Feasibility of Imaging-Based 3-Dimensional Models to Design Patient-Specific Osteosynthesis Plates and Drilling Guides

Key Points Question What is the feasibility of using a personalized approach for reconstruction of complex fractures? Findings In this case series study, 10 patients with severe acetabular fractures were surgically treated using 3-dimensional (3-D) surgical planning, patient-specific osteosynthesis plates, and drilling guides. These were designed, produced, sterilized and clinically applied within 4 days and were associated with an accurate reconstruction of the fractured articular surface and a good 1-year clinical outcome. Meaning These findings suggest that 3-D virtual surgical planning, manufacturing, and clinical application of patient-specific osteosynthesis plates and drilling guides were feasible and yielded good clinical outcomes.


Introduction
The overall incidence of acetabular fractures is estimated as 5 to 8 per 100 000 people per year, which accounts for approximately 60 000 injured individuals annually in Europe. 1,2 A pelvic injury can have major consequences for physical functioning, participating in social activities, and the ability to work. Acetabular fracture treatment consists of either nonoperative treatment (56% of patients), open reduction and internal fixation (38% of patients) or primary total hip arthroplasty (THA; 6% of patients). 3 Achieving an optimal reconstruction of the articular surface is associated with improved functional outcome and decreased risk of progressive osteoarthritis and subsequent need for THA, 4-6 but this is particularly challenging for associated-type (ie, more complex) fractures, according to the Letournel-Judet classification, with substantial displacement. 5,7,8 Postoperative computer tomography (CT) analysis of acetabular fractures has demonstrated inadequate reductions in up to 53% of patients in a study by Verbeek et al. 5 Additionally, 36% of patients with an inadequate reduction in the study by Verbeek et al 5 eventually need a conversion to THA, compared with only 10% of patients with an adequate reduction.
Despite the progress in surgical techniques and osteosynthesis plates, even experienced surgeons often do not achieve adequate reconstruction of the fractured acetabulum. 5 Currently, conventional osteosynthesis plates often require multiple intraoperative contouring maneuvers to fit the individual pelvis reasonably. Moreover, the optimal screw positions with good purchase might be hard to determine and verify with fluoroscopy. Unfortunately, a uniform osteosynthesis plate that fits the shape of each pelvis, covers all the fracture patterns, and holds the surgically reduced fracture fragments perfectly in place does not exist.
We developed an innovative surgical procedure for acetabular fractures by using 3-D virtual surgical planning and custom-made patient-specific pelvic osteosynthesis plates with drilling guides. 9 We hypothesized that this new personalized approach would be associated with optimal osteosynthesis plate fitting, accurate screw placements, and adequate fracture reductions. The patient-specific osteosynthesis plates, tailored to both the shape of the pelvis and the fracture type, could be applied to repair accurately one of the most challenging fracture types in orthopedic trauma surgery. Therefore, the aim of this study was to assess whether such a personalized approach is feasible and was associated with accurate reconstruction of associated-type acetabular fractures.

Methods
Approval for this case series was obtained from the University Medical Center Groningen institutional review board. All patients provided written informed consent. This study is reported following Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.

Patients
Eligibility criteria were patients who sustained a displaced unilateral associated-type acetabular fracture 10,11 and required surgical treatment. All consecutive patients with comminuted, displaced T-shaped, or both-column acetabular fracture types between January 1, 2017, and December 31, 2018, were included. A total of 10 patients were treated with patient-specific osteosynthesis plates according to the fast-track personalized procedures (eFigure 1 in the Supplement).

3-D Surgical Planning
Each patient's CT data (Յ1 mm slices; spatial resolution, 0.5-0.6 mm) was used to create a 3-D model ( Figure 1) using Mimics Medical software version 19.0 (Materialise). A preset threshold for bone was Pelvic radiograph, CT-scan (axial, coronal, sagittal CT slices) and 3-D model from patient 5, who had fallen from a height and sustained an associated-type bothcolumn fracture. The fracture caused medial protrusion of the femoral head and severe displacement of the anterior as well as the posterior column. This patient was treated with patient-specific osteosynthesis plates.
used for automatic segmentation of all the fracture fragments. Each fracture fragment was assigned a different color and reduced to its anatomical position by using translational and rotational tools in the planning software. The contralateral hemipelvis was mirrored and used as a template to verify the accuracy of the virtual fracture reduction. 12

Patient-Specific Osteosynthesis Plate Design
Patient-specific osteosynthesis plates tailored from the virtually reduced 3-D fracture model, to achieve optimal support and stable internal fixation of fracture fragments, were designed within 1 day after hospital admission. First, the optimal screw trajectories in relation to the fracture fragments were predetermined in the 3-D fracture model. Subsequently, the patient-specific titanium osteosynthesis plates were designed with 3-Matic software version 11.0 (Materialise), Solidworks Professional software, version 2017 (Dassault Systèmes Solidworks), and the Geomagic package for Solidworks (3D Systems). All screw lengths were predetermined as part of preoperative planning.
Drilling guides were designed to translate the virtual planning to the surgical procedure and to guide the drill bit as well as the screws in the right direction (Figure 2). The guides were provided with additional bone supporting extensions that enabled a perioperative visual check for correct placement. In addition, the guides were supplied with holes for k-wire fixation once the correct position was achieved. The synthetic guides were designed to envelope the osteosynthesis plate and

Osteosynthesis Plate Production Process
The custom-designed patient-specific osteosynthesis plates were manufactured by a regional medical company (Witec Medical) within 3 days for each patient. The osteosynthesis plate was milled out of a medical grade titanium alloy by a 5-axis milling machine. The drilling guides were 3-D laserprinted from medical certified polyamide powder (Oceanz). The osteosynthesis plates and guides were prepared for surgery with a routine 134°C autoclave steam sterilization process.

Surgical Procedure
All the patients underwent operations according to the standard of care by 2 trauma surgeons (F.F.A.IJ. and K.t.D.), each with more than 5 years of experience in pelvic surgery, which helped to avoid bias of the results due to differences in operative skills between surgeons. The fracture pattern and amount of displacement were decisive for the surgical approach. The most suitable approach (ie, anterior intrapelvic approach with or without a lateral window, Kocher-Langenbeck, or a combined approach) was left to the treating surgeon and discussed before designing the osteosynthesis plate. After exposing the fractures, standard tools were used to perform fracture reduction, during which patient-specific osteosynthesis plates were applied to put or keep the fracture fragments in place. The drilling guides with additional bone-supporting extensions, which enabled visual checks when positioning the osteosynthesis plate, were placed on top of the osteosynthesis plate (Figure 3). Intraoperative fluoroscopy was used to verify that the osteosynthesis plate was positioned according to the preoperative 3-D planning. A drill sleeve was inserted into the cylinders of the guide to aim the drill bit correctly. After drilling, the drill sleeve was removed and screws of a predetermined length were inserted 1 by 1 into their proper trajectory. After inserting the last screw, the fracture reduction and osteosynthesis material were checked with fluoroscopy. The drilling guide was removed from the osteosynthesis plate, and all wounds were closed in layers.

Postoperative CT Evaluation
A postoperative CT scan (Յ1 mm slice thickness) was performed to evaluate the accuracy of the acetabular reconstruction and screw positioning (Figure 4). Two experienced trauma surgeons (F.F.A.IJ. and K.t.D.) who were blinded to patient data assessed the quality of the reduction by measuring the greatest residual gap or step-off displacement at the acetabular dome on the postoperative CT scan in any of the axial, sagittal, or coronal plane views according to a standardized method. 13 The quality of the reduction was graded according to Matta criteria and the newly proposed CT-based criteria. 8,14,15 These images represent the preoperative planning, osteosynthesis plate designs, and drilling guides for patient 5, who underwent an associated-type bothcolumn fracture operation. Our plan was to first perform a Kocher-Langenbeck approach, reduce the posterior column with a collinear reduction clamp and apply the patient-specific posterior plate (A). The proximal screws in the posterior plate were aimed at the ilium (cream-colored fragment) by using the drilling guide (B) to not compromise the reduction of the anterior column (purple fragment). After reduction and fixation of the posterior column, the wound was closed, and an additional anterior intrapelvic approach was performed. The anterior column was stabilized with a patient-specific anterior plate (C), which provided optimal support along the quadrilateral surface (red fragment). The screws in the anterior plate were aimed in the right direction by the drilling guide placed temporarily on top of the osteosynthesis plate (D).
Furthermore, postoperative CT data were used to create a 3-D model of the reconstructed pelvis, which was matched using surface-based matching with the preoperative 3-D surgical planning to assess how accurately the preoperative plan had been executed ( Figure 4). The deviation in screw direction between the preoperative and postoperative 3-D model was determined for each screw.
Additionally, differences in screw lengths between the preoperative planning and postoperative situation were calculated to assess the feasibility of using preplanned screw lengths and of thereby saving some operation time.

Clinical Follow-up
All patients were followed up at 3 months and 1 year. The patient-reported outcomes were assessed The SMFA scores may vary from 0 to 100, with a higher score indicating a worse function. 16,17 Also, date of return to work was recorded. All complications were monitored during the follow-up.

Statistical Analysis
The Wilcoxon signed rank test was used to assess differences between the preoperative and postoperative gap and step-off by using SPSS statistical software version 23 (IBM). Furthermore, the paired samples Wilcoxon signed rank test was used to assess differences between the preinjury, 3-month, and 1-year patient-reported outcome scores (SMFA scores). The Spearman correlation coefficient was used to assess correlations between the postoperative reduction (ie, gaps and stepoffs) and clinical outcome (SMFA). P values were 2-sided, and statistical significance was set at P = .05. Data were analyzed at the 1-year follow-up.

Demographic Characteristics
Ten patients, including 9 men and 1 woman, with a median (range) age of 63 (46-79) years who sustained an associated-type both-column fracture or T-shaped fracture were included. A total of 15 patient-specific osteosynthesis plates were used. Five patients were treated with only an anterior plate, and the other 5 patients received an additional posterior plate (Table).

Surgical Procedure
The osteosynthesis plates and drilling guides were designed, fabricated, and sterilized within 4 days (Video). All plates fit well and did not need additional bending maneuvers during the operation. All the patient-specific osteosynthesis plates were a good reference for the reduction and functioned as guides for fracture reduction. Only 3 of 95 screws that were preoperatively planned could not be placed during surgery owing to the narrow space and the presence of soft tissues deep in the pelvis, which hampered the predetermined drilling angle. None of the screws penetrated into the hip joint or caused soft tissue injuries.
A total of 95 screws were placed by using a drilling guide. Per patient, the median (IQR) number of screws was 6.5 (6-7) anterior screws and 6 (5-7) posterior screws. The median (IQR) difference  between the planned screw length and the actual screw length was 1.7 (1-3) mm, and the median (IQR) difference between the planned and actual screw direction was 7.1°(7°-8°), which are within the safe zone for using personalized acetabular fracture surgery with confidence in clinical practice.

Discussion
This case series demonstrates that the application of 3-D surgical planning and patient-specific osteosynthesis plates, combined with drilling guides, was feasible and allowed for accurate operative reconstruction of complex acetabular fractures within 4 days after trauma. The application of patient-specific osteosynthesis plates and drilling guides provides the possibility to execute the preoperative plan and attain the predetermined osteosynthesis plate and screw positions, which were associated with accurate reconstruction of the articular surface and good functional patient recovery.
The use of 3-D virtual models allows the surgeon to gain more insight into the fracture pattern and treatment strategy. 18 Moreover, 3-D-printed pelvic models are used as templates for fitting and precontouring conventional osteosynthesis plates before the actual surgical procedure. [18][19][20][21] case series indicate that the use of 3-D-printed models in pelvic surgery are associated with reduced blood loss and shorter operation time. [22][23][24][25][26] Although the reported use of prebent plates, adapted to printed 3-D-models, show some benefits, little data are available on whether 3-D-printed models improve the quality of the reduction. 22,23 Our additional efforts in producing patient-specific osteosynthesis plates, with drilling guides, allowed for good alignment of the plate and screws to fit the fracture and shape variations of the pelvis, and using the plates as a reference was associated vs all types of fractures in their series) and differences in imaging modalities used for postoperative assessment (CT scan vs radiograph).
Our techniques have some benefits for surgery. First, a review of the 3-D surgical planning by a multidisciplinary team provides an opportunity for consultation and the possibility to discuss the optimal surgical approach, features of the osteosynthesis plate, and screw positions, thus following the principle of "Plan your operation-and operate your plan!" as described by Schelkun. 29 Second, guided screw placement enables tailoring the screw positions to the fracture reduction strategy. For instance, 1 patient sustained a severely displaced both-column fracture and needed a combined approach with posterior and anterior plates. The screws in the posterior plate were guided away from the anterior column and hence did not interfere with the reduction of the anterior column in the second phase of the operation. Also, the anterior plate was tailored to the fracture line in the ilium, which avoided an extra lateral window approach. Regarding another patient, the drilling guide was used to aim a lag screw through the anterior plate to indirectly reduce and fixate a large posterior wall fragment; screw accuracy planning was crucial here. Overall, the patient-specific osteosynthesis plates with guided screw insertion optimized fixation abilities and avoided additional surgical approaches in several patients.

Future Perspectives
Over the next few years, we will work on increasing the efficiency of personalized fracture care. If software applications and advanced technologies contribute to more efficient and precise surgical treatment to the benefit of the patient by improving physical functioning for years, it would be worthwhile to explore the general applicability further in the near future. A follow-up study is needed that compares patient-specific osteosynthesis with conventional osteosynthesis.

Limitations
This study has some limitations. One such limitation is that the advanced technologies are not applicable yet in all hospitals. We realize that these innovative techniques require sufficient resources, including the availability of dedicated engineers, validated software packages, and collaborative osteosynthesis plate production facilities. The cost for designing and producing the osteosynthesis plates was not part of this feasibility study. There is no selection bias, because in the interest of challenging our innovative personalized approach, the most complex acetabular fractures were eligible for this study. However, a potential confounding factor could be the experience of the surgeon. Therefore, all of the surgical procedures in this case series were performed by the same team. Another potential bias could be that the CT-based gap and step-off measurements are prone to intraobserver and interobserver variability. 30 Therefore, measurements were performed in consensus by 2 experienced surgeons, who were blinded to patient data.

Conclusions
In this case series, we describe the development and implementation of a patient-specific multidisciplinary workflow for acetabular fracture surgery that made it possible to reconstruct the acetabulum accurately and fixate the fracture fragments with custom-made osteosynthesis plates, resulting in good 1-year clinical outcome. Moreover, none of the osteosynthesis plates required intraoperative contouring maneuvers and all the screws could be placed accurately using the drilling guides.