J Child Orthop (2013) 7:25–28
DOI 10.1007/s11832-012-0459-2
CURRENT CONCEPT REVIEW
Classification of adolescent idiopathic scoliosis (AIS)
Dror Ovadia
Received: 27 June 2012 / Accepted: 20 October 2012 / Published online: 25 December 2012
 EPOS 2012
Abstract In 1983 Howard King presented his classifica-
tion system for adolescent idiopathic scoliosis (AIS) based
on the experience with Harrington rod instrumentation.
Curves were divided into five types and guidelines and
recommendations for which levels should be instrumented
were given to preserve motion as much as possible. As
segmental instrumentation systems began to gain favor
over the Harrington rods this system failed and led to the
development of a new classification system which was
presented by Lawrence Lenke in 2001. In order to define a
curve type by the Lenke classification, one must identify
the curve type, the lumbar modifier and, for the first time in
any classification system for scoliosis, the sagittal profile
was also included. The Lenke classification showed higher
inter and intra-reliability compared to the King classifica-
tion. It also provided a better and more reliable tool to
assist surgeons in choosing the best method of treatment
for each curve pattern. Although the Lenke classification is
more comprehensive and reliable than the King classifi-
cation it is still far from perfect. It does not address the
rotational component of the deformity. New technologies
which provide three-dimensional (3D) reconstruction of the
spine may serve as a basis for a truly 3D classification of
scoliosis and for new therapeutic concepts.
Keywords Adolescent idiopathic scoliosis

Classification
King
Lenke
D. Ovadia (&)
Department of Pediatric Orthopaedics,
Dana Children’s Hospital, Tel Aviv Medical Center,
6 Weizmann Street, Tel Aviv, Israel
e-mail: dovadia@tasmc.health.gov.il
In 1948 John Cobb was the first to describe a classification
system of scoliosis. He was the first to give a description of
major and minor curves, structural and non-structural curves
and set guidelines for treating these deformities accordingly.
In 1983 Howard King presented his classification system
for AIS [1]. This system was based on the experience of
John Moe in the surgical treatment of AIS patients with
Harrington rod instrumentation. Curves were divided into
five types and guidelines and recommendations for which
levels should be instrumented were given according to
those different curve types in order to preserve motion as
much as possible. King et al., gave, for the first time, some
very important definitions which some are still widely used
today: (1) Stable vertebra: the vertebra most closely
bisected by the center sacral vertical line (CSVL). (2)
Structural versus Compensatory curves: the curves were
defined according to their flexibility on side bending films.
King and Moe defined five curve types (Fig. 1):
• Type 1: an ‘‘S’’ shape deformity, in which both curves
are structural and cross the CSVL, with the lumbar
curve being larger than the thoracic one.
• Type 2: an ‘‘S’’ shape deformity, in which both curves
are structural and cross the CSVL, with the thoracic
curve being larger or equal to the lumbar one.
• Type 3: major thoracic curve in which only the thoracic
curve is structural and crosses the CSVL.
• Type 4: long ‘‘C’’ shape thoracic curve in which the
fifth lumbar vertebra is centered over the sacrum and
the forth lumbar vertebra is tilted into the thoracic
curve.
• Type 5: double thoracic curve.
As segmental instrumentation systems began to gain
favor over the Harrington rods this classification system
failed to give accurate and reliable guidelines for choosing
123
26 J Child Orthop (2013) 7:25–28

Fig. 1 The King classification

published 1983 [1]

the proper levels for fusion [2–4]. In addition, several CVSL on the apical vertebra of the lumbar curve. Lenke
articles showed low inter and intra-observer reliability of defined three lumbar modifiers:
this classification system [5, 6].
• Modifier A: when the CSVL runs between the pedicles
This lead to the development of a new classification
of the lumbar apical vertebra.
system which was presented by Lawrence Lenke in 2001
• Modifier B: when the CSVL runs between the medial
[7]. In order to define a curve type by the Lenke classifi-
border of the lumbar concave pedicle and the lateral
cation, one must identify the curve type, the lumbar mod-
margin of the apical vertebral body.
ifier and, for the first time in any classification system for
• Modifier C: when the CSVL falls completely medially
scoliosis, the sagittal profile was also included. A few new
to concave apical vertebra’s body.
definitions were made: (1) Major curve: the curve of
greatest magnitude and is always structural. (2) Minor As mentioned, a thoracic sagittal profile modifier is also
curve: a smaller curve which may be structural or non- included in the Lenke classification. Thoracic kyphosis is
structural. (3) Nonstructural curve: a curve which bends to measured between T5–T12 and defined as:
less than 25 on side bending radiographs.
• ? (plus): when thoracic kyphosis measures [40.
According to these definitions there are six different
• N (normal): when thoracic kyphosis measures between
curve types (Fig. 2):
10 and 40.
• Type 1: main thoracic (MT) is the only structural curve • - (minus): when thoracic kyphosis measures \10.
while the others (proximal thoracic and lumbar or
The Lenke classification showed higher inter and inta-
thoracolumbar) are nonstructural.
reliability compared to the King classification [7]. It also
• Type 2: double thoracic in which the MT is the major
provided a better and more reliable tool to assist surgeons in
curve, the proximal thoracic (PT) is the minor curve but
choosing the best method of treatment for each curve pattern.
is structural and the thoracolumbar (TL) or lumbar
Lenke curves types 1 and 5 can be treated either anteriorly or
(L) curves are minor and nonstructural.
posteriorly. Lenke types 2, 3, 4 and 6 should be treated
• Type 3: double major curve pattern in which the MT is
completely posteriorly. In patients with lumbar modifiers A
the major curve and the lumbar is the minor one but is
or B a selective thoracic fusion is advocated in order to avoid
structural whereas the PT is nonstructural.
as much as possible the fusion of lumbar vertebrae.
• Type 4: triple major curve pattern when the MT is the
Although the Lenke classification is more comprehen-
major curve but all three curves are structural.
sive and reliable than the King classification, it is still far
• Type 5: the TL or L curve is the major and only
from perfect. Surgeons still deviate 15 % of the time from
structural curve, with the PT and\or MT curves being
the algorithm given for selecting fusion levels [8] and the
minor and nonstructural.
classification failed to address the rotational component of
• Type 6: the TL or L curve is the major curve measuring
the deformity. Recently, investigators offered stereoradio-
at least 5 more than the MT which is the minor but
graphic measurements of spinal deformity [9, 10]. The
structural curve.
measurements utilized for the analysis were the Cobb
To these basic six curve types the lumbar spine modifier angle, apical vertebra, axial rotation of the apical vertebra
is added. This modifier is defined by the location of the and the orientation of the apical vertebra with respect of the

123
J Child Orthop (2013) 7:25–28
Fig. 2 The Lenke classification
published 2001 [7]
sagittal plane. This allows assessment of the deformity in
the coronal, sagittal and axial planes. New technologies
such as the EOS system, which provides 3D reconstruction
of the spine may serve as a basis for a truly 3D classifi-
cation of scoliosis as a basis for new concepts in thera-
peutic procedures [11].
Acknowledgments The author thanks Prof. Fritz Hefti, Basel/
Switzerland for having contributed both figures.
References
1. King HA, Moe JH, Bradford DS, Winter RB (1983) The selection
of fusion levels in thoracic idiopathic scoliosis. J Bone Joint Surg
Am 65:1302–1313
27
2. Richards BS (1992) Lumbar curve response in type II idiopathic
scoliosis after posterior instrumentation of the thoracic curve.
Spine 17:S282–S286
3. Roye DP Jr, Farcy JP, Rickert JB, Godfried D (1992) Results of
spinal instrumentation of adolescent idiopathic scoliosis by King
type. Spine 17:S270–S273
4. Knapp DR Jr, Price CT, Jones ET, Coonrad RW, Flynn JC (1992)
Choosing fusion levels in progressive thoracic idiopathic scoli-
osis. Spine 17:1159–1165
5. Cummings RJ, Loveless EA, Campbell J, Samelson S, Mazur JM
(1998) Inter-observer reliability and intra-observer reproducibil-
ity of the system of King et al. for the classification of adolescent
idiopathic scoliosis. J Bone Joint Surg Am 80:1107–1111
6. Behensky H, Giesinger K, Ogon M et al (2002) Multisurgeon
assessment of coronal pattern classification systems for adoles-
cent idiopathic scoliosis: Reliability and error analysis. Spine
27:762–767
7. Lenke LG, Betz RR, Haher TR et al (2001) Multisurgeon
assessment of surgical decision-making in adolescent idiopathic
123
28 J Child Orthop (2013) 7:25–28

scoliosis: Curve classification, operative approach, and fusion 10. Sangole AP, Aubin CE, Labelle H et al (2009) Three-dimensional
levels. Spine 26:2347–2353 classification of thoracic scoliosis curves. Spine 34:91–99
8. Newton PO, Faro FD, Lenke LG et al (2003) Factors involved in 11. Illés T, Tunyogi-Csapó M, Somoskeöy S (2011) Breakthrough in
the decision to perform a selective versus nonselective fusion for three-dimensional scoliosis diagnosis: significance of horizontal
Lenke 1B and 1C (King-Moe II) curves in adolescent idiopathic plane view and vertebra vectors. Eur Spine J 20:135–143
scoliosis. Spine 28:S217–S223
9. Stokes IA, Sangole AP, Aubin CE (2009) Classification of
scoliosis deformity three-dimensional spinal shape by cluster
analysis. Spine 34:584–590

123