Please click here to receive a FREE SCOLIOSIS TREATMENT INFORMATION KIT ASAP.
Scoliosis generally experiences it's most rapid curve progression during a patients adolescents until they reach skeletal maturity. Often parents and patients are told that curvature won't continue to progress after that point. Unfortunately, that generally isn't the case and many adolescent idiopathic scoliosis patients do experience continued curve progression throughout life (average 1-3 degrees per year). This is particularly true in cases in which the curvature was over 25 degrees when the patient reached skeletal maturity (as discussed in the following study). It is absolutely imperative that a patient engaged in an effective early stage scoliosis intervention program that works to keep the spinal curvature under 20 degrees while the patient is still growing and allows them to reach skeletal maturity with the curvature measuring less than 25 degrees.
Curve Progression in Idiopathic Scoliosis
Follow-up Study to Skeletal Maturity
Ken-Jin Tan, MBBS, MMed, MRCS,* Maung Maung Moe, MBBS,*
Rose Vaithinathan, MBBS, MMed,† and Hee-Kit Wong, MBBS, FRCS*
Study Design. This is a follow-up study to skeletal maturity
on a cohort of students screened for a 1-year prospective
epidemiological prevalence study for scoliosis.
Objectives. This study aims to identify the prognostic
factors for curve progression to a magnitude of 30° at
skeletal maturity in skeletally immature patients with adolescent
idiopathic scoliosis.
Summary of Background Data. The natural history of
idiopathic scoliosis is not well understood. Previous reports
have focused on the characteristics of curve progression
where progression has been predefined at specific angles of
5° to 6°. However, the absolute curve magnitude at skeletal
maturity is more predictive of long-term curve behavior
rather than curve progression of a defined magnitude over
shorter periods of skeletal growth. It is generally agreed that
curves less than 30° are highly unlikely to progress after
skeletal maturity. Hence, defining the factors that influence
curve progression to an absolute magnitude of more than
30° at skeletal maturity would more significantly aid clinical
practice.
Methods. One hundred eighty-six patients who fulfilled
the study criteria were selected from an initial 279 patients
with idiopathic scoliosis detected by school screening, and
who were followed-up till skeletal maturity. The initial age,
gender, pubertal status, and initial curve magnitude were
used as risk factors to predict the probability of curve progression
to more than 30° at skeletal maturity.
Results. Curve magnitude at first presentation was the
most important predictive factor for curve progression to a
magnitude of more than 30° at skeletal maturity. An initial
Cobb angle of 25° had the best receiver-operating characteristic
of 0.80 with a positive predictive value of 68.4% and
a negative predictive value of 91.9% for curve progression
to 30° or more at skeletal maturity.
Conclusion. Initial Cobb angle magnitude is the most
important predictor of long-term curve progression and behavior
past skeletal maturity. We suggest an initial Cobb
angle of 25° as an important threshold magnitude for longterm
curve progression. Initial age, gender, and pubertal
status were less important prognostic factors in our study.
Key words: adolescent idiopathic scoliosis, natural
history, Cobb angle, curve progression, skeletal maturity.
Spine 2009;34:697–700
Knowledge of the natural history and factors that influence
curve behavior over time is critical in the evaluation,
prognostication, and management of patients with scoliosis.
Particularly, in the case of school-based screening
programs, the surgeon is often faced with the management
of a large number of patients many of whom have
relatively small curves at presentation.
There have been relatively few studies on the natural
history of adolescent idiopathic scoliosis, and these have
largely been limited to the analysis of defined units of
curve progression.1–3 Some factors that have been identified
to be related to progression include the magnitude
of the curve, the patient's age at presentation, the Risser
sign, and the patient's menarchal status.1–3
Although previous authors have identified the risk of
curve progression by a stipulated magnitude of usually
5° and the associated factors, this does not allow the
physician to predict the longer-term behavior of the
curve at skeletal maturity and into adulthood.
An understanding of curve behavior into skeletal maturity
and adulthood is important as it is now established
that curves because of idiopathic scoliosis do not necessarily
stop progressing after skeletal maturity. In a longterm
follow-up study of patients with idiopathic scoliosis,
Collis and Ponseti4 found that curves of larger degree
did increase after skeletal maturity. In a separate study
with an average follow-up of 40 years, Weinstein and
Ponseti5,6 also found that a significant number of idiopathic
curves increased after skeletal maturity. They reported
that in thoracic curves, the Cobb angle, apical
vertebral rotation, and the Mehta angle were important
prognostic factors. For lumbar curves, the degree of apical
vertebral rotation, the Cobb angle, the direction of
the curve, and the relationship of the fifth lumbar vertebra
to the intercrest line were of prognostic value. However,
they also observed that curves that were less than
30° at skeletal maturity tended not to progress regardless
of curve pattern.
In this study, we report on a group of 279 patients
with idiopathic scoliosis detected by school screening,
and who were followed-up until skeletal maturity. One
hundred eighty-six patients fulfilled the study criteria
and for these adolescents, the initial age, gender, pubertal
status, and initial curve magnitude were used as risk
factors to predict the probability of curve progression to
more than 30° at skeletal maturity.
Materials and Methods
Our study population was derived from a cohort of school
children screened for scoliosis in 1997 the results of which were
From the *Department of Orthopaedic Surgery, National University
Hospital, National University of Singapore, Singapore; and †Health
Promotion Board, School Health Service, Singapore.
Acknowledgment date: May 12, 2008. Acceptance date: October 21,
2008.
The manuscript submitted does not contain information about medical
device(s)/drug(s).
No funds were received in support of this work. No benefits in any
form have been or will be received from a commercial party related
directly or indirectly to the subject of this manuscript.
Address correspondence and reprint requests to Ken-Jin Tan, MBBS,
MMed, MRCS, Department of Orthopaedic Surgery, National University
Hospital, 5 Lower Kent Ridge Road, Main Building 1, Singapore
119074, Singapore; E-mail: This e-mail address is being protected from spambots. You need JavaScript enabled to view it
697
previously reported.7 In this study, 72,699 children aged 6 to
14 years out of a total enrolment of 152,000 schoolchildren
were screened for a 1-year prospective epidemiological prevalence
study for scoliosis. Of the 72,699 screened, 279 were
diagnosed to have idiopathic scoliosis on the basis of a single
standing radiograph, which showed a Cobb angle equal to or
greater than 10°. Of the 279, 17 were either uncontactable or
did not attend more than once.
The selection criteria for the present study on curve progression
were that the adolescent must be skeletally immature at the
beginning of the study and be skeletally mature at the last
follow-up appointment. Skeletal maturity was defined as an
age of 15 years or a Risser grade of 4 or 5.5 Patients with
incomplete data on maturity or on factors studied were also
excluded.
Of the 262 patients, 36 patients were skeletally mature at
presentation and excluded from the study. The remaining 226
patients were observed-up at regular intervals ranging from 3
to 6 months over the following years. The follow-up period
ranged from 1 to 8 years. During the period of follow-up, some
patients were referred for bracing or surgical fusion of the
curve. Their final Cobb angle at the last follow-up appointment
was used for the study.
At the end of the study, 9 patients had not attained skeletal
maturity and 31 patients had incomplete data or were lost to
follow-up. These patients were excluded from the study. The
remaining 186 patients formed the study group.
84.9% of the patients were girls, with 37.3% being of prepubertal
status. The initial age of the patients ranged from 7 to
14 years of age. The median age was 12 years with a mean
(standard deviation) of 12 (1.50) years with 34.9% less than 12
years of age. Of the 186 patients, 38 (20.4%) progressed to a
Cobb angle of 30° or more at skeletal maturity.
The study parameters analyzed were the age at first appointment,
gender, initial or presentation Cobb angle and
initial pubertal status. Both univariate and multivariate
analyses were performed. The 2 test was used for univariate
analysis and a logistic regression model was used for multivariate
analysis. Receiver operating characteristic (ROC)
analysis was used to determine the optimal cut-off for Cobb
angle at presentation.
Results
ROC analysis demonstrated that the Cobb angle at
presentation was the single most important factor that
predicted further curve progression to a Cobb angle of
30° or more at skeletal maturity. Age, pubertal status,
and gender were not found to be of predictive value
(Figure 1).
Furthermore, a Cobb angle of 25° at presentation had
the best overall predictive value for curve progression to
a Cobb angle of 30° or more at skeletal maturity. The
ROC value for a Cobb angle of 25° at presentation was
0.80. This was associated with a positive predictive value
of 68.4% and a negative predictive value of 91.9%for
curve progression to 30° or more at skeletal maturity
(Figure 2). These findings on ROC analysis closely
agreed with the results of univariate and multivariate
analysis.
Using a cut-off presentation Cobb angle of 25° as
determined by ROC analysis, 68.4% of patients with a
Cobb angle of 25° or more at presentation progressed
to a final Cobb angle of 30° or more. This compared
with 8.1% of patients with a Cobb angle of less than
25° at presentation. This difference was highly significant
on both univariate and multivariate analyses
(P 0.001).
Patients who had an initial age less than 12 years
(26.2%) compared with patients aged 12 years and
above (17.4%) had a final Cobb angle of 30° or more.
This difference was not found to be of significance
(P 0.158).
Relatively similar percentages, 23.0% (17 patients)
who were prepubertal initially and, 18.8% (21 patients)
who had attained puberty initially, progressed to a final
Cobb angle of 30° or more (P 0.485).
The difference in gender between the patients that
progressed and did not progress to a final Cobb angle
of 30° was slightly more pronounced than that for age
Figure 1. ROC plots for the various factors analyzed. The Cobb
angle at presentation was the single factor that had an ROC curve
that was significantly different from the reference line. The ROC
plots for age, pubertal status, and gender all lie close to the
reference line.
Figure 2. The ROC plots for 16° to 29° cut-offs for the presentation
Cobb angle are presented here. The 25° cut-off had the
highest ROC.
698 Spine • Volume 34 • Number 7 • 2009
and pubertal status. 22.8% of girls compared with
7.1% of boys progressed to a final Cobb angle of 30°
or more. This difference was close to statistical significance
(P 0.076). Table 1 shows the univariate and
multivariate analyses of age being less than 12 years,
gender, pubertal status, and Cobb angle at presentation
on curve magnitude of 30° or more at skeletal
maturity.
When factors were combined and analyzed by logistic
regression, we found that a prepubertal girl less than 12
years of age with a Cobb angle of 25° or more at presentation
had the highest chance of progression (82.23%) to
a Cobb angle of 30° or more (Table 2).
On the other hand, a boy who was postpubertal
with an age of 12 years or more and an initial Cobb
angle of less than 25° was least likely to progress to a
Cobb angle of 30°. This probability was only 2.39%.
Conversely, the same male patient would have a
97.61% probability of his curve not progressing to a
Cobb angle of 30° or more (Table 2).
For logistic regression analysis, the probability of progression
to a final Cobb angle of 30° or more was calculated
by the following equation.
Probability of final Cobb angle 30 1/
(1 exp (z))
where z 3.709 0.931 (gender) 0.825 (pubertal
status) 3.314 (initial Cobb angle) 0.171 (age at
presentation).
Where girl 1, boy 0; before puberty 1, after
puberty 0, initial Cobb angle 25 0, initial Cobb
angle 25 1; age at presentation 12 0, age at
presentation 12 1.
Discussion
The care of adolescents with mild curves because of idiopathic
scoliosis is a constant clinical dilemma. Management
decisions should ideally be based on accurate
prediction of long-term curve behavior and not on risk
for curve progression of a defined magnitude over
shorter durations of skeletal growth.
Studies on curve progression include those by Lonstein
and Carlson and Soucacos et al, which examined
for curve progression as defined by a limited increase in
curve magnitude of 5 to 6°.1,2 Lonstein and Carlson1
found that the strongest predictive factors for curve progression
were the curve magnitude, the Risser sign and
the chronological age. Similarly, Soucacos et al observed
a strong association between the incidence of curve progression
and sex of the child, curve pattern, maturity
(pubertal status), age, and curve magnitude.2
In addition, a recent French study, which retrospectively
reviewed 205 patients with idiopathic scoliosis,
found that curve pattern, Cobb angle at onset of puberty
and curve progression velocity to be strong prognostic
factors of curve progression during pubertal growth.3 In
the above study, the authors found that juvenile scoliosis
with magnitude more than 30° increased rapidly during
pubertal growth and had a 100% prognosis for surgery.3
Weinstein and Ponseti, after following-up 102 patients
over more than 4 decades, reported that curves
that were less than 30° at skeletal maturity tended not to
progress regardless of curve pattern. This was unlike
curves with a magnitude above 30°, which had a higher
Final Cobb refers to the Cobb angle at skeletal maturity.
Curve Progression in Idiopathic Scoliosis • Tan et al 699
tendency to progress, even after skeletal maturity. In this
study, skeletal maturity was defined as a Risser grade of
4 or 5.7
Hence, it would be more useful to have data on the
risk of curve progression to a magnitude of 30° or more
and the factors associated with this. This information
would enable a more accurate prediction of curve behavior
in the long-term and at the same time reducing the
number of patients requiring follow-up. Available data
suggests that curves with a magnitude of less than 30° at
skeletal maturity do not continue to progress, whereas
those of magnitude 30° or more behave differently and
have a propensity for further progression. Our study
aims to address this need by following-up adolescents
with idiopathic scoliosis detected in our national screening
program to skeletal maturity and analyzing the progression
risk and associated risk factors.
The main prognostic factor that we identified in our
study was curve magnitude at presentation. Age, gender,
and pubertal status had much less prognostic
value than curve magnitude. There has been no agreement
as to the most important factors that influence
curve progression in previous studies. However, age,
gender, and pubertal status have previously been
found to be important factors.
A possible reason why age, gender, and pubertal status
were much less important in our study could be the
different aspect of curve behavior studied. Therefore, it is
likely that the factors involved would reflect the growth
potential of the patient during the likely shorter period of
curve progression of 5° to 6°. However, we followed-up
the patients' curves to skeletal maturity regardless of the
magnitude of curve progression. Hence, prognostic factors
in our study may be less dependent on shorter periods
of curve progression but instead be more reflective of
the risk of final curve progression to a defined threshold
magnitude of 30°.
Our findings of a critical Cobb angle of 25° at first
presentation suggests that regardless of previous curve
magnitude or curve progression during skeletal growth,
the absolute value of the curve at presentation is the most
important factor in long-term prognostication. Our logistic
regression model also identified that a Cobb angle
at presentation of 25° or more when combined with a
girl less than 12 years of age and prepubertal status gave
the highest risk for curve progression to a Cobb angle of
30° or more at skeletal maturity. This was in contrast to
a boy who was postpubertal with an age of 12 years or
more and a Cobb angle at presentation of less than 25°.
This profile was associated with only a 2.39% probability
of the curve progressing to 30° or more at skeletal
maturity.
Key Points
● Cobb angle on initial presentation is the most
important predictor of long-term curve progression.
● Curves with a Cobb angle of 25° or more have a
68.4% probability of progressing to 30° or more at
skeletal maturity. On the other hand, curves with a
Cobb angle of less than 25° have a 91.9% probability
of not progressing to 30° or more at skeletal
maturity.
● A prepubertal girl less than 12 years of age with
a Cobb angle of 25° or more at presentation had an
82.23% chance of progression to a Cobb angle of
30° or more.
● Conversely, a postpubertal boy, 12 years old or
older, with a Cobb angle of less than 25° at presentation
had only a 2.39% chance of progression to a
Cobb angle of 30° or more.
References
1. Lonstein JE, Carlson JM. The prediction of curve progression in untreated
idiopathic scoliosis during growth. J Bone Joint Surg Am 1984;66:1061–71.
2. Soucacos PN, Zacharis K, Gelalis J, et al. Assessment of curve progression in
idiopathic scoliosis. Eur Spine J 1998;7:270–77.
3. Charles YP, Daures JP, Rosa VD, et al. Progression risk of idiopathic juvenile
scoliosis during pubertal growth. Spine 2006;31:1933–42.
4. Collis DK, Ponseti IV. Long-term follow-up of patients with idiopathic scoliosis
not treated surgically. J Bone Joint Surg Am 1969;51:425–45.
5. Weinstein SL, Ponseti IV. Curve progression in idiopathic scoliosis. J Bone
Joint Surg Am 1983;65:447–55.
6. Weinstein SL, DC Zavala, Ponseti IV. Idiopathic scoliosis: long-term follow-
up and prognosis in untreated patients. J Bone Joint Surg Am 1981;63:
702–12.
7. Wong HK, Hui JH, Rajan U, et al. Idiopathic scoliosis in Singapore schoolchildren:
a prevalence study 15 years into the screening program. Spine 2005;
30:1188–96.
700 Spine • Volume 34 • Number 7 • 2009