial but
useful and necessary parameter in pediatric orthopae-
dics
1,2
. Decisions regarding epiphysiodesis or spinal ar-
throdesis as well as predictions of limb-length discrepancy and
standing height are influenced by skeletal age and are usually
made during puberty.
Many different methods are used to assess skeletal ma-
turity
3-10
. Each method involves very different parts of the skel-
eton. The Oxford method of Acheson
4
is based on the pelvis;
the method of Pyle and Hoerr
5
, on the knee; the method of
Hoerr et al.
6
, on the foot; and those described by Tanner et al.
7
,
Semp and Pavia
8
, De Roo and Schrder
9
, and Greulich and
Pyle
10
are based on the hand and wrist.
The Greulich and Pyle atlas
10
is the most widely used
method based on the hand and wrist. Individual ossification
centers are used as indicators of maturity as they tend to occur
and develop regularly in a definite order. The chronological
age when the various indicators appear was determined in
order to assign a skeletal age to a typical developmental status.
In clinical practice, these morphological characteristics on
posteroanterior radiographs of the left hand and wrist are com-
pared with references in the Greulich and Pyle atlas
10
. Never-
theless, the atlas has some disadvantages during the time of
the pubertal growth spurt. It is not regularly divided in six-
month intervals. There are no standards that correspond to
the skeletal ages of 11.5 and 12.5 years in girls as well as 14.5
years in boys. During the phase of increased growth velocity,
between eleven and thirteen years of age in girls and between
T
Io,o
THE JOURNAL OF BONE & JOI NT SURGERY J BJ S. ORG
VOLUME 87-A NUMBER 8 AUGUST 2005
ACCURACY OF THE SAUVEGRAI N METHOD I N DETERMI NI NG
SKELETAL AGE DURI NG PUBERTY
thirteen and fifteen years in boys, the morphological changes
of the hand and wrist are neither obvious nor important. The
beginning of puberty is marked by the appearance of the sesa-
moid of the first ray. The end is characterized by advanced fu-
sion in the epiphyses of the distal phalanx of the thumb and
the first metacarpal. Other developmental signs described by
Greulich and Pyle
10
, such as the shape of the radial and ulnar
epiphyses related to the metaphysis, the outline of the hamu-
lus of the hamate, or the wideness of the metacarpal epiphy-
ses, present many nuances and are difficult to evaluate.
Cundy et al.
11
demonstrated that the Greulich and Pyle
atlas
10
had a large interobserver error, which is problematic in
the assessment of skeletal maturity during pubertal growth.
Little et al.
12
demonstrated that the use of the Greulich and
Pyle atlas could not improve accuracy in the prediction of
limb-length inequality.
In order to compensate for the difficulties with the Greu-
lich and Pyle atlas
10
, Sauvegrain et al.
13
, in 1962, developed a
method to determine skeletal age from elbow radiographs. The
elbow is characterized by clear developmental sequences of its
ossification centers beginning at nine years of age in girls and
eleven years in boys. Fusion of the elbow growth centers is com-
plete at thirteen years in girls and fifteen years in boys. The
method has been used successfully in France
14
and in Belgium
15
for the past forty years.
The purposes of this study were to describe this method,
to demonstrate its reliability, and to investigate its capacity to
replace or complement information provided by the Greulich
and Pyle atlas
10
.
Materials and Methods
he method of Sauvegrain et al.
13
determines skeletal age
from anteroposterior and lateral radiographs of the left
elbow. It is based on a 27-point scoring system and takes into
account the following four anatomical structures of the elbow:
the lateral condyle (1 to 9 points), the trochlea (1 to 5 points),
the olecranon apophysis (1 to 7 points), and the proximal ra-
dial epiphysis (1 to 6 points). As shown in Figure 1, each score
corresponds to the specific developmental status of these ana-
tomical landmarks. Once each of these structures has been
rated, the scores are summed and the total score is placed on a
graph (Fig. 2). The skeletal age is then determined. The graph
represents a mean curve of points corresponding to character-
istic scores for each age. It contains two separate curvesone
for boys and one for girls. However, this method can only be
used shortly before the onset and during the first two years of
puberty, which are marked by an increased growth velocity
and the development of secondary sexual characteristics
1,2
.
In order to test this method, assessments of skeletal age
were performed on sixty boys and sixty girls during the puber-
tal period by three observers, including two experienced pedi-
atric orthopaedists (A.D. and V.dR.) and an orthopaedic
resident (Y.P.C.). Their training consisted only of the previous
routine use of the method of Sauvegrain et al. in the pediatric
orthopaedic clinics; no other advice or training was given. All
three observers worked independently, and no time limit was
imposed on the review. Skeletal age determination was per-
formed twice by each observer. There was an interval of four
weeks between both grading sessions, and all radiographs
were mixed before the second analysis. The radiographs were
blinded with respect to name and age by an individual who
did not participate in the review and who transferred the data
to a separate file in order to compare chronological and skele-
tal age after all observers had completed the second review.
This information was not accessible to any of the observers
during the analysis of the radiographs. T
Fig. 1
The scoring system for the four ossification centers of the elbow. Characteristic scores, such as
3.5 points for the trochlea, 6.5 points for the olecranon apophysis, and 5.5 points for the proxi-
mal radial epiphysis, were added to the score developed by Sauvegrain et al., after radiographic
and statistical review, to give the method greater reliability.
Io,I
THE JOURNAL OF BONE & JOI NT SURGERY J BJ S. ORG
VOLUME 87-A NUMBER 8 AUGUST 2005
ACCURACY OF THE SAUVEGRAI N METHOD I N DETERMI NI NG
SKELETAL AGE DURI NG PUBERTY
These skeletal ages were then compared with those de-
termined from posteroanterior radiographs of the left hand
and wrist with use of the Greulich and Pyle atlas
10
. These radi-
ographs were from the same sixty boys and sixty girls. They
were made at the same time as the elbow radiographs. An
identical review protocol was applied to the hand and wrist ra-
diographs so as to determine the degree of correlation be-
tween the two types of skeletal age assessment.
The children were all healthy and had no history of
systemic illnesses or endocrinopathies. They presented with
abnormalities such as idiopathic limb-length discrepancies,
idiopathic scoliosis, or sequelae of epiphyseal trauma of the
lower limb. In clinical practice, we usually determine other
factors, such as measurement of sitting and standing height
and the secondary sexual characteristics, to provide a better
assessment of puberty. In this study, these parameters were
not taken into account.
In the course of the first and second analyses, we observed
the degree of correlation among the three observers with re-
spect to the use of the methods of Sauvegrain et al.
13
and Greu-
lich and Pyle
10
by calculating intraclass correlation coefficients.
Intraclass correlation coefficients were also calculated
for each observers grading obtained throughout the first and
the second review in order to evaluate the reproducibility of
the two methods of skeletal age assessment.
Statistical Analysis
Statistical analysis was performed the same way for both
groups of subjects. The data obtained were treated with use of
the intraclass correlation coefficient for quantitative or scale
data. The value of the results was expressed with use of the
Kohr classification: correlation was considered to be excellent
if r > 0.80, good if r = 0.61 to 0.80, fair if r = 0.41 to 0.60, and
poor if r 0.40.
Results
Method of Sauvegrain et al.
he mean correlation coefficient among the three observ-
ers was r = 0.93, which indicated an excellent correlation
for the method of Sauvegrain et al.
13
. A detailed analysis of the
27-point scoring system used for the elbow demonstrated ex-
cellent correlation between the various observers (r = 0.96);
excellent correlation was also found with respect to the scores
for the lateral condyle (r = 0.93), the olecranon apophysis (r =
0.95), and the trochlea (r = 0.93). For the proximal radial epi-
physis, the mean interobserver correlation coefficient was r =
0.71. As shown in Table I, no major discrepancy in the correla-
tion coefficients was found between the first and second re-
view or between boys and girls, when the results of the integral
scoring system as well as those of the four anatomical areas
were analyzed separately. Nevertheless, some intermediate im-
ages with scores of between 6 and 7 points for the olecranon
T
Fig. 2
Graph of the method of Sauvegrain et al. for determining skeletal age.
TABLE I Intraclass Correlation Coefficients Between Observers
Girls Boys
First Review Second Review First Review Second Review
Age method of Sauvegrain et al. 0.96 0.96 0.88 0.90
Point method of Sauvegrain et al. 0.97 0.97 0.93 0.95
Points for lateral condyle 0.95 0.95 0.90 0.90
Points for olecranon apophysis 0.96 0.97 0.92 0.93
Points for proximal radial epiphysis 0.77 0.79 0.64 0.62
Points for trochlea 0.92 0.92 0.94 0.95
Age method according to atlas of Greulich and Pyle 0.89 0.93 0.79 0.89
Io,:
THE JOURNAL OF BONE & JOI NT SURGERY J BJ S. ORG
VOLUME 87-A NUMBER 8 AUGUST 2005
ACCURACY OF THE SAUVEGRAI N METHOD I N DETERMI NI NG
SKELETAL AGE DURI NG PUBERTY
apophysis were frequently observed and were a potential
source of error. When these scores were extracted and ana-
lyzed separately, the mean interobserver agreement (r = 0.82)
was lower. The same was true for scores of 5 and 6 points for
the proximal radial epiphysis (r = 0.42) as well as scores of 3
and 4 points for the trochlea (r = 0.81).
Moreover, the mean correlation coefficient of r = 0.96
showed an excellent intraobserver reproducibility between the
first and second analysis for the complete method of Sauvegrain
et al.
13
. Table II indicates the intraclass correlation coefficients
obtained for each observer and for each group of subjects.
Atlas of Greulich and Pyle
An excellent correlation between observers of r = 0.88, on the
average, was noted. An equally excellent intraobserver repro-
ducibility (r = 0.92) was seen between the first and second
analyses (Tables I and II).
Comparison of the Method of Sauvegrain et al.
and the Atlas of Greulich and Pyle
Skeletal age determination based on radiographs of the hand
and wrist and based on radiographs of the elbow demon-
strated a strong correlation between the two methods, with a
mean coefficient of r = 0.85. In Table III, all correlation coeffi-
cients calculated by category (observer, cycle of analysis, and
gender) are shown.
When the overall agreement of both methods was ana-
lyzed, the observers differed by two years or more for three
(5%) of the sixty girls and for five (8%) of the sixty boys ac-
cording to the atlas of Greulich and Pyle
10
. No disagreement of
two years or more occurred when the method of Sauvegrain et
al.
13
was used.
When the two methods were compared, the method of
Sauvegrain et al.
13
was found to be more accurate because it al-
lowed a clearer differentiation of skeletal age in six-month in-
tervals. The atlas of Greulich and Pyle
10
is made up of single
annual serial determinations for girls, and the six-month age
increments for boys are incomplete. There are no references
corresponding to 14.5 years in boys and 11.5 and 12.5 years in
girls. Agreement between these two methods can therefore be
evaluated only in terms of complete years: thirteen, fourteen,
and fifteen years of age in boys and eleven, twelve, and thir-
teen years of age in girls.
Fig. 3-A
Recalibrated graph for girls, allowing more accurate skeletal age
determination.
Fig. 3-B
Recalibrated graph for boys.
TABLE II Intraobserver Reproducibility During
First and Second Assessment
Comparison
of First and
Second Review
Girls Boys
Observer 1
Method of Sauvegrain et al. 0.92 0.98
Method of Greulich and Pyle 0.89 0.94
Observer 2
Method of Sauvegrain et al. 0.93 0.97
Method of Greulich and Pyle 0.96 0.92
Observer 3
Method of Sauvegrain et al. 0.98 0.96
Method of Greulich and Pyle 0.89 0.91
Io,,
THE JOURNAL OF BONE & JOI NT SURGERY J BJ S. ORG
VOLUME 87-A NUMBER 8 AUGUST 2005
ACCURACY OF THE SAUVEGRAI N METHOD I N DETERMI NI NG
SKELETAL AGE DURI NG PUBERTY
Comparison of Data for
Chronological and Skeletal Age
Chronological age and skeletal age were considered to be the
same if the difference between them was not greater than six
months. These ages were the same for twenty (33%) of the
sixty boys and for twenty-one (35%) of the sixty girls, accord-
ing to the method of Sauvegrain et al.
13
. When the atlas of
Greulich and Pyle
10
was used, the chronological and skeletal
ages for eighteen (30%) of the sixty boys and seventeen (28%)
of the sixty girls ranged within this margin. A statistical corre-
lation analysis between chronological and skeletal ages was
not performed because skeletal maturity may be advanced or
delayed in some patients.
Discussion
he method of Sauvegrain et al.
13
was found, on the basis of
our results, to have advantages over the atlas of Greulich
and Pyle
10
. The method is dynamic because morphological
transformations apparent on elbow radiographs are clear and
easily assessed. At the onset of puberty (eleven years of age in
girls and thirteen years in boys), the elbow is still largely carti-
laginous. Two years later, fusion of the elbow growth centers is
complete. This period is critical since it is marked by the pu-
bertal growth spurt. It is an accurate method and was found to
be more detailed than the Greulich and Pyle atlas. It allows the
evaluation of skeletal age in six-month intervals. This is most
valuable during puberty when events occur in quick succes-
sion. It is a practical method because, with practice, it is possi-
T
Fig. 4-A
Fig. 4-B
Figs. 4-A, 4-B, and 4-C Intermediate scores for the system of Sau-
vegrain et al. Fig. 4-A Intermediate score of 3.5 points for the tro-
chlea. Fig. 4-B Intermediate score of 6.5 points for the olecranon
apophysis.
Fig. 4-C
Intermediate score of 5.5 points for the proximal radial epiphysis.
Io,
THE JOURNAL OF BONE & JOI NT SURGERY J BJ S. ORG
VOLUME 87-A NUMBER 8 AUGUST 2005
ACCURACY OF THE SAUVEGRAI N METHOD I N DETERMI NI NG
SKELETAL AGE DURI NG PUBERTY
ble to interpret skeletal age in less than a minute. It is also a
reproducible method.
Nevertheless, the method of Sauvegrain et al.
13
has its
limitations. It is restricted to the period of the pubertal growth
spurt and to the year preceding this phase, that is, between ten
and thirteen years of age in girls and between twelve and fif-
teen years in boys. Prior to the prepubertal period, the elbow
is mainly cartilaginous and changes in the ossification centers
cannot clearly be differentiated at this stage.
The method does provide, however, the possibility of di-
viding puberty into two phases: acceleration and decelera-
tion. Acceleration in the velocity of growth occurs between
eleven and thirteen years of skeletal age in girls, who gain an
average of 12 cm in standing height, and between thirteen and
fifteen years of skeletal age in boys, who gain an average of 14
cm in standing height during this time. This period is called
the ascending phase of pubertal growth. Growth centers of
the elbow are open and ossify progressively during this time.
The beginning of this phase is determined by the radiographic
appearance of a second growth center in the olecranon apo-
physis. The Risser sign
16
is still grade 0 within this portion of
growth. Deceleration of the growth rate occurs between thir-
teen and sixteen years of skeletal age in girls and between fif-
teen and eighteen years of skeletal age in boys. Physeal closure
at the elbow marks the beginning of the phase known as the
descending phase of pubertal growth, and the Risser sign of
1 usually appears six months after elbow physeal closure.
Nevertheless, several weak points in the use of the
method of Sauvegrain et al.
13
were identified as a result of our
work. The curves established by Sauvegrain et al.
13
were diffi-
cult to read especially in the upper part (see Fig. 2). These
curves were recalibrated so as to facilitate the mapping of
scores on a more accessible graph (Figs. 3-A and 3-B). The
morphology of certain ossification centers illustrated by Sau-
vegrain et al. in their original study proved to be a source of
ambiguity and errors in the interpretation of data. When the
available image failed to correspond precisely to the outline of
the landmark, we suggested an intermediate score. A particu-
lar morphological characteristic should be inserted between
the olecranon apophysis classified as 6 and 7 points and given
a score of 6.5 points. Likewise for the trochlea between scores
3 and 4 and for the proximal radial epiphysis between 5 and 6
points (Figs. 1, 4-A, 4-B, and 4-C). These further precisions
provide the method with greater reliability. For instance, the
score of 5.5 points given to the proximal radial epiphysis will
greatly improve the interobserver correlation coefficient espe-
Fig. 5
Characteristic morphology of the olecranon apophysis during the pubertal growth spurt between
the ages of eleven and thirteen years in girls and between the ages of thirteen and fifteen years
in boys.
TABLE III Correlation Between Methods of Sauvegrain et al. and Greulich and Pyle
Girls Boys
First Review Second Review First Review Second Review
Observer 1 0.93 0.94 0.89 0.84
Observer 2 0.83 0.88 0.84 0.73
Observer 3 0.78 0.84 0.81 0.85
Io,,
THE JOURNAL OF BONE & JOI NT SURGERY J BJ S. ORG
VOLUME 87-A NUMBER 8 AUGUST 2005
ACCURACY OF THE SAUVEGRAI N METHOD I N DETERMI NI NG
SKELETAL AGE DURI NG PUBERTY
cially for girls between twelve and thirteen years of age.
As a result of our experience with this method, we were
able to pay particular attention to osseous maturation of the
olecranon apophysis. During the critical pubertal period,
morphological changes of the olecranon apophysis are very
characteristic. Five images were identified: two ossification
nuclei were seen at the age of eleven years in girls and thirteen
years in boys; a half-moon image was seen at the age of 11.5
years in girls and 13.5 years in boys; a rectangular aspect, at
the age of twelve years in girls and fourteen years in boys; the
beginning of fusion, at the age of 12.5 years in girls and 14.5
years in boys; and complete fusion, at the age of thirteen years
in girls and fifteen years in boys (Fig. 5).
The interpretation of skeletal age performed on the ole-
cranon apophysis alone is a simplified but very practical
method in clinical practice
1,2
. When a more thorough and de-
tailed assessment is required, the integral method of Sauve-
grain et al.
13
, including the morphology of all four areas, is
recommended.
Skeletal age assessment is an important exercise. Cundy
et al.
11
showed that the interpretations of skeletal age by four
radiologists who used the atlas of Greulich and Pyle
10
differed
by more than two years for six of sixty children. This margin
of error was smaller in the present study, with such a differ-
ence noted for three of sixty girls and five of sixty boys. There
was no instance in which the method of Sauvegrain et al.
13
led
to a difference of two years or more. Little et al.
12
stated that
the skeletal age data of Greulich and Pyle
10
could not increase
the accuracy over serial chronological age data when predict-
ing the outcome of limb-length inequality. The comparison of
chronological and skeletal age in our study showed that these
two parameters only corresponded to each other for one-third
of the patients. In children in whom skeletal maturity is ad-
vanced or delayed, this would substantially affect the timing of
epiphysiodesis, especially if only chronological age or a single
skeletal age determination were used. We agree that there is no
ideal method. In complex cases, the best approach is to match
the clinical landmarks of puberty and the skeletal ages with
use of several methods such as Greulich and Pyle
10
, Sauvegrain
et al.
13
, Tanner et al.
7
, and Acheson
4
. For assessments made
during the pubertal growth spurt, the method of Sauvegrain
et al.
13
proved to be very valuable, permitting a reduction in
the margin of error.
Skeletal age determination is also useful in idiopathic
scoliosis to assess the remaining growth and the risk for curve
progression. Duval-Beaupre et al.
17
demonstrated the impor-
tance of skeletal age, annual growth velocity, and secondary
sexual characteristics in the identification of the beginning of
the pubertal growth spurt, which is a decisive period. The first
two years of puberty represent the period of peak growth ve-
locity as well as the main spinal growth rate
18
. Little et al.
19
confirmed these observations, showing that peak height veloc-
ity was a useful contemporary clinical marker to predict the
cessation of growth (occurring, on the average, 3.6 years after
the peak height velocity). This phase of peak height velocity
corresponds to the time frame of the method of Sauvegrain et
al.
13
. It may complement the lack of information from the
Greulich and Pyle atlas
10
and aid in the identification of pa-
tients with idiopathic scoliosis who are at high risk of progres-
sion. Fusion of the elbow physes indicates the end of peak
height velocity when the adolescent is entering the decelerat-
ing phase of pubertal growth.
Our main objective was to obtain a detailed evaluation
of skeletal age without confining ourselves to a rigid mathe-
matical formula. In clinical practice, biological maturity can
best be evaluated by determining skeletal age in the context of
a critical analysis of the annual growth rate and secondary sex-
ual characteristics. To choose an adequate date for epiphysiod-
esis or spinal arthrodesis, we now use the modified score and
graphs of Sauvegrain et al.
13
to complement the Greulich and
Pyle atlas
10
in order to obtain the most exact and valuable skel-
etal age data.
Alain Dimglio, MD
Yann Philippe Charles, MD
Vincenzo de Rosa, MD
Boniface Kabor, MD
Service dOrthopdie Pdiatrique, Hpital Lapeyronie, 371 Avenue du
Doyen G. Giraud, F-34295 Montpellier Cedex 5, France. E-mail address
for A. Dimglio: [email protected]
Jean-Pierre Daures, PhD, MD
Institut Universitaire de Recherche Clinique, 641 Avenue du Doyen G.
Giraud, F-34093 Montpellier Cedex 5, France
The authors did not receive grants or outside funding in support of their
research or preparation of this manuscript. They did not receive pay-
ments or other benefits or a commitment or agreement to provide such
benefits from a commercial entity. No commercial entity paid or di-
rected, or agreed to pay or direct, any benefits to any research fund, foun-
dation, educational institution, or other charitable or nonprofit
organization with which the authors are affiliated or associated.
doi:10.2106/JBJS.D.02418
References
1. Dimglio A. Growth in pediatric orthopaedics. In: Morrissy RT, Weinstein SL,
editors. Lovell and Winters pediatric orthopaedics. 5th ed. Volume 1. Philadel-
phia: Lippincott Williams and Wilkins; 2000. p 33-62.
2. Dimglio A. Growth in pediatric orthopaedics. J Pediatr Orthop. 2001;
21:549-55.
3. Kemperdick H. [Skeletal age determination in children]. Radiologe.
1986;26:216-21. German.
4. Acheson RM. The Oxford method of assessing skeletal maturity. Clin Orthop
Relat Res. 1957;10:19-39.
5. Pyle SI, Hoerr NL. Radiographic atlas of skeletal development of the knee.
Springfield: Charles C. Thomas; 1955.
6. Hoerr NL, Pyle SI, Francis CC. Radiographic atlas of skeletal development
of the foot and ankle, a standard of reference. Springfield, IL: Charles C.
Thomas; 1962.
Io,o
THE JOURNAL OF BONE & JOI NT SURGERY J BJ S. ORG
VOLUME 87-A NUMBER 8 AUGUST 2005
ACCURACY OF THE SAUVEGRAI N METHOD I N DETERMI NI NG
SKELETAL AGE DURI NG PUBERTY
7. Tanner JM, Whitehouse RH, Marshall WA, Healy MJR, Goldstein H. Assess-
ment of skeletal maturity and prediction of adult height (TW2 Method). London:
Academic Press; 1975.
8. Semp M, Pavia C. Atlas de la maturation squelettique: ossification squen-
tielle du poignet et de la main. Paris: SIMEP; 1979.
9. De Roo T, Schrder HJ. Atlas van de skeletale leeftijd. Dordrecht, Holland:
Intercontinental Graphics; 1977.
10. Greulich WW, Pyle SI. Radiographic atlas of skeletal development of the hand
and wrist. 2nd ed. Stanford: Stanford University Press; 1959.
11. Cundy P, Paterson D, Morris L, Foster B. Skeletal age estimation in leg length
discrepancy. J Pediatr Orthop. 1988;8:513-5.
12. Little DG, Nigo L, Aiona MD. Deficiencies of current methods for the timing
of epiphysiodesis. J Pediatr Orthop. 1996;16:173-9.
13. Sauvegrain J, Nahum H, Bronstein H. [Study of bone maturation of the
elbow]. Ann Radiol (Paris). 1962;5:542-50. French.
14. Craviari T, Semp M, Chotel F, Bouchard M, Berard J. Evaluation de la
maturation squelettique au niveau du coude en orthopdie. Biom Hum et
Anthropol. 2001;19:13-27.
15. Fabry G, De Waele J. The timing of epiphysiodesis. In: De Pablos J, editor.
Surgery of the growth plate. Madrid: Ergon; 1998. p 223-8.
16. Risser JC. The Iliac apophysis: an invaluable sign in the management of
scoliosis. Clin Orthop Relat Res. 1958;11:111-9.
17. Duval-Beaupre G, Dubousset J, Queneau P, Grossiord A. [A unique theory
on the course of scoliosis]. Presse Med. 1970;78:1141-6 passim. French.
18. Dimglio A, Bonnel F. Le rachis en croissance. Paris: Springer; 1990.
19. Little DG, Song KM, Katz D, Herring JA. Relationship of peak height velocity
to other maturity indicators in idiopathic scoliosis in girls. J Bone Joint Surg Am.
2000;82:685-93.