Ankylosing spondylitis (AS) is a chronic inflammatory disease characterized by gradual ossification of spinal ligaments, fibrous rings, and intervertebral joints as the disease progresses (1, 2). With the progressive development of thoracolumbar kyphotic deformity, compensatory backward rotation of the pelvis occurs to maintain overall sagittal balance. In advanced stages, symptoms such as low back pain, difficulty maintaining upright posture, and inability to gaze forward may develop (1, 3).

Pedicle subtraction osteotomy (PSO) is widely used to treat thoracolumbar kyphotic deformities in AS. This surgical method effectively reconstructs spinal sagittal alignment and significantly improves patients' quality of life (4–6). However, previous studies have indicated that PSO is associated with postoperative mechanical complications such as rod fracture, pseudarthrosis, and proximal junctional kyphosis (7, 8). Ponte osteotomy, first described by Ponte in 1987, involves complete resection of the thoracic articular processes, partial laminae, and entire ligamentum flavum. It is used to correct thoracic kyphotic deformities, achieving an average correction of approximately 30° (9).

Although both surgical techniques are widely applied in clinical practice, systematic comparative studies regarding their biomechanical characteristics remain limited. Previous research has primarily concentrated on clinical efficacy and imaging outcomes, insufficiently addressing critical aspects such as stress distribution at osteotomy sites, biomechanical responses of adjacent segments, and internal fixation load alterations. Clarifying biomechanical differences between these two osteotomies during deformity correction is crucial for optimizing surgical decision-making and reducing postoperative internal fixation failure and adjacent segment degeneration risks. Therefore, this study aims to compare the biomechanical characteristics of Ponte and PSO osteotomies in treating AS with thoracolumbar kyphotic deformity. Biomechanical analysis was conducted to investigate spinal force lines, stress distribution at osteotomy segments, and stability of internal fixation systems. This provides a biomechanical foundation for clinical selection, aiming to enhance surgical safety and long-term outcomes.

AS is a chronic rheumatic autoimmune disease primarily affecting the sacroiliac joints, spine, and peripheral joints. It significantly reduces the patient's quality of life and daily functioning (15, 16). Early AS symptoms typically include dull lower back pain and morning stiffness, which can improve after exercise. As the disease progresses, severe cases may involve ligament ossification, muscle stiffness, muscular atrophy, spinal rigidity, and kyphosis (17). Thoracolumbar kyphotic deformity in AS can result from minor trauma, characterized by spinal sagittal imbalance. Clinical manifestations include muscle dysfunction, fatigue, pain, and difficulty lying flat or walking normally. Thus, effective treatment of AS-related thoracolumbar kyphosis has become an important challenge in orthopedic surgery. Spinal osteotomy can effectively reconstruct sagittal spinal balance, restore horizontal gaze, and relieve organ compression in AS patients (18, 19). Surgeons may select osteotomy techniques according to the specific characteristics of spinal kyphosis, location of cervical vertebral curvature, degree of anterior longitudinal ligament calcification (bone bridge formation), and kyphotic angle magnitude (20). PSO is a progressive surgical method. It can preserve basic middle-column closure and anterior-column closure, or it can be modified into an anterior-column open osteotomy to achieve greater kyphosis correction. Therefore, PSO is widely applied in AS correction (21). Compared with higher-level osteotomies, Ponte osteotomy has the advantages of simpler operation, fewer osteotomy sites, and shorter operation time (9). However, limited reports exist on the biomechanical efficacy of these two osteotomies in treating AS with thoracolumbar kyphosis. Thus, this study established a finite element model of the human T3–T8 spine. PSO and Ponte osteotomies were simulated to analyze their biomechanical stability and provide a biomechanical basis for clinical decision-making.

Wang et al. (22) conducted a finite element analysis of titanium rods following surgical correction of Lenke type 5 adolescent idiopathic scoliosis (AIS) from T3 to L5. Their results showed that the titanium rod experienced maximum stress at T11 for thoracic curves and at T12 for lumbar curves, specifically at the curve apex vertebrae. Vertebral stress distribution results revealed that the two surgical methods have different effects on spinal force transmission. In this study, Ponte osteotomy (M1) significantly reduced stress at T3–T5 compared to M0, while stress at T6 remained largely unchanged (Table 6). This suggests Ponte osteotomy achieves spinal correction through multi-level stress dispersion without significant stress concentration. Ponte osteotomy selectively releases facet joints, allowing gradual correction at multiple segments and avoiding excessive stress on a single vertebra. Conversely, PSO osteotomy (M2) substantially reduced stress in the osteotomized segments (T4–T5), but stress significantly increased distally at T7–T8 (Table 6). This finding suggests PSO may cause distal stress concentration due to changes in spinal force transmission. Mechanistically, PSO corrects spinal curvature via vertebral osteotomy, abruptly altering spinal alignment and increasing compensatory load on distal vertebrae. This likely explains the biomechanical basis for the increased clinical risk of adjacent segment degeneration after PSO (23). Therefore, in patients with long-segment kyphotic deformity, clinicians should carefully assess distal vertebral conditions after PSO, potentially extending the fixation segments to distribute stress effectively.

PSO was first reported in 1985. It involves extensive surgical resection across three spinal columns, preserving only the anterior longitudinal ligament. Currently, it is indicated for kyphotic deformities below 40° (24). Zhao et al. (25) reported 11 cases of rod breakage among 123 AS patients with thoracolumbar kyphosis treated by PSO, with an incidence rate of 8.9%. Stress and displacement outcomes of the internal fixation system further highlight biomechanical differences between the two surgical techniques. Internal fixation stress in the Ponte osteotomy group was significantly lower than in the PSO osteotomy group, while displacement was greater. This finding is directly associated with the distinct mechanical correction mechanisms of the two surgeries. PSO requires substantial distraction-compression forces via the internal fixation system to maintain wedge closure after vertebral osteotomy. This results in higher axial and bending loads on fixation devices, increasing the risk of rod breakage and screw loosening. These findings align with previous studies indicating a high rate of mechanical complications following PSO. Conversely, Ponte osteotomy achieves gradual correction by releasing the facet joints. Internal fixation primarily provides stabilization rather than active correction, leading to lower stress and a reduced risk of internal fixation failure. In this study, internal fixation displacement in the PSO group was smaller than that in the Ponte group (0.623 < 0.774, 0.339 < 0.435, 0.524 < 0.667, 0.471 < 0.586, 0.619 < 0.773, and 0.388 < 0.516). Because of small differences in resection angles, displacement differences were not substantial. Additionally, the moderate ROM preserved by Ponte osteotomy slightly increases internal fixation displacement. However, this displacement is within a physiologically acceptable range and does not compromise overall stability. Instead, it may reduce internal fixation fatigue fractures associated with overly rigid fixation.

This study has several limitations: ① It involved finite element analysis of a single case. Thus, model parameters may not fully reflect individual differences. ② The model did not account for dynamic factors such as muscle forces, and ligaments were simulated as linear elastic materials rather than incorporating their nonlinear, hyperelastic properties. These simplifications, while common in comparative FE studies, likely result in an overestimation of segmental motion and an underestimation of overall spinal stiffness. Consequently, the absolute values of vertebral stress and internal fixation stress might be influenced. However, since these simplifications were applied consistently across all models (M0, M1, M2), the core comparative conclusions regarding the biomechanical differences between Ponte and PSO osteotomies are expected to remain valid. ③ During model construction, other factors such as material homogeneity, mesh division, and boundary conditions were simplified, which might also cause deviations between the model and the actual anatomical structures (26). Future studies could use a variety of patient-specific models or parametric analysis methods to verify the universality of the results in the population to validate the long-term mechanical stability of both surgical methods across varying deformity severities.

In conclusion, Ponte osteotomy and PSO osteotomy each have distinct biomechanical advantages. Ponte osteotomy distributes stress effectively and involves low-load fixation, making it suitable for patients with moderate deformities or high surgical risks. PSO osteotomy offers strong correction capability and high stability, making it preferable for sagittal reconstruction of severe deformities. These findings provide critical biomechanical evidence for individualized surgical procedure selection, helping to minimize postoperative complications and enhance long-term outcomes.