Key words: 3-dimension print, preoperative plan,pelvic, old fracture, surgery

Abstract

Background:Old fractures of pelvic still are among themost challenging fractures to treat because of complex anatomy, involvedsurgical access to fracture sites and the relatively low incidence of theselesions. Proper evaluation and surgical planning is necessary to achieve pelvicring symmetry and stable fixation of fractures. The goalof this study was to explore the value of 3D printing techniquein surgical management of old pelvic fracture.

Methods:First, sixteen humandry cadaver pelvis study were conducted to confirm the anatomical accuracy of3D printed models produced from radiographic data. Next, a series of nineclinical cases between January 2009 and April 2013 was used for an initialevaluation of surgical reconstruction based on these 3D printing models . Thepelvic injuries were type C in nine patients, average time frominjury to the reconstruction therapy:11 weeks (8 to 17weeks). The workflowincluded following steps: (1) Based on 3D printing technique to make apatient-specific bone model from preoperative computed tomography scans, (2)interactive virtual fracture reduction with printed 3-Dimensional AnatomicTemplate, (3) virtual fracture fixation using K-wire and (4) measurement of osteotomyand implant position relative to landmarks preoperatively according to thevirtually defined deformation. (5) These models served as an aid ofcommunication between surgical team members during the surgical procedure. Thistechnique was validated comparing the preoperative planning and intraoperativeprocedure .

Results:Accuracyof 3D printed models was confirmed to be within specification. From standard CTDICOM data to produce a model taking only 7 hours(6 to 9h).Printed3-Dimensional preoperative planning was feasible in all cases. In 9 casessuperimposition of preoperative planning and postoperative follow-up X-rayinspection showed a good correlation. The patients were followed up 3-29 months(median, 5 months). The fracture healing time was 9-17weeks (mean, 10 weeks). Delayedhealing of incison, wound infection and nonunion of fracture were not obtainedin 9 cases. Aaccording to Majeed score, the results wereexcellent in 2cases, good in 5, and poor in 2.

Conclusion: The presented 3D printing planning techniquefor pelvic surgery was successfully integrated in a clinical workflow toimprove patient-specific preoperative planning, giving visual and hapticinformation about the injury and allowing a patient-specific adaptation ofosteosynthesis implants to the virtually reduced pelvis.

The treatment of old pelvic fracture has always been achallenge for surgeons, because of substantial intraoperative difficulties andpostoperative uncertainties. With few cases and literature reports associated withthe treatment, the old pelvic fracture has no widely accepted and unifiedtreatment strategy. Primary goals for pelvic surgery are restoring integralityand symmetry of the pelvic ring with attention to careful soft tissuemanagement, facilitating rapid postoperative recovery with early rehabilitationand a long-term functioning hip joint^[1]. Proper evaluation andsurgical planning is necessary to achieve these goals. To optimize the outcomes,a wide range of preoperative planning techniques can be applied, some of whichprovide surgeons with an opportunity to practice and refine the plannedprocedure, which could improve efficiency in the operating room, shortenoperative time, and reduce the incidence of iatrogenic complications. Classic2-dimensional imaging often fail when applied to complex pelvic, whereas an advanced3-dimensional informational technology and software interface could push thecapabilities of planning and surgical practice alike. Proprietary hardware,software, and service products can be costly, it is important and necessary tocombine open reduction and internal fixation, along with 3D printingtechnology, to economically and effectively prepare for surgical reconstructionof complex pelvic deformities^[2]. In this report, we describe atechnique that we used to prepare for the surgical reconstruction of deformed pelvis.This technique provided us with an affordable and reproducible, personalized,3D printed template of the pelvic before surgical intervention, and allowed usto practice and refine our surgical approach in the preoperative setting.

Methods

Productionand assessment of the 3D Printing Model

The CT scans of the patient’s pelvic were processed usingopen-source software Blender (DICOM image processingsoftware for Windows XP). The resultant files were uploaded and printed using CANVBOT- D900MN (Figure 1), a commercial company(Guangruide TechnologyCo., Ltd.) providing public access to 3D printing. The 3D print processingchain is as fellow: CT scanning, Blender processing, Mesh Processing and 3D printing.Having performed the file processing ourselves, we used a company based in Beijing: CANVBOT - D900MN.

Sixteen human dry cadaver pelvics study were conducted toconfirm the anatomical accuracy of 3D printed models produced from radiographicdata. A section of a real cadaver pelvic was scanned using CT. A 3Dprinted model was produced of the open section of the pelvic, The modelsproduced were then validated using digital electronic vernier calipers to takemultiple measurements at defined intervals (from the tip of ischial spine to thetip of anterior inferior spine) along the model bones and at the samepoints on the real cadaver pelvic. Two observers measured the same segmentsindependently on three separate occasions (Figure 2). The measurements wereanalysed for a statistically significant difference with a paired Student’st-test.

Patients and Data Acquisition

Between January 2009 and April 2013, 9 patients (7 maleand 2 female) with a median age of 47 years (range: 14 to 51 years) wereprospectively included. According to Tile classification, the pelvic injurieswere type C in nine patients (C3 in four, 2 cases were C2 and had acetabular fracture,2 cases were C1 and had acetabular fracture, 1 case were C3 and had a T-shapedfracture ), Informed consent was obtained from all patients. All patientsunderwent a whole pelvic CT scan (Sensation 64, Siemens Medical Solutions, Forchheim, Germany)onthe day of admission according to standardized trauma protocol. Near-isotropicaxialoriented CT images with a slice thickness of 1 mm were reconstructed usinga bone kernel for sharp depiction of bone fragment edges. Data were transferredto a picture archiving and communication system, according to the 3Dprint processing workflow, we produced model for operativeapplication.

Surgicalreconstruction guided by 3D printed model preoperative planning andintraoperative navigation

The virtual mobile fragments of the model were reducedvia translation and rotation operations to change position and alignment, andthe reduction was assessed for “best fit” apposition and restoration of theanatomic architecture. Surgery was simulated with the 3D print template. Once asatisfactory correction was achieved in the skeletal geometry, measurements ofthe shape of the elements to be resected were analyzed for subsequent 3D templatingand in vivo surgical application. In so doing, we were able use the virtualskeletal model to determine the position and angulation of osteotomy requiredto create the optimal wedge resection geometry necessary to surgically correctthe deformed pelvic, before ever taking the patient to the operating room.

Printed anatomic templates were used to select the optimalincision placement, instrumentation, osteotomy and placement of internalfixation devices. Kirschner wire axis guides were inserted to replicate the apicesof the resection geometry identified. Osteotomies were performed from the anterioror posterior approaches, directed toward the central axis guidewires, asdetermined by the measured landmarks. Once simulated surgical reduction wasreproduced on the physical anatomic template, osteosynthesis plating system wasassembled with reference to the printed template. After the optimal platingconstruct had been determined, the internal fixation implant was sterilized andstored for subsequent application to the patient intraoperatively. Afterpracticing on the printed template models of the patient’s pelvic, the patientwas subsequently taken to the operating room where the surgical plan was executed.The approach that had been decided upon was marked on the involved pelvic withthe aid of fluoroscopic image intensification localization of the underlying boneand joint structures. Dissection was then carried out to expose the malunionand nounion site. Once the deformity were exposed, the planned osteotomy wasmarked, once again with the aid of fluoroscopic guidance. Axis guide pins,consisting of 2.0-mm Kirschner wires, were placed at the distal and proximal apicesof the planned osteotomy, in accordance with the information gleaned from thepreoperative computer simulations and physical bone model manipulations.Thereafter, the planned osteotomy was performed and the anterior ringreduced into anatomic alignment in all 3 planes relative to the posterior ring.The osteotomy was refined by means of reciprocal planning to ensure goodbone-to-bone contact, after which interfragmental compression screw fixation wasapplied across the osteotomy interface. After placement of the internalfixation screws, which were aimed at stabilizing the reduced osteotomy, thepreviously constructed and sterilized plate was applied.

Evaluation

We recorded The P values with a pairedStudent’s t-test for measurement to assess the anatomical accuracy of 3Dprinted models. Time needed for producing 3D printing model (building thepatient-specific models from CT datasets, 3D printing) , as well as the fractureunion time was measured. In all patients a follow-up CT was performed 2 to 4days after surgery. Hip function was determined according to Majeed score.Qualitative visual analysis of the accuracy of internal fixation was done bymeans of hybrid renderings of the postoperative CT and respective preoperativeplanning, after manually registering the pelvic bones into the same space.Placement of osteosynthesis implants was then compared on these renderings.

Results

TheP values (P = 0.7943) showed no statistical difference in dimensions betweenthe printed models and the original cadaver pelvic, so accuracy of 3D printedmodels was confirmed to be within specification. From standard CT DICOM data toproduce a model taking only 7 hours (6 to 9h) . Printed 3-Dimensional preoperativeplanning was feasible in all cases. In 9 cases, selected surgical approach, osteotomyand placement of the fixation plate on the pelvic ring shows a very good match betweenplanning and actual execution, comparing the postoperative follow-up CT scansto respective preoperative planning, a good correlation was found. Additionally,anatomical landmark based measurements were helpful for intraoperativenavigation. The patients were followed up 3-29 months (median, 5 months). Thefracture healing time was 9-17weeks (mean, 10-7 weeks). Delayed healing ofincison, wound infection and nonunion of fracture were not obtained in 9 cases.At last follow-up, according to Majeed score, the results were excellent in 2cases,good in 5, and poor in 2.

Illustrative Case

A 46- year-old female patient, 8 months after the caraccident injury. The CT scan showed a vertical displaced fracture involving theiliac blade starting 3 cm below the iliac crest and extending forward, reachingup to the acetabular roof and triradiate cartilage, involving both anterior andposterior columns. There was a mild protrusion of the femoral head and the fractureline extension was present till the superior pubic rami (Figure 3A, B, C).

The preoperative planning before surgery of the pelviccomprised sequential steps: 3-D reconstruction and segmentation of CT scandata, surgical simulation, template design, sizing and alignment of the implantand production of the templates using the 3D printing technology [Figure 3 D].CT scanning of all sections was done with 1-mm-thick slices. For thepreoperative planning process, template was used to contour a 3.5-mm-thick reconstructionplate. The screw sizes were determined preoperatively and the position of theplate and holes was also decided and marked with indelible ink on the 3D model[Figure 3 E,F]. An ilioinguinal approach was used for anteriorly exposing thefracture site. The total surgical time required was 4 h 10 min. Of this, theinstrumentation took only 20 min. The blood loss during the procedure was 1000ml and the patient was transfused six unit of whole blood.

Discussion

Post pelvic fracture deformity is a commonproblem, the correction of deformity for old pelvic fractures is very difficult,because on the one hand the multiple areas around fracture line of pelvic ringneed to be reopened or osteotomized, on the other hand, severe contracturedsoft tissue requires extensive lysis. Many nerves and blood vessels and otherimportant structure surrounding pelvic area further increases the operationdifficulty and treatment risk. Due to lack of sufficient cases and the difficultyin treating this kind of injury, the surgical outcome is not very satisfactory.Kanakaris et al did a collective analysis of 25 papers on old fracture ofpelvis between 1965 and 2008, with a total of 437 cases of pelvic malunion ornonunion cases, the results of the total healing rate was 86.1%, the pain reliefrate was 93%, with a satisfaction rate of 79%, only 50% of the patientsreturned to their pre- injury activity level; there was a higher incidence ofcomplications, including 5.3% nerve injury, deep venous thrombosis 5%, 1.9%pulmonary embolism and 1.6% deep wound infection. Post pelvicfracture deformity usually is a complex three-dimensional deformity, includingrotation deformity and shortening deformity in different planes, threedimensional anatomy of the pelvis is difficult to understand .In order to fullyunderstand the deformity, routine imaging exams including anteroposterior,inlets/outlets X-ray exams, CT scan and 3D CT reconstruction，but for old pelvic fractures of complicatedthree-dimensional deformity, these conventional imaging may have been unable tomeet the need for preoperative planning. It is obvious that strict preoperativeplanning is a crucial step in pelvic surgery. Therefore, 3D printing technologyhas been developed as the application of computer based technology to assistthe surgeon to improve the precision of the operative procedure. Weused 3D printing technology assisted preoperative planning for old pelvicfractures from 2007, this paper introduces our initial results of this clinicalapplication.

3D printing technology provides more choices for doctors. Basedon imaging examination data, 3D printing technology can provide for the doctorsthe real sized model, making the preoperative planning more direct andaccurate. Since 1990s, 3D model has been used for preoperative planning of complexcranio maxillofacial surgery. In addition, there are reports of the3D printing technology used for intracranial aneurysm, livertransplantation of living donor in preoperative planning. Therewere also reports on orthopedic surgery using 3D printing technology. Cimermanet al. introduced a surgical planning software for pelvic and acetabularfractures with a mouse-based CAD-style interface. Hurson had 20cases of acetabular fracture. He obtained 3D printed model based on 3D CT data,comparing to the traditional imaging examination (anteroposterior, Judetoblique, CT), the consistency of acetabular fracture classification based onthe 3D printing model has increased. It indicates that this technology ishelpful for doctors to understand the morphological characteristic of complexfractures. Despite the rapid advances in the operativeapplication of 3D printing technique in the lastyears, surgical simulation and planning based on 3D printing technique israrely used in clinical routine. There are different reasons for the slowadoption of such technologies. One important factor may be thereservation of surgeons to explore new technologies as they are devoted totheir technical skills and performance，the main problem of the program was segmentation process,which should be done by computer engineers and controlled by radiologists.The other reason is that the proprietary nature of these solutions can resultin a product that is quite costly, and third party payers are unlikely toreimburse this expense; therefore the cost would have to be passed on topatients, care facilities, or even absorbed by the surgeon, and application ofthis type of technology outside of the research team is limited. In 2009 wedeveloped together with computer engineers from Guangruide Inc. an experimentalcomputer program which enables performance of virtual operation of injuredpelvis, the new software where all the steps can be performed by a surgeon .Thepurpose of the software was to perform all the steps of 3D printing procedureon standard PC computer. Combined application of open-source software, freewaresoftware, proprietary software and the hardware (the printer used tomanufacture the physical model of pelvic). the method that we have described canbe undertaken with minimal costs in financial terms. All the steps can becarried out on a regular personal computer by the surgeon who is doing thepreoperative planning. This is the complete novelty since segmentation can becarried out by the surgeon. In that way all the fracture lines are studied in3D during segmentation process. The procedure is quick and easy. Now we usethis program routinely in all pelvic and in most difficult articular fracturesas well. We also educated three surgeons to use the program independently. Thelearning time for the education was one day course composed of theoretical and practicalexercises.

In this study, the planned fracturefixation was followed completely in nine cases with an acceptable hip functionaccording to Majeed score in all cases. In the cases a good correlation betweenpreoperative planning and respective postoperative follow-up CT scans wasfound. Approach, osteotomy and placement of the fixation plate on the pelvic ring shows a very good match between planning and actual execution, We found no serious postoperative complications such as deep infections or failure of osteosynthesis implants. Physical 3D models produced through 3D Printing can besuccessfully applied pre-operatively to determine the amount, position and extent offractures,determine the level and direction of the osteotomy siteand the osteotomy line. plan the surgical procedure and determine whether only an anterior or both posterior andanterior surgical entries would be required.These models can also be successfullyapplied during the surgical procedure to serve as an aid of communication betweensurgical team members, serve as an aid of orientation relative to patient anatomy, assist in bending reconstruction plates. Theapplication of the 3D print model resulted in a definite reduction of surgery time,reduced fatigue of thesurgeons, and reduced blood loss.

The limitation of this study is the limited number of patients and the absence of long-term follow-up, Also due to the variability of injury patterns, it is difficult to make definite quantitative conclusions. Thisstudy therefore only is able to show initial experiences and a larger patient population is requested to further assess the presented tool.

In preoperativeplanning process for old pelvic fractures, 3D printing technology is a usefulsupplement to conventional imaging examination. It can help the operating doctorto understand the complicated deformity of pelvic fracture by osteotomy simulationoperation and fixation, to understand the feasibility of operation scheme,shorten the operation time, improve the final fracture reduction and fixationeffect, and ultimately to improve the operation effect of patients.

  参考文献（略）