26.1 Overcoming Obstacles to the Study of Early Onset Scoliosis

Early onset scoliosis is a rare diagnosis that affects a heterogeneous patient population; in regard to etiology, scoliosis may be related to congenital, neuromuscular, syndromic, or idiopathic conditions. The natural history of this disease spans decades and can lead to devastating pulmonary outcomes. Arguably the most important conceptual development in the field of early onset scoliosis was the description of thoracic insufficiency syndrome. In a milestone paper, Campbell et al in 2003 defined thoracic insufficiency syndrome as the inability of the thorax to support normal respiration or lung growth. ¹ This article formalized the concept of spine and thoracic developmental interdependence, and it altered the considerations for the treatment of early onset scoliosis. It led to studies investigating the radiographic evaluation of thoracic and spinal deformity, emphasized the importance of pulmonary outcomes, and contributed to new treatments for early onset scoliosis.

Because of the rarity and heterogeneity of early onset scoliosis, and the nature of its course, there exist significant obstacles to undertaking meaningful research. Furthermore, the treatment methods are evolving simultaneously with the elucidation of disease characteristics, so that another layer of complexity is added to the study of early onset scoliosis.

To overcome these obstacles, two study groups based in North America have been formed to study early onset scoliosis: the Growing Spine Study Group (GSSG) and the Chest Wall and Spine Deformity (CW&SD) Study Group. Founded in 2002 and 2005, respectively, both study groups maintain national registries including data from nearly 50 medical centers serving the population of patients with early onset scoliosis. Patients are prospectively enrolled with the collection of de-identified demographic data, such as age and sex, as well as clinical data, such as radiographs and complications. The patients’ courses are tracked over time, and data are entered into the registry for potential inclusion into ongoing prospective or future retrospective studies. As a result of the establishment and maintenance of these two national registries, research efforts by their members have produced significant advances, not the least of which is an identification of the methodologic challenges to producing high-level evidence for early onset scoliosis. These challenges include a lack of the following:

• 
  

  A disease-specific measure of patient outcome;

  
  
• 
  

  An evaluation of clinical equipoise;

  
  
• 
  

  A classification system that delineates patient subgroups for communication.

To address the need for an outcome measure specific for early onset scoliosis, a quality-of-life assessment was developed with contributions from sites across North America. The need for such an outcome measure was identified through studies that used existing generic assessments to examine the quality of life of patients undergoing surgery for early onset scoliosis. ² In addition to these studies, literature review and expert opinion yielded elements of an assessment specific for early onset scoliosis that were presented to physicians, patients, and their caregivers. Based on their input, and after multiple iterations, the Early Onset Scoliosis 24-Item Questionnaire (EOSQ-24) was produced. The EOSQ-24 covers 11 domains, including General Health, Pain / Discomfort, Physical Function / Transfer, Pulmonary Function, Daily Living, Fatigue / Energy Level, Emotion, Parental Burden, Financial Burden, Patient Satisfaction, and Parent Satisfaction, which can produce an overall and domain-specific scaled score ranging from 0 to 100. Through validation studies, the EOSQ-24 has been shown to be sensitive to quality-of-life and burden-of-care issues relevant to early onset scoliosis. Currently and in the future, investigators will be able to employ the EOSQ-24 as a robust outcome measure in studies of early onset scoliosis. ^{3 ,​ 4}

To evaluate clinical equipoise, a formal study of members of both the GSSG and the CW&SD Study Group was conducted to identify areas of uncertainty among providers of treatment for early onset scoliosis. Case scenarios were presented in a treatment preference survey to 11 surgeons, who proposed treatment options for each case at two points in time separated by a 6-month interval. Numerous scenarios were identified as areas of equipoise, including the optimal management of patients who have completed their lengthening course, the ideal lengthening intervals, and the comparative advantages of using rib- or spine-based proximal fixation. ⁵ This study was the basis for a prospective study of rib- vs. spine-based proximal anchors in which the resources of the GSSG and the CW&SD Study Group were used as conduits for research.

To facilitate communication and collaboration among providers of treatment for patients with early onset scoliosis, multiple classification schemes have been developed, including the Classification for Early Onset Scoliosis (C-EOS), the Complication Classification System in Growing Spine Surgery, and a system for classifying the treatments for early onset scoliosis. The C-EOS is a consensus-based classification system that was developed through iterative voting processes by a group of 14 experienced providers of treatment for early onset scoliosis. In its final iteration, it has variables for etiology, Cobb angle, and kyphosis, and an optional progression modifier (Fig. 26.1). ⁶ Its validity is now being studied through its application to clinically salient scenarios, such as its ability to predict proximal fixation failure in vertical expandable prosthetic titanium rib (VEPTR) surgery and its prognostic potential in patients with 5 years of follow-up. The Complication Classification System in Growing Spine Surgery was developed by group consensus in a method similar to that used for the C-EOS, and studies are currently in development to validate its use. It distinguishes between device- and disease-related complications and categorizes them by levels of severity according to the need for unplanned surgery or an alteration in the treatment course. ⁷ Furthermore, Skaggs and investigators have developed a system for classifying the treatment methodologies used in this field that divides the approaches to treatment into distraction-based, guided-growth, and compression techniques. ⁸

These efforts over the past decade have produced the infrastructure necessary to perform high-level evidence studies. For example, out of the equipoise study, the multicenter prospective study of rib- vs. spine-based proximal anchors was conceived. It uses the EOSQ-24 as an outcome measure, and once the enrollment requirement has been met through the CW&SD Study Group and GSSG registries, it will categorize patients with the C-EOS and report complications with the Complication Classification System in Growing Spine Surgery. This joint effort by the two study groups is the first of many collaborations; the two database registries were merged in 2013.