AUTHOR=Chen Zhen , Zhang Jianqun , Liu Xiaoyin , Ma Rong , Liang Simin , Ge Zhaohui TITLE=For severe and rigid adult idiopathic scoliosis, does an optimal extent of posterior intervertebral release exist? A finite element analysis JOURNAL=Frontiers in Bioengineering and Biotechnology VOLUME=Volume 13 - 2025 YEAR=2025 URL=https://www.frontiersin.org/journals/bioengineering-and-biotechnology/articles/10.3389/fbioe.2025.1691245 DOI=10.3389/fbioe.2025.1691245 ISSN=2296-4185 ABSTRACT=BackgroundThe surgical management of severe, rigid adult idiopathic scoliosis (AdIS) is challenging. While posterior intervertebral release (PIVR) is used to enhance correction, the optimal number of release segments is unknown. This study aimed to identify the optimal PIVR strategy by evaluating the biomechanical effects of varying release levels.MethodsA patient-specific finite element (FE) model of a Lenke 2A + AdIS spine (T1-sacrum, main thoracic curve 84°) was created and validated. Six surgical scenarios were simulated: instrumentation-only (M1), Ponte osteotomy alone (M2), and M2 combined with four different PIVR strategies (M3-M6). Simulated corrective forces were applied, and outcomes, including Cobb angle correction and von-Mises stress on vertebrae and implants, were analyzed and compared.ResultsA targeted 4-level PIVR centered on the apex (M5) achieved the greatest main thoracic curve correction, reducing the Cobb angle from 84° to 34.88° (a 58.5% correction rate). A more extensive 6-level release (M6) proved less effective (38.99°, 53.6% correction rate). Model M5 also produced the most significant reduction in peak von-Mises stress on vertebrae (15.9% decrease vs. M2) and pedicle screws (32.9% decrease vs. M2).ConclusionA selective, 4-level PIVR strategy provides superior deformity correction and a more favorable stress environment than a more extensive release. These findings challenge the “more is better” paradigm, providing a biomechanical rationale for an “optimal” rather than “maximal” release approach in severe rigid AdIS.