Reliability of functional imaging in assessing instability of lumbar degenerative spondylolisthesis

Surgical management of lumbar degenerative spondylolisthesis-whether decompression alone or combined with fusion-often relies on instability assessment with standing flexion-extension radiographs. However, these evaluations are influenced by low back pain and other factors, introducing variability into instability determination and surgical planning. Despite widespread use, the reproducibility of quantitative parameters remains unclear. This study evaluated the reproducibility of commonly used quantitative indicators from lumbar functional imaging.

We retrospectively analyzed 112 patients with L4-5 degenerative spondylolisthesis who underwent two sets of standing flexion-extension radiographs within six months. Five parameters were measured: Changes in lumbar lordosis during flexion [ΔLL(flex)], changes during extension [ΔLL(ext)], sagittal translation (ST), posterior opening (PO), and segmental angulation (SA). Instability was defined as ST ≥ 3 mm or ≥ 8% of the upper vertebral body width, PO ≥ 5°, or SA ≥ 20°, and diagnostic consistency was examined. Two raters independently assessed all radiographs. Test-retest, intra-rater, and inter-rater reliabilities were determined using intraclass correlation coefficients (ICC [1, 1] ICC [1, 2]). Bland-Altman analysis tested fixed and proportional bias.

Diagnostic discrepancies between imaging sessions occurred in 25.0-27.7%. ΔLL(flex) showed poor test-retest reliability (ICC = 0.23, 0.32), whereas ΔLL(ext) and PO demonstrated higher reliability. ST and SA yielded low ICC values. Intra-rater and inter-rater reliability was consistently good (ICC > 0.75), and no fixed or proportional bias was detected.

Variability in flexion influenced ST and SA, potentially affecting instability diagnosis. Standardized imaging protocols and clearer patient instructions are essential to improve diagnostic accuracy and reliability.