Introduction: Spinal deformity in the pediatric population can pose a significant detriment to quality of life and development¹. Spinal deformity surgery can be performed to correct coronal and sagittal deformities such as scoliosis and kyphosis. Unfortunately, pediatric spinal deformity surgery is frequently complicated by the need for reoperation, with a revision rate of nearly 10%². However, there is limited literature on outcomes following primary vs. revision deformity surgery in the pediatric population. In addition, no prior study has evaluated for a difference in physician reimbursement rates between primary and revision pediatric spinal deformity surgery. The purpose of this study is to compare the relative value units (RVUs) per minute and 30-day outcomes between primary and revision pediatric spinal deformity surgery.

Methods: This study was performed using data obtained from the American College of Surgeons National Surgical Quality Improvement Program Pediatric database. Patients between 10 and 18 years of age who underwent posterior spinal deformity surgery between 2012 and 2019 were identified using Current Procedural Terminology codes 22800, 22802, and 22804. Patients undergoing anterior-only or concurrent anterior-posterior fusions were excluded to ensure patient cohort homogeneity. Exclusion criteria also included missing demographic or clinical data, malignancy, active infection, and emergent procedure.

Univariate analysis was performed using Pearson Chi-square tests to compare baseline demographic and clinical characteristics. Univariate and multivariate regression models were used to assess the independent impact of revision surgery on RVUs and postoperative outcomes. Backwards stepwise regression model was utilized to account for all baseline differences in demographics and comorbidities, with entry at p=0.05 and removal at p=0.1.

Results: The study cohort included 15,055 patients with 358 patients who underwent revision pediatric spinal deformity surgery. Patients in the revision group were most likely to be younger (13.7 vs. 14.3 years, p<0.001) and male sex (38.0% vs. 30.4%, p=0.002). Multiple comorbidities were more common in the revision group as detailed in Table 1. Revision surgery more commonly required osteotomy (13.7% vs. 8.3%, p=0.002). All variables included in Table 1 were adjusted for in the subsequent regression models.

Univariate analysis revealed higher total RVUs (71.09 vs. 60.51, p<0.001), RVUs per minute (0.27 vs. 0.23, p<0.001), readmission rate (6.7% vs. 4.0%, p=0.012), and reoperation rate (7.5% vs. 3.3%, p<0.001) for the revision surgery group. Morbidity rates were found to be statistically similar between revision and primary surgery (65.3% vs. 69.3%, p=0.108). In addition, deep surgical site infection (2.2% vs. 0.9%, p=0.010), pulmonary embolism (0.3% vs. <0.1%, p=0.010), and urinary tract infection (1.7% vs. 0.7%, p=0.038) were more common in the revision group (Table 2). After multivariate regression, the differences in total RVUs, RVUs per minute, reoperation rate, and rate of pulmonary embolism between primary and revision surgery remained statistically significant (Table 3).

Conclusion: Revision pediatric spinal deformity surgery was found to be assigned appropriately higher mean total RVUs and RVUs per minute corresponding to the higher operative complexity compared to primary surgery. Revision surgery was also associated with poorer 30-day outcomes, including higher frequencies of reoperation and pulmonary embolism.