In‐house virtual surgical planning and guided mandibular reconstruction is less precise, but more economical and time‐efficient than commercial procedures

To compare an in‐house and a commercially available surgical planning solution for mandibular reconstruction in terms of postoperative reconstruction accuracy and economic benefit.


| INTRODUCTION
Oral squamous cell carcinoma (OSCC) accounts for the majority of malignant head and neck tumors and is one of the most common cancers worldwide. 1In advanced tumor stages, OSCC invasion may affect the adjacent mandible, requiring resection of the affected areas to achieve curative oncologic treatment. 2Tumor progression and surgical resection are associated with loss of function such as chewing, swallowing, and speaking, impaired facial aesthetics, and correspondingly reduced postoperative health-related quality of life (HRQOL). 3Therefore, reconstruction of plastic and functional defects is an important part of surgical planning. 4Reconstruction of these defects often requires free microvascular anastomosed bone grafts, mainly from the fibula, scapula, and iliac crest. 5This complex procedure relies heavily on surgical skill.There is no uniform standard for shaping the bone graft. 6The development of computer-aided virtual surgical planning (VSP), combined with computer-aided design and computer-aided manufacturing (CAD/CAM) for the production of surgical guides for guided surgery, is now considered standard practice for mandibular reconstruction. 7These procedures offer the potential for a more precise and customized surgical approach, precise positioning of the vascularized free tissue graft, reduced surgical and graft ischemia time, and improved postoperative aesthetic results. 8,9VSP and the production of surgical guides for guided surgery can be performed either in-house, using software-based planning solutions such as the Mimics Innovation Suite (Materialise, Leuven, Belgium) or the opensource project Blender (Blender ® ; Blender Foundation and Institute; Amsterdam, The Netherlands), or by a commercial partner. 10For in-house planning, a 3Dprinted mandible can be created based on the virtual reconstruction result, to which prefabricated osteosynthesis plates can be manually fitted as accurately as possible. 11,12In commercial surgical planning, a patient-specific implant (PSI) is virtually modeled and manufactured in an additive or subtractive fabrication process to precisely match the shape and size of the patient's reconstructed mandible. 13However, commercial manufacturing processes can increase healthcare costs and can be time-consuming, potentially leading to delays in tumor therapy. 8,14The aim of this study was to compare an in-house and a commercially available surgical planning solution for guided mandibular reconstruction after ablative surgery in the head and neck region in terms of postoperative reconstruction accuracy and economic benefit.

| Patients
This study retrospectively analyzed and compared the mandibular reconstructions of two cohorts of OSCC patients.Fifteen patients underwent in-house VSP and guided surgery, while 14 patients underwent commercial partner planning that included fabrication and integration of a PSI (N = 29, total).All patients underwent surgery consecutively between December 2017 and May 2022.There was no randomized assignment of patients to one of the two surgical planning solutions.Initially, all patients were treated with the in-house planning solution, followed by the commercial planning solution cohort.Preoperative VSP data sets and postoperative CT control data sets were used for morphometric analysis.Differences between the in-house and commercial groups were analyzed for the following morphometric parameters: comparison of preoperative and postoperative bone segment volume and length, comparison of preoperative and postoperative condylar distance, comparison of planned and actual osteotomy angle, comparison of overall reconstruction deviation (total mandible; VSP vs. postoperative reconstruction result), comparison of intersegmental gap volume and length.Moreover, intersegmental ossification and the incidence of wound healing disorders were evaluated over the clinical follow-up.Finally, a cost-benefit comparison of the two planning procedures was performed to determine the economically superior procedure.The study was approved by the Ethics Committee of the University Medical Center Goettingen, Germany (approval no.14/7/19).

| In-house VSP and CAD/CAM-based fabrication of surgical guides
CT scans of the head-neck-thorax region performed during initial tumor staging and angio-CT (CTA) scans for evaluation of vascular supply to the lower extremity (each with a slice thickness of 0.6 mm) were used for VSP.DICOM data sets of the skull including the mandible and the transplant region were segmented and converted into 3D anatomical models.All segmentations and VSP steps were performed using the Mimics Innovation Suite software (Materialise).A virtual osteotomy of the affected region of the mandible was performed with a safety margin of 1 cm anteriorly and posteriorly.The 3D model of the bone transplant was fitted into the bone defect to properly reconstruct the external contour of the original mandible segment.The individual bone transplant segments were virtually osteotomized.A 3D model of the entire reconstructed neomandible was printed using a Monoprice Inventor IIIP 3D printer (Monoprice, Brea, CA) with PLA material (Polymaker, Houston, TX).A 2.3 or 2.7 mm conventional reconstruction plate (KLS Martin, Tuttlingen, Germany) was customized to fit the segments and fixed to the printed model with osteosynthesis screws (KLS Martin) (Figure 1A).The osteosynthesis plate was digitally segmented from a Cone Beam Computed Tomography (CBCT) DICOM data set and the resulting 3D model was inserted into the VSP (Figure 1A).Subsequently, two surgical guides for mandibular resection (one for the anterior and one for the posterior resection plane) and a guide for bone transplant harvesting were fabricated using an in-house CAD/CAM-based procedure.The final surgical guides were 3D printed from surgical guide resin (FormLabs, Somerville, MA) using the For-mLabs Form 3B+ 3D printer (FormLabs) and sterilized.The pre-bent osteosynthesis plates were sterilized and prepared for surgery.

| Commercial surgery planning solution
In 14 cases, surgical planning was performed by a certified commercial partner.For this purpose, CT scans of the head-neck-thorax region and CTA scans of the lower extremities obtained during the staging examinations were transmitted in encrypted form via the partner's online platform.The entire planning process, including segmentation and creation of the 3D models, as well as virtual resection and subsequent modeling of the patientspecific implant (PSI; Figure 1B) and surgical guides, was performed by a trained 3D designer employed by the commercial partner.Upon completion of the initial design, an online meeting was held to implement any suggested changes.The final design was approved by the surgeon and the manufacturing process was initiated.

| Morphometric evaluation
All morphometric measurements were performed using the 3-matic and Mimics Suite software (Materialise).The distance between the medial edges of the mandibular condyles was measured virtually to determine the differences in intercondylar distance between the preoperative and postoperative mandibles (Figure 2A).This allowed a reliable and reproducible point to be defined.For volume comparison, each individual segment of the preoperative and postoperative data sets was filled to create gapless massive fillers.The mass of the volume of a segment was specified in cubic millimeters in the 3-matic software when the segment was selected.To compare the length of the preoperative and postoperative segments, the longest and shortest sections of each segment were measured and the mean was calculated (Figure 2A,B).The osteotomy angles were measured from medial to lateral at the distal and mesial ends, respectively (Figure 3).At the cutting edge, the angle leg was placed centrally through the flat cross-section of the segment.A heat map was used to compare the deviation of the preoperative and postoperative segments (Figure 4A) and the entire mandible (Figure 4B).For the most accurate matching, the largest portion of the resected mandible was selected as the reference part using the Mimics Suite software and overlaid with a matching point distance of 0.2 mm using the Global Registration command.A heat map was then generated using the "Create Part Comparison Analysis" command in the 3-matic software.Statistical evaluation provided percentiles, median, mean, standard deviation and root mean square (RMS) values.

| Evaluation of intersegmental ossification and wound healing disorders
To assess the postoperative bone union of the reconstruction segments to each other and to the adjacent mandibular bone, a bone union score was applied as previously described. 15,16For each patient, the postoperative followup CT data set furthest from the surgery was evaluated.This resulted in a mean follow-up of 22.58 months (SD 14.91 months).Each individual osteotomy was scored separately on the CT scans.A score of 0-2 was assigned, with 0 indicating no callus formation (Figure 5A), 1 indicating complete callus formation only on the labial or lingual side (Figure 5B), and 2 indicating complete callus formation on both the labial and lingual sides (Figure 5C).Thus, a three-segment reconstruction resulted in four scored osteotomies, two segments resulted in three osteotomies, and one segment resulted in two osteotomies.The mean of the individual osteotomy scores was calculated for the final statistical analysis.
Patients' clinical records were reviewed for possible postoperative wound healing disorders.Wound healing abnormalities included avascular graft tissue, exposed osteosynthesis material, exposed bone, and fractures of the osteosynthesis material.

| Cost-benefit analysis
To determine which surgical planning procedure (inhouse vs. commercial) provides economic benefit, the costs of both planning procedures were determined on a patient-by-patient basis.For the commercial surgical planning procedure, the individual invoices provided with the product were used, including the cost of the PSI, the transplant guide, the resection guides, and the anatomical models.A further breakdown of costs was not possible due to commercial secrets.To determine the planning costs for the in-house procedure, the costs for individualized plate osteosynthesis were first determined.This included the cost of materials (screws and plate), the cost of customizing the plate (labor time), and the cost of fabricating the planning model.The costs of modeling and manufacturing the transplant and osteotomy guides using the CAD/CAM process were determined.The virtual planning steps were included in each manufacturing cost.In addition, the number of working days from the start of planning to the completion of the product or shipment of the product was determined as a potential benefit.

| Statistical analysis
Continuous data were reported as mean with standard deviation and confidence interval.All data were tested for normal distribution using the Kolmogorov-Smirnov

| Patients' clinical baseline characteristics
A total of 29 consecutive patients with extensive tumor growth and indication for hemi-mandibulectomy, who underwent surgery between December 2017 and May 2022, were included in this study.The baseline clinical characteristics of the patients are summarized in Table 1.Surgical planning was performed in-house in 15 cases (52%) and by an external commercial partner in 14 cases (48%).Most bone defects were reconstructed with two segment bone grafts (n = 18), in five cases the defect was reconstructed with one bone segment and in six cases with three bone segments.The heterogeneity in defect size and extent was balanced between the two treatment groups (in-house vs. commercial planning solution).There were no relevant gender differences between the two groups.

| Comparison of reconstruction accuracy
All results of the morphometric comparison are presented in Table 2 and Figures 6 and 7. When comparing the volume of each bone segment, significant differences were found between the pre-and postoperative transplant segments for both the in-house planning group (p = 0.0431; Figure 6A) and the commercial planning group (p < 0.0001; Figure 6B).No significant differences could be observed for the in-house and commercial planning groups when comparing preoperative and postoperative bone segment lengths (Figure 6C,D).Similarly, no significant differences could be found between preand postoperative intercondylar distances for either group (Figure 6E,F).When comparing the planned and postoperative osteotomy angles, significant differences were found in the in-house group (p = 0.0391; Figure 6G), while no significant differences were found in the commercially planned group (p = 0.6544; Figure 6H).Deviation analysis of individual bone segments between VSP and postoperative reconstruction results showed no significant differences between the two planning methods (Figure 7A).However, deviation analysis for the entire mandible revealed a significantly more accurate overlay (smaller deviation) in favor of commercial planning compared to in-house planning (p = 0.0217; Figure 7B).A comparison of the cumulative postoperative volume of segmental gaps demonstrated a significant difference, with commercial planning having smaller intersegmental gaps than in-house planning (p = 0.0035; Figure 7C).This result could also be confirmed for the length of the intersegmental gap in favor of the commercial planning group (p = 0.0007; Figure 7D).

| Comparison of intersegmental ossification and wound healing disorders
Analysis of postoperative ossification of the reconstruction segments on the CT data sets using a bone union score revealed lower mean score values (MV 1.3, SD 0.55) for the in-house planned mandibular reconstructions compared to the commercially planned mandibular reconstructions (MV 1.51, SD 0.72).However, this difference was not statistically significant (Figure 7E).Postoperative wound healing disorders were noted in both groups (in-house vs. commercial).In the commercially planned mandibular reconstructions (n = 14), postoperative wound healing disorders were documented in six cases, whereas in the in-house planned cases (n = 15), postoperative wound healing disorders were documented in five cases.All wound healing disorders were successfully treated.The difference was not statistically significant (Figure 7F).

| Economic benefit comparison
Both surgical planning solutions (in-house vs. commercial) were compared in terms of their economic advantage.Table 3 provides an overview of the total gross costs and processing times for both procedures.costs and working times for in-house VSP and CAD/CAM fabrication of the surgical guides and customization of the osteosynthesis.A further breakdown of the individual cost components was not possible for the commercial planning solution due to commercial secrets.Statistical comparison of individual costs revealed that the costs of osteosynthesis individualization ( p < 0.0001; Figure 8A), surgical guide production (p < 0.0001; Figure 8B), and anatomical model fabrication ( p < 0.0001; Figure 8C) were significantly lower with in-house planning compared to the commercial planning procedure.Total costs were significantly lower with in-house planning compared to commercial surgical planning ( p < 0.0001; Figure 8D).In addition, the processing time was significantly shorter for the in-house planning group compared to the commercial planning group ( p < 0.0001; Figure 8E).

| DISCUSSION
Destruction of the bony mandible is a milestone in the progression of OSCC and can significantly affect patients' speech, chewing, and swallowing functions, as well as  facial aesthetics and health-related quality of life. 17econstruction of the resected areas of the mandible is a challenging and time-consuming surgical procedure.
Today, it is performed with the aid of preoperative VSP and CAD/CAM-based fabrication of surgical guides for a guided surgical approach. 11This makes the procedure more efficient and the reconstruction result more predictable and precise. 18he surgical planning process is available from commercial vendors worldwide, but can also be implemented in an in-house procedure.In the present study, we compared both surgical planning and implementation solutions (in-house vs. commercial) in terms of postoperative reconstruction accuracy, functional aspects (bone union and wound healing disorders), and economic benefit.Twenty-nine consecutive OSCC patients with an indication for mandibular resection and reconstruction were included, which is a representative cohort for the complexity of this surgical procedure compared to others. 19n a 3D deviation analysis of the VSP and postoperative mandibular reconstruction, we found a more accurate reconstruction result in the commercial planning group than in the in-house group.Graft segment length comparisons showed no significant differences from preoperative VSP to postoperative reconstruction results in either the in-house or commercial planning groups, as previously reported by others. 20,21A comparative analysis of preoperative and postoperative intercondylar distances  showed no significant differences between the two planning solutions, as has been previously published. 20,22,23orrect implementation of the planned osteotomy angles is a critical step in the fit of the reconstruction segments.We found a significant deviation from the preoperative planned to the actual osteotomy angle in the in-house group, while there was no significant difference in the commercial planning group.Other studies have not shared these results. 20,24,25However, Yang and colleagues also reported superiority of the commercial planning solution in terms of correct osteotomy angle implementation. 23A possible explanation for this could be the correct positioning of the surgical guides on the mandible and graft bone, which could be compromised by non-professional in-house fabrication procedures and requires further investigation.
In both groups, we found a significant volume difference between the virtually planned and the postoperative reconstructed bone graft segments, which has never been reported in the literature.A possible explanation could be the positioning of the graft guide over the fibular bone during surgery.While in VSP the guide is typically designed approximately 7-8 cm from the upper ankle joint of the fibula, the guide often has to be positioned more cranially due to a deviated position of the skin perforators, which significantly changes the volume of the transplanted bone segments.This raises the question of whether new high-resolution imaging techniques will allow skin perforators to be incorporated into VSP for more accurate planning and will be the subject of future analysis.
From the data and literature presented here, it is known that the integration of a PSI is associated with more precise mandibular reconstruction compared to individualized plate osteosynthesis.Intersegmental volumes and gap widths are critical in this regard, and we were able to show that they were significantly smaller in the commercial planning group compared to the in-house planning group.Knitschke and colleagues confirmed this finding for intersegmental volumes and observed earlier ossification of the segments, which is critical for adequate graft healing. 26ran et al. also measured smaller gaps in a commercial planning solution that integrated PSIs. 27Compared to the in-house solution, the commercial planning solution offers the advantage of virtually modeling and fabricating a PSI in a laser sintering or milling process that precisely fits the patient's reconstructed mandible.Larger intersegmental gaps, which can interfere with new bone formation and induce fibrotic remodeling, 28 are less likely to occur, eliminating the need for potentially necessary secondary surgical procedures with additional patient burden and increased healthcare costs.Although we did see less intersegmental ossification in the in-house planning group compared to the commercial planning group, this difference was not statistically significant, and we did not see differences in the incidence of postoperative wound healing disorders or the need for secondary surgical procedures.Therefore, a potential long-term economic advantage of the commercial planning solution must be viewed critically at this time.
The economic comparison of the two planning solutions showed that the in-house approach required significantly less processing time than the commercial approach and had a significant cost advantage.Both the manufacturing costs of the individual components used in the reconstruction and the cumulative total costs were significantly lower with the in-house planning than with the commercial planning solution.A comparative costbenefit analysis such as the one presented here has not yet been published.In a prospective study design, Tarsitano and colleagues compared the costs of in-house VSP and guided mandibular reconstruction with traditional freehand reconstruction in 40 consecutive OSCC patients.The authors concluded that the cost of in-house surgical planning did not exceed the total cost of freehand reconstruction. 29In a study of 31 OSCC patients, Rommel et al. described an alternative and cost-effective planning approach using 3D-printed anatomical models of the mandible and fibula that were manually processed and surgical guides fabricated from a moldable composite. 30roduction costs were significantly lower than a control group using commercially available surgical planning solutions. 30These two studies are the only studies in the literature to date on the topic of planning costs in mandibular reconstruction.However, it should be noted that freehand mandibular reconstruction is rarely performed in clinics today, and a planning procedure such as that presented by Rommel et al. is difficult to implement due to changes in Medical Device Act.The regulatory environment for medical devices, such as those used in VSP and guided mandibular reconstruction, plays a critical role in ensuring patient safety and procedural efficacy. 31,32The European Union (EU) and the United States have adopted specific guidelines and regulations for medical devices to ensure strict standards of safety and effectiveness. 33,34These regulations are essential to the approval and use of VSP tools, as they help to standardize and harmonize the quality and performance of these products, 32 but also make it difficult to use in-house planning solutions.
While the use of commercially available VSP and guided surgery approaches are associated with a high degree of reconstructive accuracy and reduced operative time, 8,9 the drawbacks can be seen in the longer fabrication time, which can significantly delay oncologic therapy in patients with advanced tumor stages. 35,36In addition, the primary costs of these procedures can have a significant impact on overall healthcare costs, although the financing of healthcare services can be highly dependent on local taxation, which can lead to significant inequalities between different regions. 37ommercial surgical planning solutions for mandibular reconstruction have been shown to provide significantly more accurate implementation of VSP than in-house solutions.The potential impact of larger intersegmental gap widths in in-house planning procedures with potentially compromised bone healing and the need for second surgical procedures needs to be further evaluated in large clinical trials and case numbers.To date, inhouse solutions offer an economic advantage and can be performed in a shorter time, allowing patients with advanced OSCC to benefit from a faster implementation of their oncological treatment.However, current medical device regulations have made it difficult to use an inhouse planning solutions.

| CONCLUSIONS
In-house surgical planning solutions for mandibular reconstruction are associated with lower postoperative accuracy than commercial planning solutions, but are more economical and faster to implement.It remains to be determined whether patients with advanced cancer will benefit more from a more accurate reconstruction result or from a faster implementation of cancer therapy with a concomitant reduction in healthcare costs.

F I G U R E 1
Illustration of mandibular reconstruction.(A) In-house surgical planning procedure with customized conventional reconstruction plate.(B) Commercial surgical planning with incorporation of a laser sintered PSI.[Color figure can be viewed at wileyonlinelibrary.com]

F
I G U R E 2 Illustration of morphometric analysis of intercondylar distance and bone segment volumes and lengths.(A) Preoperative VSP.(B) Postoperative reconstruction result.[Color figure can be viewed at wileyonlinelibrary.com]F I G U R E 3 Illustration of morphometric evaluation of osteotomy angles.[Color figure can be viewed at wileyonlinelibrary.com]test, the Shapiro-Wilk test, and the Anderson-Darling test.Metric values were compared using the 2-tailed t test.All statistical analyses were performed at a significance level of α = 5% using Prism 9.5 (GraphPad, La Jolla, CA).A p-value less than 0.05 was considered statistically significant.

F
I G U R E 4 Representation of the deviation comparison (with heat map) by superimposing the preoperative VSP and the postoperative reconstruction result.The color of the heatmap between red and green indicates the deviations of the two 3D data sets between 0 (green) and 10 (red) in millimeters.(A) Bone segment analysis.(B) Analysis for the entire mandible.[Color figure can be viewed at wileyonlinelibrary.com]F I G U R E 5 Postoperative CT data set for the evaluation of intersegmental ossification according to a bone union score of 0 to 2. (A) 0 (absent callus formation).(B) 1 (complete callus formation only on the labial or lingual side).(C) 2 (complete callus formation on both the labial and lingual side).[Color figure can be viewed at wileyonlinelibrary.com] 3 | RESULTS

F I G U R E 6
Illustration of the match between the preoperative VSP and postoperative reconstruction result.(A) Volume comparison of individual bone segments, in-house group.(B) Commercial group.(C) Length comparison of individual bone segments, in-house group.(D) Commercial group.(E) Intercondylar distance comparison, in-house group.(F) Commercial group.(G) Osteotomy angle comparison, inhouse group.(H) Commercial group.p-values from 2-tailed Student's t test.[Color figure can be viewed at wileyonlinelibrary.com]

F I G U R E 8
Cost differences for the manufacturing process of the parts used for mandibular reconstruction between in-house and commercial planning solution.(A) Cost of manufacturing the osteosynthesis (individualized osteosynthesis plate vs. PSI).(B) Cost of designing and manufacturing the surgical guides.(C) Cost of manufacturing anatomical models using 3D printing technique.(D) Overall cost.(E) Processing time from design to product finalization or product assembly.p-values from 2-tailed Student's t test.[Color figure can be viewed at wileyonlinelibrary.com]

Table 4
Baseline clinical characteristics of patients.Results of morphometric evaluation of reconstruction accuracy between virtual surgical planning (VSP) and postoperative implementation as well as intersegmental ossification in in-house and commercial planning procedures.
shows theT A B L E 1 T A B L E 4 Cost calculation for in-house VSP, surgical guide production, and customized osteosynthesis.