Reproducibility of sagittal radiographic parameters in adolescent idiopathic scoliosis-a guide to reference values using serial imaging.

Background Context: Knowledge of sagittal radiographic parameters in adolescent idiopathic scoliosis (AIS) patients has not yet caught up with our understanding of their roles in patients with adult spinal deformity. It is likely that more emphasis will be placed in restoring sagittal parameters for AIS patients in the future. Therefore, we need to understand how these parameters may vary in AIS to facilitate management plans.

Purpose: This study aimed to determine the reproducibility of sagittal spinal parameters on lateral film radiographs in patients with AIS.

Study Design/setting: This was a retrospective, comparative study conducted in a tertiary health-care institution from January 2013 to February 2016 (3-year period).

Patient Sample: All AIS patients who underwent deformity correction surgery from January 2013 to February 2016 and had two preoperative serial lateral radiographs taken within the time period of a month were included in the study.

Outcome Measures: Radiographic sagittal spinal parameters including sagittal vertical axis (SVA), cervical lordosis (CL), thoracic kyphosis (TK), thoracolumbar alignment (TL), lumbar lordosis (LL); standard spinopelvic measurements such as pelvic incidence (PI), pelvic tilt (PT), sacral slope (SS); as well as end and apical vertebrae of cervical, thoracic, and lumbar curves were the outcome measures.

Methods: All patient data were pooled from electronic medical records, and X-ray images were retrieved from Centricity Enterprise Web. Averaged X-ray measurements by two independent assessors were analyzed by comparing two radiographs of the same patients performed within a 1-month time period. Chi-squared and Wilcoxon signed-rank tests were used for categorical and continuous variables.

Results: The study cohort comprised 138 patients, 28 men and 110 women, with a mean age of 15 years (range 11-20). Between the two lateral X-rays, there was a mean difference of 0.79 cm in SVA (p<.001), 0.70° in LL (p=.033), and 0.73° in PT (p=.010). In the combined Lenke 1 and 2 subgroup, there was a similar 0.77 cm (p=.002), 0.79° (p=.009), and 1.49° (p=.001) mean difference in SVA, LL, and PT, respectively. Additionally, there was also a 1.85° (p=.009) and 1.76° (p=.006) mean difference seen in TL and SS, respectively. The overall profile of the sagittal curves remained largely similar, with only the lumbar apex shifting from L3 to L4 during the first and the second X-rays, respectively (p<.001). This occurred for the combined Lenke 1 and 2 subgroup as well (p<.001).

Conclusion: Most radiographic sagittal spinal parameters in AIS patients are generally reproducible with some variations up to a maximum of 4°. This natural variation should be taken into account when interpreting these radiographic sagittal parameters so as to achieve the most accurate results in surgical planning.