The use of three-dimensional navigation and advanced intraoperative imaging in minimally invasive pelvic and acetabular fracture fixation: A systematic review

Abstract

Pelvic and acetabular fractures are challenging injuries to treat. This review evaluates three-dimensional intraoperative imaging and navigation-guided percutaneous SI, trans-iliac, trans-sacral, and acetabular screw placement versus conventional methods performed with C-arm imaging. A systematic search of MEDLINE, Embase, and Cochrane was performed. Two reviewers independently extracted data into a collaborative data form designed a priori and piloted prior to its use. Descriptive statistics are presented where applicable. Summary statistics analysis was presented based on the fracture type. Continuous data such as fluoroscopic and operative time were compared with unpaired Student t-test and pooled data of revision rate and complications were compared with chi-square analysis with an alpha set at 0.05. The rate of complications using conventional fluoroscopy was 11.3% (26/230) compared to three-dimensional navigation (6.7% (40/597), X² (DF: 1, N = 827) = 4.79, p = .028.) which translated to a higher rate of revision surgeries in the conventional fluoroscopy group (10.9% vs. 0.8%) X² (DF: 1, N = 827) = 47.8, p ≤.001. Average fluoroscopic time was lower for studies using three-dimensional navigation (28.8 ± 14.3 s, n = 71) compared to conventional fluoroscopy (57.8 ± 4.2 s, n = 38, p ≤.001). Three-dimensional navigation during minimally invasive pelvis and acetabular fracture fixation may have some benefits. Level of evidence: IV.

Keywords

Pelvis acetabulum three-dimensional navigation fracture fixation

Introduction

Pelvic and acetabular fractures are serious injuries with high morbidity and mortality rates (between 9% and 30%).^1,2 Many patients who survive do not return to baseline activity levels due to long-term issues including chronic pain (30%–85%).^3,4

Management of pelvic and acetabular fractures includes temporization using skeletal traction or external fixation; definitive management includes open reduction and internal fixation (ORIF) or image-guided percutaneous fixation.^5,6 Internal fixation with plates and screws requires extensive surgical exposure and risks damage to neurovascular structures.^6,7 Image-guided fixation techniques may minimize the need for soft tissue dissection, decreasing pain and infection risk.^7,8 Due to the complex anatomy of the pelvis and acetabulum, there are risks of screw malposition and neurovascular injuries.^9,10 During percutaneous posterior pelvic ring fixation, screw malposition occurs in 12% of patients with normal sacral anatomy and 28% in those with a dysmorphic sacrum.¹¹

Computer-assisted navigation has been introduced as a potential way to improve accuracy and outcomes following percutaneous pelvic and acetabular fixation. Iso-C three-dimensional (3D) and ARCADIS Orbic 3D are intraoperative image-based navigation devices used in trauma and elective cases. The 190° arc C-arm provides 3D reconstruction based on multiple fluoroscopic images.¹² Several studies found Iso-C 3D exhibits greater sensitivity and specificity for evaluating hardware position and reduction in acetabular fractures.^6,10,13,14 O-arm is a 2D fluoroscopic and 3D multi-planar volumetric imaging system used to provide intraoperative CT imaging.¹⁵ These tools may support optimal hardware position, reducing the risk of neurovascular injury.¹⁶ It has been used in spinal surgeries to guide decompression and in trauma surgery for ORIF.^17–20 Studies show promising results for intraoperative 3D navigation with good accuracy (97%).^9,16 Coste et al.²⁰ evaluated the use of O-arm in osteosynthesis of pelvic fractures, concluding that O-arm offers safe, precise margins during the surgery and is not associated with any malpositioning or secondary screw displacement.

Intraoperative image-based navigation also has drawbacks related to implementation and radiation exposure. Theologis et al.¹⁹ performed a study on 10 cadavers, comparing the use of C- versus O-arm for left-sided S1 screw fixation. The set-up time for O-arm was longer and radiation exposure was higher compared to cases done with standard C-arms alone, carrying health implications for operating room personnel and patients.

To better understand how 3D imaging systems compare to their 2D counterparts during percutaneous pelvic and acetabular SI screw fixation, we performed a systematic review to assess 3D intraoperative imaging and navigation-guided percutaneous SI, trans-iliac, trans-sacral, and acetabular screw placement to conventional methods alongside conventional C-arm imaging. Specifically, we aimed to evaluate the accuracy of placement, quality of reduction, duration of radiation exposure, and operative time. We also analyzed complication types and rates.

Methods

This systematic review was conducted in accordance with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines for conducting and reporting systematic reviews.²¹ The study protocol was registered prospectively on The International Prospective Register of Systematic Reviews (PROSPERO) (ID: CRD42020222285). No data was derived from experimental or clinical observations in human or animal subjects.

Source of funding

No funding was received for this study.

Search strategy

A systematic search of MEDLINE, Embase, and Cochrane was performed on 21 November 2020, including all studies from inception to the search date. Search terms included “pelvi$,” “acetabul,*” “percutaneous screw,” “3D imaging,” and “Navigation” (see the Appendix for a list of MESH terms).

Eligibility criteria

The research question and study eligibility criteria were established a priori. All studies evaluating acetabular, sacral, and pelvic ring fractures treated with S1 or S2 percutaneous screws in skeletally mature patients were included. Exclusion criteria were as follows: surgical procedures such as open reduction or adjunct fixation, sawbones or cadaveric studies, non-human studies, studies lacking adequate clinical information, conference proceedings or abstracts without an available full-text, review papers, and case reports describing five cases or less.

Study screening

Two independent reviewers (LN and KL) assessed the titles and abstracts and applied the eligibility criteria using the online software Rayyan (2010, Qatar Computing Research Institute, Doha, Qatar). Any discrepancies at this stage were automatically included in the full-text review. Full texts were then screened for duplicates; discrepancies were discussed between the reviewers and resolved by consensus. References of included studies were reviewed manually to identify any additional relevant studies for inclusion.

Data extraction

Two reviewers (LN, KL) independently extracted data from the included studies into a collaborative data extraction form (Google Sheets, Google, CA, USA) which was designed a priori and piloted prior to its use. Relevant collected data included study characteristics, injury and patient demographics, operative outcomes, accuracy of screw placement, and complications.

Risk of bias of included studies

The Methodological Index for Non-Randomized Studies (MINORS) was used to assess the quality of included non-randomized studies. Each paper was assigned a score from 0 to 2 against 12 validated criteria for non-randomized studies. The maximum score was 16 for non-comparative studies and 24 for comparative studies.²² We categorized the quality of evidence a priori based on a previous systematic review: < 5 indicated very low-quality evidence, 6–9 low quality, 10–14 fair quality, and > 14 good quality.²³ Randomized controlled studies were evaluated with the revised Cochrane risk-of-bias tool (RoB2).²⁴ RoB2 assesses numerous study domains as either low risk, some concerns, or high risk of bias based on set criteria.

Statistical analysis

The Cohen kappa statistic (κ) was calculated at each screening stage to evaluate the level of inter-reviewer agreement and categorized as follows: substantial agreement: κ > 0.60; moderate agreement: 0.21 < κ < 0.60; and slight agreement: κ < 0.21.²⁵ Inter-class correlation (ICC) was calculated to determine agreement between reviewers’ risk of bias assessments. Descriptive statistics including means, standard deviations, and ranges are presented where applicable. A separate summary statistics analysis was presented based on fracture type (pelvic ring vs. acetabular), as was defined a priori. Due to the literature's heterogeneity and inconsistency in reported outcomes, we were unable to perform a formal meta-analysis. Finally, continuous data such as fluoroscopic and operative time were compared with unpaired Student t-test, and pooled data of revision rate and complications were compared with chi-square analysis with an alpha set at 0.05.

Results

Study characteristics

The initial search identified 195 unique studies; 20 met the inclusion criteria (Figure 1). There were two randomized controlled trials, two prospective comparative studies, two retrospective comparative studies, and 14 case series that met eligibility criteria.^11,26–44 The included studies were published between 1992 and 2017. Sixteen studies investigated the use of 3D imaging on percutaneous pelvic ring screws; four studies focused on acetabular screws. A total of 789 patients were included across the 20 studies and 322 of which were female.

Figure 1.

PRISMA flowchart for the systematic search strategy utilized.

Study quality

The included studies were mostly of level IV evidence. There were two studies of each level I, II, and III evidence. There was substantial agreement between reviewers at the title and abstract screening (κ = 0.72) and excellent agreement at full-text screening (κ =0.82). Similarly, there was substantial agreement between assessors for quality assessment scores using the MINORS criteria (ICC = 0.643). Included non-randomized studies had an average MINORS score of 13 ± 4 which indicates an overall fair quality of evidence. The randomized study was assessed using the RoB tool. Matityahu et al.¹¹ generally had a low risk of bias as the process of the randomization and selection of reported results were well done with minimal loss of data. It is important to mention that this study was multi-centered with a good sample size. Some concerns regarding the measurement of the outcome were noted as a pre-determined and strict definition for screw misplacement was used and they lacked information regarding blinding.

Results for pelvic ring fractures

Of the 16 studies which included pelvic ring fractures, there were 713 included patients.^{11,27–36,40–44} The mean age of 44.6 ± 15.6 years (range: 13–94). A total of 38% (n = 271) were female. The mean length of time between injury and surgery was 9 days (range: 1–217). Table 1 summarizes these studies.

Table 1.

Demographic data of included studies.

Author (year)	Country	Study Design	Level of Evidence	Fracture type	Sample Size	Female patients	Mean Age ± SD (years)	Age Range (years)	Time from injury to surgery (days)	Mean length of follow up (months)	Follow up range (months)	Average MINORS score
He (2016)	China	Prospective comparative study	II	Acetabulum	22	7	35.6	NR	4.4 ± 1.9	26.2	12-42	19.5
Schwabe (2014)	Germany	Case series	IV	Acetabulum	12	9	65	16-80	NR	30	16-72	10
Ruan (2012)	China	Case series	IV	Acetabulum	5	3	NR	NR	2-5	12	NR	11.5
Swartman (2020)	Germany	Retrospective comparative study	III	Acetabulum	37	32	53	18-80	NR	NR	NR	19
Amoretti (2009)	France	Case series	IV	Pelvic ring	20	8	38.6	18-54	2-4	45	36-48	11.5
Fischer (2015)	Germany	Case series	IV	Pelvic ring	38	18	49.6 ± 19.5	NR	9.7 ± 5.0 (range: 3-22)	6	NR	11
Iguchi (2010)	Japan	Case series	IV	Pelvic ring	6	4	55.8	35-77	NR	14	6-21	12
Matityahu (2014)	USA	Retrospective comparative study	III	Pelvic ring	130	NR	NR	NR	NR	NR	NR	20.5
Mu (2009)	China	Case series	IV	Pelvic ring	8	2	34	19-76	1-7	16.1	12-20	11.5
Dewelius (1992)	USA	Case series	IV	Pelvic ring	13	8	25.5	16-39	6.1 (range: 1-14)	12.6	7-24	11.5
Ebraheim (1994)	USA	Case series	IV	Pelvic ring	19	7	34.6	15-75	NR	24	6-60	10.5
Jacob (1997)	Switzerland	Case series	IV	Pelvic ring	13	4	41	18-66	9 (range: 2-213)	13	NR	12
Li (2015)	China	Retrospective comparative study	III	Pelvic ring	81	45	35.8 ± 11.9	NR	8.4 ± 4.0	12.7	6-24	11
Peng (2013)	Taiwan	Case series	IV	Pelvic ring	13	3	41.2 ± 15.9	21-75	NR	None	None	8.5
Pieske (2015)	Germany	Case series	IV	Pelvic ring	71	29	51.4 ± 22.3	16-91	9.5 ± 7.8	29.1 ± 19.1	6-65	11.5
Sciulli (2007)	USA	Case series	IV	Pelvic ring	23	13	NR	13-69	NR	NR	NR	11
Zwingmann (2010)	Germany	Retrospective comparative study	II	Pelvic ring	141	93	44 ± 21	NR	NR	NR	NR	19
Reuther (2014)	Germany	Case series	IV	Pelvic ring	135	15	77	20-94	NR	NR	NR	9.5
Wang (2017)	China	RCT	I	Pelvic ring	30	18	39.4	NR	NR	NR	NR	-
Takeba (2017)	Japan	Case series	IV	Pelvic ring	10	4	56	35-80	NR	NR	NR	9.5

Of the four comparative studies which investigated the use of 3D navigation on percutaneous pelvic fixation, three compared the Iso-C 3D C-arm to conventional 2D fluoroscopy.^11,29,33 The additional comparative study compared a robotic system (TiRobot) performed autonomously to robotic-assisted screw fixation and conventional 2D fluoroscopy.³⁵ Most of the case series described CT scan-guided percutaneous pelvic fixation.^{27,28,30–32,34,40–42,44} Mu et al.⁴³ were the only ones to describe the results of Iso-C 3D C-arm guided fixation.

Screw misplacement

Complications rates were higher in procedures that used conventional fluoroscopy (11.3% (26/230)) compared to 3D navigation (6.7% (40/597), X² (DF: 1, N = 827) = 4.79, p = .028) (Table 2). Higher complication rate also translated into a higher revision rate in the conventional group (10.9% vs. 0.8%) X² (DF: 1, N = 827) = 47.8, p ≤ .001).

Table 2.

Comparison of complication and revision rates between conventional fluoroscopy and 3D navigation.

Study, year publication	Sample sizeNumber digits	Flexor Zones	Method of Rehabilitation	Splint TypeTotal Follow Up	Suture Repair Method	Functional Outcome (P Value)	Complication
May, 1991 (9)	140 patients 159 digits	Zone II	Gr 1: Modified kleinert: 54 digits Gr 2: Modified kleinert + passive control: 51 digits Gr 3: Dynamic flexion traction + nighttime extension: 54 digits	Gr 1: dorsal plaster splint w/ wrist in 30° - 45° flexion, MCP in 50° - 70° flexion and IP extension. Gr 2: dorsal plaster splint w/ wrist in 30° - 45° flexion, MCP in 50° - 70° flexion, and IP full extension. Group 3: dorsal plaster splint extend to PIP, w/ wrist 30° - 45° flexion, MCP 50° - 70° Flexion, IP extension. Foll-up at 6 mo and 1 y	FDP: modified Kessler 4-0 w/ braided polyester 6-0 epitendinous stitch FDS: mattress sutures 4-0 braided polyester 2-strand repair	Strickland: Gr 1: 72% good-excellent Gr 2: 62% good-excellent Gr 3: 83% good-excellent	Gr1: 2 ruptures 15 extension deficit Gr 2: 1 rupture 14 extension deficit Gr 3: 2 ruptures 3 extension deficit
Bainbridge, 1994 (10)	Group 1: 52 patients, 68 digits Group 2: 56 patients, 67 digits	Zone I,II	Group 1: passive flexion – active extension Group 2: controlled active motion	Group 1: dorsal splint w/ wrist 30° flexion, MCP 90° flexion, and IP neutral Gr 2: dorsal splint W/ wrist 30° flexion, MCP 90° flexion, and IP full extension Foll-up: 4 mo	FDP: modified Kessler of 3-0 / 4-0 Prolene w/ epitendinous 6-0 nylon / Prolene FDS: 4-0 or 5-0 horizontal mattress sutures 2-strand repair	Buck-Gramcko: Gr 1: zone I: 90% good-excellent zone II: 50% good–excellent Gr 2: zone I: 89% good–excellent zone II: 90% good–excellent Strickland: Gr 1: zone I: 10(90%) good –excellent zone II: 31(53%) good-excellent Gr 2: zone I: 16(89%) good–excellent zone II: 46(94%) good–excellent ASSH: Gr 1: zone I: 7(70%) good –excellent zone II: 27(46%) good-excellent Gr 2: zone I: 16(89%) good–excellent zone II: 47(92%) good–excellent Zone 1: P>0.05 Zone 2: P<0.001	Gr 1: 2 ruptures 27 extensor deficits Gr 2: 5 ruptures 7 extensor
Elliot, 1994 (11)	233 patients 137 tendons	Zone I,II	Controlled active motion	Thermoplastic dorsal splint w/ 30° wrist flexion, 30° MCP flexion, and IP neutral Foll-up: 3 mo	FDP: 3-0 / 4-0 Prolene and 6-0 Prolene / nylon epitendinous FDS: horizontal mattress of 4-0 / 5-0 nylon 2-strand repair	Strickland: 77% good–excellent	18 ruptures
Silverskiold, 1994 (12)	46 patients 55 digits	Zone II	Active extension and passive/ active flexion	Dorsal splint PIP, w/ wrist at 0°, MCP in 50° - 70° flexion, and IP fully extend foll-up: 6 wks and 6 mo	FDP: modified Kessler—4-0 polyester and 6-0 polypropylene epitendinous 2-strand repair	DIP and PIP had 82% and 88% ROM Strickland: Excellent-good: 53%	2 ruptures
Baktir, 1996 (13)	77 patients 88 digits	Zone II	Gr 1: 33 digits: Kleinert-rubber band passive flexion/active extension method Gr 2: 38 digits: early active mobilization	Gr 1: dorsal splint W/ wrist 30° - 40° flexion, MCP in 70°- 90° flexion, and IP full extension Gr 2: dorsal splint w/ wrist 0° flexion, MCP 70°- 90° flexion, and IP full extension Foll-up: 1 yr	FDP: modified Kessler—4-0 braided polyester and epitendinous 6-0 Prolene FDS: horizontal mattress sutures— 4-0 braided polyester 2-strand repair	Strickland: Gr 1: 78% good–excellent and 84% mean grip strength Gr 2: 85% good–excellent 90% mean grip strength	Gr 1: 2 re-ruptures Gr 2: 2 re-ruptures Flexion contracture: 10 in Gr 1 5 in Gr 2
Schenk, 1996 (14)	33 patients 44 digits	Zone II	The Washington regimen: passive flexion& extension, followed active extension& passive flexion, active extension & active flexion	Dorsal splint thermoplastic Foll-up: 3 and 6 mo	Gr 1: kleinert et al modification of bunnel suture tech, 4.0 nylon- 31 finger Gr 2: modified Kessler grasping suture, 6.0 nylon – 13 finger	Strickland and glogovac: Excellent After 3 mo – 3 patients After 6 mo – 4 patients 36% total excellent	No complication
Peck, 1998 (15)	52 patients 52 digits 92 tendons	Zone II	Gr 1: controlled active motion - 26 digits Gr 2: modified Kleinert: 26 digits	Gr 1: thermoplastic dorsal splint wrist 40° flexion MCP 60° flexion, IP neutral Gr 2: thermoplastic dorsal splint wrist 40°flexion, MCP 60° flexion, IP neutral Foll-up: 12 wks	FDP: modified Kessler 3-0 Prolene,6-0 nylon / Prolene epitendinous FDS: horizontal mattress 6-0 prolene. 2-strand repair	Strickland: Gr 1: 85% good - excellent Gr 2: 69% good- excellent P >0.05	Gr 1: 12 re-ruptures Gr 2: 2 re-ruptures
Cetin, 2001 (16)	37 patients 74 digits	Zone I-V	Controlled mobilization foll. w/ combined (modified Kleinert and modified Duran) Kleinert splint with a palmar pulley	Dorsal splint wrist 30° - 35° flexion and MCP flexion 50° - 60° Foll-up: 12.9 wks	Modified Kessler 4-0 Prolene with 6-0 nylon epitendinous 2-strand repair	Buck-Gramcko: 97% good–excellent	1 rupture 13 extensor deficits
Galanakis, 2003 (17)	19 patients 23 digits	Zone II	Passive flexion-active extension	Dorsal splint, full extension PIP, DIP, 30⁰ palmar flexion, 70-90⁰ MCP flexion Foll-up: 20 wks	Modified Kessler 4.0 nylon repair core suture 2 strand repair	Strickland: 15 sample: excellent 5 sample: good 2 sample: fair	2 ruptures
Silva, 2005 (18)	82 patients 136 digits	Zone II	Early active mobilization	N/A Foll-up: 12-36 wks	Modified Kessler 3-0 nylon with 5-0 epitendinous 2-strand repair	IFSSH and Strickland criteria: Long fingers: 98% good – excellent (Strickland); 82% good (IFSSH) Thumb: 96% good–excellent (Strickland); 96% good– excellent (IFSSH)	5 ruptures
Osada, 2006 (19)	21 patients 27 digits	Zone II	Controlled active mobilization (active extension and passive/active flexion)	Dorsal splint, wrist neutral, MCP joint 30-40⁰ flexion, IP joint full extended Foll-up range 13 mo	Gr 1: Y1 tech.-6 strands repair- 11 digits Gr 2: Tsuge suture + looped thread with modified Kessler 2-strand repair-8 digits	Strickland criteria: Gr 1: Excellent – 11 digits Good –7 digits Fair – 1 digits Gr 2: Excellent – 6 digits Good –2 digits P > 0.05 ROM: avg flex 62⁰ DIP, 91⁰ PIP	0 ruptures
Yen, 2008 (20)	20 patients	Zone II	Active extension, active place an hold- 10 digits Kleinert method- 10 digits	Dorsal splint with wrist in 30° flexion, MCPs in 70° flexion, and IPs in full extension Foll-up: 4 mo	4-0 Prolene core sutures plus 6-0 Prolene circumferential sutures 4-strand repair	Mayo Wrist Score: Active motion: 70% good–excellent Kleinert splint: 0% good–excellent	Active place and hold: 0 complications Kleinert splint: 1 rupture
Kitis, 2009 (21)	192 patients 263 digits	Zone II	Gr 1: modified Kleinert (Washington regimen)— 137 digits) Gr 2: controlled passive movement—126 digits	Group 1: 6 wks in Kleinert splint with wrist minus 20° full flexion, MCPs 10° to 20°flexion, IP neutral Group 2: 5 wk in dorsal splint wrist 20° flexion, MCP 50° flexion, IP full extension Foll-up range, 6–20 mo	Modified Kessler 4-0 nylon w/ 6-0 nylon epitendinous 2-strand repair	Buck-Gramcko excellent: Gr 1: 87% Gr 2: 75% P 0.01 Grip strength Gr 1: 89% Gr 2: 81% P 0.08 DASH Gr 1: mean 29.6 Gr 2: mean 41.7 P 0.01	Gr 1: 16 extension deficits 0 ruptures Gr 2: 26 extension deficits 1 rupture
Trumble, 2010 (22)	103 patients 119 digits	Zone II	Gr 1: Passive motion—51 patients with 58 digits Gr 2: Active motion with place and- hold—52 patients with 61 digits	N/A Foll-up: 6, 12, 36, 52 wks	FDP: Strickland Method, 2 core sutures 3-0 polyester and 6-0 Prolene epitendinous FDS: simple Kessler with 3-0 polyester 4-strand repair	ROM: Gr 1-52 wks: 128⁰± 22⁰ Gr 2- 52 wks: 156⁰± 25⁰ P<0.05 DASH: Gr 1: 3.1 ± 4.3 Gr 2: 2.0 ± 3.7 P<0.05 Satisfaction score: Gr 1: 8.2 ± 3.9 Gr 2: 9.4 ± 4.3	Gr 1: Passive motion: 2 ruptures Gr 2: Active motion: 2 ruptures Flexion contracture: Gr 1: 28⁰±13⁰ Gr 2: 15⁰± 8⁰ P < 0.05
Frueh, 2014 (23)	132 patients 159 digits	Zone I, II	Gr 1: Early passive mobilization (EPM)- 138 Gr 2: Controlled active motion (CAM) – 21 Other combination protocol - 72	Gr 1: EPM with Dorsal 4 finger orthosis (wrist 30°flexion, intrinsic plus position of finger) foll. Rubber band for passive finger flexion Gr 2: CAMdorsal forearm orthosis Foll-up: 4 and 12 wks	FDP & FDS 4 strands core suture with locking loops	Total active motion Wk 4-CAM : EPM = P<0.001 Wk 12-CAM : EPM P = 0.75 Functional grades (AASH) in 12 wks: EPM: excellent 8 pt (53%) CAM: good 65 pt (63%) P > 0.05	Rupture rate: EPM: 10 rupture (7%) CAM: 1 rupture (5%) P > 0.05 Other protocol: 8% rupture
Sabour, 2018 (24)	26 patients 36 tendons	Zone I,II,III	Gr 1: early passive rehabilitation (modified kleinert – passive flexion and active extension)- 20 pt Gr 2: early active rehabilitation (place and hold)- 16	Dorsal blocking splint wrist 20⁰ flexion, MCP 70⁰ flexion IP joint full extension Foll-up: 12 wks	Modified Kessler technique with 3.0	Patient satisfaction (analog scale): Gr 1: median: 4 (range 2-7) Gr 2: median: 6 (range 4-8.5) DASH score Gr 1: median: 15 (range 10-15) Gr 2: median: 23 (range 2-26) P=0.62	Flexion deformity DIP Gr 1: 10 tendon (50%) Gr 2: 1 tendon (6.2%) P=0.004 Flexion deformity PIP Gr 1: 8 tendon (40%) Gr 2: 3 tendon (18.7%) P=0.1 Flexion deformity DIP-PIP P=0.02 Rupture rate Gr 1: 2 (10%) Gr 2: 1 (6.25%) P>0.05

Matityahu et al.¹¹ performed a subgroup analysis of previously randomized patients. They included 130 patients that were either treated with conventional fluoroscopy (n = 58), 2D navigation-based imaging (n = 18), or 3D navigation (n = 54). After a blind review of the 109 patients with the non-dysmorphic sacrum, they found nine misplaced screws in the conventional group (total of 73 screws, 12%), six in the 2D navigation (total of 22 screws, 27%), and 0/66 (0%) in the 3D navigation group. The observed difference was even higher among dysmorphic sacrums with 7/22 (32%) misplaced screws for the conventional group and 0/12 (0%) in the 3D group.

In a comparative cohort study, Zwingman et al.³³ reported the results of consecutive patients treated before (n = 87, 131 screws) and after (n = 54, 63 screws) the implementation of 3D navigation. As mentioned by the authors, there was high variability among surgeons which makes direct comparison between the conventional and navigation groups difficult. Fourteen different surgeons had individual revision rates varying from 0% (0/40) to 29% (7/24); most were low-volume surgeons (< 15 SI screws over an 8-year period). 3D navigation technology lowered the incidence of screw malpositioning (19% with navigation vs. 58% conventional). They also saw a reduction in revision for screw perforation (1/63, 1.6% vs. 13/131, 9.9%). Screw perforations were seen on post-operative CT scans. The authors did not report cases of neurologic deficit/symptoms (Table 3).

Table 3.

Screw perforation and complications for the percutaneous placement of pelvic screws.

Author (year)	Comparison with a conventional method	Technology used	No of patients	No of screws	Average operative time (min)	No (%) of screw perforation or misplacement	Complications	Neurologic complications
Amoretti (2009)	No	CT-guided	20	40	65	NR	0	0
Fischer (2015)	No	CT-guided	38	100	84	2 (2%)	0	0
He (2016)	Yes	Iso-C 3D C-arm	10	10	34	0	0	0
		Conventional 2D C-arm	12	12	35.7	2 (17%)	0	0
Iguchi (2010)	No	CT-guided	6	11	92.5	0	0	0
Matityahu (2014)	Yes	Conventional fluoroscopy	58	95	NR	16 (17%)	NR	NR
		Fluoroscopy + 2D navigation (brainLab software)	18	25	NR	8 (32%)	NR	NR
		CT or Iso-C 3D C-arm	54	78	NR	0 (0%)	NR	NR
Mu (2009)	No	Iso-C 3D C-arm + Stealthstation	8	15	57	0 (0%)	1 LLD	0
Dewelius (1992)	No	CT-guided	13	13	144	1 (7.7%)	2 LLD	0
Ebraheim (1994)	No	CT-guided	19	19	80	0 (0%)	1 superficial infection, 2 large rotational deformity (20°)	No
Jacob (1997)	No	CT-guided	13	27	86	2 (7.4%)	8: 1 screw loosing, 1 broken screw, 1 superficial wound infection, 1 pneumonia, 2 necrotizing cholecystitis, 1 PE, 1 non-union with hypotrophic pseudo-arthrosis	1 mild power deficit
Li (2015)	Yes	Conventional fluoroscopy	38	61	30.9	9 (14.8%)	0	1 mild numbness
		Iso-C 3D	43	69	19.5	2 (2.9%)	0	0
Peng (2013)	No	CT-guided	13	13	NR	NR	0	0
Pieske (2015)	No	CT-guided	71	136	88	4/136, (3%)	0	0
Sciulli (2007)	No	CT-guided	23	29	80	NR	2 loss of reduction	0
Zwingmann (2010)	Yes	Conventional fluoroscopy	87	131	NR	76 (58%)	23 reoperations: 14 perforations, 8 loosening, 2 pseudarthrosis	0
		Vectorvision + Iso-C 3D	54	63	NR	12 (19%)	1 reoperation for perforation	0
Reuther (2014)	No	CT-guided	135	182	Unilateral: 23, Bilateral: 35	NR	12/85 loss of reduction; 1 gluteal bleed	NR
Schwabe (2014)	No	Arcadis Orbic3D and Vectorvision	12	22	NR	0	0	0
Ruan (2012)	No	Arcadis 3D	5	7	20	0	0	0
Wang (2017)	Yes	Conventional fluoroscopy	15	22	104	NR	0	0
		Robotic (TiRobot)	15	23	150	NR	0	0
Swartman (2020)	Yes	Conventional fluoroscopy	13	13	NR	NR	NR	NR
		CT-guided	24	24	NR	NR	NR	NR
Takeba (2017)	No	O-arm Stealthstation	10	17	71	0	1 wound infection	0

Operative time and radiation exposure

In 122 cases, the average operative time for percutaneous pelvic fixation using 3D navigation was 134.6 ± 31.3 min. Average fluoroscopic time was lower for studies using 3D navigation (28.8 ± 14.3 s, n = 71) compared to conventional fluoroscopy (57.8 ± 4.2 s, n = 38, p ≤.001).

Wang et al.³⁵ compared 22 cases of SI screws placed freehand using fluoroscopy and 23 robot-assisted cases. While they showed no difference in complications or revisions, they reported longer mean operative time with the robot (150.0 min (range: 75–230 min) vs. 104.0 min (60–154 min), p = .158). However, both the number of guide-wire insertion attempts (average 1.0 vs. 7.0, p < .001) and the radiation exposure were significantly lower with the use of the robot (mean = 6 s vs. 36 s, p < .001).

Using Iso-C 3D, Li et al.²⁹ also reported lower fluoroscopy time in their navigation group (46.4 s ± 2.5 vs. 75.4 s ± 6.9, p < 0.001) and SI screw placement time (19.5 min ± 2.2 vs. 30.9 min ± 5.7, p < .001) compared to conventional fluoroscopy.

Acetabular percutaneous screws

Although less common, advanced imaging is also used for acetabular percutaneous fixation. This includes a combination of anterior column, posterior column, anterograde, and retrograde screws. Swartman et al.³⁹ performed a retrospective study of 24 cases of CT navigated acetabular screw placement and 13 conventional cases using 7.3 mm cannulated screws. There was no significant difference between the two methods in terms of the quality of the fracture reduction (pre- and postoperatively) and the position of the screws relative to the hip joint and in the bone corridor. He et al.²⁶ compared a small cohort of patients receiving percutaneous fixation with 6.5 mm cannulated screws through Iso-C 3D (n = 10) and C-arm fluoroscopy (n = 12). They reported similar operative time and post-operative outcomes; radiation exposure was significantly less in the navigation group (26.9 s ± 2.1 vs. 80.7 s ± 3.0, p < .001). Schwabe et al.³⁷ described a case series of 22 cases using the ARCADIS Orbic (3D C-arm). They reported good clinical and radiologic results but lacked comparison to a conventional method.

Bleeding and infection rates

Overall, infection rates appeared to be higher with 3D versus conventional navigation (6/597 patients and 0/230 patients, respectively). Likewise, the incidence of bleeding was also higher with 3D navigation (1/597 vs. 0/230).

Timing of post-operative CT

The timing of post-operative CT was variable and not consistently reported.

Discussion

This study identified a trend towards improved precision and lower rates of perforation with the use of 3D navigation. This is based on the limited evidence from six higher-quality studies that compared outcomes from Iso-C 3D C-arm or robotics with conventional fluoroscopy. Included navigation-assisted case series showed rates of screw misplacement that ranged from 0% to 8%, which are lower than reported misplacement rates ranging from 12% to 23% using conventional techniques.^45–47 Although these ranges cannot be directly compared due to variations in the methodology in referenced studies, they suggest that 3D navigation may reduce misplacement rates. Differences suggested in comparative studies were greater in cases that presented a greater technical challenge for the surgeons, including patients with dysmorphic sacrum. This suggests that there may be a benefit for those surgeons who are less comfortable with the procedure.

Ionizing radiation is a classified carcinogen by the WHO, with a likelihood for long-term sequelae.⁴⁸ One potential benefit of 3D navigation is decreased fluoroscopic time and radiation exposure for patients. The trend towards lower fluoroscopic time using 3D navigation compared to 2D fluoroscopy in the pooled analysis is seen in the comparative trials, suggesting a potential reduction in radiation exposure for patients and operating room staff using 3D navigation. Operative time stands as an indication of resource utilization and surgeon proficiency and is useful in determining the value of new techniques. The average operative time for pooled cases using 3D navigation (134 min) was slightly higher than the average operative time seen in cases using conventional fluoroscopy in the randomized control trial (RCT) performed by Wang et al. (104 min, p ≤ .001).³⁵ While it is difficult to directly compare these figures due to the small sample size of conventional cases, variability in anatomy, number of screws implanted, and surgeon experience, these findings suggest that operative time is relatively similar between the two methodologies. Maintenance of operative time in addition to the increased technological costs associated with using navigation suggests a significant financial burden associated with the implementation of navigation in minimally invasive pelvic fixation.

Of additional importance is a complication. No included studies demonstrated fewer complications or reoperations with navigation. Only Wang et al.³⁵ directly compared rates of complications and revisions between fluoroscopy and navigation, showing no difference. However, pooled analysis of included studies showed the rate of revision surgeries in the 3D navigation group was 0.8% (5/597) compared to 11% (25/230) in the fluoroscopy group. All conventional fluoroscopy cases in which a revision was required were from a single study.³³ While the trends towards lower cut-out rates and radiation exposure are promising, there lacks evidence that 3D navigation reduces rates of reoperation and additional high-powered comparative studies are needed to better elucidate this relationship.

As only four studies investigated the use of 3D navigation in acetabular fracture fixation, two of which were comparative studies, it is difficult to draw any meaningful conclusions. The only significant difference seen in these studies was a trend toward a lower dose of radiation exposure in the navigation group. Included case studies reported good clinical and radiologic outcomes. More research is needed to identify the meaningful benefits of using 3D navigation in percutaneous fixation of acetabular fractures.

There is often a learning curve associated with the implementation of new technology, involving a period of skill acquisition associated with increased operative time and complication rates.⁴⁹ In Zwingman et al.'s³³ retrospective comparative study, all fluoroscopy cases were performed prior to incorporating navigation technology. This introduces a potential for bias within the results, as the surgeons may have been in the skill-acquisition phase during early fluoroscopic cases. When 3D navigation was implemented, the surgeons may already have developed proficiency, inflating the results of the navigation-assisted procedures. This bias has the potential to influence additional included studies, depending on the surgeon's comfort and experience using 3D navigation. Future studies may account for this learning curve by excluding the first set of navigation-assisted cases.

While the current literature on the use of O-arm in orthopedics focuses on spine surgery, the clinically significant outcomes are promising. Wang and colleagues demonstrated operative time for posterior lumbar interbody fusion in elderly patients was significantly shorter when O-arm was used compared to fluoroscopy.⁵⁰ It has been well established that prolonged operative time is associated with higher rates of infection in orthopedic surgeries, so this significant decrease in operative time may also represent decreased rates of infection.⁵¹ Furthermore, a recent retrospective review showed that patients who underwent a transforaminal lumbar interbody fusion with O-arm navigation had significantly lower visual analogue scale leg and Oswestry disability index scores at 6 months post-operatively compared to patients who underwent the procedure using fluoroscopy.⁵² It is suggested that these improved clinical outcomes are due to improved surgical accuracy, so would likely translate to pelvic and acetabular trauma in which outcomes are similarly reliant on accurate screw placement.

This systematic review was strengthened by its rigorous methodology, involving a comprehensive search of major databases. Reviewer bias was minimized by blindly completing the screening; conflicts were resolved via consensus with assistance from senior authors. This resulted in the inclusion of 713 patients. The range of demographic factors makes the results broadly applicable to the general population. To the authors’ knowledge, this study is the first of its kind to compare 3D navigation and conventional fluoroscopy in percutaneous fixation of pelvic and acetabular fractures.

The findings are limited by the low-quality evidence of included studies. As most of the included studies were level 4 evidence, with only one RCT, there is a risk of bias within the results. Since only four studies directly compare 3D navigation with conventional fluoroscopy, it is difficult to determine the true differences that exist between the two. There was significant heterogeneity in the methodology of included studies, involving varying fracture patterns and screw placements. This prevented the authors from performing a formal meta-analysis.

The paucity of high-quality evidence emphasizes the need for rigorous RCTs to elucidate the benefits of 3D navigated minimally invasive pelvic and acetabular fixation compared to conventional fluoroscopy. While 3D navigation has the potential to improve screw accuracy and reduce radiation exposure, large studies that are sufficiently powered to detect differences in rates of complications and revision surgeries are imperative to guide practice. 3D navigation is costly, with the O-arm capital costs estimated to be $700,000 CAD (∼ $575,000 USD) plus annual operating and maintenance costs.⁵³ Based on this study's equivocal results, it seems unlikely that O-arm's high up-front costs are worthwhile for pelvis and acetabular surgery alone but may be more acceptable if used for additional patient populations and injury patterns. Additionally, once the benefits of navigation on screw cut-out and revision rates are better understood, a cost-effectiveness analysis should be performed to evaluate the financial feasibility of implementing this technology.

Conclusion

This review demonstrates there may be limited evidence to support using 3D navigation during minimally invasive pelvic fracture fixation. Based on current evidence, there is no clinical impact of navigation on cortical perforations or neurovascular complications. The paucity of high-quality trials limits the applicability of these findings clinically. Research is needed to characterize the implications of navigation on complication and revision rates, radiation exposure, and cost-effectiveness before implementing this technology into routine practice.

Footnotes

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.

ORCID iDs

Kathy Liu

Julien Chapleau

Appendix. MESH Search Terms

	MEDLINE	Embase	Cochrane
Search strategy	Sacral Sacrum Sacroiliac Posterior pelvic ring Acetabul* Pelvi$ Screw fixation Percutaneous screw Percutaneous fixation 3D imaging Navigation O-arm STEALTHSTATION CT TiRobot Placement Radiation Complication Reduction Revision Operative time Fluoroscopic time Infection 1 OR 2 OR 3 OR 4 OR 5 OR 6 7 OR 8 OR 9 10 OR 11 OR 12 OR 13 OR 14 OR 15 16 OR 17 OR 18 OR 19 OR 20 OR 21 OR 22 OR 23 24 AND 25 AND 26 AND 27	Sacral Sacrum Sacroiliac Posterior pelvic ring Acetabul* Pelvi$ Screw fixation Percutaneous screw Percutaneous fixation 3D imaging Navigation O-arm STEALTHSTATION CT TiRobot Placement Radiation Complication Reduction Revision Operative time Fluoroscopic time Infection 1 OR 2 OR 3 OR 4 OR 5 OR 6 7 OR 8 OR 9 10 OR 11 OR 12 OR 13 OR 14 OR 15 16 OR 17 OR 18 OR 19 OR 20 OR 21 OR 22 OR 23 24 AND 25 AND 26 AND 27	((Sacral) OR (Sacrum) OR (Sacroiliac) OR (Posterior pelvic ring) OR (Acetabul*)) OR (Pelvi$) ((Screw fixation) OR (Percutaneous screw) OR (Percutaneous fixation)) ((3D imaging) OR (Navigation) OR (O-arm) OR (STEALTHSTATION) OR (CT)) OR (TiRobot) ((Placement) OR (Radiation) OR (Complication) OR (Reduction) OR (Revision)) ((Operative time) OR (Fluoroscopic time) OR (Infection)) (#1 OR #2) AND (#3) AND (#4 OR #5) AND (#6 OR #7)
Number of articles retrieved	164	263	10

References

Hauschild

Strohm

Culemann

, et al. Mortality in patients with pelvic fractures: results from the German pelvic injury register. J Trauma 2008; 64: 449–455.

Ooi

Goh

Tay

, et al.

Patients with pelvic fracture: what factors are associated with mortality?

Int J Emerg Med 2010; 3: 299–304.

Brennemann

Katyal

Boulanger

, et al. Long-term outcomes in open pelvic fractures. J Trauma 1997; 42: 773–777.

Papasotiriou

Prevezas

Krikonis

, et al. Recovery and return to work after a pelvic fracture. Saf Health Work 2017; 8: 162–168.

Dunbar

. Diseases & conditions: pelvic fractures: OrthoInfo from Am. Acad. Orthop. Surg, https://orthoinfo.aaos.org/en/diseases–conditions/pelvic-fractures/ (2016).

Keil

Beisemann

Schnetzke

, et al. Intraoperative assessment of reduction and implant placement in acetabular fractures-limitations of 3D-imaging compared to computed tomography. J Orthop Surg Res 2018; 13: 78.

Hung

, et al. Minimally invasive treatment for anterior pelvic ring injuries with modified pedicle screw-rod fixation: a retrospective study. J Orthop Surg Res 2018; 13: 238.

Lin

Guo

Dong

, et al. Acetabular lateral view: effective fluoroscopic imaging to evaluate screw penetration intraoperatively. Med Sci Monit 2019; 25: 5953–5960.

Ghisla

Napoli

Lehoczky

, et al. Posterior pelvic ring fractures: intraoperative 3D-CT guided navigation for accurate positioning of sacro-iliac screws. Orthop Traumatol Surg Res 2018; 104: 1063–1067.

10.

Kendoff

Garaner

Citak

, et al. Value of 3D fluoroscopic imaging of acetabular fractures comparison to 2D fluoroscopy and CT imaging. Arch Orthop Trauma Surg 2008; 128: 599–605.

11.

Matityahu

Kahler

Krettek

, et al. Three-dimensional navigation is more accurate than two-dimensional navigation or conventional fluoroscopy for percutaneous sacroiliac screw fixation in the dysmorphic sacrum: a randomized multicenter study. J Orthop Trauma 2014; 28: 707–710.

12.

Atesok

Finkelstein

Khoury

, et al. CT (ISO-C-3D) image based computer assisted navigation in trauma surgery: a preliminary report. Injury Extra 2008; 39: 39–43.

13.

Kendoff

Citak

Gardner

, et al. Intraoperative 3D imaging: value and consequences in 248 cases. J Trauma 2009; 66: 232–238.

14.

Eckardt

Lind

Toendevold

. Open reduction and internal fixation aided by intraoperative 3-dimensional imaging improved the articular reduction in 72 displaced acetabular fractures. Acta Orthop 2015; 86: 684–689.

15.

Prod'homme

Sans-Merce

Pitteloud

, et al. Intraoperative 2D C-arm and 3D O-arm in children: a comparative phantom study. J Child Orthop 2018; 12: 550–557.

16.

Chen

Liu

, et al.

Does pedicle screw fixation assisted by O-arm navigation perform better than fluoroscopy-guided technique in thoracolumbar fractures in percutaneous surgery?

Clin Spine Surg 2020; 33: 247–253.

17.

Tjardes

Shafizadeh

Rixen

, et al. Image-guided spine surgery: state of the art and future directions. Eur Spine J 2010; 19: 25–45.

18.

Sembrano

Yson

Theismann

. Computer navigation in minimally invasive spine surgery. Curr Rev Musculoskelet Med 2019; 12: 415–424.

19.

Theologis

Burch

Pekmezci

. Placement of iliosacral screws using 3D image-guided (O-arm) technology and stealth navigation: comparison with traditional fluoroscopy. Bone Joint J 2016; 98-B: 696–702.

20.

Coste

Asloum

Marcheix

, et al. Percutaneous iliosacral screw fixation in unstable pelvic ring lesions: the interest of O-ARM CT-guided navigation. Orthop Traumatol Surg Res 2013; 99: S273–S278.

21.

Moher

Shamseer

Clarke

, et al. Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement. Syst Rev 2015; 4.

22.

Slim

Nini

Forestier

, et al. Methodological index for non-randomized studies (MINORS): development and validation of a new instrument. ANZ J Surg 2003; 73: 712–716.

23.

Ekhtiari

Horner

Bedi

, et al. The learning curve for the Latarjet procedure: a systematic review. Orthop J Sports Med 2018; 6: 2325967118786930.

24.

Sterne

JAC

Savovic

Page

, et al. Rob 2: a revised tool for assessing risk of bias in randomised trials. Br Med J 2019; 366: l4898.

25.

McGinn

Wyer

Newman

, et al. Tips for learners of evidence-based medicine: 3. Measures of observer variability (kappa statistic). CMAJ 2004; 171: 1369–1373.

26.

Tan

Zhou

, et al. Comparison of isocentric C-arm 3-dimensional navigation and conventional fluoroscopy for percutaneous retrograde screwing for anterior column fracture of acetabulum: an observational study. Medicine (Baltimore) 2016; 95: e2470.

27.

Ebrahim

Coombs

Jackson

, et al. Percutaneous computed tomography-guided stabilization of posterior pelvic fractures. Clin Orthop Relat Res 1994; 307: 222–228.

28.

Jacob

Messmer

Stock

, et al. Posterior pelvic ring fractures: closed reduction and percutaneous CT-guided sacroiliac screw fixation. Cardiovasc Intervent Radiol 1997; 20: 285–294.

29.

Zhu

, et al. Comparison of 3D C-arm fluoroscopy and 3D image-guided navigation for minimally invasive pelvic surgery. Int J Comput Assist Radiol Surg 2015; 10: 1527–1534.

30.

Peng

Hsu

, et al. Intraoperative computed tomography with integrated navigation in percutaneous iliosacral screwing. Injury 2013; 44: 203–208.

31.

Pieske

Landersdorfer

Trumm

, et al. CT-guided sacroiliac percutaneous screw placement in unstable posterior pelvic ring injuries: accuracy of screw position, injury reduction and complications in 71 patients with 136 screws. Injury 2015; 46: 333–339.

32.

Sciulli

Daffner

Altman

, et al. CT-guided iliosacral screw placement: technique and clinical experience. AJR Am J Roentgenol 2007; 188: W181–W192.

33.

Zwingman

Konrad

Mehlhorn

, et al. Percutaneous iliosacral screw insertion: malpositioning and revision rate of screws with regards to application technique (navigated vs. conventional). J Trauma 2010; 69: 1501–1506.

34.

Reuther

Rohner

Will

, et al. CT-guided screw fixation of vertical sacral fractures in local anaesthesia using a standard CT. Rofo 2014; 186: 1134–1139.

35.

Wang

Feng

, et al. Percutaneous sacroiliac screw placement: a prospective randomized comparison of robot-assisted navigation procedures with a conventional technique. Chin Med J (Engl) 2017; 130: 2527–2534.

36.

Takeba

Umakoshi

Kikuchi

, et al. Accuracy of screw fixation using the O-arm® and StealthStation® navigation system for unstable pelvic ring fractures. Eur J Orthop Surg Traumatol 2018; 28: 431–438.

37.

Schwabe

Altintas

Schaser

, et al. Three-dimensional fluoroscopy-navigated percutaneous screw fixation of acetabular fractures. J Orthop Trauma 2014; 28: 700–706.

38.

Ruan

Luo

Zeng

, et al. Percutaneous screw fixation for the acetabular fracture with quadrilateral plate involved by three-dimensional fluoroscopy navigation: surgical technique. Injury 2012; 43: 517–521.

39.

Swartman

Pelzer

Beisemann

, et al. Fracture reduction and screw position after 3D-navigated and conventional fluoroscopy-assisted percutaneous management of acetabular fractures: a retrospective comparative study. Arch Orthop Trauma Surg 2021; 141: 593–602.

40.

Amoretti

Marcy

Hauger

, et al. Percutaneous screw fixation of a vertebral pedicle fracture under CT-guidance: a new technique. Eur J Radiol 2012; 81: 591–593.

41.

Fischer

Vogl

Marzi

, et al. Percutaneous cannulated screw fixation of sacral fractures and sacroiliac joint disruptions with CT-controlled guidewires performed by interventionalists: single center experience in treating posterior pelvic instability. Eur J Radiol 2015; 84: 290–294.

42.

Iguchi

Ogawa

Doi

, et al. Computed tomography fluoroscopy-guided placement of iliosacral screws in patients with unstable posterior pelvic fractures. Skeletal Radiol 2010; 39: 701–705.

43.

Wang

Zhou

, et al. Computer navigated percutaneous screw fixation for traumatic pubic symphysis diastasis of unstable pelvic ring injuries. Chin Med J (Engl) 2009; 122: 1699–1703.

44.

Duwelius

Van Allen

Bray

, et al. Computed tomography-guided fixation of unstable posterior pelvic ring disruptions. J Orthop Trauma 1992; 6: 420–426.

45.

Grossterlinden

Rueger

Catala-Lehnen

, et al. Factors influencing the accuracy of iliosacral screw placement in trauma patients. Int Orthop 2011; 35: 1391–1396.

46.

Pishnamaz

Dienstknecht

Hoppe

, et al. Assessment of pelvic injuries treated with ilio-sacral screws: injury severity and accuracy of screw positioning. Int Orthop 2016; 40: 1495–1501.

47.

Tonetti

Carrat

Blendea

, et al. Clinical results of percutaneous pelvic surgery. Computer assisted surgery using ultrasound compared to standard fluoroscopy. Comput Aided Surg 2001; 6: 204–211.

48.

Hamada

Fujimichi

. Classification of radiation effects for dose limitation purposes: history, current situation and future prospects. J Radiat Res 2014; 55: 629–640.

49.

Hopper

Jamison

Lewis

. Learning curves in surgical practice. Postgrad Med J 2007; 83: 777–779.

50.

Wang

Chen

, et al. Comparison between free-hand and O-arm-based navigated posterior lumbar interbody fusion in elderly cohorts with three-level lumbar degenerative disease. Int Orthop 2019; 43: 351–357.

51.

Ravi

Jenkinson

O'Heireamhoin

, et al. Surgical duration is associated with an increased risk of periprosthetic infection following total knee arthroplasty: a population-based retrospective cohort study. EClinicalMedicine 2019; 16: 74–80.

52.

Singhatanadgige

Pholprajug

Songthong

, et al. Comparative radiographic analyses and clinical outcomes between O-arm navigated and fluoroscopic-guided minimally invasive transforaminal lumbar interbody fusion. Int J Spine Surg 2022; 16: 151–158.

53.

Dea

Fisher

Batke

, et al. Economic evaluation comparing intraoperative cone beam CT-based navigation and conventional fluoroscopy for the placement of spinal pedicle screws: a patient-level data cost-effectiveness analysis. Spine J 2016; 16: 23–31.