Preoperative hounsfield unit measurements predict non-infectious complications following posterior spinal instrumentation for traumatic thoracolumbar fractures.

Authors

Federico Cardahi, Ali Fazlollahi, Kaveh Mozafari-Lorestani, Paolo Brigato, Ahmed Aoude, Peter Jarzem, Ruldof Reindl, Ouellet Jean, Carlo Santaguida

Published in

European spine journal : official publication of the European Spine Society, the European Spinal Deformity Society, and the European Section of the Cervical Spine Research Society. Aug 04, 2026. Epub Aug 04, 2026.

Abstract

Surgical management of traumatic vertebral fractures remains debated. Bone quality influences fixation stability and postoperative complications, yet X-ray absorptiometry (DXA) is rarely available in acute trauma. Computed tomography (CT), routinely obtained during trauma evaluation, enables bone quality assessment. Hounsfield Unit (HU) measurements derived from CT serve as a surrogate for bone mineral density and predictor of complications in degenerative spine surgery; however, their role in traumatic thoracolumbar fracture instrumentation remains unclear.
To evaluate whether preoperative HU measurements predict non-infectious complications following posterior spinal instrumentation for traumatic thoracolumbar fractures.
Retrospective cohort study.
Patients undergoing spinal instrumentation for traumatic thoracolumbar fractures between 2019 and 2024 at the Montreal General Hospital, Canada.
Incidence of non-infectious postoperative complications, including instrumentation failure.
Patients were retrospectively reviewed (mean follow-up 18 months). Data included demographics, preoperative HU values, instrumentation and augmentation techniques, AO fracture classification, levels instrumented, and history of diffuse idiopathic skeletal hyperostosis (DISH) or ankylosing spondylitis (AS). HU values were measured on preoperative CT using three axial slices per vertebral body, averaged across three spine surgeons. Standing radiographs assessed postoperative segmental alignment immediately postoperatively and at one year. Logistic regression identified predictors of non-infectious complications.
Among 218 patients (141 males, 64.7%; mean age 54.9 years, range 18-92), the most common complications were instrumentation failure (8.7%), infection (3.7%), loss of reduction (3.2%), and proximal junctional kyphosis (1.4%). Each one-unit HU decrease increased complication odds by 2.4% (OR 0.976; 95% CI: 0.959-0.994), corresponding to ~ 21% higher risk per 10-HU decreases from the cohort mean of 177.56 HU (SD 70.07) at the vertebral level above the fracture.
Lower preoperative HU values are associated with increased risk of non-infectious complications following instrumentation for traumatic thoracolumbar fractures. HU measurements may aid surgical planning and risk stratification and should be interpreted alongside other clinical and fracture-specific factors.

PMID:
42547608
Bibliographic data and abstract were imported from PubMed on 04 Aug 2026.

Read full publication at:
Please sign in to see all details.

Stats

1-terrible, 9-excellent. How would you rate this publication? Sign in in to submit your rating.

Post a comment

You need to be signed in to post comments. You can sign in here.

Comments

There are no comments yet.