IMPACT OF UNTREATED ADOLESCENT IDIOPATHIC SCOLIOSIS IN ADULTHOOD: A TEN-YEAR ANALYSIS

Abstract

Title: Impact of Untreated Adolescent Idiopathic Scoliosis in Adulthood: A Ten-Year Analysis Background: Long-term studies of adolescent idiopathic scoliosis (AIS) in adulthood demonstrate most patients function well, though some have increased disability. The Oswestry Disability Index (ODI) and SRS-22r are validated questionnaires for assessing back disability and quality of life respectively. Our purpose was to further establish the impact of untreated AIS in adulthood. The rationale for this project was to improve understanding of the natural course of AIS and the usefulness of HRQoL measures in determining treatment decisions. We predict that increasing ODI scores will correlate with age, curve size, curve location and progression to surgery. Additionally, we predicted that SRS-22r scores in our study population would show no difference between age-gender matched controls. Methods: All unoperated adult patients with a diagnosis of AIS seen at a tertiary deformity clinic from 2008-2018 were identified using ICD coding and reviewed. Variables collected included: general demographics, cardiac, pulmonary, endocrine, and psychological comorbidities, family history of scoliosis, curve size, curve location, curve major, visual analog pain score, ODI and SRS-22r scores, and previous treatment. ODI and SRS-22r scores were analyzed across three separate age groups: 20-39yrs (G1), 40-59yrs (G2), and ≥60yrs (G3). Continuous variables were analyzed and compared using means, standard deviations, and variances and categorical variables were compared using frequencies. For ODI analysis we used ANOVA for comparison between groups, Pearson correlation to assess linear relationship with patient characteristics, and chi-square analysis to determine score frequency of ≥30 within groups. For SRS-22r, mean domain scores were analyzed for each age-gender stratified group (G1-G3). ANOVA and Kruskal-Wallis tests were used to compare continuous variables between all three groups and one sample t-tests were used to compare our sample means to published normative data. Analysis comparing presurgical SRS-22r domain scores and ODI in patients who did and did not go on to have surgery was also performed using T ≥50o and/or TL ≥40o as a cutoff for surgical indication. Results: A total of 249 patients (84% female and 16% male) met inclusion and exclusion criteria and were identified for analysis. 214 patients (84% female) had an ODI score. Mean ODI score increased with each age group (p.05). Pain was worse in all age-gender matched domains (p<.05). Of the 249 patients in the study, approximately 10% went on to have surgery. In patients with surgical sized curves noted in the methods, no difference was seen in age or curve size between surgical and non-surgical patients, however ODI and SRS-22r scores (excluding mental health) were significantly worse (p≤.01). Conclusions: Patients with AIS have SRS-22r scores that are lower than age-gender matched controls in most domains. ODI had a positive linear correlation with age, body mass index, and curve size. Furthermore, 10%, of adults with surgical sized curves (Thoracic ≥50o; Thoracolumbar ≥40o) who sought evaluation for scoliosis pursued surgery. Patients who did go on to have surgery, reported worse preoperative HRQoL scores than their non-surgical counterparts. These results can help healthcare providers when counseling patients and families concerning treatment decisions.