Risk Factors for Poor Matching of Intervertebral Fusion Devices After Posterior Vertebral Column Resection in Patients With Kyphosis or Kyphoscoliosis.

Objectives: Numerous studies have reported the manifestations and influencing factors of poor matching of intervertebral fusion devices (IFDs) in patients with cervical and lumbar degenerative diseases. However, there is currently no research addressing the use of IFDs matching in posterior vertebral column resection (PVCR) procedures. The purpose of this retrospective radiographic study was to analyze the risk factors associated with poor matching of IFDs following PVCR.

Methods: Data from 92 patients using IFDs following PVCR between June 2006 and July 2024 were reviewed. IFDs implantation failure, adjacent vertebral fractures, subsidence greater than 5 mm, angle formation exceeding 10°, and malposition (defined as one-third of the IFDs exceeding the outer edge of the matching interface) were used as screening indicators to divide patients into poor matching and matching groups. Potential risk factors of poor matching were assessed through univariate and multivariate logistic regression analysis. The multiple regression model was evaluated by the area under the receiver operating characteristic curve (AUC).

Results: Among the 92 patients, 37 (40.2%) experienced poor matching. Univariate logistic regression analysis revealed that the preoperative and postoperative sagittal Cobb angles, the angle between osteotomy surfaces, total instrumented vertebrae, the number of vertebrae resected, the height of IFDs, and the use of titanium mesh or artificial vertebral bodies were potential risk factors for poor matching. Backward stepwise multivariate logistic regression analysis indicated that the preoperative sagittal Cobb angle (OR = 1.053, p = 0.001), the angle between osteotomy surfaces (OR = 1.152, p = 0.003), and the height of IFDs (OR = 1.058, p = 0.033) were independent risk factors for poor matching. The overall predictive performance of this multiple regression model (AUC = 0.872) for poor matching was deemed satisfactory.

Conclusion: The use of IFDs in PVCR was associated with a high rate of immediate poor matching. The preoperative sagittal Cobb angle, the angle between osteotomy surfaces, and the height of IFDs are independent risk factors for poor matching.