Achieving the Needed Correction in Vertebral Body Tethering: The Relationship Between Preoperative Flexibility,...

Vertebral body tethering (VBT) is a nonfusion surgical treatment for scoliosis. Recent data have shown that intraoperative correction is critical for successful curve correction over time. This study aims to evaluate the relationship between preoperative, intraoperative, and postoperative correction. We hypothesize that preoperative flexibility will match first erect imaging, intraoperative correction will overestimate postoperative correction, and correction has improved over time with increased surgeon experience.

This study is a retrospective review of patients who underwent VBT at a single center between 2015 and 2023. Patient radiographs were reviewed preoperatively (standing and fulcrum bending), intraoperatively, and postoperatively first erect.

Our study included 157 patients: 127 females (81%) with a mean age of 13.2 years (range: 9 to 17) at surgery and a mean of 7.9 vertebrae instrumented (range: 5 to 12 vertebrae). Twenty-three had 2 curves instrumented (85% single curve). One hundred eight patients had thoracic curves (69%), 33 thoracolumbar (21%), and 16 lumbar (10%). Preoperative curve magnitude averaged 51 degrees (range: 36 to 72). The mean bending radiograph curve measurement was 24 degrees (53% correction). The mean intraoperative curve magnitude was 15 degrees (72% correction). Intraoperative curve magnitude and correction were significantly different between curve types, with increased correction in lumbar curves (7 degrees, 86% correction) and less correction in thoracic curves (16 degrees, 68% correction). On first erect postoperative imaging (mean 8 d post-op, range: 1 to 44), the curve magnitude was 26 degrees (49% correction). The R2 correlation of the first erect radiograph was 0.209 for preoperative bending and 0.554 for intraoperative measurements. The mean difference in first erect curve magnitude was +2 degrees from preoperative flexibility radiographs and +11 degrees from intraoperative radiographs. Pre-2020, the intraoperative curve averaged 18 degrees, and >2020 averaged 13 degrees ( P =0.001). Pre-2020, the first erect averaged 30 degrees, and >2020 improved to 24 degrees ( P <0.001), despite equivalent preoperative metrics (both 51 degrees curves, P =0.98, with 54 vs. 52% flexibility, P =0.31).

Our study indicates that preoperative bending films can provide a practical approximation of the correction on first erect imaging. Intraoperative correction has the strongest relationship and averages 11 degrees less than the postoperative standing curve magnitude. Further, our site's intraoperative and postoperative correction has improved over time with increased surgeon experience.

Level III-retrospective cohort.