The association between preoperative cervical degeneration in adult patients with spinal deformity and poor postoperative cervical sagittal alignment after thoracolumbar deformity correction

Why clinicians should know about it

• Picked for Spine Surgery (paper of the day, 3 November 2026): Adult deformity correction and cervical alignment outcomes

Abstract

OBJECTIVE: Thoracolumbar (TL) deformity correction in adolescent idiopathic scoliosis has been shown to promote favorable compensatory cervical alignment changes, often restoring cervical lordosis. However, the influence of cervical degeneration on postoperative cervical alignment in adult spinal deformity (ASD) remains unclear. This study investigated the effect of preoperative cervical disc degeneration on postoperative cervical sagittal alignment (CSA) following TL deformity correction in ASD. METHODS: One hundred eighty-one patients with ASD who underwent deformity correction from the cervicothoracic junction (upper instrumented vertebra [UIV] between C7 and T4) to the lumbosacral spine (lowest instrumented vertebra between L1 and S1 or the ilium) with a minimum 2-year follow-up were analyzed. Preoperative cervical MRI was evaluated using the Miyazaki classification system to grade intervertebral disc degeneration. Patients were stratified into 3 groups: group 1, minimal degeneration; group 2, 1-2 severely degenerated discs (grade ≥ IV); and group 3, ≥ 3 severely degenerated discs. CSA (kyphosis vs lordosis), C2-7 sagittal vertical axis (cSVA), and T1 slope (T1S) were measured preoperatively, immediately postoperatively, and at 1- and 2-year follow-up. Optimal CSA was defined as within 20° of the T1S (T1S-CSA mismatch < 20°). RESULTS: At baseline, groups 1-3 demonstrated no significant differences in CSA, T1S, cSVA, or TL alignment. Postoperatively and through 2 years of follow-up, groups 2 and 3 demonstrated significantly worse CSA compared with group 1. Patients in group 1 maintained improved cervical alignment and T1S-CSA matching, whereas groups 2 and 3 either preserved or worsened their preoperative malalignment. On multivariable logistic regression, the odds of achieving T1S-CSA mismatch < 20° were 81% lower in group 2 (OR 0.198, p < 0.05) and 91% lower in group 3 (OR 0.087, p < 0.0001; area under the receiver operating characteristic curve = 0.88). On multivariable linear regression, each severely degenerated cervical disc (grade ≥ IV) was associated with a 1.4° increase in T1S-CSA mismatch (β = 1.4, R2 = 0.75, p < 0.0001). CONCLUSIONS: Patients with ASD and minimal cervical degeneration experienced sustained reciprocal improvements in cervical alignment for up to 2 years after TL deformity correction that included a UIV between C7 and T4. In contrast, patients with preexisting cervical degeneration were more likely to maintain or worsen their malalignment. Preoperative cervical MRI should be routinely performed to assess degeneration during surgical planning.

Abstract as published, via PubMed.