PEDIATRIC REHABILITATION, APRIL – JUNE 2003, VOL. 6, NO.

2, 111–118

Incidence of surgery in conservatively treated

patients with scoliosis
HANS-RUDOLF WEISS, GRITA WEISS and
HANS-JÜRGEN SCHAAR

Accepted for publication: described in literature. Conclusions: The conservative treat-

ment of scoliosis (which includes outpatient physiotherapy,
Keywords Idiopathic scoliosis, brace treatment, physio- inpatient intensive scoliosis rehabilitation (SIR) and high-
therapy, inpatient intensive rehabilitation correction-effect bracing based on plastercasting) shows an
obvious reduction in the rate of surgery in patients with idio-
Summary pathic scoliosis compared with untreated scoliosis patients.
Thus it can be seen that the conservative treatment of scoliosis
The effectiveness of bracing and the effectiveness of conser- may reduce the incidence of surgery, which not only causes
vative treatment of scoliosis as a whole is widely denied and considerable medical expenses but also carries with it the risk
rejected. We therefore carried out the present study in order to of repeated operations and further complications. In our
establish whether the work involved in the conservative treat- study, we also found highly significant differences by compar-
ment of scoliosis is reasonable and worthwhile, including as it ison with the work of a bracing centre in the USA. Therefore,
does the education of patients, techniques of brace moulding, the quality of the bracing itself and the effects of primary
time-consuming follow-ups and the strain put on patients correction must receive greater consideration in future
by the treatment itself. Study design: A retrospective analysis scientific studies on the subject of ‘conservative’ treatment
of the incidence of surgery in patients with scoliosis presenting of scoliosis.
every kind of aetiology. To improve comparison with data
already published by other centres, this group of patients
was matched with the study design of the control group.
Materials and methods: Since 1992, in addition to the intensive
inpatient rehabilitation programme developed at our centre, Introduction
we have also been practising moulding techniques according
to Chêneau in order to construct scoliosis orthoses of high On the one hand the literature confirms the effec-
correction. For the purposes of this study we chose a number tiveness of certain orthoses in the treatment of scoliosis
of scoliosis patients from our data bank who had had conser- [1, 2] and yet on the other hand their effectiveness is
vative treatment at our centre between 1993 and 1996. All the generally denied and the treatment is widely rejected
patients, like those of the untreated control group, were at
least 15 years of age at the time they were last investigated [3, 4]. Although conservative methods of treatment
or questioned. The incidence of surgery in our group with physiotherapy and braces are generally accepted
was compared statistically with data published by other in central Europe [5, 6], conservative treatment in the
centres. Results: Three hundred and forty-three patients USA is largely discounted. In addition to the language
(females only) with a curvature of 33.4
(SD ¼ 18.9) were barrier, which is so prejudicial to the English-speaking
included in our study and could be followed up under the
described pre-conditions. Forty-one patients (11.95%) had countries in that it reduces the availability there of
had surgery. In patients with adolescent idiopathic scoliosis, up-to-date information about conservative methods
we found an incidence of surgery of only 7.3% compared of treatment of scoliosis, a number of mistaken and
with 28% in the control group with the same diagnosis. unhelpful ideas even exist in the practice of this treat-
Statistically, therefore, the incidence of surgery in our ment which, far from being beneficial to patients, may
group was highly significantly lower than the incidence of
surgery in the control group from Ireland and other centres actually put strain on them [7].
In Germany, conservative treatment (including brace
treatment and physiotherapy) is regarded as effective
if a curvature shows no further progression (even
Authors: Hans-Rudolf Weiss (author for correspondence; when further growth is expected) or if there is a clear
e-mail: [email protected]), Grita Weiss and
Hans-Jürgen Schaar, Asklepios Katharina-Schroth-Klinik, tendency towards a decreasing curvature. Hanks et al.
Spinal Deformities Rehabilitation Centre, Korczakstr. 2, [7] nevertheless speak of successful brace treatments in
D-55566 Bad Sobernheim, Germany. the USA, provided that the increase of the curvature

Pediatric Rehabilitation ISSN 1363–8491 print/ISSN 1464–5270 online # 2003 Taylor & Francis Ltd
http://www.tandf.co.uk/journals
DOI: 10.1080/13638490310001593446
 H. Weiss et al.

is no greater than 10
[7]. However, that study fails to of conservative treatment in general, and of scolioses
draw attention to the corrective effect of an orthosis, an orthoses in particular, is very complex and cannot be
aspect that its authors do not even mention. Moreover, answered on a one-dimensional basis. By contrast, the
other widely used ideas for treatment with orthoses, for treatment of scoliosis with the Chêneau brace is
example the Milwaukee brace [8], have proved to be currently most practised in Germany [12]. The clinical
ineffective by comparison with the effectiveness of the histories of individuals enjoying excellent corrective
braces used in central Europe because their corrective effects and favourable outcomes are encouraging the
effect has been too low to show any positive end result use of this method there (figure 1). We must emphasize
[9]. Based on our own review of the literature, we were nevertheless that how ever well adjusted an orthosis
able to prove that there does indeed exist a direct may be, there do indeed exist curvatures whose progres-
positive correlation between the primary effect of an sion cannot be halted. And so we ask the following
orthosis and the end result [10]. In his study of the questions: Are our efforts reasonable and worthwhile,
treatment of idiopathic scoliosis with the aid of involving as they do the time-consuming education of
the Chêneau brace, Landauer [2] also concluded that patients, moulding techniques and follow-ups, and the
compliance on one hand and the primary effect of the strain put on patients by all the conservative methods?
brace on the other are the main parameters of successful Or should we follow the American way of scoliosis
brace treatment. treatment, i.e. wait until the scoliosis has to be corrected
In the USA, however, braces are not very often surgically?
‘custom made’. They are sometimes able to achieve a In a recent study, Goldberg et al. [4] analysed the
corrective effect of 50% in relation to the initial curva- incidence of surgery in patients with adolescent idio-
ture, but only on smaller curves, and thus may indeed pathic scoliosis at several centres in which a comparison
exercise an essential influence on the final prognosis [11]. between the incidence of surgery in braced patients and
This shows that the question of the effectiveness or not the incidence of surgery in non-braced patients was

Figure 1 (a) Cobb angle before brace treatment of 37
in 13-year-old girl with a low thoracic curve (with thoracic flatback), (b) in the brace with
an over-correction to 16
, (c) 14
directly after weaning off the brace, and (d ) 18 months after weaning off 16
at the age of 19.

112
 Incidence of surgery in conservatively treated scoliosis patients

made. The study concluded that, as far as the number of Thus, in order to be able to explain to patients whether
surgical interventions was concerned, from a statistical conservative treatment is a reasonable procedure or
point of view, patients who used a Milwaukee brace in not, we have felt it desirable to make a study of the
the years between 1950 and 1970 did not differ from effectiveness or not of an out-patient programme of
untreated patients in the 1990s. Although the method- physiotherapy, in-patient intensive scoliosis rehabili-
ology of this study showed some weaknesses [6], it has tation and brace treatment (figures 2 and 3).
enabled us to use the collective of Goldberg et al. [4] as
a control group for a retrospective study of our own.
In the non-operative treatment of scoliosis, a great Scoliosis in-patient rehabilitation (SIR)
number of individual factors play an important role, SIR employs an individualized exercise programme
as already noted by Goldberg et al. [4]. In cases of combining corrective behavioural patterns with
adolescent scoliosis, surgery is normally a cosmetic physiotherapeutic methods. The exercise programme is
indication [4], and so patients may decide for themselves based on sensomotor and kinesthetic principles, and its
what treatment they want to follow. The conservative goals are: (1) to facilitate correction of the asymmetric
route of treatment is time consuming and also demands posture; and (2) to teach the patient to maintain the
restrictions on life-style, whereas surgical intervention corrected posture in daily activities.
will not take much time and may also have excellent A 4-week minimum stay is required for the first
cosmetic results. treatment, and may be up to 6 weeks, depending on
When offered all the information available about prognosis; return treatments are 3–6 weeks in length,
conservative treatment and surgical intervention, many depending on symptoms and prognosis. Patients
adolescent patients, even those with curvatures of are admitted in groups, with the first day of the
over 40
, nevertheless opt for conservative treatment. programme devoted to diagnosis and evaluation of

Figure 2 Chêneau brace from the rear view

113
 H. Weiss et al.

between. Central to the individual and group exercise

programmes is therapist assistance, by a staff of 20
physical and sports therapists, who supervise all
exercises and provide exteroceptive stimulation needed
to obtain desired correction. Depending on individual
curve patterns, the patients are assigned to special exer-
cise groups for an additional 2 h daily. Development
and maintenance of the corrected posture is facilitated
using asymmetric standing exercises designed to employ
targeted traction to restore torso balance and mobility.
Rice-bag bolsters provide localized sustained pressure
during floor exercises for mobilization of rib pro-
minences or other torso and lumbar asymmetries.
Bracing (since 1992) is applied as indicated (depending
on actual growth velocity, curvature pattern, curve
flexibility, and Cobb angle).

Materials and methods

In order to answer the question as to whether conser-
vative methods (out-patient physiotherapy, in-patient
intensive rehabilitation (SIR) and brace treatment—in
this case mainly the Chêneau brace) are effective,
we compared the incidence of surgery in patients
presenting various kinds of aetiology treated at this
centre with the incidence of surgery as that reported
Figure 3 Deflection/derotation brace from our centre with improved by Goldberg et al. [4] in patients without conservative
correction effect. management. In the Goldberg et al. [4] study, the
criteria for inclusion were:
1. Diagnosis of adolescent idiopathic scoliosis;
the three-dimensional deformity, supervised by seven 2. Minimum age of 15 years at last check-up (not
staff physicians (two orthopaedic surgeons, and seven younger than 10 years at first check-up);
general practitioners or specialists in physical medicine 3. Documentation for Cobb angle without brace at
and rehabilitation), who also provide oversight for first check-up (at least 10
).
each patient’s programme. On the second day, instruc-
tion in basic human anatomy, spinal deformity and Curve pattern distribution was given for the whole
principles of postural balancing therapy is provided to collective. In girls, menarchial status and the incidence
the group. Each patient receives a detailed summary of of surgery were also known.
In order to be able to compare the patients at our
his/her own condition, and those with matching diag-
centre who had had one or more in-patient intensive
noses (based on age, degree and pattern of curvature)
courses of rehabilitation (with or without a brace; see
work together in groups. Evening and weekend social
figure 4) with the collective of Goldberg et al. [4], we
activities provide a sense of community and foster
followed their study design. All patients (girls only in
development of psychological support systems that our study, aged between 10 and 13 years) diagnosed
can be maintained after treatment is complete. with scoliosis of every kind of aetiology, and who had
The treatment programme consists of correction of attended our centre between 1993 and 1996, were
the scoliotic posture with the help of proprioceptive and included in this study. As not all patients in this
exteroceptive stimulation, and begins on the third day group had had repeated treatment at our centre or
after admission. Each weekday, after a 20-min group had had only out-patient appointments elsewhere,
warm-up session, the patients exercise in matched the study was conducted by means of a questionnaire.
groups for 2 h in the morning and 2 h in the afternoon Six hundred and eighty-nine patients were sent
and receive shorter more individual training sessions in questionnaires by mail. The number of responses

114
 Incidence of surgery in conservatively treated scoliosis patients

Figure 4 Patients exercising in special groups with similar curve pattern to learn how to control their curves by themselves. Mirror monitoring is of
importance to see and feel simultaneously the postural corrections.

was low—only 214 questionnaires were returned. We Table 1 View of diagnoses, Cobb angles and incidence of surgery in
could not, therefore, present any convincing results the whole collective for adolescent idiopathic scolioses (AIS), early-
onset scolioses (EOS), kyphotic scolioses (KS), congenital scolioses
with this group, and so all patients who failed to (CS) and other aetiologies
answer our questionnaire received a phone call and
Diagnosis n Cobb (
) SD OP (n) OP(%)
after contact were sent the questionnaire for a second
time. In this way we were able to reach 343 patients out AIS11—14 159 32.6 14.6 11 6.9
of the described group. All others had moved away AIS9/10 20 29.5 14.2 2 10.0
AISall 179 32.2 14.6 13 7.3
from their original addresses. EOS 106 35.2 21.8 22 20.8
The average angle of curvature according to Cobb KS 38 24.4 23.6 1 2.64
CS 9 52.6 38.7 2 22.2
before the first in-patient intensive course of rehabili- Others 11 37.1 13.1 3 27.3
tation was 33.4
(SD ¼ 18.9), which was comparable to
the angle of curvature evaluated by Goldberg et al. [4] Published with kind permission by Gentner Verlag, Stuttgart.
(33
). The distribution of the pattern of curvature was
as follows: Results
1. Thoracic scolioses: 35% Patients treated with a programme of intensive SIR
2. Double major scolioses: 37.2% showed an incidence of surgery of 11.95%; the patients
3. Lumbar scolioses: 10.2% in the Goldberg collective, however, showed an inci-
4. Thoracolumbar scolioses: 17.2% dence of 28.1%. The differences are statistically highly
Two hundred and forty-four (71%) of the patients significant.
studied wore braces. Of these, 235 also gave details of Out of the first 214 patients who sent back completed
how long they wore their braces for. Fifty-eight patients questionnaires, 11% had been treated surgically,
(25%) wore the brace for up to 8 h daily, 47 patients whereas out of the 343 patients who were finally
(20%) for up to 16 h, 54 patients (23%) 16–21 h, and included in the study, 41 had had surgical treatment.
76 patients (32%) more than 21 h daily. In surgically There was no statistical difference between the incidence
treated scoliosis patients (n ¼ 41) the angle of curvature rate reported in the questionnaires returned initially
according to Cobb was 50
(SD ¼ 25.4). Table 1 shows and the incidence of surgery in the whole collective.
the distribution of the whole collective into different A one-tailed t-test was used for the comparison of
types of scoliosis. two independent proportions [13]. Using this method

115
 H. Weiss et al.

of testing, highly significant differences were shown scoliosis present the less favourable form of scoliosis
(t ¼ 3.995, p<0.0001, a ¼ 0.05) by reference to compared with that of kyphoscoliosis.
Goldberg’s collective (no brace treatment, incidence of
surgery 28.1%, n ¼ 153). Comparing this value with the Discussion
Milwaukee-braced control group of Lonstein and
Winter [14] (incidence of surgery 22.4%, n ¼ 1020) a Scoliosis is a very variable condition as can be seen
highly significant difference was also evident in the attempts made to classify it according to the vari-
(t ¼ 4.7553, p<0.0001, a ¼ 0.05). ous categories. Clinically, patients present with a wide
The differences by reference to the control group spectrum of curves, particularly if they are referred from
with brace treatment as studied by Noonan et al. [15] a school screening programme, employed for early
(surgery incidence 31%, n ¼ 88) reveal a similarly high detection of potential problems in the UK and in the
significance. There was a probability of error of USA. This can lead to overloading of clinics with
p ¼ 0.0001 with a t-value of 3.6329. curves of minor magnitude, which probably do not need
treatment. Referrals to our centre, however, are from
When comparing the diagnosis-matched patient
spine centres, general orthopaedic surgeons, paediatric
groups (adolescent idiopathic scoliosis: Goldberg
physicians and general practitioners. Therefore the
n ¼ 153, incidence of surgery 28.1% vs Weiss n ¼ 179,
patient sample is pre-selected with bad prognosis.
incidence of surgery 7.3%) we found a t-value of 5.05
The comparison of the study group with a totally
with a probability of error of p<0.0001 ( ¼ 0.005). different population might seem inappropriate;
The incidence of surgery with regard to different Goldberg et al. [4], however, also compared groups
aetiologies is shown in table 1. Based on our experience from different populations (Ireland and USA) and
with more than 2800 cases of scoliosis per year, we from different decades (braced population from the
feel that adolescent idiopathic scoliosis should not be years 1950 to 1970 compared with a natural history
considered as a uniform condition. With this in mind, population without treatment from the 1990s). Our
we have been distinguishing over the past 10 years study group is from the same decade as the control
between early adolescent idiopathic scoliosis with its group and so is comparable to the Goldberg sample
first onset at the age of 9 or 10 years and later adoles- [4]. The comparison of patient samples out of different
cent idiopathic scoliosis beginning at the age of 11. For populations only seems to play a minor role and does
this reason we have made a subdivision in table 1, since not question the conclusions drawn [4].
in scolioses with early beginnings a more unfavourable In their introduction, Goldberg et al. [4] cited two
course of evolution has to be expected. references [11, 16] in which a good compliance of the
In comparing the early-onset scolioses in our own patients favoured the outcome of bracing. However,
collective (infantile idiopathic scoliosis and juvenile they omitted the important point that in the study by
idiopathic scoliosis) with the more benign patients Emans et al. [11] the actual extent of the corrective
having adolescent idiopathic scoliosis in the Goldberg effect is also described as an essential criterion in
collective, we found no significant differences (with a successful bracing. This is supported by the study by
probability of error of p ¼ 0.087). Had there been a Mellerowicz et al. [17] and the research done by
larger number of cases, however, the limit of signifi- Landauer [2]. The importance of the corrective effect
cance would also have been passed. is also confirmed by a review of the literature [10].
Patients with adolescent idiopathic scoliosis present The study by Nachemson and Peterson [1] also proves
the effectiveness of bracing. Certainly, Thulbourne and
an incidence of surgery of slightly over 7%. Patients
Gillespie [18] are right in saying that even if the progres-
with kyphoscoliosis show an incidence of surgery of
sion can be reduced by bracing, cosmetic appearance
2.6%. Thus, patients with adolescent idiopathic
and the rib hump may not always be influenced posi-
tively, nor may a successful course as shown by X-ray
always be appreciated as a successful treatment by the
Table 2 Distribution of pattern of curvature in idiopathic scoliosis
patient. Nevertheless, one must point out that neither
Thoracic Double Lumbar Thoracolumbar self-image nor other psychological factors can be
Diagnosis (%) major (%) (%) (%)
affected by surgical treatment either [19]. Bettany et al.
AIS 31 40.0 11.7 17.3 concluded that 6 months after surgery, levels of pain
EOS 32 37.7 13.2 17.9
and emotional problems connected with scoliosis
Published with kind permission by Gentner Verlag, Stuttgart. remain unaltered by most surgical techniques; and the

116
 Incidence of surgery in conservatively treated scoliosis patients

cosmetic results directly obtained by surgery are not remained clearly below the rate of incidence of surgery
necessarily stable later on either. One year after surgery in Goldberg et al. Thus no statistical test was required.
a good cosmetic initial result may deteriorate and a Distribution of the pattern of curvature of more than
reduced rib hump may become clearly evident again 70% thoracic, or double major curves (table 2), showed
[20]. Goldberg et al. [4] considered that not only has that the conservative treatment group was not the
the Cobb angle to be taken into account in cases of group with the most favourable history [24]. This is
surgical treatment, but cosmetic appearance and the also evident from a comparison of the angle of curva-
chances of obtaining a visible improvement have also ture in the control group with that of the treatment
to be appreciated. group.
It was interesting to note that the (female) patients in In comparing the collective of patients suffering
our collective who had undergone surgical treatment from early-onset scolioses (infantile idiopathic scoliosis,
had been mainly urged to do so by a physician. At juvenile idiopathic scoliosis) with Goldberg’s collective
least 30 patients out of 36 surgically treated patients suffering from adolescent idiopathic scoliosis, we found
underwent surgery because their physicians had advised no statistically significant differences. However, we did
them to do so and not because of any wish of their own. find important medical differences. Even the group of
Thus, it is seen that any decision to have surgical treat- patients with early-onset scolioses (a condition that
ment can also be influenced by the surgeons themselves. normally has the most unfavourable history) presented
We know from experience that, unfortunately, German better results thanks to conservative scoliosis treatment
patients with adolescent idiopathic scoliosis are nowa- than Goldberg et al.’s collective, which, as far as prog-
days not the only ones to be told that in the future nosis is concerned, was the more favourable group.
they will have to suffer pain and that they risk the Is the reason for the less favourable results obtained
collapse of their spines and will probably be forced in patients conservatively treated in the USA perhaps
into a wheelchair. Such patients are also often told due to the fact that there is not sufficient control of the
that their health is threatened by cor pulmonale. quality of treatment in the methods applied? In any
However, as early as 1969, Collis and Ponseti [21] case, it is obvious that conservative methods of treat-
reported that patients with idiopathic scoliosis do not ment should never be ruled out from scoliosis treatment
in fact suffer from back pain any more frequently than a because these can and do offer a viable alternative to
healthy control group. Pehrsson et al. [22] stated that those patients who cannot or do not wish to opt for
patients with adolescent idiopathic scoliosis do not in surgical treatment.
fact die earlier compared with a healthy control However, the study by Goldberg et al. [4], though
group—and so far there has been not a single case of suitable as a control group for our conservatively treat-
untreated adolescent idiopathic scoliosis recorded in ed group of patients, has to be examined more closely
the literature in which the patient ended up in a wheel- for its value as a positive statement. According to the
chair because of the curvature [23]. manual [25] used by Goldberg et al., the validity of
The collective of 343 patients that we studied showed statistical test methods should be approached more
a highly significantly lower incidence of surgery than the cautiously when confirming the H0-hypothesis (non-
untreated Goldberg group [4] or Lonstein and Winter’s significant differences). Distinction must be made
group [14] treated with the Milwaukee brace. Patients between the statistical statement itself and its medical
who had conservative treatment following our methods relevance. Assessment of this statistical data suggests
followed at least one in-patient intensive rehabilitation that the non-significant differences reported can in no
programme of 4–6 weeks and, if necessary, had brace way be regarded as proof of the ineffectiveness of the
treatment using the German quality criteria for brace method of treatment. The statement succeeds only in
treatment. proving that in that particular study and in the
Contrary to Goldberg et al. [4] in her study, we conditions described, no significant differences could
included patients in our collective who presented all be found. The statistical data of the statement are
kinds of aetiology and exclusively scolioses in girls therefore clearly distinct from any medical relevance
with a more unfavourable history in general [24]. The they may have to an evaluation of the results of the
level of incidence of surgery for all aetiologies remained study in general. Even when following this, the manual
below the level reported by Goldberg et al. [4] for [25], a different interpretation from that of Goldberg
patients with adolescent idiopathic scoliosis. The diag- et al. may be placed upon the outcome by highlighting
nosis-matched group (adolescent idiopathic scoliosis) the fact that medically relevant differences were
presented a surgery rate of slightly over 7%, and thus certainly found between the untreated control group,

117
 H. Weiss et al.

with an incidence of surgery of more than 28%, and the Zeitschrift für Orthopädie und ihre Grenzgebiete, 123: 323–337,
1985.
brace group treated with the Milwaukee brace with an 10. Weiss, H. R.: Standard der Orthesenversorgung in der
incidence of surgery of only 22.4%. This is reason Skoliosebehandlung. Medizinisch-Orthopädische Technik, 115:
enough to believe that treatment involving bracing 323–330, 1995.
11. Emans, J. B., Kaelin, A., Bancel, P. et al.: The Boston bracing
with a Milwaukee orthosis does indeed have some system for idiopathic scoliosis. Follow-up results in 295 patients.
effect. However, to conclude from a study that used Spine, 11: 792–801, 1986.
only one brace (outmoded by today’s standards!), that 12. Weiss, H. R., Rigo, M. and Che“ neau, J.: Praxis der Cheˆneau-
Korsettversorgung in der Skoliosetherapie (Stuttgart: Thieme),
bracing with any kind of bracing system should be 2000.
rejected, simply does not stand up against the scientific 13. Bland, M.: An Introduction to Medical Statistics (Oxford: Oxford
evidence. University Press), 1989.
14. Lonstein, J. E. and Winter, R. B.: The Milwaukee brace for
the treatment of adolescent idiopathic scoliosis. Review of 1020
Conclusions patients. Journal of Bone Joint Surgery [America], 76: 1207–1221,
1994.
Conservative methods of treatment with out-patient 15. Noonan, K. J., Weinstein, S. L., Jacobson, W. C. et al.: Use of
the Milwaukee brace for progressive idiopathic scoliosis. Journal
physiotherapy, in-patient intensive rehabilitation and of Bone Joint Surgery [Britain], 78: 557–567, 1996.
high-correction braces have clearly proven their effec- 16. Fernandez-Feliberti, R., Flynn, J., Ramirez, N. et al.:
tiveness in reducing the incidence of surgery in female Effectivevess of TLSO bracing in the conservative treatment of
idiopathic scoliosis. Journal for Pediatric Orthopedics, 15: 176–
patients with idiopathic scoliosis. 181, 1995.
17. Mellerowicz, H., Bo«ckel, T., Neff, G. and Frey, R.: Mittel-
und Langzeitergebnisse der Behandlung von lumbalen Skoliosen
Acknowledgement mit dem Boston-Brace. Paper presented at the 42nd Annual
The authors are thankful to Professor G. Walker, Cambridge, Meeting of the Vereinigung Süddeutscher Orthopäden e.V.,
for his editorial help. 28 April–1 May, Baden-Baden, Germany, 1994.
18. Thulbourne, T. and Gillespie, R.: The rib hump in idiopathic
scoliosis. Measurement, analysis and response to treatment.
References Journal of Bone Joint Surgery [Britain], 58: 64–71, 1976.
1. Nachemson, A. L. and Peterson, L. E.: Effectiveness of treat- 19. Bettany, J., Patridge, C. and Edgar, M.: Topographical,
ment with a brace in girls who have adolescent idiopathic scolio- kinesiological and psychological factors in the surgical man-
sis. Journal of Bone Joint Surgery [America], 77: 815–822, 1995. agement of adolescent idiopathic scoliosis. In: M. D’Amico, A.
2. Landauer, F.: Ist die Therapie mit dem Chêneau-Korsett Merolli and G. C. Santambrogio (editors) The Threedimensional
wirksam? In: A. Imhoff (editor) A (Hrsg) Fortbildung Analysis of Spinal Deformities (Amsterdam: IOS Press), pp. 321–
Orthopädie—Die ASG-Kurse der DGOT, Bd. 2: Wirbelsäule 326, 1995.
(Darmstadt: Steinkopff), pp. 31–38, 1999. 20. Bettany, J., Edgar, M., Wojick, A., Irwin, A., Maclean, I.
3. Goldberg, C. J., Dowling, F. E., Hall J. E. et al.: A statistical and Damato, M.: Are back shape changes following scoliosis
comparison between natural history of idiopathic scoliosis and surgery stable? A prospective three-year follow-up of five different
brace treatment in skeletally immature adolescent girls. Spine, surgical procedures. Paper presented at the European Spinal
18: 902–908, 1993. Deformities Society Meeting, 31 May–3 June, Birmingham, UK,
4. Goldberg, C. J., Moore, D. P., Fogarty, E. E. and Dowling, 1994.
F. E.: Adolescent idiopathic scoliosis: the effect of brace treatment 21. Collis, D. K. and Ponseti, I. V.: Long-term follow-up of
on the incidence of surgery. Spine, 26: 42–47, 2001. patients with idiopathic scoliosis not treated surgically. Journal
5. Bauer, R., Kerschbaumer, F. and Rehart, S.: Skoliose. In: C. J. of Bone Joint Surgery [America], 51: 425–445, 1969.
Wirth (editor) (Hrsg.) Praxis der Orthopädie, Bd. 1: Konservative 22. Pehrsson, K., Larsson, S., Oden A. et al.: Long-term follow-up
Orthopädie (Stuttgart: Thieme), pp. 498–502, 2001. of patients with untreated scoliosis. A study of mortality, causes
6. Weiss, H. R.: Letter to the editor. Spine, 18: 2058–2059, 2001. of death, and symptoms. Spine, 17: 1091–1096, 1992.
7. Hanks, G., Zimmer, B. and Nogi, J.: TLSO treatment of 23. Bradford, D. S., Tay, B. and Hus, S.: Adult scoliosis: surgical
idiopathic scoliosis—an analysis of the Wilmington jacket. indication; operative management, complications, and outcome.
Spine, 13: 626–629, 1988. Spine, 24: 2617–2629, 1999.
8. Blount, W. and Moe, J.: The Milwaukee-Brace, 2nd edn 24. Weinstein, S. L.: Natural history. Spine, 24: 2592–2600, 1999.
(Baltimore: Williams & Wilkins), 1980. 25. Daly, L. E. and Bourke, G. J.: Interpretation and Uses of
9. Heine, J. and Go« tze, H. G.: Endergebnisse der konservativen Medical Statistics, 5th edn (London: Blackwell Science),
Behandlung der Skoliose mit dem Milwaukee-Korsett. pp. 73–74, 2000.

118