Differential Risk Factors for Proximal Junctional Kyphosis Between T8-T10 and T11-L1 Upper Instrumented Vertebrae in Adult Spinal Deformity

Se-Jun Park; Jin-Sung Park; Dong-Ho Kang; Chong-Suh Lee

doi:10.1227/neu.0000000000003901

Differential Risk Factors for Proximal Junctional Kyphosis Between T8-T10 and T11-L1 Upper Instrumented Vertebrae in Adult Spinal Deformity

Neurosurgery. 2026 Jan 9. doi: 10.1227/neu.0000000000003901. Online ahead of print.

Authors

Se-Jun Park¹, Jin-Sung Park¹, Dong-Ho Kang¹, Chong-Suh Lee²

Affiliations

¹ Department of Orthopedic Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, South Korea.
² Department of Orthopedic Surgery, Haeundae Bumin Hospital, Busan, South Korea.

PMID: 41511115
DOI: 10.1227/neu.0000000000003901

Abstract

Background and objectives: Proximal junctional kyphosis (PJK) remains a persistent mechanical complication after adult spinal deformity surgery. While lower thoracic upper instrumented vertebrae (UIV; T8-L1) are frequently selected, some studies have treated them as a single group, potentially obscuring sublevel-specific risk profiles. Therefore, this study aims to compare risk factors for PJK between patients undergoing lower thoracic fusion with UIV at T8-T10 and those with UIV at T11-L1.

Methods: A retrospective review was performed of 334 adult spinal deformity patients (mean age, 69.6 years; 88.3% female) who underwent corrective fusion from T8-L1 to the sacrum/pelvis between 2014 and 2022 with a 2-year follow-up. Based on UIV, patients were categorized into group I (T11-L1, N = 157) and group II (T8-T10, N = 177). Demographic, surgical, and radiographic variables were compared between groups. Univariate and stepwise multivariate logistic regression analyses identified independent PJK predictors. Receiver operating characteristic analyses determined optimal cutoff values.

Results: PJK developed in 99 patients (29.7%), more frequently in group I than group II (35.7% vs 24.3%; P = .023). Preoperative proximal junctional angle was a significant predictor in both groups, with optimal cutoffs near 0° (1.0° in group I; -0.8° in group II). In group I, additional independent predictors included osteoporosis (odds ratio [OR], 6.435) and postoperative change in lumbar lordosis ≥28.7° (OR, 1.056). In group II, advanced age ≥72 years (OR, 1.138) and postoperative age-adjusted pelvic incidence minus lumbar lordosis overcorrection (OR, 2.745) were significant risk factors.

Conclusion: While preoperative proximal junctional angle is a universal predictor, surgeons should consider avoiding the thoracolumbar junction for osteoporotic patients or for those requiring large lordosis corrections, and in T8-T10 fusions, particular attention should be paid to age-adjusted alignment targets and patient age. These findings may guide UIV-level-specific surgical planning to mitigate PJK risk.

Keywords: Adult spinal deformity; Lower thoracic fusion; Proximal junctional kyphosis; Risk factor; T11-L1; T8-T10; Upper instrumented vertebra.