Classification of Congenital Scoliosis and Kyphosis: A New Approach to the Three-dimensional Classification for Progressive Vertebral Anomalies Requiring Operative Treatment

Noriaki Kawakami, MD; Taichi Tsuji, MD; Shiro Imagama, MD; Lawrence G. Lenke, MD; Rolando M. Puno, MD; Timothy R. Kuklo, MD


Spine. 2009;34(17):1756-1765. 

In This Article

Abstract and Introduction


Study Design: We reviewed three-dimensional (3D) computed tomography (CT) images of congenital spinal deformities and proposed a new classification based on the information obtained.
Objectives: The purposes of this article were to clearly illustrate the limitations of two-dimensional classification, to summarize the clinical significance of 3D analysis of congenital vertebral anomalies, and to propose a new 3D classification of congenital vertebral anomalies.
Summary of Background Data: The classification of congenital scoliosis or kyphosis were based on radiographic findings of plain radiograph images of congenital vertebral anomalies, it is sometimes difficult in classifying the large variety of anomalous vertebrae or severely twisted 3D curves.
Methods: Three-dimensional CT images of more than 150 patients with congenital spinal deformities were analyzed and compared with plain radiograph images. By developing the algorithm for the evaluation of malformed vertebrae in terms of numbers of abnormal vertebrae, type of formation failure, and type of segmentation failure in separate steps, we attempted to revise the classification of congenital spinal deformities.
Results: The images of plain radiograph cannot demonstrate the spatial relationship of each structure of the vertebrae. Three-dimensional findings in congenital-deformed vertebrae included several types of laminae and clearer definitions of each type of anomalous vertebrae. By developing an algorithm for the evaluation of congenital spinal deformity, congenital spinal deformity could be mainly classified into 4 types of congenital vertebral abnormalities: Type 1: solitary simple, Type 2: multiple simple, Type 3: complex, Type 4: segmentation failure.
Conclusion: The large volume of information that can be obtained by evaluating 3D CT images of congenitally deformed vertebrae can be a great help in developing a strategy for surgical treatment. We need to develop a new classification of congenital scoliosis based on the perspective of 3D imaging to understand the etiology and embryology, as well as to determine an operative strategy.


There are many factors by which congenital scoliosis and kyphosis can be classified. One of the distinct differences of the vertebral anomalies in congenital spinal deformities is whether or not there are defects of lamina in some of the anomalous vertebrae. Radiographically and clinically, it is very simple and useful to divide them into closed vertebral defects and open spine lesions according to these differences.

Open spine lesions are clearly distinct in morphology from closed defects. This remarkable difference may be due to the differences of initial inductive influences during development of the vertebral bodies and arches, which confer a degree of independence in their initial developmental pattern. This article focuses on the closed vertebral defects similar to those reported in classifications of congenital scoliosis and kyphosis. From the morphologic viewpoint, closed vertebral defects demonstrate great variation in the number of vertebrae involved and patterns. Winter et al[1,2] and McMaster et al[3,4] presented independent classifications of congenital scoliosis and kyphosis. Both classifications, whether they were for scoliosis or kyphosis, were based on the same concept of embryologic maldevelopment of the spine, that is, abnormalities of formation and segmentation. Winter[1] classified congenital scoliosis into failure of formation, failure of segmentation, and the mixed-type. This classifications has been widely used to describe the type of congenital abnormal vertebrae for more than 20 years, However, because all these classifications were based on the radiographic findings of front and lateral, plain x-ray images of congenital vertebral anomalies, they sometimes result in difficulty in classifying them when severe twisted three-dimensional (3D) curves are present, and might result in wrong decision-making during strategic planning of congenital spinal deformities.

We have analyzed the 3D computed tomography (CT) images of more than 150 patients with congenital spinal deformities and have described the results of this analysis in several reports.[5–7] Based on these 3D CT findings and a review of literature related to the classification of congenital spinal deformity, this review aims to (1) more clearly illustrate the problems of two-dimensional (2D) classification, (2) emphasize the clinical significance of 3D analysis of congenital vertebral anomalies, and (3) propose a new classification system for congenital vertebral anomalies, which might allow us to develop a more logical approach to the surgical strategy as well as more clearly understand the natural history and etiology.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: