The mechanism in junctional failure of thoraco-lumbar fusions. Part II: Analysis of a series of PJK after thoraco-lumbar fusion to determine parameters allowing to predict the risk of junctional breakdown.

Purpose: To identify risk factors, in 12 patients with junctional breakdown (JBD) after thoraco-sacral fusions and to test a software locating maximal bending moment on full spine EOS images.

Methods: Twelve patients underwent long fusions for lumbar degenerative pathologies. Preop EOS images were compared to first postop EOS showing JBD. Parameters analyzed were: spinopelvic parameters [pelvic incidence (PI), pelvic tilt (PT), sacral slope (SS), sagittal vertical axis (SVA), spinosacral angle (SSA), lordosis, and kyphosis], proximal junctional angle (PJA), odontoid-hip axis angle (ODHA), and CIA. A new software estimated the location of maximum bending moment (M ) before and after JBD.

Results: All patients except one had a JBD located between T10 and L1, diagnosed at average follow-up of 18.58 months. JBD was a fracture in six patients, severe adjacent disc degeneration in the remaining. Average PI was 52°. PT increased, SS decreased after JBD versus preop (p > 0.05). Average PJA was 34.5°. Global lordosis (GLL), upper lordosis (ULL), L4-S1 lordosis, and thoracic kyphosis (TK) were increased (p < 0.05). Lower lumbar lordosis (LLL), was not increased postJBD (p = 0.6). SVA, SSA, ODHA, and C7 slope were not modified (p > 0.05). CIA average value decreased by 7.5% after JBD. T1-T5 alignment was correlated to C7 slope before (R  = 0.77075) and after JBD (R  = 0.85409). ODHA decreased after JBD (p > 0.05). Most JBD occurred at or one level away from preoperative M location.

Conclusion: This study confirms the importance of harmonious distribution of lumbar (GLL, ULL, and ILL) and thoracic curves (TK, T1-T5 segment) in thoraco-sacral fusions. All patients showed an exaggerated ULL, resulting in a posterior shift and increased lever arm at the thoraco-lumbar junction, leading to JBD.