Impact of Spinal Deformity Characteristics on Patient-reported Outcome Measurement Information System Scores in Patients With Idiopathic Scoliosis Undergoing Posterior Spinal Fusion

Abstract

Introduction: The impact of posterior spinal fusion (PSF) on physical function and pain and mental health in pediatric patients as quantified by the Patient-Reported Outcomes Measurement Information System (PROMIS), developed by the National Institute of Health, is largely unknown. The purpose of this study is to report the changes of PROMIS scores for upper extremity (UE), pain interference (PI), mobility (MOB), and peer relationships (PR) after PSF in patients with idiopathic scoliosis (IS), compare postoperative changes in PROMIS PI and Scoliosis Research Society-30 pain scores, and evaluate associations between curve characteristics and PROMIS scores.

Methods: A retrospective cohort of 122 patients (<18 years old) who underwent PSF for IS was identified through electronic medical record search. PROMIS scores were obtained preoperatively and 6 weeks, 6 months, 1 years, 2 years, and 3 years postoperatively.

Results: The mean age of the cohort was 14.2 ± 1.6 years, and the mean Cobb angle was 62.9 ± 13.8° at surgery. Eighty patients had preoperative PROMIS data. UE and MOB scores were statistically lower at 6 weeks and 6 months postoperatively and returned to baseline with a longer follow-up. PI scores were significantly lower at 1 and 2 years postoperatively. PR was unchanged up to 2 years postoperatively and then showed significant improvement. There was a statistically significant negative relationships between lowest instrumented vertebra and PROMIS UE and MOB scores at 6 weeks and 1 year postoperatively, but not at a longer follow-up. There were no significant differences noted in PI and PR PROMIS scores and lowest instrumented vertebra. PROMIS scores were not statistically associated with the Lenke Classification, number of vertebral levels fused, or percentage coronal correction.

Discussion: Changes in PROMIS functional domains (UE and MOB) postoperatively normalize at longer follow-ups. Changes in PI and PR demonstrated improvements over preoperative values at 1 to 2 years postoperatively. Preoperative coronal and sagittal measures, and the percentage correction did not correlate with any PROMIS scores.