Accuracy of pedicle screw placement based on preoperative computed tomography versus intraoperative data set acquisition for spinal navigation system

Abstract

Aim:

To investigate the accuracy of pedicle screw placement based on preoperative computed tomography in comparison with intraoperative data set acquisition for spinal navigation system.

Methods:

The PubMed (MEDLINE), EMBASE, and Web of Science were systematically searched for the literature published up to September 2015. This review followed the Preferred Reporting Items for Systematic Reviews and Meta-analysis guidelines. Statistical analysis was performed using the Review Manager 5.3. The dichotomous data for the pedicle violation rate was summarized using relative risk (RR) and 95% confidence intervals (CIs) with the fixed-effects model. The level of significance was set at p < 0.05.

Results:

For this meta-analysis, seven studies used a total of 579 patients and 2981 screws. The results revealed that the accuracy of intraoperative data set acquisition method is significantly higher than preoperative one using 2 mm grading criteria (RR: 1.82, 95% CI: 1.09, 3.04, I ² = 0%, p = 0.02). However, there was no significant difference between two kinds of methods at the 0 mm grading criteria (RR: 1.13, 95% CI: 0.88, 1.46, I ² = 17%, p = 0.34). Using the 2-mm grading criteria, there was a higher accuracy of pedicle screw insertion in O-arm-assisted navigation than CT-based navigation method (RR: 1.96, 95% CI: 1.05, 3.64, I ² = 0%, p = 0.03). The accuracy between CT-based navigation and two-dimensional-based navigation showed no significant difference (RR: 1.02, 95% CI: 0.35-3.03, I ² = 0%, p = 0.97).

Conclusions:

The intraoperative data set acquisition method may decrease the incidence of perforated screws over 2 mm but not increase the number of screws fully contained within the pedicle compared to preoperative CT-based navigation system. A significantly higher accuracy of intraoperative (O-arm) than preoperative CT-based navigation was revealed using 2 mm grading criteria.

Keywords

accuracy computed tomography fluoroscopy navigation

Introduction

Pedicle screw fixation is widely used in spinal surgery for traumatic, degenerative, and oncological diseases because of its three-column stability and good fixation for the vertebral body.^1
–3 The placement of pedicle screws not only influence clinical efficacy but also relate to complications of surgical treatment. The malposition of pedicle screws may lead to severe complications, including dural tearing, neural damage, and vascular or visceral injury, especially in the cervical and thoracic spine due to their complex anatomy and pedicle dimensions.^4
–6 On account of it, surgeons try several methods to improve the accuracy of pedicle screw placement.

Conventional pedicle screw placement focused on the anatomical landmarks, entry point, and insertion angle with or without intraoperative two-dimensional (2D) imaging can be performed by experienced surgeons. With the development of computer-assisted technology, image navigation is applied to spinal operation as the latest surgical procedure. Some research demonstrate that operations with the assistance of navigation systems result in a higher accuracy of pedicle screw placement and less risk of neurovascular injuries than operations without navigation systems.^7

–11 Common navigation systems applied to assist pedicle screw insertion include computed tomography–based navigation (CT Nav), 2D fluoroscopy-based navigation (2D FluoroNav), 3D fluoroscropy-based navigation (3D FluoroNav), and O-arm-assisted navigation (O-arm Nav). The first one involves preoperative data set acquisition and the rest involve intraoperative data set acquisition. CT Nav conducts preoperative CT scan of the target vertebrae and the data are transferred to the navigational system. After matching and registration for the operation, the preoperative anatomical information is then used to assist pedicle screw insertion.¹² Conversely, 2D FluoroNav, 3D FluoroNav, and O-arm Nav acquire information during the operation. 2D FluoroNav uses the intraoperative 2D images and provides real-time intraoperative visualization of spinal anatomy.¹³ The 3D FluoroNav uses real-time intraoperatively 3D visualization.¹⁴ O-arm Nav uses a 3D CT-like navigation system through a computerized image-guidance system.^15,16

Up to now, the difference in various navigation methods does not lead to a consistent view. Few studies analyze the difference in the accuracy of pedicle screw placement using navigation-assisted methods to acquire information preoperatively or intraoperatively. In this study, we performed a systematic review and meta-analysis to investigate the accuracy of pedicle screw placement based on preoperative computed tomography in comparison with intraoperative data set acquisition for spinal navigation system.

Methods

Search strategy

PubMed (MEDLINE), EMBASE, and Web of Science were searched for the literature published up to September 2015 based on the combination of the following key words: navigation, navigated, computer assisted, computer assistance, computer aided, image guided, image guiding, image guidance, neuronavigation, and pedicle screw. Reference lists of included studies were reviewed for additional relevant studies. No language or publication date restrictions were applied. Two reviewers (HL and WKC) independently screened all citations and abstracts identified to identify potential eligible studies.

Inclusion/exclusion criteria

Studies were included in the current meta-analysis if they met the following criteria: (1) the article must include navigated pedicle screw insertion based on preoperative computed tomography and intraoperative data set acquisition technique, (2) randomized-controlled trials or comparative studies (prospective and retrospective), and (3) accuracy of pedicle screw placement evaluated by postoperative computed tomography scans. Exclusion criteria were (1) the pedicle screw inserted without navigation, (2) no postoperative accuracy assessment done, (3) cadaveric and animal studies as well as studies that used spine models and morphologic articles, and (4) repetitive studies.

Data extraction

Data abstraction was conducted independently by two reviewers (HL and WKC), with any discrepancies resolved by discussion and consensus. Information extracted on each eligible study included authors, year of publication, type of study, patient characteristics, indications of surgery, methods of pedicle screw insertion (CT-based navigation, 2D/3D fluoro-based navigation, and O-arm-assisted navigation), number of pedicle screws inserted, outcome measurements (number of misplaced screws, pedicle screw violation), and screw-related complications.

Statistical analysis

Review Manager 5.3 (Nordic Cochrane Centre, Cochrane Collaboration, Copenhagen, Sweden) was used for meta-analysis. The dichotomous data for the pedicle violation rate were summarized using relative risk (RR) and 95% confidence intervals (CIs). The level of significance was set at p < 0.05. Heterogeneity was evaluated by χ² test and I ² statistics. If heterogeneity was significant (p < 0.10 and I ² > 50%), the random-effects model was used. Alternately, the fixed-effects model was used. Funnel plots were examined for evidence of publication bias if sufficient studies were included for meta-analysis (n ≥ 10).¹⁷

Results

Study inclusion and characteristics

The results of the search criteria and selection of trials for inclusion are expressed in a flow diagram of study selection as recommended by Preferred Reporting Items for Systematic Reviews and Meta-analysis. Figure 1 shows a summary of the study identification and selection process. After screening the title and reading the abstracts and the full text, we identified seven published retrospective comparative studies^{6,12,18

–22} that met all inclusion criteria.

Figure 1.

Flow diagram of the study identification and selection process.

A total of 579 patients and 2981 screws were included in the seven studies. All pedicle screws were inserted using the spinal navigation system. CT-based navigation and 2D fluoroscopy-based navigation were compared in three studies.^18
–20 CT-based navigation and 3D fluoroscopy-based navigation were compared in two studies.^18,21 And three studies^6,12,22 compared CT-based navigation with O-arm-assisted navigation (Table 1). The number of pedicle screws in different grades according to the postoperative CT assessment and screw-related complications of the respective studies are shown in Table 2.

Table 1.

Characteristics of included studies.

Author and year	Type of study	Anatomic level	Patients	Indications	Methods of insertion	Tools and criteria for screw assessment
Kotani/2014	Retrospective comparative study	Thoracic	N = 61, pre n = 29/intra n = 32	Progressive scoliosis	CT Nav/O-arm	CT 2 mm
Wood/2011	Retrospective comparative study	Lumbar	N = 67, pre n = 24/intra n = 43	Degenerative lumbar diseases/spondylolysis	CT Nav/O-arm	CT 2 mm
Costa/2011	Retrospective comparative study	Lumbosacral L1-S1	N = 100, pre n = 50/intra n = 50	Degenerative spondylolisthesis	CT Nav/O-arm	CT 2 mm
Nottmeier/2009	Retrospective comparative study	Thoracic lumbar sacral T1-S1	N = 220, pre n = 80/intra n = 140	Degenerative spine diseases/spinal sernosis	CT Nav/3D FluoroNav	CT 2 mm
Fu/2008	Retrospective comparative study	Thoracolumbar T8-S1	N = 24, pre n = 11/intra n = 13	Vertebral fracture/degenerative spondylolisthesis/spondylolytic spondylolisthesis/tuberculous spondylitis/failed earlier back surgery/ankylosing spondylitis with pseudarthrosis	CT Nav/2D FluoroNav	CT 2 mm
Lee/2007	Retrospective comparative study	Cervical C7, thoracic T1-T2	N = 28, pre n = 9/intra n = 19	Cervical spondylotic myelopathy/trauma/kyphotic deformity/malignant disease	CT Nav/2D FluoroNav	CT 2 mm
Gruetzner/2004	Retrospective comparative study	Cervical–thoracic lumbar	N = 79, pre n = 27/intra(2D), n = 28/intra(3D), n = 24	Trauma/spinal stenosis, segmental instability/spondylolisthesis	CT Nav/2D Fluoro Nav/3D FluoroNav	CT 2 mm

Pre: preoperative data set acquisition; CT Nav: computed tomography–based navigation; intra: intraoperative data set acquisition; 2D FluoroNav: 2D fluoroscopy-based navigation; 3D FluoroNav: 3D fluoroscopy-based navigation; O-arm: O-arm-assisted navigation.

Table 2.

Grade of screw and screw-related complications of included studies.

Author (year)	Number of screws and criteria					Screw-related complications
Author (year)		I	II	III	IV	Screw-related complications
Kotani (2014)	N = 638	≤0 mm	<2 mm	2–4 mm	≥4 mm	No intraoperative complications; no wound infection was observed
	Pre: 222	188	23	11	0
	Intra: 416	351	52	13	0
Wood (2011)	N = 296	≤0 mm	<2 mm	≥2 mm		No screw-related complications
	Reg: 110	–	103	7
	Con: 186	–	183	3
Costa (2011)	N = 504	≤0 mm	<2 mm	2–4 mm	4–6 mm	No screw-related complications
	Pre: 258	237	18	2	1
	Intra: 246	234	10	1	1

Nottmeier (2009)	N = 951	≤0 mm	>0 mm			Two nerve root injuries without motor deficit, no other complications
	Pre: 314	285	29
	Intra: 637	595	42
Fu (2008)	N = 150	≤0 mm	<2 mm	≥2 mm		No postoperative complications
	Pre: 76	73	2	1
	Intra: 74	69	3	2
Lee (2007)	N = 108	≤0 mm	<2 mm	2–4 mm	>4 mm	No screw-related complications
	Pre: 45	40	5	0	0
	Intra: 63	52	10	1	0
Gruetzner (2004)	N = 334	≤0 mm	<2 mm	≥2 mm		No screw-related complications
	Pre: 112	–	107	5
	Intra(2D): 108	–	105	3
	Intra(3D): 114	–	113	1

Pre: preoperative data set acquisition; intra: intraoperative data set acquisition. I, II, III, and IV are grades of pedicle screws.

Meta-analysis results

The comparison of accuracy of pedicle screw insertion between preoperative and intraoperative data set acquisition methods according to 2 mm grading criteria in six articles is shown in Figure 2. Meta-analysis revealed that accuracy of intraoperative data set acquisition method is significantly higher than preoperative one (RR: 1.82, 95% CI: 1.09-3.04, I ² = 0%, p = 0.02). However, data analyzed in five articles with the 0 mm grading criteria (Figure 3) revealed no significant difference between the two kinds of methods (RR: 1.13, 95% CI: 0.88-1.46, I ² = 17%, p = 0.34).

Figure 2.

Pedicle screw insertion accuracy (2 mm): Preoperative versus intraoperative. Preoperative: computed tomography-based navigation. Intraoperative: two-dimensional (2D) fluoroscopy-based navigation, 3D fluoroscopy-based navigation, and O-arm-assisted navigation.

Figure 3.

Pedicle screw insertion accuracy (0 mm): Preoperative versus intraoperative. Preoperative: computed tomography–based navigation. Intraoperative: two-dimensional (2D) fluoroscopy-based navigation, 3D fluoroscopy-based navigation, and O-arm-assisted navigation.

As a respective comparison, there was a higher accuracy of pedicle screw placement in O-arm-assisted navigation than CT-based navigation method, which was analyzed at the 2 mm grading criteria in Figure 4 (RR: 1.96, 95% CI: 1.05-3.64, I ² = 0%, p = 0.03). Using the same 2-mm grading criteria, the accuracy of CT-based navigation and 2D-based navigation showed no significant difference, as shown in Figure 5 (RR: 1.02, 95% CI: 0.35-3.03, I ² = 0%, p = 0.97). The difference between CT-based navigation and 3D-based navigation was not analyzed because of the different grading criteria in two studies^18,21 (0 mm and 2 mm).

Figure 4.

Pedicle screw insertion accuracy (2 mm): CT Nav versus O-arm. CT Nav: computed tomography–based navigation, O-arm: O-arm-assisted navigation.

Figure 5.

Pedicle screw insertion accuracy (2 mm): CT Nav versus two-dimensional (2D) FluoroNav. CT Nav: computed tomography–based navigation; 2D FluoroNav: 2D fluoroscopy-based navigation.

Discussion

Several studies have reported that CT-based navigation system may provide higher accuracy of pedicle screw placement than conventional nonnavigated techniques.⁹ However, the disadvantages such as the steep learning curve, the excessive operative preparation, and the intraoperative registration compared with other navigation systems are concerned raised by surgeons. What’s more, the main limitation of the CT-based navigation the variation in the patient position from that in the preoperative CT scan (which was obtained in a supine position preoperatively, but intraoperatively in prone position). For spinal instability resulting from trauma or isthmic spondylolisthesis, the position variation may change the intersegmental relationships between the adjacent vertebral levels, resulting in potential navigational inaccuracy. In the recent year, intraoperative data set acquisition navigation techniques are used to improve the accuracy of pedicle screw placement and solve the above problems, especially the position-dependent problem.¹² Moreover, intraoperative navigations offer the possibility of timely revision surgery for misplaced screws during the surgical procedure to decrease the incidence of postoperative complications.

Ideal pedicle screw should have a maximum diameter and length without breaching the cortical layer of the pedicle or the vertebral body.²³ According to the degree of cortex violation, most research consider pedicle violation as a safe zone of pedicle perforation of smaller than 2 mm.^24,25 The grade of pedicle screw violation based on Gertzbein classification²⁶ is defined into four groups: screws fully contained within the pedicle (grade I), perforated screws with up to 2 mm of displacement (grade II), perforated screws with 2–4 mm of displacement (grade III), and perforated screws with greater than 4 mm of displacement (grade IV). However, the grading criteria evaluating the accuracy of pedicle screw placement often varies in different studies. The seven articles enrolled in our study followed three kinds of grading criteria. We statistically analyzed these articles using the two criteria (0 and 2 mm). The results revealed significant difference using 2 mm criteria and no significant difference using 0 mm criteria, comparing the accuracy of pedicle screw insertion between preoperative and intraoperative data set acquisition methods. The intraoperative data set acquisition method may decrease the incidence of perforated screws over 2 mm but not increase the number of screws fully contained within the pedicle compared to the preoperative CT-based navigation system. The decreased number of perforated screws over 2 mm in intraoperatively real-time navigated methods may lead to lower incidence rate of severe complications. No statistical difference in 0 mm criteria implies that there is still the necessity to improve the computer-assisted navigation system for more accurate placement of pedicle screws and achieve more number of screws fully contained within the pedicle.

Tian et al.²⁷ revealed a higher median placement accuracy with a CT-based navigation system (90.7% vs. 85.5%) than with the 2D fluoroscopy-based navigation on 7533 lumbar pedicle screws. However, in our study, there was no significant difference in pedicle screw insertion accuracy (2 mm), with no consideration of screw violation directions between CT-based and 2D fluoroscopy-based navigation systems. Their own advantages and disadvantages of the two navigated methods may lead to the statistical result. 2D guidance is intraoperative real-time navigation but does not provide any axial images and the quality of the images is directly dependent on the resolution of the acquired fluoroscopic projections so that image interpretation may be difficult and navigation affected adversely.¹² CT-based navigation system provides axial, coronal, and sagittal images and decreases the reliance on intraoperative imaging and reduces the use of intraoperative ionizing radiation.

The O-arm navigation, one of the latest intraoperative imaging platforms, has been well acknowledged for its superiorities in surgeries for trauma or degenerative spinal diseases.^16,25,28 A higher accuracy of intraoperative than preoperative CT-based navigation was revealed in our meta-analysis results. The main reason may be the position changes (from supine to prone) of the patients. In addition, the automatic recognition and merging of intraoperative images with surgical anatomy and real-time registration minimize the range of error.

A cortex violation, in many cases, does not correlate with neurologic, vascular, and/or visceral complication.²⁹ In our study, only one article²¹ reported two cases of nerve root injuries but no motor deficit, and other six articles reported no screw-related complications. Some studies suggest an association between the degree of cortex violation and the frequency of neurological symptoms.³⁰ The actual tendency is a continuous research to obtain better results minimizing surgical risks.

Several limitations should be interpreted in this meta-analysis. First, the types of studies enrolled in this meta-analysis were only retrospective comparative studies, of which the data included in the present study were reported from nonrandomized-controlled trials. A meta-analysis of such data leads to less powerful results compared to those obtained purely from randomized studies. However, a prospective randomized trial is difficult due to poor patient compliance and complex demographic characteristics of patients. Second, only seven articles were enrolled in this study. And there are not enough references to statistically analyze the difference between CT-based navigation and 3D-based navigation. More high-quality articles should be searched in other database systems to strengthen the quality of evidence.

Conclusion

The present systematic review provides a comprehensive evaluation of the available evidence on the accuracy of pedicle screw placement based on preoperative computed tomography in comparison with intraoperative data set acquisition for spinal navigation system. This meta-analysis found an overall significantly higher accuracy of pedicle screw placement with intraoperative data set acquisition than preoperative method using 2 mm grading criteria. However, there was no statistical difference in the number of screws fully contained within the pedicle between preoperative and intraoperative data set acquisition methods. A significantly higher accuracy of intraoperative (O-arm) than preoperative CT-based navigation was revealed using 2 mm grading criteria.

Footnotes

Authors’ Note

Hao Liu and Weikai Chen these authors are contributed equally to this work.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Patton

Morris

Kuo

. Accuracy of fluoroscopy versus computer-assisted navigation for the placement of anterior cervical pedicle screws. Spine (Phila Pa 1976) 2015; 40(7): E404–E410.

Yuan

Zhang

. Clinical evaluation of the polymethylmethacrylate-augmented thoracic and lumbar pedicle screw fixation guided by the three-dimensional navigation for the osteoporosis patients. Eur Spine J 2015; 24(5): 1043–1050.

Luther

Iorgulescu

Geannette

. Comparison of navigated versus non-navigated pedicle screw placement in 260 patients and 1434 screws: screw accuracy, screw size, and the complexity of surgery. J Spinal Disord Tech 2015; 28(5): E298–E303.

Panjabi

Takata

Goel

. Thoracic human vertebrae. Quantitative three-dimensional anatomy. Spine (Phila Pa 1976) 1991; 16(8): 888–901.

Vaccaro

Rizzolo

Balderston

. Placement of pedicle screws in the thoracic spine. Part II: An anatomical and radiographic assessment. J Bone Joint Surg Am 1995; 77(8): 1200–1206.

Kotani

Akazawa

Sakuma

. Accuracy of pedicle screw placement in scoliosis surgery: a comparison between conventional computed tomography-based and O-arm-based navigation techniques. Asian Spine J 2014; 8(3): 331–338.

Tian

Huang

Zhou

. Pedicle screw insertion accuracy with different assisted methods: a systematic review and meta-analysis of comparative studies. Eur Spine J 2011; 20(6): 846–859.

Shin

James

Njoku

. Pedicle screw navigation: a systematic review and meta-analysis of perforation risk for computer-navigated versus freehand insertion. J Neurosurg Spine 2012; 17(2): 113–122.

Gelalis

Paschos

Pakos

. 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(2): 247–255.

10.

Tang

Zhu

Sui

. Position and complications of pedicle screw insertion with or without image-navigation techniques in the thoracolumbar spine: a meta-analysis of comparative studies. J Biomed Res 2014; 28(3): 228–239.

11.

Bourgeois

Faulkner

Bradley

. Improved accuracy of minimally invasive transpedicular screw placement in the lumbar spine with three-dimensional stereotactic image guidance: a comparative meta-analysis. J Spinal Disord Tech 2015; 28(9): 324–329.

12.

Costa

Cardia

Ortolina

. Spinal navigation: standard preoperative versus intraoperative computed tomography data set acquisition for computer-guidance system: radiological and clinical study in 100 consecutive patients. Spine (Phila Pa 1976) 2011; 36(24): 2094–2098.

13.

Briem

Windolf

Rueger

JM.

Percutaneous, 2D-fluoroscopic navigated iliosacral screw placement in the supine position: technique, possibilities, and limits. Unfallchirurg 2007; 110(5): 393–401.

14.

Wendl

von Recum

Wentzensen

. Iso-C(3D0-assisted) navigated implantation of pedicle screws in thoracic lumbar vertebrae. Unfallchirurg 2003; 106(11): 907–913.

15.

Ammirati

Salma

. 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(1): 157–162.

16.

Rivkin

Yocom

. Thoracolumbar instrumentation with CT-guided navigation (O-arm) in 270 consecutive patients: accuracy rates and lessons learned. Neurosurg Focus 2014; 36(3): E7.

17.

Ioannidis

Trikalinos

. The appropriateness of asymmetry tests for publication bias in meta-analyses: a large survey. CMAJ 2007; 176(8): 1091–1096.

18.

Gruetzner

Waelti

Vock

. Navigation using fluoro-CT technology concept and clinical experience in a new method for intraoperative navigation. Eur J Trauma 2004;30:161–170.

19.

Lee

Massicotte

Rampersaud

. Clinical accuracy of cervicothoracic pedicle screw placement: a comparison of the “open” lamino-foraminotomy and computer-assisted techniques. J Spinal Disord Tech 2007; 20(1):25–32.

20.

Wong

Tsai

. Pedicle screw insertion: computed tomography versus fluoroscopic image guidance. Int Orthop 2008; 32(4): 517–521.

21.

Nottmeier

Seemer

Young

. Placement of thoracolumbar pedicle screws using three-dimensional image guidance experience in a large patient cohort. J Neurosurg Spine 2009;10(1):33–39.

22.

Wood

Mannion

. A comparison of CT-based navigation techniques for minimally invasive lumbar pedicle screw placement. J Spinal Disord Tech 2011; 24(1): E1–E5.

23.

Cho

. The biomechanics of pedicle screw-based instrumentation. J Bone Joint Surg Br 2010; 92(8): 1061–1065.

24.

Shin

Hur

Ryu

. Prospective comparison study between the fluoroscopy-guided and navigation coupled with O-arm-guided pedicle screw placement in the thoracic and lumbosacral spines. J Spinal Disord Tech 2015; 28(6): E347–E351.

25.

Jin

Liu

. Does intraoperative navigation improve the accuracy of pedicle screw placement in the apical region of dystrophic scoliosis secondary to neurofibromatosis type I: comparison between O-arm navigation and free-hand technique. Eur Spine J 2016;25(6):1729–1737.

26.

Gertzbein

Robbins

. Accuracy of pedicular screw placement in vivo. Spine (Phila Pa 1976) 1990; 15(1): 11–14.

27.

Tian

. Image-guided pedicle screw insertion accuracy: a meta-analysis. Int Orthop 2009; 33(4): 895–903.

28.

Schouten

Lee

Boyd

. Intra-operative cone-beam CT (O-arm) and stereotactic navigation in acute spinal trauma surgery. J Clin Neurosci 2012; 19(8): 1137–1143.

29.

Roy-Camille

Saillant

Mazel

. Internal fixation of the lumbar spine with pedicle screw plating. Clin Orthop Relat Res 1986;(203):7–17.

30.

Weinstein

Spratt

Spengler

. Spinal pedicle fixation: reliability and validity of roentgenogram-based assessment and surgical factors on successful screw placement. Spine (Phila Pa 1976) 1988; 13(9): 1012–1018.