P36. Operative time and learning curve between conventional fluoroscopy, fluoroscopy-based instrument navigation, and robot-assisted instrumentation in minimally invasive transforaminal lumbar interbody fusion (MIS-TLIF)

Abstract

BACKGROUND CONTEXT Instrument-navigation modalities including CT-guided and robot-assisted methods claim both efficacy and accuracy when applied to spine surgery, yet often increase setup and operating times. PURPOSE The authors seek to study the impact of multitude of instrument navigational technologies on operative time and the learning curve associated with their use. STUDY DESIGN/SETTING Prospective, single-surgeon, single-institution. PATIENT SAMPLE Patients undergoing minimally invasive transforaminal lumbar interbody fusion with or without decompression, with or without previous hardware removal. OUTCOME MEASURES Total operative time, immediate surgical complications. METHODS Consecutive patients undergoing minimally invasive transforaminal lumbar interbody fusion (MIS-TLIF) were analyzed. Two adjunctive technologies were studied and compared with a more conventional fluoroscopic approach without navigation. Consecutive cases were done with assistance of a robot (Mazor Robotics), with the assistance of x-ray tracking (TrackX) or without navigation in three consecutive blocks of time. The cases done without assistance were used to further normalize for number of interbodies and decompressions performed as well as hardware removal if needed. Age, body mass index (BMI), sex, operative levels, laminectomy, need for hardware removal, and total operative time was recorded directly from EPIC. RESULTS A total of 120 cases (74 conventional, 14 robot-assisted, 32 Track-X) were included in analysis. There were no significant differences in age, sex, or BMI between modalities. Uncontrolled, the average operative time for conventional, robot-assisted, and TrackX was 118±36, 157±41, and 107±37 minutes, respectively, for each MIS-TLIF level (p<0.05 across each group). Once normalized for hardware removal, decompression, and number of fusion levels, the robot cases were 24% slower (added 32 minutes/case) while the x-ray tracking cases were 19% faster (subtracted 16 minutes). Operative times in no robot-assisted case were faster than the comparable control procedure, while 69% of those utilizing x-ray tracking were quicker. TrackX cases were initially slower but trended downwards through approximately 11 patients, at which point operative times were consistently quicker (R=0.63). The Mazor cases displayed no learning curve (R=0.08). None of the assisted cases were abandoned in favor of standard fluoroscopy or required hardware revision. CONCLUSIONS Enabling technology can have a significant impact to surgical efficiency. Robotic-assistance consistently negatively impacted operative times, while x-ray tracking was associated with a short learning curve and improved operative times in the vast majority of cases studied. FDA DEVICE/DRUG STATUS TrackX (Approved for this indication) Instrument-navigation modalities including CT-guided and robot-assisted methods claim both efficacy and accuracy when applied to spine surgery, yet often increase setup and operating times. The authors seek to study the impact of multitude of instrument navigational technologies on operative time and the learning curve associated with their use. Prospective, single-surgeon, single-institution. Patients undergoing minimally invasive transforaminal lumbar interbody fusion with or without decompression, with or without previous hardware removal. Total operative time, immediate surgical complications. Consecutive patients undergoing minimally invasive transforaminal lumbar interbody fusion (MIS-TLIF) were analyzed. Two adjunctive technologies were studied and compared with a more conventional fluoroscopic approach without navigation. Consecutive cases were done with assistance of a robot (Mazor Robotics), with the assistance of x-ray tracking (TrackX) or without navigation in three consecutive blocks of time. The cases done without assistance were used to further normalize for number of interbodies and decompressions performed as well as hardware removal if needed. Age, body mass index (BMI), sex, operative levels, laminectomy, need for hardware removal, and total operative time was recorded directly from EPIC. A total of 120 cases (74 conventional, 14 robot-assisted, 32 Track-X) were included in analysis. There were no significant differences in age, sex, or BMI between modalities. Uncontrolled, the average operative time for conventional, robot-assisted, and TrackX was 118±36, 157±41, and 107±37 minutes, respectively, for each MIS-TLIF level (p<0.05 across each group). Once normalized for hardware removal, decompression, and number of fusion levels, the robot cases were 24% slower (added 32 minutes/case) while the x-ray tracking cases were 19% faster (subtracted 16 minutes). Operative times in no robot-assisted case were faster than the comparable control procedure, while 69% of those utilizing x-ray tracking were quicker. TrackX cases were initially slower but trended downwards through approximately 11 patients, at which point operative times were consistently quicker (R=0.63). The Mazor cases displayed no learning curve (R=0.08). None of the assisted cases were abandoned in favor of standard fluoroscopy or required hardware revision. Enabling technology can have a significant impact to surgical efficiency. Robotic-assistance consistently negatively impacted operative times, while x-ray tracking was associated with a short learning curve and improved operative times in the vast majority of cases studied.