Abstract
Eighty consecutive complex spinal robotic cases utilizing intraoperative 3D CT imaging (E3D, Group 2) were compared to 80 age -matched controls using the Excelsius robot alone with C-arm Fluoroscopic registration (Robot Only, Group 1). The demographics between the two groups were similar—severity of deformity, ASA Score for general anesthesia, patient age, gender, number of spinal levels instrumented, number of patients with prior spinal surgery, and amount of neurologic compression. The intraoperative CT scanning added several objective factors improving patient safety. There were significantly fewer complications in the E3D group with only 3 of 80 (4%) patients requiring a return to the operating room compared to 11 of 80 (14%) patients in the Robot Only Group requiring repeat surgery for implant related problems (Chi squared analysis = 5.00, p = 0.025). There was a significant reduction the amount of fluoroscopy time in the E3D Group (36 seconds, range 4 to 102 seconds) compared to Robot only group (51 seconds, range 15 to 160 seconds) (p = 0.0001). There was also shorter mean operative time in the E3D group (257 +/- 59.5 minutes) compared to the robot only group (306 ± 73.8 minutes) due to much faster registration time (45 seconds). A longer registration time was required in the Robot only group to register each vertebral level with AP and Lateral fluoroscopy shots. The estimated blood loss was also significantly lower in Group 1 (mean 345 ± 225 ml) vs Group 2 (474 ± 397 ml) (p = 0.012). The mean hospital length of stay was also significantly shorter for Group 1 (3.77 ± 1.86 days) compared to Group 2 (5.16 ± 3.40) (p = 0.022). There was no significant difference in the number of interbody implants nor corrective osteotomies in both groups—Robot only 52 cases vs. 42 cases in E3D group. Level of Evidence: IV, Retrospective review