Sustainable high performance computing in the UK: a position paper on public investment, accountability, and the path to net zero.

BACKGROUND

Factors associated with early re-collapse after posterior fixation of thoracolumbar burst fractures have been sparsely examined in previous studies. Early re-collapse is challenging to predict due to multifactorial risk. Therefore, we aimed to investigate factors associated with vertebral re-collapse within the first 3 months following posterior fixation.

METHODS

This retrospective study used data from the Swedish Fracture Register (2015-2019) and included patients aged 18 to 66 years with a single-level thoracolumbar burst fracture between T11 and L3 treated with posterior fixation. Fractures were classified according to the load-sharing classification (LSC) and the 2013 AO Spine Injury Classification System. Bone quality was assessed using Hounsfield unit (HU) values. In addition, postoperative pedicle screw depth within the vertebral body (% depth) and the number of instrumented vertebrae were recorded. To evaluate postoperative vertebral collapse, kyphosis angle, wedge angle, and anterior/posterior vertebral height ratio were measured. Re-collapse was defined as an anterior vertebral height loss of at least 20%, and pedicle screw pull-out was also assessed. Multivariable logistic regression was used to identify factors associated with vertebral re-collapse within 3 months postoperatively.

RESULTS

This study comprised 100 patients. Sixteen patients had radiographic vertebral re-collapse, and 15 had pedicle screw pull-out. Lower HU values were significantly associated with vertebral re-collapse (OR 0.97, 95% CI 0.94-0.99, P = 0.01). Higher preoperative LSC scores were also significantly associated with an increased risk of re-collapse (OR: 2.35, 95% CI: 1.07-6.82, P = 0.02). In contrast, neither screw insertion depth nor the number of instrumented vertebrae was associated with re-collapse.

CONCLUSION

Lower HU values and higher preoperative LSC scores were associated with an increased risk of early vertebral re-collapse within 3 months after posterior fixation of thoracolumbar burst fractures. However, early re-collapse appears to be multifactorial and remains difficult to predict, warranting further investigation.

CLINICAL RELEVANCE: This study provides clinically relevant insights into early re-collapse after posterior fixation of thoracolumbar burst fractures and identifies factors associated with early mechanical failure.