Comparison of robotic AI-assisted and manual pedicle screw fixation for treating thoracolumbar fractures: a retrospective controlled trial.

OBJECTIVE

To compare the clinical efficacy and screw placement accuracy of robot artificial intelligence (AI)-assisted percutaneous screw fixation and conventional C-arm-assisted percutaneous screw fixation (manual placement) in the treatment of thoracolumbar single-segment fractures without neurological symptoms.

METHODS

This study is a single-center retrospective analysis involving patients with thoracolumbar single-segment fractures without neurological symptoms. Patients were divided into Group A (robotic AI-assisted placement) and Group B (manual placement). Clinical outcomes such as operative time, intraoperative fluoroscopy frequency, screw placement accuracy, postoperative complications, length of hospital stay, and postoperative pain were compared between the two groups.

RESULTS

Group A showed significantly better screw placement accuracy, fewer intraoperative fluoroscopy attempts, shorter fluoroscopy time, and fewer guidewire adjustments compared to Group B (P < 0.05). Additionally, Group A had shorter hospital stays, a lower incidence of postoperative complications, and short-term greater improvement in Visual Analog Scale (VAS) scores (P < 0.05). However, after 1 year of follow-up, there was no statistically significant difference between the two groups in the improvement of VAS scores.

CONCLUSION

Robotic AI-assisted placement improves pedicle screw placement accuracy, reduces intraoperative fluoroscopy frequency and time, alleviates postoperative pain, and accelerates patient recovery. This approach aligns with the principles of enhanced recovery in orthopedic surgery and holds promise for wider clinical application in the treatment of thoracolumbar fractures.