Does 3D-assisted surgery of tibial plateau fractures improve surgical and patient outcome? A systematic review of 1074 patients

Purpose The aim of this systematic review was to provide an overview of current applications of 3D technologies in surgical management of tibial plateau fractures and to assess whether 3D-assisted surgery results in improved clinical outcome as compared to surgery based on conventional imaging modalities. Methods A literature search was performed in Pubmed and Embase for articles reporting on the use of 3D techniques in operative management of tibial plateau fractures. This systematic review was performed in concordance with the PRISMA-guidelines. Methodological quality and risk of bias was assessed according to the guidelines of the McMaster Critical Appraisal. Differences in terms of operation time, blood loss, fluoroscopy frequency, intra-operative revision rates and patient-reported outcomes between 3D-assisted and conventional surgery were assessed. Data were pooled using the inverse variance weighting method in RevMan. Results Twenty articles evaluating 948 patients treated with 3D-assisted surgery and 126 patients with conventional surgery were included. Five different concepts of 3D-assisted surgery were identified: ‘3D virtual visualization’, ‘3D printed hand-held fracture models’, ‘Pre-contouring of osteosynthesis plates’, ‘3D printed surgical guides’, and ‘Intra-operative 3D imaging’. 3D-assisted surgery resulted in reduced operation time (104.7 vs. 126.4 min; P < 0.01), less blood loss (241 ml vs. 306 ml; P < 0.01), decreased frequency of fluoroscopy (5.8 vs. 9.1 times; P < 0.01). No differences in functional outcome was found (Hospital for Special Surgery Knee-Rating Scale: 88.6 vs. 82.8; P = 0.23). Conclusions Five concepts of 3D-assisted surgical management of tibial plateau fractures emerged over the last decade. These include 3D virtual fracture visualization, 3D-printed hand-held fracture models for surgical planning, 3D-printed models for pre-contouring of osteosynthesis plates, 3D-printed surgical guides, and intra-operative 3D imaging. 3D-assisted surgery may have a positive effect on operation time, blood loss, and fluoroscopy frequency. Supplementary Information The online version contains supplementary material available at 10.1007/s00068-021-01773-2.


Introduction
Intra-articular fractures of the tibial plateau are usually composed of complex fracture patterns including multiple fracture fragments, which are displaced and rotated in multiple directions. Achieving normal knee alignment and an optimal reconstruction of the articular surface decreases the risk of progressive osteoarthritis [1]. However, due to the complexity of these fractures, the goals of surgery cannot always be achieved. Recently, it has been shown that up to 30% of the surgically treated tibial plateau fractures resulted in a suboptimal reduction [2]. Assessment of the fracture is essential to fully understand the fracture pattern and to choose the optimal treatment strategy. Clinical decisionmaking and preoperative planning is mostly based on conventional imaging modalities, including plain radiographs, two-dimensional (2D) fluoroscopy and 2D CT images [3]. With these modalities, it is difficult to fully comprehend the true extent of these injuries, since the fracture fragments are often displaced and rotated in multiple directions. 3D visualization and printing modalities have the potential to provide the physician with a better understanding of the fracture pattern and could improve treatment strategy and patient outcome [4,5].
The growing popularity and expansion across industries providing 3D printing resources has substantially decreased costs, increased access, and led to multiple applications in orthopaedic trauma surgery [6,7]. Early results on the clinical application of 3D printing improved levels of understanding into complex fractures for both surgeons and patients and strengthened the informed consent process [8]. Also, 3D technologies may be valuable for teaching students about fracture morphology or explaining residents about the surgical plan [9]. 3D-assisted surgery encompasses the use of 3D technology to pre-plan the operation and guide the surgeon to the planned outcome during surgery. This includes a spectrum of modalities such as 3D visualization, 3D printing and patient-specific surgical guides or implants. However, the potential advantages of 3D-assisted surgery in tibial plateau fracture management are still subject of debate.
Despite the rapid advances in technology and an increasing number of publications on the applications of 3D technologies, a comprehensive overview of the current evidence for the application of 3D-assisted surgery of tibial plateau fractures is still lacking. Therefore, the purpose of this systematic review is to provide a complete and comprehensive overview of the currently used concepts of 3D-assisted surgery in patients receiving surgical treatment for their tibial plateau fracture by including both observational and intervention studies. The aim is to answer the following clinical research questions: (1) Does the clinical application of 3D-assisted surgery for tibial plateau fractures improve intra-operative results in terms of operation time, blood loss, fluoroscopy time and intra-operative surgical revisions compared to conventional surgery? (2) Does the application of 3D-assisted surgery improve postoperative results in terms of patient functional outcome compared to conventional surgery?

Materials and methods
This systematic review was performed according to the Preferred Reporting Items for Systematic Reviews (PRISMA) [10]. The protocol of this systematic review is registered in the international PROSPERO-database (CRD42021235524). Ethical approval was not required for this study.

Search strategy
The Pubmed and Embase libraries were searched on the 1st of February 2021 for articles published on state-of-the-art 3D technology between January 2010 until January 2021. The search string was developed in collaboration with a medical librarian. The exact search string for the different libraries is shown in the online supplementary (Appendix 1 in Supplementary file 1).

Study selection
Eligible studies for inclusion were randomized controlled trials, prospective and retrospective observational studies, descriptive studies, and case reports reporting on the use of 3D techniques in the management of tibial plateau fractures in orthopaedic trauma patients. Studies were excluded in case of: (1) paediatric fractures; (2) fracture classification studies; (3) animal or cadaveric studies; (4) review articles, letters to the editor or conference abstracts; and (5) studies in another language than English, German, French, Spanish or Dutch.
All articles were imported into Rayyan QCRI, a webbased sorting tool for systematic literature reviews [11]. The study selection was then performed in two phases: first two reviewers (NA, FIJ) independently screened the articles for eligibility based on the titles and abstracts using the Rayyan QCRI tool. Second, all articles which were considered eligible, were subsequently screened in full text by the same reviewers. Disagreement was resolved by discussion according to the Cochrane Handbook for Systematic Reviews of Interventions [12].

Quality check and data extraction
Methodological quality and risk of bias of the included studies were independently assessed by NA and FIJ according to the guidelines of the McMaster University Occupational Therapy Evidence-Based Practice Research Group [13]. Any continued disagreements were solved during a consensus meeting with NA, FIJ and IR. The McMaster critical appraisal consists of eight categories including: (1) study purpose; (2) literature review; (3) study design; (4) study sample; (5) study outcome; (6) study intervention; (7) study results; and (8) conclusions and implications. Scores were giving with 'yes = 1 point', 'no = 0 points', 'not addressed (NS)', and 'not applicable (NA)'. The total score reflects the methodological quality with a maximum score of 16 for RCTs and 14 for other designs. The definitive score is expressed as a percentage that may vary from 0 to 100%, with a higher score indicating a higher methodological quality. Scores between 90 and 100% were considered as excellent quality, studies between 75 and 89% as good quality studies and studies < 75% as moderate quality studies.
The data extraction was independently conducted (NA, FIJ) using a precompiled extraction file (Microsoft Excel version 14.0; Microsoft Inc., Redmond, WA, USA). Information on study characteristics, fracture classification, 3D technologies and outcome measures were extracted. In case data regarding the reported outcomes was missing, authors were contacted to retrieve raw data or means with their standard deviations.

Outcome measures
All parameters describing the operation were determined to assess the effect of 3D-assisted surgery on intra-operative results. These parameters include operation time, blood loss, fluoroscopy time, and the number of intra-operative revisions of the fracture reduction or implant position as a result of intra-operative 3D imaging. Second, Patient-Reported Outcome Measures (PROMs) were recorded to evaluate the effect of 3D-assisted surgery on postoperative functional outcome.

Statistical analysis
Analysis of the extracted data was performed using Rev-Man (version 5.4.1). Continuous variables were presented as means with standard deviation (SD) and dichotomous variables as frequencies and percentages. Continuous outcomes were pooled using the inverse variance weighting method and were presented as weighted mean difference (WMD) with the corresponding 95% confidence interval (95%CI). Heterogeneity between studies was assessed for all reported outcomes by the I 2 statistic for heterogeneity. The I 2 statistic was interpreted according to the benchmarks of the Cochrane Handbook for Systematic Reviews of Interventions, which considered < 40% as irrelevant, 30-60% as moderate heterogeneity, 50-90% as substantial heterogeneity, and > 75% as considerable heterogeneity [12]. A P value of < 0.05 was considered to indicate statistical significance.

Identified 3D applications in tibial plateau fracture surgery
Within this search, five different concepts of 3D-assisted surgery in the management of tibial plateau fractures were identified over the past decade: '3D virtual fracture visualization', '3D printed hand-held models', 'Pre-contouring of osteosynthesis plates', '3D printed surgical guides' and 'Intra-operative 3D imaging'. Figure 2 depicts a representation of these concepts.

3D virtual fracture visualization
Two studies reported about the use of 3D virtual visualization of the fracture before surgery [30,34]. Suero et al. used the VoXim software (IVS Solutions AG, Chemnitz, Germany) to create a 3D reconstruction of the fracture from which the surgeon determined the surgical plan [30], whereas Zhang et al. used the Mimics software (Materialise, Leuven, Belgium) to determine a preoperative plan in which the reduction procedure was simulated [34]. Using the 3D software, the required elevation of the depressed articular surface was measured and the surgical procedure was virtually planned.

3D-printed hand-held models
The majority of the studies reported on the use of 3D-printed models of tibial plateau fractures [8, 20-22, 24, 27, 29, 32, 33]. In these studies, a 3D-printed model of the fractured tibial plateau was used to determine the surgical plan and to guide the surgeon during surgery (Fig. 3). Furthermore, the 3D-printed models were found to be useful for educating residents and students, and to inform patients about their injury [8].

Pre-contouring of osteosynthesis plates
Two studies reported on the use of pre-contoured osteosynthesis plates [18,25] using either a 3D-printed contralateral mirrored tibia or a virtually reduced fracture model. Using the printed models, implants were (pre-)operatively bended for optimized fitting along the contour of the proximal tibia.

3D printed surgical guides
Three studies described the use of 3D-printed surgical guides [23,26,31]. In two studies, the directions of the screws were virtually predetermined. A surgical guide was designed to translate the predetermined screw trajectories to the actual surgical procedure [23,26]. Another application of a surgical guide was found in the operative correction of a malunited tibial plateau fracture by Wang et al. [31]. First the osteotomy was performed using a guide, which helped the surgeon to perform the (virtually) predetermined osteotomy. Secondly, a reduction guide was applied to help the surgeon to reduce the fragment to its original anatomical position.

Intra-operative 3D imaging
Four studies reported on the use of intra-operative 3D images [16,17,19,28]. These studies investigated the use of an intra-operative 3D imaging system, which was used to verify whether the achieved surgical reduction was satisfactory. Using this technology, the surgeon was able to make prompt perioperative decisions based on 3D instead of 2D fluoroscopy images. In case of dissatisfaction with the articular reduction or the position of the screws or implant,

Effect of 3D-assisted surgery on clinical outcome
To answer the first research question, the effect of 3D-assisted surgery on intra-operative results in terms of operation time, blood loss, fluoroscopy time and intra-operative revisions was assessed. The second research question concerns the effect of 3D-assisted surgery on postoperative results in terms of functional outcome.

Blood loss
Six studies reported on blood loss [20,21,24,27,29,34]. However, Giannetti et al. did not report the standard deviation and was, therefore, excluded from further analysis [20], leaving five studies, including one excellent quality study and four moderate quality studies. Four studies reported that 3D-printed model-assisted fracture surgery led to significantly less blood loss in comparison with conventional surgery [21,24,27,29]. Zhang et al. reported that the use of a preoperative 3D virtual model resulted in significantly less blood loss compared to conventional surgery [34]. The blood loss was significantly less in the 3D-assisted group in comparison with the conventional group (WMD 73.1 ml, 95% CI −102.8 to −43.5) (Fig. 5). The heterogeneity was considerable within these studies (I 2 = 96%).

Fluoroscopy frequency
Four studies reported on the frequency of use of fluoroscopy [21,24,27,29], including one study of excellent quality and three of moderate quality. The use of fluoroscopy was significantly reduced in the 3D-assisted group in comparison with the conventional group (WMD 3.5 times used, 95% CI Fig. 3 a Fluoroscopy of an intra-articular fracture of the tibial plateau. b 3D-printed hanheld model of the tibial plateau fracture. c Intra-operative fracture assessment using the 3D-printed hand-held model  (Fig. 6). The heterogeneity was considerable within these studies (I 2 = 96%).

Intra-operative revision rates
Three studies reported on immediate intra-operative revision rates resulting from intra-operative 3D imaging [16,19,28], including two good quality and one moderate study. These articles reported on a total of 698 patients of which 183 (26.2%) patients had an instant intra-operative revision of the fracture reduction or implant position as a result of intra-operative 3D imaging (Table 2).

Patient-reported functional outcome
Three studies reported on functional outcome [21,24,29], of which one was of excellent quality. All studies used the Hospital for Special Surgery (HSS) scoring system. The HSS scoring system is based on a total of 100 points. A HSS score of ≥ 85 points is considered excellent, 70-84 points is good, 60-69 points is fair, and ≤ 59 points is poor [35]. Two studies reported the actual HSS score [24,29], whereas another study provided the rating of the score [21]. The weighted HSS score was 88.6 (86.1-90) in the 3D-assisted group and 82.8 (79.1-85) in the conventional group. Guo et al. reported no relevant differences in HSS score between 3D printing assisted and conventional surgery [21].

Discussion
The rationale for applying 3D technology in tibial plateau fracture surgery is that it may optimize preoperative planning, potentially improves fracture reduction and eventually benefits the patients' recovery. This systematic review aimed to provide an overview of the current concepts of 3D-assisted tibial plateau fracture surgery and their relation to clinical outcome. The search was not limited to study design, which provides a complete overview of all 3D applications for tibial plateau fracture surgery published over the last decade. Five different concepts of 3D-assisted surgery were identified including '3D virtual visualization', '3D printed hand-held models', 'Pre-contouring of osteosynthesis plates', '3D printed surgical guides', and 'Intraoperative 3D imaging'. Pooled analysis of studies, concerning mainly the use of 3D-printed models, showed to have a positive effect on operation time, blood loss, and fluoroscopy frequency. This review revealed that the majority of the studies (nine) used 3D-printed hand-held fracture models in clinical practice. Converting a CT-scan into a hand-held 3D-printed model could provide valuable insights for the pre-operative planning of the fracture reduction and fixation. Care should be taken regarding the soft tissue injuries which cannot be taken into account in the 3D model. These models could be sterilized and used in theatre to guide the surgeon during the operation. From an educational perspective, these models allow surgical trainees to accurately plan the surgery ahead of time, and subsequently discuss their plan with a senior. Moreover, a 3D-printed model may help in providing patient information during clinical consultation [8]. One could argue that most of these benefits could also be achieved with only 3D virtual visualization of the fracture [36]. Besides that it saves the cost of printing (€ 50 -100,-for a proximal tibia), it is instantly available and has no environmental impact. Yet, in this review only two articles were identified that described the use of a 3D virtual model for surgical planning [30,34]. It should be noted that 3D visualization and printing itself has a learning curve, and it takes time to become familiar with the software. Virtual preoperative planning and discussing a new case may easily take up to two hours, of which a significant part is spent on the process of segmenting the CT-scan into a 3D model, virtually reducing the fracture fragments, and predetermining the implant positions.

Fluoroscopy frequency (number of times)
Guo et al. [21] 3D printed fracture models 14 14 2.7 ± 0.4 4.7 ± 0.6 < 0.05* Lou et al. [24] 3D printed fracture models 34 38 5.3 ± 0.2 7.1 ± 0.2 < 0.001* Ozturk et al. [27] 3D printed fracture models 10 10 10.7 ± 2 18.5 ± 2.2 < 0.05* Shen et al. [29] 3D printed fracture models 20 22 6.5 ± 1. to small case series. The full potential of this technique should therefore be further explored. The use of 3D-printed surgical guides should be considered another 3D technique, which aims at translating a pre-operative plan to the patient [23,26,31]. Three case series introduced this concept for tibial plateau fractures and showed that 3D-printed guides may help the surgeon to accurately adhere to the pre-determined surgical plan. 3D-printed surgical guides are widely used in clinical practice and have been successfully applied in neurosurgery, dental surgery, spinal surgery and maxillofacial surgery [38]. In spinal surgery for instance, the use of 3D-printed drill guides led to accurate vertebral screw insertion with a mean deviation of 1.4 mm and 6.7° from the planned entry point and screw trajectory, respectively [39].
Several studies assessed the use of intra-operative 3D imaging to verify fracture reduction, implant position, and screw trajectories and lengths. These studies showed instant intra-operative revision rates up to 27% as a consequence of the 3D imaging [16,19,28]. However, these studies evaluated only the intra-operative acts resulting from the 3D imaging and not the clinical outcome. Downsides of this technique are the radiation exposure and increased operation time, where in more than 70% of the patients the intra-operative 3D imaging did not lead to any adjustments in the achieved surgical reduction. It should therefore be evaluated which fractures might benefit from this technique, and which not.
The main research questions concerned the effects of 3D-assisted surgery of tibial plateau fractures on intra-and postoperative outcomes. Surgery assisted by 3D visualization or prints resulted in improved intra-operative results in terms of operation time, blood loss and frequency of fluoroscopy. This is in line with previous findings regarding the use of 3D printing techniques in orthopaedic trauma fracture care [5,38]. 3D technology provides the surgeon the ability to extensively prepare the surgery. This benefits the workflow in the operating room leading to a reduction in operation time and the frequency of fluoroscopy. A possible explanation for the decrease in blood loss could be the efficiency during the operation and a smaller incision size due to improved preoperative planning. Zhang et al. showed that the 3D-assisted group had a significant smaller incision length [34]. Studies included in this review indicate that 3D-assisted surgery might improve functional outcome. It could be hypothesized that 3D-assisted surgery leads to improved preoperative planning and eventually better reduction of the fracture. This assumption is still a matter of debate since no post-operative CTs were available in any of the studies. The effect of the 3D technique on the fracture reduction should, therefore, be further assessed.
This review has some strengths and some limitations. First, this review provides a clinically question-driven overview about the ongoing debate whether these advanced 3D technologies contribute to operation results and patientrecovery. To present a complete overview of the stare-ofthe-art 3D technologies applied for tibial plateau fracture surgery we were forced to not restrict our search to solely RCTs. Inevitably, the included studies therefore encompass a wide range of study designs including case series, observational studies and retro-and prospective cohort studies. Due to the wide range of the methodological quality and heterogeneity between these studies, the pooled analysis of operation time (I 2 = 88%), blood loss (I 2 = 96%) and fluoroscopy frequency (I 2 = 96%) should be interpret with caution. Moreover, some studies suffered from a limited sample size. Lastly, different concepts of 3D technologies were aggregated under the term "3D-assisted surgery". However, the studies used for the pooled analysis mainly concerned the use of 3D-printed models and 3D virtual visualization. This hampers the generalizability of the results and therefore these should be interpreted with caution. High-quality randomized controlled trials for each of the 3D application are, therefore, recommended to fully explore the potential benefits of these rapid developing advanced technologies.

Conclusion
Over the last decade, five different concepts of 3D-assisted surgical management of tibial plateau fractures emerged: '3D virtual visualization', '3D printed hand-held models', 'Pre-contouring of osteosynthesis plates', '3D printed surgical guides', and 'Intra-operative 3D imaging'. Several studies indicate that 3D-assisted surgery had a positive effect on operation time, blood loss, frequency of fluoroscopy, and functional outcome. However, 3D technologies also come with a price in preparation time and production costs (i.e. software, materials, printing devices). The potential benefits should be further investigated in high-quality studies before widespread clinical use.