- Email: [email protected]
Comparing Next-Generation Robotic Technology with 3-Dimensional Computed Tomography Navigation Technology for the Insertion of Posterior Pedicle Screws
Original Article
Comparing Next-Generation Robotic Technology with 3-Dimensional Computed Tomography Navigation Technology for the Insertion of Posterior Pedicle Screws Asham Khan1,3, Joshua E. Meyers1,3, Samantha Yavorek2, Timothy E. O’Connor1,3, Ioannis Siasios4, Jeffrey P. Mullin1,3, John Pollina1,3
-
OBJECTIVE: To study the differences between robotguided (Mazor X, Mazor Robotics Ltd., Caesarea, Israel) and 3-dimensional (3D) computed tomography (CT) navigation (O-arm Surgical Imaging System, Medtronic, Minneapolis, Minnesota, USA) for the insertion of pedicle screws.
shorter hospital stay than 3D-CT navigation. Further studies are warranted to verify our results.
METHODS: We reviewed the charts of 50 patients who underwent robot-guided pedicle screw insertion (between May 2017eOctober 2017), and 49 patients who underwent 3D-CT navigation pedicle screw insertion (between September 2015eAugust 2016). Variables included were age, sex, body mass index, blood loss, length of stay, lumbar level(s), operation time, fluoroscopy time, radiation dose, accuracy, and time-per-screw placement.
INTRODUCTION
-
RESULTS: Mean ages were 59.3 years in the robotic group and 58.2 years in the 3D-CT navigation group. Mean was 30.7 kg/m2 in the robotic group and 32.1 kg/m2 in the 3D-CT navigation group. Mean time-per-screw placement was 3.7 minutes for the robotic group and 6.8 minutes for the 3D-CT navigation group, P < 0.001. In the robotic group, 189 of 190 screws were placed with Ravi grade I accuracy, and 1 was grade II. In the 3D-CT navigation group, 157 of 165 screws were Ravi grade I, and 8 were grade II (P [ 0.11). Fluoroscopy time (P < 0.001), time-per-screw placement (P < 0.001), and length of stay (P < 0.001) were significantly lower in the robotic group.
-
CONCLUSIONS: Both technologies are safe and accurate. Robotic technology exposed patients to less fluoroscopy time, decreased time-per-screw placement and
-
Key words - 3D-CT navigation - Mazor X - O-arm - Pedicle screws - Robotic guidance - Spinal stabilization Abbreviations and Acronyms 3D: 3-dimensional CT: Computed tomography K-wire: Kirschner wire ROI: Region of interest
WORLD NEUROSURGERY -: e1-e8, - 2018
S
pinal fusion proceeded by the insertion of posterior pedicle screws is performed routinely to establish spinal stabilization. Pedicle screws can be inserted by several distinct techniques: freehand, fluoroscopy-assisted, computed tomographic (CT)-guided navigation, and robotic-assisted navigation.1-4 Newer navigation technologies, including CT-guided navigation, 3-dimensional (3D)-fluoroscopy, and robot-assisted navigation, were introduced in an attempt to increase accuracy rates because of the risk of neurologic and vascular compromise associated with misplaced screws.1,5-7 In conjunction with the recent introduction of the newer generation Mazor X robot (Mazor Robotics Ltd., Caesarea, Israel), we aimed to compare the differences between this robotic technology and CT-navigation technology (O-arm Surgical Imaging System, Medtronic, Minneapolis, Minnesota, USA) for the insertion of pedicle screws. MATERIALS AND METHODS Patient Population A retrospective review was conducted of the charts of 50 consecutive patients who underwent robotic-guided pedicle screw insertion between May 2017eApril 2018, as well as 49 consecutive patients in whom CT-guided navigation was performed between September 2015eAugust 2016 at Buffalo General Medical Center, which is an academic teaching hospital. Some of the cases in each
From the 1Department of Neurosurgery, 2Jacobs School of Medicine and Biomedical Sciences, University at Buffalo; 3Department of Neurosurgery, Buffalo General Medical Center, Kaleida Health, Buffalo, New York, USA; and 4Department of Neurosurgery, Papageorgiou General Hospital, Thessaloniki, Greece To whom correspondence should be addressed: John Pollina, M.D. [E-mail: [email protected]] Citation: World Neurosurg. (2018). https://doi.org/10.1016/j.wneu.2018.11.190 Journal homepage: www.journals.elsevier.com/world-neurosurgery Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2018 Elsevier Inc. All rights reserved.
www.journals.elsevier.com/world-neurosurgery
e1
ORIGINAL ARTICLE ASHAM KHAN ET AL.
ROBOTIC TECHNOLOGY VS CT NAVIGATION
group have been reported in other articles.2,8 The University at Buffalo institutional review board determined this study to be of minimal risk, and a Health Insurance Portability and Accountability Act waiver was awarded to conduct this study. Informed consent was obtained from the patients before the operative procedures were performed. Patients in both groups were treated for degenerative disc disease with or without spondylolisthesis with lumbar interbody fusion. Since obtaining the new robotic platform in May 2017, we routinely perform all pedicle screw insertions using this technology. All robotics surgeries were performed by a single surgeon (senior author [J.P.]), whereas 3D-CT navigation surgeries were performed by 1 of 2 surgeons (1 of whom is the senior author [V.D.]). Robotic Procedure Registration. Two distinct methods can be used for registration and planning for pedicle screw insertion with the Mazor X: preoperative planning and scan-and-plan. The preoperative planning method uses a preoperative CT scan and the proprietary system software to create a surgical plan with optimal screw position, length, and skin incisions. The registration process for this technique requires anteroposterior and oblique fluoroscopic images. The C-arm is fitted with a specialized fiducialized array (called a star marker) connected to the robot arm with 4 points along the region of interest (ROI). The scan-and-plan technique involves taking an intraoperative CT scan with the O-arm with a 4point fiducialized array connected to the robotic arm, obviating the need for intraoperative fluoroscopy. We found scan-and-plan to be more efficient and primarily used that technique in our cases; preoperative planning was utilized for the first 8 cases and scan-and-plan for the subsequent 42 cases. Robotic Guidance (Scan-and-Plan). We positioned the patient prone on the Jackson fusion table with a chest plate and hip and thigh pads. A specialized robotic attachment was secured to the caudal end of the table. The robotic arm was then attached and draped. A bone mount bridge was secured to the effector arm, and the array was secured to the side of that arm. A Schanz pin was then drilled into the right posterior superior iliac spine, and the O-arm optical array was placed in the left posterior superior iliac spine for utilization of Stealth navigation (Medtronic PLC, Louisville Colorado, USA). A towel was then placed over the surgical ROI and lights were directed away from the field to eliminate glare while an optical camera built in the robotic arm scanned the field and created a 3D environment to avoid collisions with surroundings. A reference stylus was then placed over the ROI, which brought the robotic arm to the approximate location of screw placement prior to registration. The star marker was connected to the effector arm and positioned in the ROI. The O-arm was brought into the field for a CT scan. All 4 beads on the star marker must be seen for registration to be successful. Surgical planning then took place on the robotic computer interface. The robotic arm was sent to the first lumbar level, and incisions were made with a no. 11 blade that is customized to fit through the effector arm. A cannulated dilator was placed through the surgical arm, followed by a drill guide with tines, and docked securely on top of the bony surface to prevent skiving. A 3 mm drill was placed through the drill guide and used to make a pilot hole into the pedicle. A reducing cannula was
e2
www.SCIENCEDIRECT.com
placed through the drill guide followed by a Kirschner wire (Kwire). The outer cannula (the dilator) was placed through the drill guide. The reducing cannula was removed over the wire. The Kwire was pulled through a small opening in the side of the effector arm and secured to the drapes, allowing the robot to move to the next spinal level. After placement of all K-wires, screws were placed manually using a Triton-powered driver (Medtronic PLC) under Stealth navigation based on an intraoperative CT scan obtained at the beginning of the procedure, as this process allows precise visualization of the pedicles in the axial, coronal, and sagittal planes for insertion of pedicle screws. This process is repeated at each spinal level. CT Navigation Procedure The 3D-CT navigation procedures were performed as previously described by several authors (I.S., A.K., J.P.) of this current study.2 Data Collection Data collected for the study included patient demographics, spinal level(s), pedicle screw insertion time (defined as when the robot moved to the defined trajectory to when the screw was inserted in the robotic group; in the CT-navigation group, it was determined taking the total time to place the K-wires and screws and dividing it by the total number of screws inserted), pedicle screw position (breach and deviation) grading, American Society of Anesthesiologists (ASA) class,9 fluoroscopy time, radiation dose, blood loss, duration of the operation (this included planning of the screws in the robotic cohort), and length of hospital stay. Accuracy of the pedicle screw placement was determined by reviewing the immediate postoperative CT scan and applying the scale developed by Ravi et al.10 If the screw is completely in the pedicle, it is considered Grade I; a <2 mm deviation is grade II; a 2 mme4 mm deviation is grade III; and a >4 mm deviation is considered grade IV. Breach direction (superior, inferior, lateral, or medial) was also recorded. Screw placement data were recorded by a single author (A.K.). Statistical Methods Analysis was completed using the Statistical Analysis Software 9.4 package (SAS Institute Inc., Cary, North Carolina, USA). A 2sample Student t test was used for comparison of continuous variables between the 2 groups. A c2 test was used to compare categorical variables between the 2 groups. Alpha was set at 0.05 with P values < 0.05 considered statistically significant. Data missing from the patients’ charts were excluded from final statistical analysis. RESULTS Robotic Group Ten of the 50 patients (20%) had a smoking history, and 31 were women (62%). The mean age of the robotic cohort was 59.3 11.7 years (range 28e77 years). The diagnoses were degenerative disc disease with spondylolisthesis in 34 patients, and without spondylolisthesis in 16 patients. The mean body mass index was 30.7 5.1 kg/m2 (range 19.5e42.3 kg/m2). Demographic data are summarized in Table 1. A total of 190 pedicle screws were placed across 61 lumbar levels with a mean 3.1 screws placed per level. Mean operative time
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2018.11.190
ORIGINAL ARTICLE ASHAM KHAN ET AL.
ROBOTIC TECHNOLOGY VS CT NAVIGATION
Table 1. Demographic Data of Study Cohort Parameter Age (years), mean (SD)
Robotic Group (n [ 50)
3D CT Navigation Group (n [ 49)
59.3 11.7
58.1 10.5
Female (no. patients, %)
31 (62)
24 (48.9)
BMI, mean (SD), (kg/m2)
30.7 5.1
32.1 5.9
10 (20)
10 (20.4)
Smoking history (no. patients, %) Comorbidities (no. patients)
Lung disease* ¼ 3 Diabetes ¼ 8 Osteoarthritis ¼ 4 Heart diseasey ¼ 7
Lung disease* ¼ 10 Diabetes ¼ 10 Osteoarthritis ¼ 8 Heart diseasey ¼ 6
Previous surgery (no. patients)
Anterior fusion ¼ 2 Posterior fusion ¼ 1 Laminectomy ¼ 3 Microdiscectomy ¼ 4
Anterior fusion ¼ 1 Posterior fusion ¼ 5 Laminectomy ¼ 10 Microdiscectomy ¼ 9
Diagnosis
DDD with spondylolisthesis ¼ 34 DDD without spondylolisthesis ¼ 16
DDD with spondylolisthesis ¼ 33 DDD without spondylolisthesis ¼ 16
3D, 3-dimensional; CT, computed tomography; SD, standard deviation; BMI, body mass index; DDD; degenerative disc disease. *Lung disease included chronic obstructive pulmonary disease and asthma. yHeart disease included previous myocardial infarction, pericarditis, and cardiac conduction disorders.
(from skin incision to skin closure) was 153.9 54.7 minutes (range 64e287 minutes). Mean fluoroscopy time was 19.2 23.9 seconds (range 4.47e121.1 seconds), and the mean radiation dose was 58.5 90.5 milligray (range 6.9e189). Operative data are summarized in Table 2. Mean time-per-screw placement was 3.7 1.8 minutes (this parameter was only recorded for 61 screw placements). Pedicle screws placement by lumbar level can be seen in Table 3. Of the 190 screws inserted into the pedicles, 189 (99.5%) were inserted directly (Ravi grade I) (Figure 1) and 1 deviated medially <2 mm (Ravi grade II) owing to skiving; this was assessed by intraoperative CT and repositioned using stereotactic navigation. An accurate visualization of the insertion of pedicle screws can be seen in Figure 2.
8 (4.9%) were Ravi grade II, which is considered acceptable (i.e., no need to remove and replace the pedicle screw).10 Of the 8 screws that were grade II, 7 were laterally deviated and 1 was medially deviated and required revision intraoperatively; all deviations were <2 mm. Robotic versus 3D-CT Navigation When comparison across the 2 groups was made, statistically significant results were observed for time-per-screw placement (P < 0.001), fluoroscopy time (P < 0.001), and length of hospital stay (P ¼ 0.002). Comparison data across all variables can be seen in Table 2. Visualization of the accuracy of screw placement between the 2 groups can be seen in Figure 1. DISCUSSION
3D-CT Navigation Group Ten of the 49 patients (20.4%) had a smoking history, and 24 patients (48.9%) were women. The mean age of this cohort was 58.1 10.5 years (range 34e78 years). The diagnoses were degenerative disc disease with spondylolisthesis in 33 patients and degenerative disc disease without spondylolisthesis in 16 patients. The mean body mass index was 32.1 5.9 kg/m2 (range 20.5e 49.06 kg/m2). Demographic data are summarized in Table 1. A total of 165 pedicle screws were inserted across 74 lumbar levels with a mean of 2.3 screws placed per level. The mean operative time was 162.1.6 69.4 minutes (range 55e319 minutes). The mean fluoroscopy time was 89.2 98.7 seconds (range 5.9e515 seconds). The mean radiation dose was 75.8 139.9 milligray (range 2.7e 900.5). Operative data are summarized in Table 2. The mean time-per-screw placement was 6.8 0.9 minutes (this parameter was only recorded for 75 screw placements). Pedicle screw insertion by lumbar level can be seen in Table 3. Of the 165 inserted pedicle screws, 157 (95.1%) were Ravi grade I and
WORLD NEUROSURGERY -: e1-e8, - 2018
Robotic Guidance Robotic guidance provides many intraoperative utilities compared to the freehand technique and other navigation technologies. When comparing robotic-assistance to CT navigation for pedicle screw insertion, it is important to note that robotic technology eliminates 4 of 6 degrees of freedom of the human arm. This unique feature of robotic guidance has potential to increase accuracy11,12 and reduce complications.11,13,14 Navigation Technology Accuracy Navigation technology has been gaining widespread application during spinal stabilization owing to an increased level of accuracy when placing pedicle screws.3 The accuracy of pedicle screw placement is of utmost importance because misplaced screws can lead to both neurologic and vascular compromise.5,15 Misplacement using the freehand technique is reported in up to 14% of cases.3,16-18 Navigation technologies have lower rates of
www.journals.elsevier.com/world-neurosurgery
e3
ORIGINAL ARTICLE ASHAM KHAN ET AL.
ROBOTIC TECHNOLOGY VS CT NAVIGATION
Table 2. Operative Data of Study Cohort Parameter Operation time (skin to skin) (mean minutes, entire set/minutes, SD)
Robotic Group
3D-CT Navigation Group
Statistical Test
P value
153.9 54.7
162.1 69.4
2-sample t test
0.5
Fluoroscopy time (mean, seconds, SD)
19.2 23.9
89.2 98.7
2-sample t test
<0.001*
Radiation dose (mean, mGy, SD)
58.5 90.5
75.8 139.9
2-sample t test
0.47
Blood loss (mean, mL, SD)
85.2 94.6
100.3 143.5
2-sample t test
0.5
Length of stay (mean, SD) (days)
1.7 1.3
3.0 1.4
2-sample t test
0.002*
Screw time (mean time-per-screw placement, minutes, SD)y
3.7 1.8
6.8 0.9
2-sample t test
<0.001*
Breach grade (According to Ravi et al.)10
Grade I ¼ 189 Grade II ¼ 1
Grade I ¼ 157 Grade II ¼ 8
Fisher exact test
0.14
ASA classþ (Saklad)9
Class II ¼ 35 Class III ¼ 15
Class II ¼ 31 Class III ¼ 18
N/A
N/A
3D, 3-dimensional; CT, computed tomography; SD, standard deviation; mGy, milligray; ASA, American Society of Anesthesiologists; N/A, not applicable. *Statistically significant. yTime-per-screw placement was only recorded for 61 screw placements in the robotic group and 75 screw placements in the 3D-CT navigation group.
misplacement, between 0 and 11% reported in the literature.3,19 The accuracy of pedicle screw insertion using computer-assisted navigation was studied extensively in a meta-analysis performed by Kosmopoulos and Schizas.3 They found an 11% increase in accuracy in lumbar pedicle screw insertion with the use of navigation versus without navigation. These results were later confirmed by both Verma et al.20 and Aoude et al.21 who reported superior accuracy rates for navigation technology compared with freehand technique. Rivkin and Yocum22 analyzed the accuracy of lumbar and thoracic pedicle screw insertion using CT-guided navigation in 270 patients. Applying the breach classification scale developed by Mirza et al.,23 they found an overall breach rate of 5.3%, with varying distributions of grades 1e3.22 When lumbar levels were isolated from their analysis, a total of 1148 screws were inserted in the L1-L5 vertebral levels and a total of 72 (6.3%) were misplaced. Best et al.24 published a study involving 674 screws inserted with computer-assisted spinal navigation and found no misplaced screws on postoperative radiographs, although 1 screw (0.14%) backed out of the pedicle, which was observed during a follow-up visit. Our results are comparable with the current literature on CT-navigation technology in that we report an accuracy rate of 95. 1% across 165 screws with the deviation of 8 screws being clinically acceptable according to the classification scheme of Ravi et al.10 Robotic Accuracy Robotic guidance for the placement of pedicle screws has been previously studied in the literature, primarily with older generation platforms such as the SpineAssist (Mazor Robotics, Ltd.) and Renaissance (Mazor Robotics, Ltd.).4,11-13,15,25-32 Accuracy rates ranging from 85%e99% have been reported.4,11,28,31-33 Keric et al.13 reported an accuracy of 96.7% in the placement of 2067 screws. Laudato et al.32 compared the accuracy of pedicle screw insertion with CT navigation, robotic guidance, and freehand technique. According to a scale by Rampersaud et al.34 that defines severely misplaced screws as >4 mm pedicle wall breach, Hu et al.4
e4
www.SCIENCEDIRECT.com
reported an accuracy rate of 98.9% across 960 screws, with 11 (1. 1%) screws misplaced secondary to skiving. Solomiichuk et al.15 studied accuracy rates for robotic versus fluoroscopy-guided pedicle screw insertion in patients with metastatic disease in the thoracic, lumbar, and sacral spine. Of the 192 screws inserted using the SpineAssist, 129 (67.2%) were Gertzbein-Robbins grade A, 0 mm deviation, and 33 (17.2%) were Gertzbein-Robbins grade B, <2 mm deviation. Grades A and B are considered clinically acceptable, according to the widely used Gertzbein-Robbins classification scale.35 When comparing these results to their fluoroscopy-guided cohort, the robotic group showed slightly superior accuracy rates (84.4% robotic, 83.6% fluoroscopy-guided). We reported an accuracy rate of 99.5% without any complications across 190 pedicle screws insertions, which is comparable to the results associated with older generation platforms. Table 3. Screw Time Comparison of Each Group Across Lumbar Level Lumbar Level
Robotic Group
3D-CT Navigation Group
L1 (minutes, SD)
0
7.3 0.6 (n ¼ 3)
L2 (minutes, SD)
0
7.3 0.8 (n ¼ 7)
L3 (minutes, SD)
4.6 2.2 (n ¼ 10)
7.1 1.2 (n ¼ 12)
0.006*
L4 (minutes, SD)
3.7 1.5 (n ¼ 25)
6.7 0.9 (n ¼ 22)
<0.001*
P value
L5 (minutes, SD)
3.4 1.9 (n ¼ 24)
6.7 0.6 (n ¼ 22)
<0.001*
S1 (minutes, SD)
2.5 0.7 (n ¼ 2)
6.3 0.9 (n ¼ 9)
<0.001*
Total (minutes, SD)y
3.6 1.8 (n ¼ 61)
6.8 0.9 (n ¼ 75)
<0.001*
3D, 3-dimensional; CT, computed tomography; SD, standard deviation. *Statistically significant. yA total of 190 screws were placed into our robotic cohort; screw placement times were recorded for 61 of these screw insertions. A total of 165 screws were inserted into our CT-navigation cohort, and 75 screw placement times were recorded.
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2018.11.190
ORIGINAL ARTICLE ASHAM KHAN ET AL.
Figure 1. Bar graph comparing the percentages for the accuracy of screws inserted with computed tomography navigation and screws inserted using robotic guidance. Grades I and II are from the grading scale determined by Ravi et al.10
Robotic Guidance and CT Navigation Differences between robotic guidance and CT navigation have been studied, albeit with older generation robotic platforms.12,32 Laudato et al.32 (using an earlier robotic system from Mazor Surgical Technologies Ltd.) reported grade A accuracy to be 70.4% in the freehand group, 69.6% in the CT-navigation group, and 78.8% in the robotic group with severely misplaced screws in 2.55% of the
Figure 2. An accurately placed screw within the pedicle in axial (A), coronal (B), and sagittal (C) view on
WORLD NEUROSURGERY -: e1-e8, - 2018
ROBOTIC TECHNOLOGY VS CT NAVIGATION
freehand cohort, 2.62% of the O-arm cohort, and 1.56% of the robotic cohort. In that study, radiologic evaluation was done using the scale developed by Rampersaud et al.34 When comparing this scale to the Ravi scale we used for analysis, severe misplacement is equivalent to Ravi grade IV; nevertheless, the results obtained by Laudato et al.32 in their comparison of the different surgical techniques were not statistically significant. Roser et al.12 conducted a randomized controlled trial comparing accuracy rates with the freehand, CT-navigation, and robotic (SpineAssist) techniques as well. They reported accuracy rates to be 97.5% in the freehand cohort, 92% in the CT-navigation cohort, and 99% in the robotic cohort. This study used the Gertzbein-Robbins classification scale.35 These 2 studies laid the foundation for comparing CT navigation and robotic guidance, although their results are conflicting in terms of accuracy. Our results of 95.1% accuracy in the CT-navigation cohort and 99.5% in the robotic cohort are similar to those reported by Roser et al.12 However, it should be noted that our study used the newest generation robotic technology, and classification scales were different among our study and previously reported studies. Time-Per-Screw Placement Time-per-screw placement is a variable that we studied to determine the intraoperative effectiveness between robotic and CT-navigation technologies. We found that the mean time-perscrew placement differed significantly between our 2 groups with robotic technology reducing the duration of screw insertion by almost a mean of 3 minutes (Table 2). This statistically significant finding of time-per-screw placement between our 2 groups is similar to previously reported data,
postoperative computed tomography images. CT, computed tomography.
www.journals.elsevier.com/world-neurosurgery
e5
ORIGINAL ARTICLE ASHAM KHAN ET AL.
ROBOTIC TECHNOLOGY VS CT NAVIGATION
although no authors made this direct comparison within the scope of the same study.2,17,26,28,30 Siasios et al.2 determined the average time-per-screw placement to be 6.9 minutes per screw using 3D-CT navigation for pedicle screw insertion. Shin et al.17 reported an average time-per-screw placement of 4.5 minutes using CT navigation. With regards to robotic guidance (Renaissance platform), Hyun et al.30 determined that time-per-screw placement decreased between their first 15 cases and subsequent 15 cases from 5.5 minutes to 4.0 minutes, respectively. Using the SpineAssist platform, Pechlivanis et al.28 determined time-per-screw placement to be 3 minutes per screw. Urakov et al.26 (Renaissance platform) determined that it took 5.7 minutes to insert screws percutaneously and 3.6 minutes if the midline open approach was used. Operative Time Skin-to-skin operative time is an important variable to consider when using new technologies, because it provides a measure of the effectiveness of the surgical tool being used. Using ROSA spine robot technology (Zimmer-Biomet, Warsaw, Indiana, USA), Lonjon et al.36 found a significantly longer mean operative time in the robotic cohort than the freehand cohort, and attributed this increased time to both learning curve and set-up time for robotic use. Hyun et al.30 found equal operative times for robotic guidance and freehand technique. Khanna et al.37 compared CT navigation to freehand technique and reported a slight decrease in operative time in the CT-navigation group. The average operative times in the literature are notably longer than those of our own. In our patients, operative times were equivalent when comparing robotic and CT-navigation techniques. Radiation Radiation exposure to the patient has been previously studied regarding navigation technology and robotic guidance. Hyun et al.30 reported average fluoroscopic exposure per screw placement of 3.5 seconds in robot-assisted cases, and 13.3 seconds in fluoroscopic-assisted cases. Kantelhardt et al.11 reported intraoperative x-ray exposure to be 34 seconds for robot-assisted cases and 77 seconds for conventional cases. Roser et al.12 found that robot-assisted cases were associated with less radiation exposure than freehand cases, but more than navigation cases. Our study reports a statistically significant reduction in fluoroscopy time in the robotic cohort compared to the CT-navigation cohort. This also correlated with less radiation exposure in the robotic cohort, compared to the CT-navigation cohort, although this was not statistically significant. Length of Hospital Stay Length of hospital stay also significantly differed between our 2 patient populations with patients in the robotic group being discharged on average 1.3 days earlier than patients who underwent CT navigation insertion of pedicle screws. Hyun et al.30 reported the average length of stay to be 2.5 days less in the robotic group than the freehand technique group, whereas Lonjon et al.36 found no difference between the 2 groups. Xiao et al.38 reported an average length of stay of 4.7 days after CT navigation, although the diagnoses varied in this large patient
e6
www.SCIENCEDIRECT.com
population. Shorter hospital stays were an advantage associated with the use of robotic technology in our study. Complications and Revisions Misplaced pedicle screws can lead to vascular injury, neurologic deficit, fractures, and dural injury with cerebrospinal fluid leaks,5,15 which can necessitate revision surgery. Intraoperative complications pertaining to robotic guidance usually occur owing to skiving of the screw. Devito et al.25 reported a postoperative rate of neurologic complications of 0.7% in 593 patients using the SpineAssist platform. Adverse events were observed at a rate of 4.7% in a cohort of 112 patients by Kantelhardt et al.11 (who also used the SpineAssist platform), which was almost half as much as the 9.1% rate they observed in the freehand cohort. Staartjes et al.39 conducted a systematic review and meta-analysis of revision rates for freehand, CT-navigation and robotic guidance techniques. Those authors found that intraoperative screw revisions occurred for 8.8% of screw insertions in both robotic and CT navigation technique. Postoperative revisions occurred in 0. 82% of insertions using robotic guidance and 1.07% using the CT-navigation technique. In the robotic cohort in our study, 1 screw required intraoperative revision owing to skiving (which was assessed intraoperatively) and a low entry point (assessed postoperatively on retrospective review of the planning); this screw was repositioned without incidence. In the CT-navigation cohort, 1 intraoperative revision occurred without incidence. No postoperative revision surgery was necessary in either cohort. Future Perspectives Robotic assistance for spine surgery offers mechanical guidance for drilling pilot holes and other linear trajectories. Although robotic assistance spine surgery is done with superior accuracy compared to the freehand technique, robotic technologies are semi active, and the surgeon must perform the drilling and screw insertion. In our opinion, superior clinical outcomes must be demonstrated with robotic systems or these systems must provide the surgeon with more intraoperative utility to achieve widespread adoption. Utilities might include decompression of bone and soft tissues during disc preparation and more advanced capabilities, such as prediction of the biomechanics of the implant construct, personalization of implants, conserving or correcting spinal alignment, and prevention of proximal junctional kyphosis. Study Limitations Our study has the inherent limitations of a single center retrospective review along with a small sample size, not recording time-per-screw placement data for all extrapolations of that parameter, and not having independent radiologists review our accuracy data. Furthermore, we would use a repeated measure model for statistical analysis with the F-test for continuous variables, but because of the nature of our data and because this model uses a highly specialized correlation, our sample size was unable to support its application. CONCLUSIONS We present a preliminary comparison of CT navigation and new generation robotic guidance. Both technologies had high
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2018.11.190
ORIGINAL ARTICLE ASHAM KHAN ET AL.
ROBOTIC TECHNOLOGY VS CT NAVIGATION
accuracy rates with percutaneous approaches to insert pedicle screws. We found that robotic technology led to a decrease in time-per-screw placement, decreased fluoroscopy time, and shorter hospital stays. Further studies with larger patient populations and direct comparison of different surgical techniques are warranted to obtain generalizability of the results of our study.
REFERENCES 1. Castro WH, Halm H, Jerosch J, Malms J, Steinbeck J, Blasius S. Accuracy of pedicle screw placement in lumbar vertebrae. Spine (Phila Pa 1976). 1996;21:1320-1324. 2. Siasios ID, Pollina J, Khan A, Dimopoulos VG. Percutaneous screw placement in the lumbar spine with a modified guidance technique based on 3D CT navigation system. J Spine Surg. 2017;3: 657-665. 3. Kosmopoulos V, Schizas C. Pedicle screw placement accuracy: a meta-analysis. Spine (Phila Pa 1976). 2007;32:E111-E120. 4. Hu X, Ohnmeiss DD, Lieberman IH. Roboticassisted pedicle screw placement: lessons learned from the first 102 patients. Eur Spine J. 2013;22: 661-666. 5. Faraj AA, Webb JK. Early complications of spinal pedicle screw. Eur Spine J. 1997;6:324-326. 6. Nasser R, Yadla S, Maltenfort MG, et al. Complications in spine surgery. J Neurosurg Spine. 2010; 13:144-157. 7. Schulze CJ, Munzinger E, Weber U. Clinical relevance of accuracy of pedicle screw placement: a computed tomographic-supported analysis. Spine (Phila Pa 1976). 1998;23:2215-2220 [discussion: 2021]. 8. Khan A, Meyers J, Siasios I, Pollina J. Nextgeneration robotic spine surgery: first report on feasibility, safety, and learning curve [e-pub ahead of print]. Oper Neurosurg (Hagerstown). https://doi. org/10.1093/ons/opy280, Accessed November 20, 2018. 9. Saklad M. Grading of patients for surgical procedures. Available at: http://anesthesiology.pubs.asahq. org/Article.aspx?articleid¼1973837. Accessed November 20, 2018. 10. Ravi B, Zahrai A, Rampersaud R. Clinical accuracy of computer-assisted two-dimensional fluoroscopy for the percutaneous placement of lumbosacral pedicle screws. Spine (Phila Pa 1976). 2011;36: 84-91. 11. Kantelhardt SR, Martinez R, Baerwinkel S, Burger R, Giese A, Rohde V. Perioperative course and accuracy of screw positioning in conventional, open robotic-guided and percutaneous roboticguided, pedicle screw placement. Eur Spine J. 2011;20:860-868. 12. Roser F, Tatagiba M, Maier G. Spinal robotics: current applications and future perspectives. Neurosurgery. 2013;72(suppl 1):12-18.
ACKNOWLEDGMENTS The authors thank Paul H. Dressel for preparation of the illustrations and W. Fawn Dorr and Debra J. Zimmer for editorial assistance. Authors’ contributions: Conception and design, data acquisition, data analysis and interpretation, critically revising the manuscript, and final approval of the manuscript: all authors. Drafting the manuscript: A.K. and J.E.M.
13. Keric N, Doenitz C, Haj A, et al. Evaluation of robot-guided minimally invasive implantation of 2067 pedicle screws. Neurosurg Focus. 2017;42:E11. 14. Schroder ML, Staartjes VE. Revisions for screw malposition and clinical outcomes after robotguided lumbar fusion for spondylolisthesis. Neurosurg Focus. 2017;42:E12. 15. Solomiichuk V, Fleischhammer J, Molliqaj G, et al. Robotic versus fluoroscopy-guided pedicle screw insertion for metastatic spinal disease: a matchedcohort comparison. Neurosurg Focus. 2017;42:E13. 16. Miekisiak G, Kornas P, Lekan M, Dacko W, Latka D, Kaczmarczyk J. Accuracy of the free-hand placement of pedicle screws in the lumbosacral spine using a universal entry point: clinical validation. J Spinal Disord Tech. 2015;28:E194-198. 17. Shin MH, Ryu KS, Park CK. Accuracy and safety in pedicle screw placement in the thoracic and lumbar spines: comparison study between conventional C-arm fluoroscopy and navigation coupled with O-arm guided methods. J Korean Neurosurg Soc. 2012;52:204-209. 18. Silbermann J, Riese F, Allam Y, Reichert T, Koeppert H, Gutberlet M. Computer tomography assessment of pedicle screw placement in lumbar and sacral spine: comparison between free-hand and O-arm based navigation techniques. Eur Spine J. 2011;20:875-881. 19. Gelalis ID, Paschos NK, Pakos EE, et al. Accuracy of pedicle screw placement: a systematic review of prospective in vivo studies comparing free hand, fluoroscopy guidance and navigation techniques. Eur Spine J. 2012;21:247-255. 20. Verma R, Krishan S, Haendlmayer K, Mohsen A. Functional outcome of computer-assisted spinal pedicle screw placement: a systematic review and meta-analysis of 23 studies including 5,992 pedicle screws. Eur Spine J. 2010;19:370-375. 21. Aoude AA, Fortin M, Figueiredo R, Jarzem P, Ouellet J, Weber MH. Methods to determine pedicle screw placement accuracy in spine surgery: a systematic review. Eur Spine J. 2015;24: 990-1004. 22. Rivkin MA, Yocom SS. Thoracolumbar instrumentation with CT-guided navigation (O-arm) in 270 consecutive patients: accuracy rates and lessons learned. Neurosurg Focus. 2014;36:E7. 23. Mirza SK, Wiggins GC, Kuntz CT, et al. Accuracy of thoracic vertebral body screw placement using standard fluoroscopy, fluoroscopic image guidance, and computed tomographic image guidance: a cadaver study. Spine (Phila Pa 1976). 2003; 28:402-413.
WORLD NEUROSURGERY -: e1-e8, - 2018
24. Best NM, Sasso RC, Garrido BJ. Computer-assisted spinal navigation using a percutaneous dynamic reference frame for posterior fusions of the lumbar spine. Am J Orthop. 2009;38:387-391. 25. Devito DP, Kaplan L, Dietl R, et al. Clinical acceptance and accuracy assessment of spinal implants guided with SpineAssist surgical robot: retrospective study. Spine (Phila Pa 1976). 2010;35: 2109-2115. 26. Urakov TM, Chang KH, Burks SS, Wang MY. Initial academic experience and learning curve with robotic spine instrumentation. Neurosurg Focus. 2017;42:E4. 27. Onen MR, Simsek M, Naderi S. Robotic spine surgery: a preliminary report. Turk Neurosurg. 2014; 24:512-518. 28. Pechlivanis I, Kiriyanthan G, Engelhardt M, et al. Percutaneous placement of pedicle screws in the lumbar spine using a bone mounted miniature robotic system: first experiences and accuracy of screw placement. Spine (Phila Pa 1976). 2009;34: 392-398. 29. Kim HJ, Jung WI, Chang BS, Lee CK, Kang KT, Yeom JS. A prospective, randomized, controlled trial of robot-assisted vs freehand pedicle screw fixation in spine surgery. Int J Med Robot. 2017;13. https://doi.org/10.1002/rcs.1779. Accessed November 20, 2018. 30. Hyun SJ, Kim KJ, Jahng TA, Kim HJ. Minimally invasive robotic versus open fluoroscopic-guided spinal instrumented fusions: a randomized controlled trial. Spine (Phila Pa 1976). 2017;42: 353-358. 31. Ringel F, Stuer C, Reinke A, et al. Accuracy of robot-assisted placement of lumbar and sacral pedicle screws: a prospective randomized comparison to conventional freehand screw implantation. Spine (Phila Pa 1976). 2012;37:E496-501. 32. Laudato PA, Pierzchala K, Schizas C. Pedicle screw insertion accuracy using O-arm, robotic guidance or freehand technique: a comparative study. Spine (Phila Pa 1976). 2018;43:E373-378. 33. Lieberman IH, Togawa D, Kayanja MM, et al. Bone-mounted miniature robotic guidance for pedicle screw and translaminar facet screw placement: part I–technical development and a test case result. Neurosurgery. 2006;59:641-650 [discussion: 50]. 34. Rampersaud YR, Pik JH, Salonen D, Farooq S. Clinical accuracy of fluoroscopic computerassisted pedicle screw fixation: a CT analysis. Spine (Phila Pa 1976). 2005;30:E183-190.
www.journals.elsevier.com/world-neurosurgery
e7
ORIGINAL ARTICLE ASHAM KHAN ET AL.
35. Gertzbein SD, Robbins SE. Accuracy of pedicular screw placement in vivo. Spine (Phila Pa 1976). 1990; 15:11-14. 36. Lonjon N, Chan-Seng E, Costalat V, Bonnafoux B, Vassal M, Boetto J. Robot-assisted spine surgery: feasibility study through a prospective casematched analysis. Eur Spine J. 2016;25:947-955.
ROBOTIC TECHNOLOGY VS CT NAVIGATION
39. Staartjes VE, Klukowska AM, Schroder ML. Pedicle screw revision in robot-guided, navigated, and freehand thoracolumbar instrumentation: a systematic review and meta-analysis. World Neurosurg. 2018;116:433-443.e8.
Conflicts of interest statement: Dr. Pollina is involved with surgical training for Medtronic/NuVasive (no compensation received). The other authors have no personal, financial, or institutional interest in the materials or devices described in this manuscript.
38. Xiao R, Miller JA, Sabharwal NC, et al. Clinical outcomes following spinal fusion using an intraoperative computed tomographic 3D imaging system. J Neurosurg Spine. 2017;26:628-637.
Ethical considerations: The University at Buffalo institutional review board reviewed the protocol for this study (STUDY00002057), and it has been given an exempt determination.
www.SCIENCEDIRECT.com
Received 4 September 2018; accepted 20 November 2018 Citation: World Neurosurg. (2018). https://doi.org/10.1016/j.wneu.2018.11.190
37. Khanna AR, Yanamadala V, Coumans JV. Effect of intraoperative navigation on operative time in 1level lumbar fusion surgery. J Clin Neurosci. 2016;32: 72-76.
e8
Previous presentation: This article was presented as a Poster at the 2018 Annual Meeting of the Congress of Neurological Surgeons, October 6e10, Houston, Texas.
Journal homepage: www.journals.elsevier.com/worldneurosurgery Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2018 Elsevier Inc. All rights reserved.
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2018.11.190
Comparing Next-Generation Robotic Technology with 3-Dimensional Computed Tomography Navigation Technology for the Insertion of Posterior Pedicle Screws Asham Khan1,3, Joshua E. Meyers1,3, Samantha Yavorek2, Timothy E. O’Connor1,3, Ioannis Siasios4, Jeffrey P. Mullin1,3, John Pollina1,3
-
OBJECTIVE: To study the differences between robotguided (Mazor X, Mazor Robotics Ltd., Caesarea, Israel) and 3-dimensional (3D) computed tomography (CT) navigation (O-arm Surgical Imaging System, Medtronic, Minneapolis, Minnesota, USA) for the insertion of pedicle screws.
shorter hospital stay than 3D-CT navigation. Further studies are warranted to verify our results.
METHODS: We reviewed the charts of 50 patients who underwent robot-guided pedicle screw insertion (between May 2017eOctober 2017), and 49 patients who underwent 3D-CT navigation pedicle screw insertion (between September 2015eAugust 2016). Variables included were age, sex, body mass index, blood loss, length of stay, lumbar level(s), operation time, fluoroscopy time, radiation dose, accuracy, and time-per-screw placement.
INTRODUCTION
-
RESULTS: Mean ages were 59.3 years in the robotic group and 58.2 years in the 3D-CT navigation group. Mean was 30.7 kg/m2 in the robotic group and 32.1 kg/m2 in the 3D-CT navigation group. Mean time-per-screw placement was 3.7 minutes for the robotic group and 6.8 minutes for the 3D-CT navigation group, P < 0.001. In the robotic group, 189 of 190 screws were placed with Ravi grade I accuracy, and 1 was grade II. In the 3D-CT navigation group, 157 of 165 screws were Ravi grade I, and 8 were grade II (P [ 0.11). Fluoroscopy time (P < 0.001), time-per-screw placement (P < 0.001), and length of stay (P < 0.001) were significantly lower in the robotic group.
-
CONCLUSIONS: Both technologies are safe and accurate. Robotic technology exposed patients to less fluoroscopy time, decreased time-per-screw placement and
-
Key words - 3D-CT navigation - Mazor X - O-arm - Pedicle screws - Robotic guidance - Spinal stabilization Abbreviations and Acronyms 3D: 3-dimensional CT: Computed tomography K-wire: Kirschner wire ROI: Region of interest
WORLD NEUROSURGERY -: e1-e8, - 2018
S
pinal fusion proceeded by the insertion of posterior pedicle screws is performed routinely to establish spinal stabilization. Pedicle screws can be inserted by several distinct techniques: freehand, fluoroscopy-assisted, computed tomographic (CT)-guided navigation, and robotic-assisted navigation.1-4 Newer navigation technologies, including CT-guided navigation, 3-dimensional (3D)-fluoroscopy, and robot-assisted navigation, were introduced in an attempt to increase accuracy rates because of the risk of neurologic and vascular compromise associated with misplaced screws.1,5-7 In conjunction with the recent introduction of the newer generation Mazor X robot (Mazor Robotics Ltd., Caesarea, Israel), we aimed to compare the differences between this robotic technology and CT-navigation technology (O-arm Surgical Imaging System, Medtronic, Minneapolis, Minnesota, USA) for the insertion of pedicle screws. MATERIALS AND METHODS Patient Population A retrospective review was conducted of the charts of 50 consecutive patients who underwent robotic-guided pedicle screw insertion between May 2017eApril 2018, as well as 49 consecutive patients in whom CT-guided navigation was performed between September 2015eAugust 2016 at Buffalo General Medical Center, which is an academic teaching hospital. Some of the cases in each
From the 1Department of Neurosurgery, 2Jacobs School of Medicine and Biomedical Sciences, University at Buffalo; 3Department of Neurosurgery, Buffalo General Medical Center, Kaleida Health, Buffalo, New York, USA; and 4Department of Neurosurgery, Papageorgiou General Hospital, Thessaloniki, Greece To whom correspondence should be addressed: John Pollina, M.D. [E-mail: [email protected]] Citation: World Neurosurg. (2018). https://doi.org/10.1016/j.wneu.2018.11.190 Journal homepage: www.journals.elsevier.com/world-neurosurgery Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2018 Elsevier Inc. All rights reserved.
www.journals.elsevier.com/world-neurosurgery
e1
ORIGINAL ARTICLE ASHAM KHAN ET AL.
ROBOTIC TECHNOLOGY VS CT NAVIGATION
group have been reported in other articles.2,8 The University at Buffalo institutional review board determined this study to be of minimal risk, and a Health Insurance Portability and Accountability Act waiver was awarded to conduct this study. Informed consent was obtained from the patients before the operative procedures were performed. Patients in both groups were treated for degenerative disc disease with or without spondylolisthesis with lumbar interbody fusion. Since obtaining the new robotic platform in May 2017, we routinely perform all pedicle screw insertions using this technology. All robotics surgeries were performed by a single surgeon (senior author [J.P.]), whereas 3D-CT navigation surgeries were performed by 1 of 2 surgeons (1 of whom is the senior author [V.D.]). Robotic Procedure Registration. Two distinct methods can be used for registration and planning for pedicle screw insertion with the Mazor X: preoperative planning and scan-and-plan. The preoperative planning method uses a preoperative CT scan and the proprietary system software to create a surgical plan with optimal screw position, length, and skin incisions. The registration process for this technique requires anteroposterior and oblique fluoroscopic images. The C-arm is fitted with a specialized fiducialized array (called a star marker) connected to the robot arm with 4 points along the region of interest (ROI). The scan-and-plan technique involves taking an intraoperative CT scan with the O-arm with a 4point fiducialized array connected to the robotic arm, obviating the need for intraoperative fluoroscopy. We found scan-and-plan to be more efficient and primarily used that technique in our cases; preoperative planning was utilized for the first 8 cases and scan-and-plan for the subsequent 42 cases. Robotic Guidance (Scan-and-Plan). We positioned the patient prone on the Jackson fusion table with a chest plate and hip and thigh pads. A specialized robotic attachment was secured to the caudal end of the table. The robotic arm was then attached and draped. A bone mount bridge was secured to the effector arm, and the array was secured to the side of that arm. A Schanz pin was then drilled into the right posterior superior iliac spine, and the O-arm optical array was placed in the left posterior superior iliac spine for utilization of Stealth navigation (Medtronic PLC, Louisville Colorado, USA). A towel was then placed over the surgical ROI and lights were directed away from the field to eliminate glare while an optical camera built in the robotic arm scanned the field and created a 3D environment to avoid collisions with surroundings. A reference stylus was then placed over the ROI, which brought the robotic arm to the approximate location of screw placement prior to registration. The star marker was connected to the effector arm and positioned in the ROI. The O-arm was brought into the field for a CT scan. All 4 beads on the star marker must be seen for registration to be successful. Surgical planning then took place on the robotic computer interface. The robotic arm was sent to the first lumbar level, and incisions were made with a no. 11 blade that is customized to fit through the effector arm. A cannulated dilator was placed through the surgical arm, followed by a drill guide with tines, and docked securely on top of the bony surface to prevent skiving. A 3 mm drill was placed through the drill guide and used to make a pilot hole into the pedicle. A reducing cannula was
e2
www.SCIENCEDIRECT.com
placed through the drill guide followed by a Kirschner wire (Kwire). The outer cannula (the dilator) was placed through the drill guide. The reducing cannula was removed over the wire. The Kwire was pulled through a small opening in the side of the effector arm and secured to the drapes, allowing the robot to move to the next spinal level. After placement of all K-wires, screws were placed manually using a Triton-powered driver (Medtronic PLC) under Stealth navigation based on an intraoperative CT scan obtained at the beginning of the procedure, as this process allows precise visualization of the pedicles in the axial, coronal, and sagittal planes for insertion of pedicle screws. This process is repeated at each spinal level. CT Navigation Procedure The 3D-CT navigation procedures were performed as previously described by several authors (I.S., A.K., J.P.) of this current study.2 Data Collection Data collected for the study included patient demographics, spinal level(s), pedicle screw insertion time (defined as when the robot moved to the defined trajectory to when the screw was inserted in the robotic group; in the CT-navigation group, it was determined taking the total time to place the K-wires and screws and dividing it by the total number of screws inserted), pedicle screw position (breach and deviation) grading, American Society of Anesthesiologists (ASA) class,9 fluoroscopy time, radiation dose, blood loss, duration of the operation (this included planning of the screws in the robotic cohort), and length of hospital stay. Accuracy of the pedicle screw placement was determined by reviewing the immediate postoperative CT scan and applying the scale developed by Ravi et al.10 If the screw is completely in the pedicle, it is considered Grade I; a <2 mm deviation is grade II; a 2 mme4 mm deviation is grade III; and a >4 mm deviation is considered grade IV. Breach direction (superior, inferior, lateral, or medial) was also recorded. Screw placement data were recorded by a single author (A.K.). Statistical Methods Analysis was completed using the Statistical Analysis Software 9.4 package (SAS Institute Inc., Cary, North Carolina, USA). A 2sample Student t test was used for comparison of continuous variables between the 2 groups. A c2 test was used to compare categorical variables between the 2 groups. Alpha was set at 0.05 with P values < 0.05 considered statistically significant. Data missing from the patients’ charts were excluded from final statistical analysis. RESULTS Robotic Group Ten of the 50 patients (20%) had a smoking history, and 31 were women (62%). The mean age of the robotic cohort was 59.3 11.7 years (range 28e77 years). The diagnoses were degenerative disc disease with spondylolisthesis in 34 patients, and without spondylolisthesis in 16 patients. The mean body mass index was 30.7 5.1 kg/m2 (range 19.5e42.3 kg/m2). Demographic data are summarized in Table 1. A total of 190 pedicle screws were placed across 61 lumbar levels with a mean 3.1 screws placed per level. Mean operative time
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2018.11.190
ORIGINAL ARTICLE ASHAM KHAN ET AL.
ROBOTIC TECHNOLOGY VS CT NAVIGATION
Table 1. Demographic Data of Study Cohort Parameter Age (years), mean (SD)
Robotic Group (n [ 50)
3D CT Navigation Group (n [ 49)
59.3 11.7
58.1 10.5
Female (no. patients, %)
31 (62)
24 (48.9)
BMI, mean (SD), (kg/m2)
30.7 5.1
32.1 5.9
10 (20)
10 (20.4)
Smoking history (no. patients, %) Comorbidities (no. patients)
Lung disease* ¼ 3 Diabetes ¼ 8 Osteoarthritis ¼ 4 Heart diseasey ¼ 7
Lung disease* ¼ 10 Diabetes ¼ 10 Osteoarthritis ¼ 8 Heart diseasey ¼ 6
Previous surgery (no. patients)
Anterior fusion ¼ 2 Posterior fusion ¼ 1 Laminectomy ¼ 3 Microdiscectomy ¼ 4
Anterior fusion ¼ 1 Posterior fusion ¼ 5 Laminectomy ¼ 10 Microdiscectomy ¼ 9
Diagnosis
DDD with spondylolisthesis ¼ 34 DDD without spondylolisthesis ¼ 16
DDD with spondylolisthesis ¼ 33 DDD without spondylolisthesis ¼ 16
3D, 3-dimensional; CT, computed tomography; SD, standard deviation; BMI, body mass index; DDD; degenerative disc disease. *Lung disease included chronic obstructive pulmonary disease and asthma. yHeart disease included previous myocardial infarction, pericarditis, and cardiac conduction disorders.
(from skin incision to skin closure) was 153.9 54.7 minutes (range 64e287 minutes). Mean fluoroscopy time was 19.2 23.9 seconds (range 4.47e121.1 seconds), and the mean radiation dose was 58.5 90.5 milligray (range 6.9e189). Operative data are summarized in Table 2. Mean time-per-screw placement was 3.7 1.8 minutes (this parameter was only recorded for 61 screw placements). Pedicle screws placement by lumbar level can be seen in Table 3. Of the 190 screws inserted into the pedicles, 189 (99.5%) were inserted directly (Ravi grade I) (Figure 1) and 1 deviated medially <2 mm (Ravi grade II) owing to skiving; this was assessed by intraoperative CT and repositioned using stereotactic navigation. An accurate visualization of the insertion of pedicle screws can be seen in Figure 2.
8 (4.9%) were Ravi grade II, which is considered acceptable (i.e., no need to remove and replace the pedicle screw).10 Of the 8 screws that were grade II, 7 were laterally deviated and 1 was medially deviated and required revision intraoperatively; all deviations were <2 mm. Robotic versus 3D-CT Navigation When comparison across the 2 groups was made, statistically significant results were observed for time-per-screw placement (P < 0.001), fluoroscopy time (P < 0.001), and length of hospital stay (P ¼ 0.002). Comparison data across all variables can be seen in Table 2. Visualization of the accuracy of screw placement between the 2 groups can be seen in Figure 1. DISCUSSION
3D-CT Navigation Group Ten of the 49 patients (20.4%) had a smoking history, and 24 patients (48.9%) were women. The mean age of this cohort was 58.1 10.5 years (range 34e78 years). The diagnoses were degenerative disc disease with spondylolisthesis in 33 patients and degenerative disc disease without spondylolisthesis in 16 patients. The mean body mass index was 32.1 5.9 kg/m2 (range 20.5e 49.06 kg/m2). Demographic data are summarized in Table 1. A total of 165 pedicle screws were inserted across 74 lumbar levels with a mean of 2.3 screws placed per level. The mean operative time was 162.1.6 69.4 minutes (range 55e319 minutes). The mean fluoroscopy time was 89.2 98.7 seconds (range 5.9e515 seconds). The mean radiation dose was 75.8 139.9 milligray (range 2.7e 900.5). Operative data are summarized in Table 2. The mean time-per-screw placement was 6.8 0.9 minutes (this parameter was only recorded for 75 screw placements). Pedicle screw insertion by lumbar level can be seen in Table 3. Of the 165 inserted pedicle screws, 157 (95.1%) were Ravi grade I and
WORLD NEUROSURGERY -: e1-e8, - 2018
Robotic Guidance Robotic guidance provides many intraoperative utilities compared to the freehand technique and other navigation technologies. When comparing robotic-assistance to CT navigation for pedicle screw insertion, it is important to note that robotic technology eliminates 4 of 6 degrees of freedom of the human arm. This unique feature of robotic guidance has potential to increase accuracy11,12 and reduce complications.11,13,14 Navigation Technology Accuracy Navigation technology has been gaining widespread application during spinal stabilization owing to an increased level of accuracy when placing pedicle screws.3 The accuracy of pedicle screw placement is of utmost importance because misplaced screws can lead to both neurologic and vascular compromise.5,15 Misplacement using the freehand technique is reported in up to 14% of cases.3,16-18 Navigation technologies have lower rates of
www.journals.elsevier.com/world-neurosurgery
e3
ORIGINAL ARTICLE ASHAM KHAN ET AL.
ROBOTIC TECHNOLOGY VS CT NAVIGATION
Table 2. Operative Data of Study Cohort Parameter Operation time (skin to skin) (mean minutes, entire set/minutes, SD)
Robotic Group
3D-CT Navigation Group
Statistical Test
P value
153.9 54.7
162.1 69.4
2-sample t test
0.5
Fluoroscopy time (mean, seconds, SD)
19.2 23.9
89.2 98.7
2-sample t test
<0.001*
Radiation dose (mean, mGy, SD)
58.5 90.5
75.8 139.9
2-sample t test
0.47
Blood loss (mean, mL, SD)
85.2 94.6
100.3 143.5
2-sample t test
0.5
Length of stay (mean, SD) (days)
1.7 1.3
3.0 1.4
2-sample t test
0.002*
Screw time (mean time-per-screw placement, minutes, SD)y
3.7 1.8
6.8 0.9
2-sample t test
<0.001*
Breach grade (According to Ravi et al.)10
Grade I ¼ 189 Grade II ¼ 1
Grade I ¼ 157 Grade II ¼ 8
Fisher exact test
0.14
ASA classþ (Saklad)9
Class II ¼ 35 Class III ¼ 15
Class II ¼ 31 Class III ¼ 18
N/A
N/A
3D, 3-dimensional; CT, computed tomography; SD, standard deviation; mGy, milligray; ASA, American Society of Anesthesiologists; N/A, not applicable. *Statistically significant. yTime-per-screw placement was only recorded for 61 screw placements in the robotic group and 75 screw placements in the 3D-CT navigation group.
misplacement, between 0 and 11% reported in the literature.3,19 The accuracy of pedicle screw insertion using computer-assisted navigation was studied extensively in a meta-analysis performed by Kosmopoulos and Schizas.3 They found an 11% increase in accuracy in lumbar pedicle screw insertion with the use of navigation versus without navigation. These results were later confirmed by both Verma et al.20 and Aoude et al.21 who reported superior accuracy rates for navigation technology compared with freehand technique. Rivkin and Yocum22 analyzed the accuracy of lumbar and thoracic pedicle screw insertion using CT-guided navigation in 270 patients. Applying the breach classification scale developed by Mirza et al.,23 they found an overall breach rate of 5.3%, with varying distributions of grades 1e3.22 When lumbar levels were isolated from their analysis, a total of 1148 screws were inserted in the L1-L5 vertebral levels and a total of 72 (6.3%) were misplaced. Best et al.24 published a study involving 674 screws inserted with computer-assisted spinal navigation and found no misplaced screws on postoperative radiographs, although 1 screw (0.14%) backed out of the pedicle, which was observed during a follow-up visit. Our results are comparable with the current literature on CT-navigation technology in that we report an accuracy rate of 95. 1% across 165 screws with the deviation of 8 screws being clinically acceptable according to the classification scheme of Ravi et al.10 Robotic Accuracy Robotic guidance for the placement of pedicle screws has been previously studied in the literature, primarily with older generation platforms such as the SpineAssist (Mazor Robotics, Ltd.) and Renaissance (Mazor Robotics, Ltd.).4,11-13,15,25-32 Accuracy rates ranging from 85%e99% have been reported.4,11,28,31-33 Keric et al.13 reported an accuracy of 96.7% in the placement of 2067 screws. Laudato et al.32 compared the accuracy of pedicle screw insertion with CT navigation, robotic guidance, and freehand technique. According to a scale by Rampersaud et al.34 that defines severely misplaced screws as >4 mm pedicle wall breach, Hu et al.4
e4
www.SCIENCEDIRECT.com
reported an accuracy rate of 98.9% across 960 screws, with 11 (1. 1%) screws misplaced secondary to skiving. Solomiichuk et al.15 studied accuracy rates for robotic versus fluoroscopy-guided pedicle screw insertion in patients with metastatic disease in the thoracic, lumbar, and sacral spine. Of the 192 screws inserted using the SpineAssist, 129 (67.2%) were Gertzbein-Robbins grade A, 0 mm deviation, and 33 (17.2%) were Gertzbein-Robbins grade B, <2 mm deviation. Grades A and B are considered clinically acceptable, according to the widely used Gertzbein-Robbins classification scale.35 When comparing these results to their fluoroscopy-guided cohort, the robotic group showed slightly superior accuracy rates (84.4% robotic, 83.6% fluoroscopy-guided). We reported an accuracy rate of 99.5% without any complications across 190 pedicle screws insertions, which is comparable to the results associated with older generation platforms. Table 3. Screw Time Comparison of Each Group Across Lumbar Level Lumbar Level
Robotic Group
3D-CT Navigation Group
L1 (minutes, SD)
0
7.3 0.6 (n ¼ 3)
L2 (minutes, SD)
0
7.3 0.8 (n ¼ 7)
L3 (minutes, SD)
4.6 2.2 (n ¼ 10)
7.1 1.2 (n ¼ 12)
0.006*
L4 (minutes, SD)
3.7 1.5 (n ¼ 25)
6.7 0.9 (n ¼ 22)
<0.001*
P value
L5 (minutes, SD)
3.4 1.9 (n ¼ 24)
6.7 0.6 (n ¼ 22)
<0.001*
S1 (minutes, SD)
2.5 0.7 (n ¼ 2)
6.3 0.9 (n ¼ 9)
<0.001*
Total (minutes, SD)y
3.6 1.8 (n ¼ 61)
6.8 0.9 (n ¼ 75)
<0.001*
3D, 3-dimensional; CT, computed tomography; SD, standard deviation. *Statistically significant. yA total of 190 screws were placed into our robotic cohort; screw placement times were recorded for 61 of these screw insertions. A total of 165 screws were inserted into our CT-navigation cohort, and 75 screw placement times were recorded.
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2018.11.190
ORIGINAL ARTICLE ASHAM KHAN ET AL.
Figure 1. Bar graph comparing the percentages for the accuracy of screws inserted with computed tomography navigation and screws inserted using robotic guidance. Grades I and II are from the grading scale determined by Ravi et al.10
Robotic Guidance and CT Navigation Differences between robotic guidance and CT navigation have been studied, albeit with older generation robotic platforms.12,32 Laudato et al.32 (using an earlier robotic system from Mazor Surgical Technologies Ltd.) reported grade A accuracy to be 70.4% in the freehand group, 69.6% in the CT-navigation group, and 78.8% in the robotic group with severely misplaced screws in 2.55% of the
Figure 2. An accurately placed screw within the pedicle in axial (A), coronal (B), and sagittal (C) view on
WORLD NEUROSURGERY -: e1-e8, - 2018
ROBOTIC TECHNOLOGY VS CT NAVIGATION
freehand cohort, 2.62% of the O-arm cohort, and 1.56% of the robotic cohort. In that study, radiologic evaluation was done using the scale developed by Rampersaud et al.34 When comparing this scale to the Ravi scale we used for analysis, severe misplacement is equivalent to Ravi grade IV; nevertheless, the results obtained by Laudato et al.32 in their comparison of the different surgical techniques were not statistically significant. Roser et al.12 conducted a randomized controlled trial comparing accuracy rates with the freehand, CT-navigation, and robotic (SpineAssist) techniques as well. They reported accuracy rates to be 97.5% in the freehand cohort, 92% in the CT-navigation cohort, and 99% in the robotic cohort. This study used the Gertzbein-Robbins classification scale.35 These 2 studies laid the foundation for comparing CT navigation and robotic guidance, although their results are conflicting in terms of accuracy. Our results of 95.1% accuracy in the CT-navigation cohort and 99.5% in the robotic cohort are similar to those reported by Roser et al.12 However, it should be noted that our study used the newest generation robotic technology, and classification scales were different among our study and previously reported studies. Time-Per-Screw Placement Time-per-screw placement is a variable that we studied to determine the intraoperative effectiveness between robotic and CT-navigation technologies. We found that the mean time-perscrew placement differed significantly between our 2 groups with robotic technology reducing the duration of screw insertion by almost a mean of 3 minutes (Table 2). This statistically significant finding of time-per-screw placement between our 2 groups is similar to previously reported data,
postoperative computed tomography images. CT, computed tomography.
www.journals.elsevier.com/world-neurosurgery
e5
ORIGINAL ARTICLE ASHAM KHAN ET AL.
ROBOTIC TECHNOLOGY VS CT NAVIGATION
although no authors made this direct comparison within the scope of the same study.2,17,26,28,30 Siasios et al.2 determined the average time-per-screw placement to be 6.9 minutes per screw using 3D-CT navigation for pedicle screw insertion. Shin et al.17 reported an average time-per-screw placement of 4.5 minutes using CT navigation. With regards to robotic guidance (Renaissance platform), Hyun et al.30 determined that time-per-screw placement decreased between their first 15 cases and subsequent 15 cases from 5.5 minutes to 4.0 minutes, respectively. Using the SpineAssist platform, Pechlivanis et al.28 determined time-per-screw placement to be 3 minutes per screw. Urakov et al.26 (Renaissance platform) determined that it took 5.7 minutes to insert screws percutaneously and 3.6 minutes if the midline open approach was used. Operative Time Skin-to-skin operative time is an important variable to consider when using new technologies, because it provides a measure of the effectiveness of the surgical tool being used. Using ROSA spine robot technology (Zimmer-Biomet, Warsaw, Indiana, USA), Lonjon et al.36 found a significantly longer mean operative time in the robotic cohort than the freehand cohort, and attributed this increased time to both learning curve and set-up time for robotic use. Hyun et al.30 found equal operative times for robotic guidance and freehand technique. Khanna et al.37 compared CT navigation to freehand technique and reported a slight decrease in operative time in the CT-navigation group. The average operative times in the literature are notably longer than those of our own. In our patients, operative times were equivalent when comparing robotic and CT-navigation techniques. Radiation Radiation exposure to the patient has been previously studied regarding navigation technology and robotic guidance. Hyun et al.30 reported average fluoroscopic exposure per screw placement of 3.5 seconds in robot-assisted cases, and 13.3 seconds in fluoroscopic-assisted cases. Kantelhardt et al.11 reported intraoperative x-ray exposure to be 34 seconds for robot-assisted cases and 77 seconds for conventional cases. Roser et al.12 found that robot-assisted cases were associated with less radiation exposure than freehand cases, but more than navigation cases. Our study reports a statistically significant reduction in fluoroscopy time in the robotic cohort compared to the CT-navigation cohort. This also correlated with less radiation exposure in the robotic cohort, compared to the CT-navigation cohort, although this was not statistically significant. Length of Hospital Stay Length of hospital stay also significantly differed between our 2 patient populations with patients in the robotic group being discharged on average 1.3 days earlier than patients who underwent CT navigation insertion of pedicle screws. Hyun et al.30 reported the average length of stay to be 2.5 days less in the robotic group than the freehand technique group, whereas Lonjon et al.36 found no difference between the 2 groups. Xiao et al.38 reported an average length of stay of 4.7 days after CT navigation, although the diagnoses varied in this large patient
e6
www.SCIENCEDIRECT.com
population. Shorter hospital stays were an advantage associated with the use of robotic technology in our study. Complications and Revisions Misplaced pedicle screws can lead to vascular injury, neurologic deficit, fractures, and dural injury with cerebrospinal fluid leaks,5,15 which can necessitate revision surgery. Intraoperative complications pertaining to robotic guidance usually occur owing to skiving of the screw. Devito et al.25 reported a postoperative rate of neurologic complications of 0.7% in 593 patients using the SpineAssist platform. Adverse events were observed at a rate of 4.7% in a cohort of 112 patients by Kantelhardt et al.11 (who also used the SpineAssist platform), which was almost half as much as the 9.1% rate they observed in the freehand cohort. Staartjes et al.39 conducted a systematic review and meta-analysis of revision rates for freehand, CT-navigation and robotic guidance techniques. Those authors found that intraoperative screw revisions occurred for 8.8% of screw insertions in both robotic and CT navigation technique. Postoperative revisions occurred in 0. 82% of insertions using robotic guidance and 1.07% using the CT-navigation technique. In the robotic cohort in our study, 1 screw required intraoperative revision owing to skiving (which was assessed intraoperatively) and a low entry point (assessed postoperatively on retrospective review of the planning); this screw was repositioned without incidence. In the CT-navigation cohort, 1 intraoperative revision occurred without incidence. No postoperative revision surgery was necessary in either cohort. Future Perspectives Robotic assistance for spine surgery offers mechanical guidance for drilling pilot holes and other linear trajectories. Although robotic assistance spine surgery is done with superior accuracy compared to the freehand technique, robotic technologies are semi active, and the surgeon must perform the drilling and screw insertion. In our opinion, superior clinical outcomes must be demonstrated with robotic systems or these systems must provide the surgeon with more intraoperative utility to achieve widespread adoption. Utilities might include decompression of bone and soft tissues during disc preparation and more advanced capabilities, such as prediction of the biomechanics of the implant construct, personalization of implants, conserving or correcting spinal alignment, and prevention of proximal junctional kyphosis. Study Limitations Our study has the inherent limitations of a single center retrospective review along with a small sample size, not recording time-per-screw placement data for all extrapolations of that parameter, and not having independent radiologists review our accuracy data. Furthermore, we would use a repeated measure model for statistical analysis with the F-test for continuous variables, but because of the nature of our data and because this model uses a highly specialized correlation, our sample size was unable to support its application. CONCLUSIONS We present a preliminary comparison of CT navigation and new generation robotic guidance. Both technologies had high
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2018.11.190
ORIGINAL ARTICLE ASHAM KHAN ET AL.
ROBOTIC TECHNOLOGY VS CT NAVIGATION
accuracy rates with percutaneous approaches to insert pedicle screws. We found that robotic technology led to a decrease in time-per-screw placement, decreased fluoroscopy time, and shorter hospital stays. Further studies with larger patient populations and direct comparison of different surgical techniques are warranted to obtain generalizability of the results of our study.
REFERENCES 1. Castro WH, Halm H, Jerosch J, Malms J, Steinbeck J, Blasius S. Accuracy of pedicle screw placement in lumbar vertebrae. Spine (Phila Pa 1976). 1996;21:1320-1324. 2. Siasios ID, Pollina J, Khan A, Dimopoulos VG. Percutaneous screw placement in the lumbar spine with a modified guidance technique based on 3D CT navigation system. J Spine Surg. 2017;3: 657-665. 3. Kosmopoulos V, Schizas C. Pedicle screw placement accuracy: a meta-analysis. Spine (Phila Pa 1976). 2007;32:E111-E120. 4. Hu X, Ohnmeiss DD, Lieberman IH. Roboticassisted pedicle screw placement: lessons learned from the first 102 patients. Eur Spine J. 2013;22: 661-666. 5. Faraj AA, Webb JK. Early complications of spinal pedicle screw. Eur Spine J. 1997;6:324-326. 6. Nasser R, Yadla S, Maltenfort MG, et al. Complications in spine surgery. J Neurosurg Spine. 2010; 13:144-157. 7. Schulze CJ, Munzinger E, Weber U. Clinical relevance of accuracy of pedicle screw placement: a computed tomographic-supported analysis. Spine (Phila Pa 1976). 1998;23:2215-2220 [discussion: 2021]. 8. Khan A, Meyers J, Siasios I, Pollina J. Nextgeneration robotic spine surgery: first report on feasibility, safety, and learning curve [e-pub ahead of print]. Oper Neurosurg (Hagerstown). https://doi. org/10.1093/ons/opy280, Accessed November 20, 2018. 9. Saklad M. Grading of patients for surgical procedures. Available at: http://anesthesiology.pubs.asahq. org/Article.aspx?articleid¼1973837. Accessed November 20, 2018. 10. Ravi B, Zahrai A, Rampersaud R. Clinical accuracy of computer-assisted two-dimensional fluoroscopy for the percutaneous placement of lumbosacral pedicle screws. Spine (Phila Pa 1976). 2011;36: 84-91. 11. Kantelhardt SR, Martinez R, Baerwinkel S, Burger R, Giese A, Rohde V. Perioperative course and accuracy of screw positioning in conventional, open robotic-guided and percutaneous roboticguided, pedicle screw placement. Eur Spine J. 2011;20:860-868. 12. Roser F, Tatagiba M, Maier G. Spinal robotics: current applications and future perspectives. Neurosurgery. 2013;72(suppl 1):12-18.
ACKNOWLEDGMENTS The authors thank Paul H. Dressel for preparation of the illustrations and W. Fawn Dorr and Debra J. Zimmer for editorial assistance. Authors’ contributions: Conception and design, data acquisition, data analysis and interpretation, critically revising the manuscript, and final approval of the manuscript: all authors. Drafting the manuscript: A.K. and J.E.M.
13. Keric N, Doenitz C, Haj A, et al. Evaluation of robot-guided minimally invasive implantation of 2067 pedicle screws. Neurosurg Focus. 2017;42:E11. 14. Schroder ML, Staartjes VE. Revisions for screw malposition and clinical outcomes after robotguided lumbar fusion for spondylolisthesis. Neurosurg Focus. 2017;42:E12. 15. Solomiichuk V, Fleischhammer J, Molliqaj G, et al. Robotic versus fluoroscopy-guided pedicle screw insertion for metastatic spinal disease: a matchedcohort comparison. Neurosurg Focus. 2017;42:E13. 16. Miekisiak G, Kornas P, Lekan M, Dacko W, Latka D, Kaczmarczyk J. Accuracy of the free-hand placement of pedicle screws in the lumbosacral spine using a universal entry point: clinical validation. J Spinal Disord Tech. 2015;28:E194-198. 17. Shin MH, Ryu KS, Park CK. Accuracy and safety in pedicle screw placement in the thoracic and lumbar spines: comparison study between conventional C-arm fluoroscopy and navigation coupled with O-arm guided methods. J Korean Neurosurg Soc. 2012;52:204-209. 18. Silbermann J, Riese F, Allam Y, Reichert T, Koeppert H, Gutberlet M. Computer tomography assessment of pedicle screw placement in lumbar and sacral spine: comparison between free-hand and O-arm based navigation techniques. Eur Spine J. 2011;20:875-881. 19. Gelalis ID, Paschos NK, Pakos EE, et al. Accuracy of pedicle screw placement: a systematic review of prospective in vivo studies comparing free hand, fluoroscopy guidance and navigation techniques. Eur Spine J. 2012;21:247-255. 20. Verma R, Krishan S, Haendlmayer K, Mohsen A. Functional outcome of computer-assisted spinal pedicle screw placement: a systematic review and meta-analysis of 23 studies including 5,992 pedicle screws. Eur Spine J. 2010;19:370-375. 21. Aoude AA, Fortin M, Figueiredo R, Jarzem P, Ouellet J, Weber MH. Methods to determine pedicle screw placement accuracy in spine surgery: a systematic review. Eur Spine J. 2015;24: 990-1004. 22. Rivkin MA, Yocom SS. Thoracolumbar instrumentation with CT-guided navigation (O-arm) in 270 consecutive patients: accuracy rates and lessons learned. Neurosurg Focus. 2014;36:E7. 23. Mirza SK, Wiggins GC, Kuntz CT, et al. Accuracy of thoracic vertebral body screw placement using standard fluoroscopy, fluoroscopic image guidance, and computed tomographic image guidance: a cadaver study. Spine (Phila Pa 1976). 2003; 28:402-413.
WORLD NEUROSURGERY -: e1-e8, - 2018
24. Best NM, Sasso RC, Garrido BJ. Computer-assisted spinal navigation using a percutaneous dynamic reference frame for posterior fusions of the lumbar spine. Am J Orthop. 2009;38:387-391. 25. Devito DP, Kaplan L, Dietl R, et al. Clinical acceptance and accuracy assessment of spinal implants guided with SpineAssist surgical robot: retrospective study. Spine (Phila Pa 1976). 2010;35: 2109-2115. 26. Urakov TM, Chang KH, Burks SS, Wang MY. Initial academic experience and learning curve with robotic spine instrumentation. Neurosurg Focus. 2017;42:E4. 27. Onen MR, Simsek M, Naderi S. Robotic spine surgery: a preliminary report. Turk Neurosurg. 2014; 24:512-518. 28. Pechlivanis I, Kiriyanthan G, Engelhardt M, et al. Percutaneous placement of pedicle screws in the lumbar spine using a bone mounted miniature robotic system: first experiences and accuracy of screw placement. Spine (Phila Pa 1976). 2009;34: 392-398. 29. Kim HJ, Jung WI, Chang BS, Lee CK, Kang KT, Yeom JS. A prospective, randomized, controlled trial of robot-assisted vs freehand pedicle screw fixation in spine surgery. Int J Med Robot. 2017;13. https://doi.org/10.1002/rcs.1779. Accessed November 20, 2018. 30. Hyun SJ, Kim KJ, Jahng TA, Kim HJ. Minimally invasive robotic versus open fluoroscopic-guided spinal instrumented fusions: a randomized controlled trial. Spine (Phila Pa 1976). 2017;42: 353-358. 31. Ringel F, Stuer C, Reinke A, et al. Accuracy of robot-assisted placement of lumbar and sacral pedicle screws: a prospective randomized comparison to conventional freehand screw implantation. Spine (Phila Pa 1976). 2012;37:E496-501. 32. Laudato PA, Pierzchala K, Schizas C. Pedicle screw insertion accuracy using O-arm, robotic guidance or freehand technique: a comparative study. Spine (Phila Pa 1976). 2018;43:E373-378. 33. Lieberman IH, Togawa D, Kayanja MM, et al. Bone-mounted miniature robotic guidance for pedicle screw and translaminar facet screw placement: part I–technical development and a test case result. Neurosurgery. 2006;59:641-650 [discussion: 50]. 34. Rampersaud YR, Pik JH, Salonen D, Farooq S. Clinical accuracy of fluoroscopic computerassisted pedicle screw fixation: a CT analysis. Spine (Phila Pa 1976). 2005;30:E183-190.
www.journals.elsevier.com/world-neurosurgery
e7
ORIGINAL ARTICLE ASHAM KHAN ET AL.
35. Gertzbein SD, Robbins SE. Accuracy of pedicular screw placement in vivo. Spine (Phila Pa 1976). 1990; 15:11-14. 36. Lonjon N, Chan-Seng E, Costalat V, Bonnafoux B, Vassal M, Boetto J. Robot-assisted spine surgery: feasibility study through a prospective casematched analysis. Eur Spine J. 2016;25:947-955.
ROBOTIC TECHNOLOGY VS CT NAVIGATION
39. Staartjes VE, Klukowska AM, Schroder ML. Pedicle screw revision in robot-guided, navigated, and freehand thoracolumbar instrumentation: a systematic review and meta-analysis. World Neurosurg. 2018;116:433-443.e8.
Conflicts of interest statement: Dr. Pollina is involved with surgical training for Medtronic/NuVasive (no compensation received). The other authors have no personal, financial, or institutional interest in the materials or devices described in this manuscript.
38. Xiao R, Miller JA, Sabharwal NC, et al. Clinical outcomes following spinal fusion using an intraoperative computed tomographic 3D imaging system. J Neurosurg Spine. 2017;26:628-637.
Ethical considerations: The University at Buffalo institutional review board reviewed the protocol for this study (STUDY00002057), and it has been given an exempt determination.
www.SCIENCEDIRECT.com
Received 4 September 2018; accepted 20 November 2018 Citation: World Neurosurg. (2018). https://doi.org/10.1016/j.wneu.2018.11.190
37. Khanna AR, Yanamadala V, Coumans JV. Effect of intraoperative navigation on operative time in 1level lumbar fusion surgery. J Clin Neurosci. 2016;32: 72-76.
e8
Previous presentation: This article was presented as a Poster at the 2018 Annual Meeting of the Congress of Neurological Surgeons, October 6e10, Houston, Texas.
Journal homepage: www.journals.elsevier.com/worldneurosurgery Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2018 Elsevier Inc. All rights reserved.
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2018.11.190