Possible Correlation Between Kyphosis of Lumbar Osteoporosis Fractures and the Spinal Signal Intensity Ratio (SSIR)

Background The aim of this study was to examine the correlation between the risk of increasing kyphosis as well as collapse of the osteoporotic vertebral body fractures and the intensity of the bone edema in magnetic resonance imaging (MRI) scans. Inclusion criteria included the following: age >18 years and osteoporotic vertebral body fracture grade I–IV according to OF classification. Exclusion criteria included the following: other pathological fractures due to primary tumors or metastasis, OF grade V fractures, and AO type B or C fractures.

Methods This was a retrospective study from pseudonymized data of a tertiary spine center. No additional imaging were performed. Measurements of bisegmental kyphosis angle of the fracture for involvement of both endplates and monosegmental angle for involvement of 1 endplate, as well as vertebral body height loss in initial radiographs and at follow-ups after 3 and 6 months have been performed. Also, the initial signal intensity of the vertebral body edema was measured using integrated tool of the DICOM viewer (Impax V6.5 Agfa, Brentford, UK) in addition to the signal intensity of the cerebrospinal fluid (CSF) as reference for T1, T2, and separate target illumination radar (STIR) sequences of the MRI scans. A quotient from the signal intensity of the vertebral body edema and the reference (CSF) has been generated. Patients have been divided to 4 groups according to the ratio (<1, 1–2, 2–3, >3) and compared in regards to the results of the degree of kyphosis and vertebral collapse at follow-ups and final examination. The statistical analysis was performed using linear regression using statistic software SPSS version 26.

Results and Conclusions Forty-four patients have been included: 9 males and 35 females with an average age of 71.5 years. The analysis showed a significant correlation between the increasing kyphosis at follow-ups and the quotient of the signal intensity for STIR and T2 weighing with P = .002 (SD ±2.664) for STIR and P = .001 (SD ±1.616) for T2 sequences. Furthermore, there was only a correlation between the intensity ratio and kyphosis for STIR weighting at last examination (P = .017; SD ±1.360). There was no correlation between the height loss and the signal intensity.

Level of Evidence Level 2