Comparison of accuracy, revision, and perioperative outcomes in robot-assisted spine surgeries: systematic review and meta-analysis

J Neurosurg Spine 41:519–531, 2024

Pedicle screw placement guidance is critical in spinal fusions, and spinal surgery robots aim to improve accuracy and reduce complications. Current literature has yet to compare the relative merits of available robotic systems. In this review, the authors aimed to 1) assess the current state of spinal robotics literature; 2) conduct a meta-analysis of robotic performance based on accuracy, speed, and safety; and 3) offer recommendations for robotic system selection.

METHODS Following PRISMA guidelines, the authors conducted a systematic literature review across PubMed, Embase, Cochrane Library, Web of Science, and Scopus as of April 28, 2022, for studies on approved robots for placing lumbar pedicle screws. Three reviewers screened and extracted data relating to the study characteristics, accuracy rate, intraoperative revisions, and reoperations. Secondary performance metrics included operative time, blood loss, and radiation exposure. The authors statistically compared the performance of the robots using a random-effects model to account for variation within and between the studies. Each robot was also compared with performance benchmarks of traditional techniques including freehand, fluoroscopic, and CT-navigated insertion. Finally, we performed a Duval and Tweedie trim-and-fill test to assess for the presence of publication bias.

RESULTS The authors identified 46 studies, describing 4670 patients and 25,054 screws, that evaluated 4 different robotic systems: Mazor X, ROSA, ExcelsiusGPS, and Cirq. The weighted accuracy rates of Gertzbein-Robbins classification grade A or B screws were as follows: ExcelsiusGPS, 98.0%; ROSA, 98.0%; Mazor, 98.2%; and Cirq, 94.2%. No robot was significantly more accurate than the others. However, the accuracy of the ExcelsiusGPS was significantly higher than that of traditional methods, and the accuracies of the Mazor and ROSA were significantly higher than that of fluoroscopy. The intraoperative revision rates were Cirq, 0.55%; ROSA, 0.91%; Mazor, 0.98%; and ExcelsiusGPS, 1.08%. The reoperation rates were Cirq, 0.28%; ExcelsiusGPS, 0.32%; and Mazor, 0.76% (no reoperations were reported for ROSA). Operative times were similar for all robots. Both the ExcelsiusGPS and Mazor were associated with significantly less blood loss than the ROSA. The Cirq had the lowest radiation exposure. Robots tended to be more accurate and generally their use was associated with fewer reoperations and less blood loss than freehand, fluoroscopic, or CT-navigated techniques.

CONCLUSIONS Robotic platforms perform comparably based on key metrics, with high accuracy rates and low intraoperative revision and reoperation rates. The spinal robotics publication rate will continue to accelerate, and choosing a robot will depend on the context of the practice.

A Retrospective Analysis of Pedicle Screw Placement Accuracy Using the ExcelsiusGPS Robotic Guidance System

Operative Neurosurgery 24:242–247, 2023

Robotic guidance has become widespread in spine surgery. Although the intent is improved screw placement, further system-specific data are required to substantiate this intention for pedicle screws in spinal stabilization constructs.

OBJECTIVE: To determine the accuracy of pedicle screws placed with the aid of a robot in a cohort of patients immediately after the adoption of the robot-assisted surgery technique.

METHODS: A retrospective, Institutional Review Board–approved study was performed on the first 100 patients at a single facility, who had undergone spinal surgeries with the use of robotic techniques. Pedicle screw accuracy was graded using the Gertzbein– Robbins Scale based on pedicle wall breach, with grade A representing 0 mm breach and successive grades increasing breach thresholds by 2 mm increments. Preoperative and postoperative computed tomography scans were also used to assess offsets between the objective plan and true screw placements.

RESULTS: A total of 326 screws were analyzed among 72 patients with sufficient imaging data. Ages ranged from 21 to 84 years. The total accuracy rate based on the Gertzbein– Robbins Scale was 97.5%, and the rate for each grade is as follows: A, 82%; B, 15.5%; C, 1.5%; D, 1%; and E, 0. The average tip offset was 1.9 mm, the average tail offset was 2.0 mm, and the average angular offset was 2.6°.

CONCLUSION: Robotic-assisted surgery allowed for accurate implantation of pedicle screws on immediate adoption of this technique. There were no complications attributable to the robotic technique, and no hardware revisions were required.

Ninety-day complication, revision, and readmission rates for current-generation robot-assisted thoracolumbar spinal fusion surgery

J Neurosurg Spine 36:841–848, 2022

Robotics is a major area for research and development in spine surgery. The high accuracy of robot-assisted placement of thoracolumbar pedicle screws is documented in the literature. The authors present the largest case series to date evaluating 90-day complication, revision, and readmission rates for robot-assisted spine surgery using the current generation of robotic guidance systems.

METHODS An analysis of a retrospective, multicenter database of open and minimally invasive thoracolumbar instrumented fusion surgeries using the Mazor X or Mazor X Stealth Edition robotic guidance systems was performed. Patients 18 years of age or older and undergoing primary or revision surgery for degenerative spinal conditions were included. Descriptive statistics were used to calculate rates of malpositioned screws requiring revision, as well as overall complication, revision, and readmission rates within 90 days.

RESULTS In total, 799 surgical cases (Mazor X: 48.81%; Mazor X Stealth Edition: 51.19%) were evaluated, involving robot-assisted placement of 4838 pedicle screws. The overall intraoperative complication rate was 3.13%. No intraoperative implant-related complications were encountered. Postoperatively, 129 patients suffered a total of 146 complications by 90 days, representing an incidence of 16.1%. The rate of an unrecognized malpositioned screw resulting in a new postoperative radiculopathy requiring revision surgery was 0.63% (5 cases). Medical and pain-related complications unrelated to hardware placement accounted for the bulk of postoperative complications within 90 days. The overall surgical revision rate at 90 days was 6.63% with 7 implant-related revisions, representing an implant-related revision rate of 0.88%. The 90-day readmission rate was 7.13% with 2 implant-related readmissions, representing an implant-related readmission rate of 0.25% of cases.

CONCLUSIONS The results of this multicenter case series and literature review suggest current-generation robotic guidance systems are associated with low rates of intraoperative and postoperative implant-related complications, revisions, and readmissions at 90 days. Future outcomes-based studies are necessary to evaluate complication, revision, and readmission rates compared to conventional surgery.

 

Neurological events due to pedicle screw malpositioning with lateral fluoroscopy–guided pedicle screw insertion

J Neurosurg Spine 33:806–811, 2020

The risk of novel postoperative neurological events due to pedicle screw malpositioning in lumbar fusion surgery is minimized by using one of the several image-guided techniques for pedicle screw insertion. These techniques for guided screw insertion range from intraoperative fluoroscopy to intraoperative navigation. A practical technique consists of anatomical identification of the screw entry point followed by lateral fluoroscopy used for guidance during insertion of the screw. This technique is available in most clinics and is less expensive than intraoperative navigation. However, the safety of lateral fluoroscopy–guided pedicle screw placement with regard to novel postoperative neurological events due to screw malposition has been addressed only rarely in the literature. In this study the authors aimed to determine the rate of novel postoperative neurological events due to intraoperative and postoperatively established screw malpositioning during lateral fluoroscopy–assisted screw insertion.

METHODS Included patients underwent lateral fluoroscopy–assisted lumbosacral screw insertion between January 2012 and August 2017. The occurrence of novel postoperative neurological events was analyzed from patient files. In case of an event, surgical reports were screened for the occurrence of intraoperative screw malposition. Furthermore, postoperative CT scans were analyzed to identify and describe possible screw malposition.

RESULTS In total, 246 patients with 1079 screws were included. Novel postoperative neurological events were present in 36 patients (14.6%). In 8 of these 36 patients (3.25% of the total study population), the neurological events could be directly attributed to screw malposition. Screw malpositioning was caused either by problematic screw insertion with immediate screw correction (4 patients) or by malpositioned screws for which the malposition was established postoperatively using CT scans (4 patients). Three patients with screw malposition underwent revision surgery without subsequent symptom relief.

CONCLUSIONS Lateral fluoroscopy–assisted lumbosacral screw placement results in low rates of novel postoperative neurological events caused by screw malposition. In the majority of patients suffering from novel postoperative neurological events, these events could not be attributed to screw malpositioning, but rather were due to postoperative neurapraxia of peripheral nerves, neuropathy, or intraoperative traction of nerve roots.

Instrumentation of the Posterior Thoracolumbar Spine: From Wires to Pedicle Screws

Instrumentation of the Posterior Thoracolumbar Spine- From Wires to Pedicle Screws

Operative Neurosurgery 10:497–505, 2014

Over the past 120 years, spinal stabilization has advanced immensely. An updated review highlighting these advancements has not been performed in the past 20 years.

The objective of this report is to provide a historical assessment of the decades outlining various key innovators, their techniques, and instrumentation.

It is important to provide new generations of surgeons and students with historical evidence of the value of developing new techniques and instrumentation to improve patient care and outcomes.

Placement of thoracolumbar pedicle screws using O-arm-based navigation

Placement of thoracolumbar pedicle screws using O-arm-based navigation- technical note on controlling the operational accuracy of the navigation system

Neurosurg Rev (2013) 36:157–162

Suboptimal placements of pedicle screws may lead to neurological and vascular complications. Computer-assisted image guidance has been shown to improve accuracy in spinal instrumentation.Checking the accuracy of the navigation system during pedicle screw placement is fundamental.

We describe a novel technique of using continuous accuracy check of the navigation system during O-arm based neuronavigation to instrument the thoracolumbar region.

Forty thoracic and 42 lumbar screws were inserted in 12 patients. The Mirza evaluation system was used to evaluate the accuracy of the inserted screws. There was no neurological injury and no need to reposition any screw. The accuracy of the screws placement was excellent.

Our technique of continuous at will operational accuracy check of the neuronavigation system is associated with extreme accuracy of screw placement, no need to bring a patient back to the operating room to reposition a pedicle screw, and with excellent outcome.

Posterior Lumbar Interbody Fusion: Comparison of Single Intervertebral Cage and Single Side Pedicle Screw Fixation versus Bilateral Cages and Screw Fixation

Minim Invas Neurosurg 2009; 52: 132 – 136. DOI 10.1055/s-0029-1224097

Introduction: The efficacy and economy of an alternative sparing method for posterior lumbar interbody fusion (PLIF) using a single cage fixed with pedicle screws placed on a single side (SS group, n = 22) was compared to that of a standard bilateral protocol using two cages and pedicle screws placed bilaterally (BL group, n = 15).

Methods: All PLIFs were non-compensation cases done at a single level by a single surgeon and were similar in most background characteristics. Significant differences were not found between the two groups in fusion rates, complications or in 2-year prospectively collected outcomes including percent improvement in back and leg pain (visual analog scales) and the Oswestry disability index.

Results: Perioperative results significantly favored the SS group: BL patients lost 81 % more blood, used 74 % more time for surgery, stayed in hospital 1.7 days longer, and the hospital-related cost per procedure was twice as high. Currently, the SS procedure typically averages less than 1 hand blood loss less than 50 mL. In summary, the BL and SS groups had similar outcomes while the SS procedure provided substantially superior efficiency and economy.

Conclusion: In conclusion, the results of this retrospective comparative level III study warrant further studies on the SS protocol which may lead to the adoption of this minimally invasive protocol in the standard practice of PLIF in selected cases.