RADIOGRAPHIC MORPHOMETRIC PREDICTION OF NEW VERTEBRAL COMPRESSION FRACTURES AFTER VERTEBROPLASTY
DOI:
https://doi.org/10.15674/0030-59872026143-52Keywords:
Percutaneous vertebroplasty, new vertebral compression fractures, radiographic morphometry, discriminant functions analysis, osteoporosisAbstract
New vertebral compression fractures (NVCF) following percutaneous vertebroplasty (PVP) remain a clinical challenge. Methods. A retrospective cohort study (2023–2025) was conducted at the Institute of Spine and Joint Pathology, Ukraine, involving 26 patients (24 females, 2 males; mean age (69.04 ± 2.04) years) with osteoporotic vertebral compression fractures treated with PVP. Morphometric parameters of 99 vertebrae w ere a ssessed o n d igital r adiographs ( frontal/sagittal, RadiAnt DICOM Viewer, precision 0.1 mm / 0.1°): anterior (ha), middle (hm), posterior (hp) heights, relative compression (%), wedge index, local kyphotic angle, thoracic kyphosis, lumbar lordosis, and scoliosis. Results. NVCF occurred in 46.2 % of cases (45 fractures in 12 patients). Significant differences between NVCF and non-NVCF groups were observed for ha compression (20.69 ± 1.16 mm vs. 23.89 ± 0.78 mm, p = 0.0338), hm (17.84 ± 1 mm vs. 21.31 ± 0.61 mm, p = 0.0021), hp (26.97 ± 0.81 mm vs. 29.61 ± 0.51 mm, p = 0.0073), lumbar lordosis (44,4° ± 1,52° v s. 3 8.28° ± 1.46°, p = 0.01), a nd s coliosis (7.53° ± 0.56° vs. 5.90° ± 0.48°, p = 0.022). The linear discriminant functions model, based on hm (canonical correlation = –0.863) and relative Ha compression (canonical correlation = 0.139), achieved Wilks’ Λ = 0.8 (χ² = 15.77, p = 0.000376), classification accuracy of 76.25 % (sensitivity 76.9 %, specificity 75.9 %), a nd A UC = 0 ,754 ± 0 ,058. A djusted f unction: F = 6,07 – 0,28×hm (mm) –0,035×Ha (%); F > 0 indicates NVCF risk. Other parameters were excluded due to low discriminatory power or collinearity. Conclusion. This two-parameter model, using hm and relative Ha compression, offers moderate predictive accuracy for NVCF post-PVP. Its simplicity suits resource-limited Ukrainian clinics. External validation and inclusion of confounders (e. g., BMD, therapy) are required for broader adoption.
Downloads
How to Cite
Issue
Section
License
Copyright (c) 2026 Andrii Popov, Mykyta Moloduk, Volodymyr Kutsenko, Ruslan Zlatnik, Maryna Nessonova
This work is licensed under a Creative Commons Attribution 4.0 International License.
The authors retain the right of authorship of their manuscript and pass the journal the right of the first publication of this article, which automatically become available from the date of publication under the terms of Creative Commons Attribution License, which allows others to freely distribute the published manuscript with mandatory linking to authors of the original research and the first publication of this one in this journal.
Authors have the right to enter into a separate supplemental agreement on the additional non-exclusive distribution of manuscript in the form in which it was published by the journal (i.e. to put work in electronic storage of an institution or publish as a part of the book) while maintaining the reference to the first publication of the manuscript in this journal.
The editorial policy of the journal allows authors and encourages manuscript accommodation online (i.e. in storage of an institution or on the personal websites) as before submission of the manuscript to the editorial office, and during its editorial processing because it contributes to productive scientific discussion and positively affects the efficiency and dynamics of the published manuscript citation (see The Effect of Open Access).
