Safety and efficacy of robot-assisted technology in severe complex and non-severe spinal deformity correction: a retrospective comparative study.

BACKGROUND

Surgical correction of severe complex spinal deformities (SCD) poses significant technical challenges and carries a high risk of neurological complications. In recent years, robot-assisted (RA) technology has been increasingly adopted in spinal surgery, with theoretical advantages in enhancing surgical precision and safety, particularly in anatomically complex cases. However, its clinical outcomes in the correction of SCD have not been thoroughly evaluated.

METHODS

A retrospective analysis was conducted on patients who underwent RA scoliosis correction surgery between January 2021 and April 2024. Patients were classified into non-severe deformities (NSD) and SCD groups based on the criteria of Cobb angle > 80°, flexibility < 25%, or focal kyphosis > 90°. Demographic data, surgical parameters, and imaging outcomes were collected and compared between the two groups. Primary outcomes included pedicle screw placement accuracy, major curve correction rates, operative time, intraoperative blood loss, hospital stay, and postoperative complications. A p-value < 0.05 was considered statistically significant.

RESULTS

A total of 38 patients were included (NSD: 21, SCD: 17). No significant differences were observed between the groups in terms of age, sex distribution, BMI, or major curve region (p > 0.05). The SCD group exhibited significantly greater preoperative Cobb angles (93.69° ± 23.20° vs. 51.05° ± 14.33°, p < 0.0001) and kyphosis angles (102.65° ± 36.52° vs. 37.77° ± 21.57°, p < 0.0001) compared to the NSD group. There were no significant differences between the groups in the number of screws placed (18.35 ± 7.00 vs. 16.86 ± 7.77, p = 0.5414), operative time (277.35 ± 68.81 min vs. 247.00 ± 80.40 min, p = 0.2256), intraoperative blood loss (379.41 ± 241.78 ml vs. 319.05 ± 197.15 ml, p = 0.4019), or transfusion rates (23.53% vs. 23.81%, p = 1.0000). However, the SCD group had significantly longer hospital stays compared to the NSD group (27.65 ± 9.64 days vs. 21.19 ± 7.24 days, p = 0.0239). The complication rates were similar between the two groups (5.88% vs. 4.76%, p = 1.0000). Although the perfect screw placement rate in the SCD group was slightly lower than in the NSD group, the difference was not statistically significant (90.60% ± 4.90% vs. 92.18% ± 6.29%, p = 0.3997). Major curve correction rates also showed no significant difference between the groups (66.34% ± 19.91% vs. 71.05% ± 16.08%, p = 0.5134).

CONCLUSION

This study demonstrates that RA technology achieves comparable pedicle screw placement accuracy and major curve correction rates in SCD patients relative to NSD patients, with no significant increase in complication rates. However, SCD patients may require longer hospital stays. These findings suggest that RA technology is a safe and effective strategy for SCD management, though broader implementation requires validation through larger prospective trials. Further large-scale, prospective studies are warranted to validate these results.