Introducing the Early Onset Scoliosis Classification System
Purpose: Early Onset Scoliosis (EOS) is a complex and debilitating pathology, the care for which is hindered by a lack of standardization of management and low level of evidence supporting existing treatment strategies. Classification systems are utilized widely to rectify these issues by providing a framework to simplify communication between providers and facilitate the research process. Unfortunately, classification systems developed within the surgical field are often only representative of individual opinion and rarely validated. Given the complexity of the EOS population, we sought to develop an EOS classification system driven by consensus opinion from a large panel of experienced surgeons.
Methods: 14 senior members of the Chest Wall & Spine Deformity Study Group and the Growing Spine Study Group took part in an extended Nominal Group Technique consisting of 2 group meetings and iterative surveying. Participants reviewed project goals, proposed a draft classification structure based on literature review findings, and created a list of potential variables. All identified variables were rated on a 3-point Likert scale, and then ranked based on content validity ratio and sum of ranks. Consensus was achieved regarding overall classification structure, final variables, and number of variable subgroups and cut points. 30 EOS patient cases were presented to the study group in an online survey and inter-observer reliability analysis was performed.
Results: Thirteen potential variables were identified. The highest ranking variables in order were: Etiology, Cobb angle, Kyphosis, Age, and Curve Progression. Open ended survey responses and group discussions indicated that no more than 4 variables should be included in the classification. Further discussion determined that Curve Progression should be included as an optional modifier. Figure 1 displays the subgroups and cut points selected based on consensus voting. Kappa analysis of reliability testing revealed near perfect agreement for Cobb Angle and Kyphosis, substantial agreement for Etiology, and moderate agreement for Curve Progression.
Conclusion: Utilizing a formalized consensus building process among a large group of experienced pediatric spine surgeons, a novel Classification System for the field or Early Onset Scoliosis has been developed. Continuing efforts will seek to evaluate the inter and intra-observer reliability of this system in a group of individuals with varying levels of experience with EOS patients and to assess the ability of the classification to stratify patients based on specific outcomes by reviewing real patient data from prospectively gathered EOS databases.