Abstract
BACKGROUND: The aims of the total knee arthroplasty (TKA) are to restore accurately the position of components in three dimensions and the alignment of lower limb, to maintain the ligament balance of knee and to avoid the femoropatellar complications, to achieve a pain free, stable, good functional and long durable prosthesis.
OBJECTIVE: To investigate the optimization of Bone Morphine(r) system for proper restoration of lower limb mechanical axis alignment, rotational position of components and ligament balance in TKA by comparison of the initial outcomes between the computer-navigated and conventional procedures.
DESIGN: A prospective clinical comparative study.
SETTING: Department of Orthopaedic and Traumatic Surgery, Henri Mondor Hospital, France.
PARTICIPANTS: Between November 2002 and June 2003 in Henri Mondor Hospital of France, 21 patients including 14 varus knees and 7 valgus knees, aged from 64 to 79 years, were operated by an experienced senior surgeon with TKA using a Bone Morphing based image-free Ceravision navigation system (navigated group). Twenty patients, including 15 varus knees and 5 valgus knees, aged from 65 to 83 years, were operated by the same surgeon with the same type of prosthesis using conventional technique (conventional group). All patients suffered from gonarthrosis received the primary arthroplasty were enrolled in the study. The reversion cases were excluded. All patients were informed with the intervention and agreed to be operated.
METHODS: All patients underwent TKA using posterior stabilized total knee prosthesis (Hermes(r) Ceraver, France) by the basically same procedure. In the navigated group, TKA was performed under the monitoring of Bone Morphing based image-free computer-assisted Ceravision(r) system.
MAIN OUTCOME MEASURES: To compare the preoperative, intra-operative and postoperative relative image data in navigation group, and to analyze the postoperative alignment and ligament balance, and to assess the range of motion (ROM), frontal laxity, stability and patella complication of the operated knees.
RESULTS: All the patients of the two groups were included in this clinical study and analysis. All femoral, tibial and patella components were implanted in satisfactory position in all patients. The Bone Morphing system showed the accuracy of components alignment and ligaments balance. Lower limb alignments deviated within 3° varus and 3° valgus in frontal view. There was no significant difference between the two groups (P ＞ 0.05). Three months postoperative clinical check-up showed the average ROM was 115°in the navigated group, 109.4°in the conventional group. There was no significant difference between the two groups (P =0.06). Frontal laxity in the navigated group was found with internal laxity 0.27(0.2-0.5) mm/ external laxity 1.7(1.0-2.5) mm, in the conventional group was internal laxity 0.27(0.1-0.5) mm/ external laxity 2.23(1.0-3.0) mm. There was a significant difference between the two groups (P =0.03). There were no knee instability and patella complications.
CONCLUSION: Bone Morphing navigated system allows for optimization of lower axis alignment and prosthesis rotational position as well as ligament balance in TKA, more ideal frontal laxity and ligament balance impacted on initial clinical outcome.

INTRODUCTION

Since the introduction of computer-assisted orthopedic surgery (CAOS) in total knee arthroplasty (TKA), more and more comparative reports have showed the advantages of this technique, and the superior to conventional TKA not only in accurate alignment of the lower limb and osteotomy, components position, but also the assessment of ligament balance and knee kinematics[1-4]. Whether a Bone Morphing based image-free computer-assisted Ceravision(r) system can improve lower limb axis alignment, prosthesis rotational position, and ligament balance to obtain good clinical initial outcome is the objective of current comparative study.

SUBJECTS AND METHODS

Subjects
Between November 2002 and June 2003 in Henri Mondor Hospital of France, a group of 21 subjects were operated via primary TKA in 21 knees by an experienced senior surgeon using an image-free module (CT-free Knee 1.0, based on "Bone morphing") of Ceravision System (Ceraver, France), with posterior stabilized total knee prosthesis (Hermes(r) Ceraver, France), fixed polyethylene component and resurfaced patella, all components were cemented. A group of 20 subjects were operated via primary TKA in 20 knees using a conventional technique, performed by the same surgeon, and implanted the same type of prosthesis. All patients suffered from gonarthrosis failed from the medical treatment. All revision cases were excluded. Written informed consent was obtained from each patient.
In the navigated group, 21 patients included 5 men and 16 women with an average age of 72.4 (range: 64-79) years. In conventional group, 20 patients included 6 men and 14 women with an average age of 73 (range: 65-83) years. Preoperatively, in the navigated group, there were 14 varus knees (65%) and 7 valgus knee（35%）; In the conventional group, there were 15 varus knees（69%）and 5 valgus

knee（31%）. Statistic analysis did not show significant difference (P > 0.05).

Methods
Surgical technique
Ceravision system consisted of several hardwares as follows: ① a main station equipped with a passive optical sensor, located in the side of contra-lateral knee, 190 cm from the operative field, two boom-mounted infrared cameras were able to detect the infrared light reflecting from the passive sterile and disposable markers attached to the rigid bodies fixed to bones and instruments during the operation, without any connecting line with operative knee, transferring the signal to the computer station set up the software previously with the patient's name, date of birth and operating side of knee entered; ② an F shape rigid body with three markers attached to the femur; ③ a T shape rigid body with three markers attached to the tibia; ④ a P shape probe equipped with two faces of three markers, with a 2-mm diameter spherical tip that slided smoothly on bone and cartilage surfaces; ⑤ a G shape rigid body (Guide) equipped with two faces of three markers attached to cutting blocks during the navigation.
For all cases of primary TKA, anterior approach was selected. After standard knee exposure, osteophytes should be removed to prevent impingement and tethering over the capsule and ligament. Two sets of rigid body F and T with passive markers attached were fixed on the anteromedial aspect of femoral and tibial diaphysis separately, at the level of 6-8 cm from the joint line. The probe and trackers were initialized. The surgeon guided the navigated procedure, using the probe to collect the points of anatomical landmarks from knee and ankle so as to compute the knee and ankle centers. The surgeon made a circular motion with the lower limb (knee in full extension), a special algorithm of the customized software calculated the center of rotation of the hip, so as to create and show the lower limb alignment. Followed by a surface matching process in which the surgeon had to digitize up to 800-1 200 points of free choice on the bone and cartilage surface by sliding a probe over the tibial plateaus and the femoral condyles separately, the Bone Morphing algorithm would then compensate for the radius of the sphere, to reconstruct the exact surface structure of the operating knee that was specific for the patient anatomy. The reconstructed three-dimensional model and relative data showed on the screen in real time, the surgeon modified the operation via a menu and inputted the going forward and backward command using foot pedal or touching screen.
After the acquisition of the geometric and morphologic data, the computer provided the surgeon for reference and modification with a preplanning of the tibial cut generated by the system automatically. Bone cutting blocks equipped the passive markers were positioned and orientated in real time visualization on the display of the system, allowing the surgeon to check up all planes of each cut, a fine-tuning of resection planes and component orientation could be performed. After the tibial cutting, ligament balancing commenced via appropriate medial/lateral releases and components rotational position based on preoperative deformity. The surgeon modified the femoral planning suggested by the system and performed the femoral cut with the help of the cutting block, then checked up the results of implantation when the component trials implanted in place. Optimization of TKA strategy and criteria included lower limb alignment deviation within (180±3)°, rectangular symmetric extension/flexion gaps, (3-6)° external rotation position, frontal laxity within ±3° using varus/valgus stress test. All the selected components were positioned and cemented. All data, steps and results of the procedure were saved as record in a CD-Rom for patient reporting and further investigations.
Most navigated performance was similar to the conventional procedure. While the navigation system provided the surgeon with precise parameters and three-dimensional reconstructed images data for reference to perform the surgical planning, tibial cut, ligament balance and femoral cut, enabled the surgeon to control the procedure and to optimize the results step by step.

Measurement of image data
①Standard post-anterior and profile lateral knee X-ray films.② Weight charged knee flexion post-anterior view X-ray film. ③ Patellar axis view (sky view) (30°, 45°, 60°). ④Pre-operative full-length of lower limb radiograph (laying position). ⑤ Pre-operative CT scan for examination of hip knee and ankle torsion. ⑥Varus-valgus stress X-ray films under general anesthesia.

CD Rom data from Ceravision system
① Intra-operative lower limb mechanical axis alignment. ②Varus-valgus stress test. ③Post-prosthetic frontal laxity and rotational alignment.

Measurement of post-operative image data
① Standard post-anterior and lateral knee X-ray films. ②Full-length of lower limb radiograph. ③ Patellar axis view (sky view). ④Post-operative CT scan for examination of hip, ankle and knee (femoral and tibial) prosthetic components torsion.

Follow up and data analyses
Three months post-operative clinical check-up results were set for the initial outcome. Measurement for the range of motion (ROM) mobility and frontal laxity of knee was performed Statistical comparisons were conducted with a Test de Sphéricité de Bartlet, using a software of StatView 5.0. P < 0.05 was regarded as being statistically significant.

RESULTS

Quantitative analysis of the participants
All patients of the two groups were included in this clinical study and analysis. All femoral, tibial and patella components were implanted in satisfactory position in all patients.

Comparison of postoperative lower limb alignment deviation between the two groups
In the navigated group (21 knees), 14 knees were with varus (65%), and lower limb alignment deviated with valgus 0.16°(varus1°-valgus 2°) postoperatively; In the conventional group (20 knees), 15 knees were with varus (69%), and lower limb alignment deviated with varus 0.9°(varus 3.5°- valgus 3°) postoperatively. There was no significant deference (P > 0.05). In the navigated group (21 knees), 7 knees were with valgus (35%), and lower limb alignment deviated with valgus 0.14°(varus 2°-valgus 3°) postoperatively. In the conventional group (20 knees), 5 knees were with valgus (31%), and limb alignment deviated with valgus 0.25°(varus 1°-valgus 2°) postoperatively. There was no significant deference (P > 0.05).

Evaluation of short-term therapeutic effects
Mean flexion angle was measured as the range of motion (ROM) in the navigated group, 115°(range: 105°-130°), in the conventional group 109.4°(range: 90°-130°). The difference was not significant (P =0.06); Frontal laxity check-up showed internal laxity 0.27 (range: 0.2-0.5) cm/ external laxity1.7 (range: 1-2.5) cm in the navigated group, while internal laxity 0.27 (range: 0.1-0.5) cm /external laxity 2.23 (range: 1.0-3.0) cm in the conventional group. The difference was significant (P =0.03).

Adverse reactions and side effects
In the both groups, there were no any patella instability and other complications influencing the clinical outcomes observed.

DISCUSSION

To obtain successful midterm and long term outcome of TKA surgery, there are several important factors, including proper selected indications, appropriate implant alternative sizing, correct surgical technique, and effective peri-operative care [5-6]. It was reported that 5%-8% failures related to loosening and instability, 20%-40% of the cases with femoropatellar complications led to anterior knee pain and limited knee flexion-extension motion, moreover, the early revision was due to malalignment, malposition and instability [5-6]. The influences of TKA outcome are particularly accurate surgical techniques. It is accepted that ① ideal lower limb mechanical axis alignment was deviation between ±3° in frontal view [6-7], ②3°-6° external rotational alignment of the posterior axis, parallel to epicondylar axis, in order to achieve symmetric (equivalent) flexion-extension gap [6,8-10]. Mechanical alignment systems have fundamental problems that limit their ultimate accuracy. It is difficult to determine accurately, the correct location of crucial alignment landmarks (e.g., the center of the femoral head, the center of the ankle) using conventional techniques with intramedullary and extramedullary alignment instrumentation, even for experienced surgeons. Because of the complex knee structure and the limitation of the visible operative field, it is difficult to evaluate with the eyesight and the hand sense, controlling checked up with standard X ray film, so that it influences the accuracy of operation [11].
Since 1990s, with the development of autonomous control systems and digital medical image techniques, computer assisted surgery has been well developed. The high accuracy and good security of digital information, based on the principle, several systems were designed and used to guide the interventions or to visualize the invisible and indirect access structures [12]. It provides surgeons a system to help to perform operation accurately and mini-invasively. It enables surgeons to imitate the surgical process. And it offers some user-friendly systems to surgeons and is available to use easily. The system allows a precision to the order of 1 degree for angles and 1 mm for lengths, and it helps surgeons to achieve ideal operative outcomes [3,13].

Accuracy of system for lower limb axis alignment
Proper axial alignment is important for longevity of the implant. Minor malalignment can lead to early loosening, increased polyethylene wearing out and poor function. Maquet's line passes from the center of the femoral head to the center of the body of the talus. Jeffrey et al analyzed the outcome after TKA in 115 patients. They found a rate of 24% of prosthetic loosening when the mechanical axis exceeded ±3° varus/valgus deviation, while it was only 3% in those patients with an axis within a range of ±3°. They were assessed clinically and long-leg standing radiographs were taken before operation, soon after surgery and up to 12 years later. So accurate coronal alignment appears to be an important factor in prevention of loosening [7]. Some authors evaluated the alignment with HKA angle, where the normal was 180°, valgus knee was over 180°, while varus knee was under 180°[6]. In China, full leg X-film can not be taken into account ordinarily, moreover, it can not be used into the operation, so with Bone Morphing system, like Surgetics Station (Praxim, Grenoble, France), Ceravision (Ceraver, France) and VectorVision (CT-free Knee, BrainLAB, Munich, Germany) system [3,6,14], it is not needed to perform these image examinations and it can save the cost. In our previous and the present study, it showed that there was no significant difference between the lower limb alignment no matter from the system or from the long leg X-film, but the former seemed much more accurate [14].

Efficiency for symmetry flexion-extension gap and ligament balance
Ligament balance issue is also one of the most essential factors when performing TKA. In operation, settings ligament balancing requires attention to flexion-extension gap balancing and adjusting the asymmetry of the gaps. The term ligament balancing at primary and revision total knee arthroplasty can have two different or even combined meanings. Primarily, the term refers to balancing or adjusting the collateral ligaments in patients in whom varus or valgus deformities are present. Ligament balancing also may be used to regard the creation of balanced flexion and extension gaps [15]. We took varus-valgus stress X-ray films assisted by a radiological technician for the patients under general anesthesia pre-operatively, to measure the difference between the medial and lateral gaps of knees, to gain the value of the frontal laxity, which helped us to make and modify the decision of bone cutting and ligament releasing. After positioning the components trials, the frontal laxity should also be paid attention to restore a stable prosthetic knee, 3° deviation of axis is acceptable when examination performed in the position of extension and flexion during the operation. As our experience, Ceravision as a CT-free module allows intra-operative visualization of lower axis alignment, ligament balancing and knee kinematics, it keeps the components alignment deviation within 3° varus and 3° valgus. Three months post-operative clinical check-up for the initial result, navigated group seemed more stable and movable than conventional group, without femoropatellar complications in all patients.

Accuracy for components rotational alignment
As a study of Hofmann et al [16], the rotational alignment of the tibial and femoral components played an important role in modern total knee replacement surgery. After correcting frontal alignment and proper soft tissue balancing, the rotational placement of the components represents the "third dimension" in TKA surgery. Improved surgical techniques with modified instruments and better rotational component positioning would lead to better functional outcomes. As the study of Kaper et al [17], 3° rotational alignment of femoral component was better for the flexion and extension gap and the petallar kinemics, according to Akagi et al study, 3°-6° rotational alignment of femoral component to the posterior axis was better for non valgas/varus knee. Rotation of the femoral component was usually determined with the posterior condylar axis [10].
Olcott et al [9] studied a total of 100 consecutive posterior cruciate-retaining TKA, the femoral alignment necessary to create a rectangular flexion gap was determined and compared with Whiteside's line, the transepicondylar axis, and a line in 3 ° of external rotation relative to the posterior condyles of the femur. The transepicondylar axis most consistently recreated a balanced flexion space, whereas 3 ° of external rotation off the posterior condyles was least consistent, in particular in valgus knees. The transepicondylar axis was most consistent (90% of knees within 3° of flexion gap symmetry) followed by the Whiteside axis (83%) and posterior condylar axis (70%). Despite its consistency, the transepicondylar axis would not yield flexion gap symmetry within 3° in 10% of knees preoperatively in neutral or varus, and 14% of knees in valgus.
The Bone Morphing technique can provide more information of the individual morphologic data, especially in knee deformation in great varus/valgus, so the 3° to 6 ° external rotational position of femur and tibia components is easy to achieve properly, regarding a rectangular symmetrical flexion-extension gap and ligament balance. The navigation system uses an average algorithm between the determined angle of Whiteside's line and the transepicondylar axis [2]. The post-operative CT examination validated the proper components rotational alignment [14].

Efficiency of Bone Morphing system for optimization of TKA
Sharkey et al [5] studied 212 surgeries done on 203 patients (nine patients had bilateral surgeries). The reasons for failure listed in order of prevalence among the patients include polyethylene wear, aseptic loosening, instability, infection, arthrofibrosis, malalignment or malposition, deficient extensor mechanism, and so on. Fifty percent of early revision TKA in the series was related to instability, malalignment or malposition, and failure of fixation. They hoped to solve these technique problems by using the computer-assisted system.
Since the introduction of the computer assisted systems in TKA, many reports on initial clinical outcomes emerged [1-4, 6, 13, 18-21]. Bone Morphing is an accurate, fast, and user-friendly method that can provide morphologic and geometric data. These two aspects of the data are actually linked to improve the global safety of the system. Considering one of the key factors influencing the success of TKA, the soft-tissue balancing involved in the plan is the absolute priority of the system, allowing a good biomechanical and anatomic result for TKA procedure. The image-free navigation systems have many advantages. Compared to techniques using knee landmarks and kinematics data, the Bone Morphing technique offers significant improvement, because it enables the surgeon to plan a real and global tradeoff, taking into account all morphologic and functional parameters, including ligament balancing, accurate mechanical axis alignment and dynamics visualization of the knee. Compared to CT-based approaches, it is more accurate, without suffering from the standard errors of CT-based methods, achieving better management of ligament balance, and does not need the scans of the hip, knee and ankle avoiding the radiation dose to the patient, etc [6]. It is not only a navigation tool, but a measurement tool as well. It enables surgeon to perform the procedure more precisely, by imitating and showing the planning and result, optimizing and controlling each step of the performance, to realize the human-machine interaction via the computer interface, as well as to obtain abundant data for results evaluation and experience accumulation postoperatively. In our study, axis alignment and rotational position data from the Ceravision system and from full leg film and CT scan, they have coincidence [14].

Interaction, clinic experience, learn curve and evolution
In the current study of 21 knees operated with TKA using Ceravision system, we analyzed image data, pre-operative full-length radiograph (laying position), varus-valgus stress radiographs under general anesthesia, pre-operative CT scan for examination of hip knee and ankle torsion, weight charged knee flexion post-anterior X-ray film, patellar axis view; data stored in CD Rom of Ceravision system, intraoperative lower limb alignment, three-dimensional components positioning, including rotational alignment, sizing, achieving rectangular symmetry flexion-extension gap and bilateral ligaments balancing, controlled by the system intra-operatively, we found that it was more accurate than conventional eye-hand with X ray film techniques [14]. All the performance and decision made during the operation were affected by the experience and flexibility of the operator, it presented via the interaction between the surgeon and the computer. Because it was the beginning stage to perform the TKA using Ceravision system and for accumulation clinical experiences, so many image data were involved in this study, to validate the system on the one side and to do the relative research on the other side. How to choose the image data depends on each situation and each operator. As a new technique, it must have a learn curve. After a period practice, one can adapt, optimize this system and make progress [4,11]. The long-term results with this technique remain to be determined.
Conclusion
The utilization of image-free Bone Morphing system for TKA promotes high accuracy of lower limb alignment, component rotational alignment, flexion-extension gap and ligament balance, to achieve satisfactory initial outcome of TKA.

REFERENCES

1 Sparmann M, Wolke B, Czupalla H, et al. Positioning of total knee arthroplasty with and without navigation support: A prospective, randomised study. J Bone Joint Surg Br 2003;85(6):830-835
2 St?ckl B, Nogler M, Rosiek R, et al. Navigation improves accuracy of rotational alignment in total knee arthroplasty. Clin Orthop Relat Res 2004;(426):180-186
3 B?this H, Perlick L, Tingart M, et al. Alignment in total knee arthroplasty. A comparison of computer-assisted surgery with the conventional technique. J Bone Joint Surg Br 2004;86(5):682-687
4 B?this H, Perlick L, Tingart M, et al. Radiological results of image-based and non-image-based computer-assisted total knee arthroplasty. Int Orthop 2004;28(2):87-90
5 Sharkey PF, Hozack WJ, Rothman RH, et al. Insall Award paper. Why are total knee arthroplasties failing today? Clin Orthop Relat Res 2002;(404):7-13
6 Stindel E, Briard JL, Merloz P, et al. Bone morphing: 3D morphological data for total knee arthroplasty. Comput Aided Surg 2002;7(3):156-168
7 Jeffery RS, Morris RW, Denham RA. Coronal alignment after total knee replacement. J Bone Joint Surg Br 1991;73(5):709-714
8 Barrack RL, Schrader T, Bertot AJ, et al. Component rotation and anterior knee pain after total knee arthroplasty. Clin Orthop Relat Res 2001;(392):46-55
9 Olcott CW, Scott RD. The Ranawat Award. Femoral component rotation during total knee arthroplasty. Clin Orthop Relat Res 1999;(367):39-42
10 Akagi M, Yamashita E, Nakagawa T, et al. Relationship between frontal knee alignment and reference axes in the distal femur. Clin Orthop Relat Res 2001;(388):147-156
11 Stulberg SD, Loan P, Sarin V. Computer-assisted navigation in total knee replacement: results of an initial experience in thirty-five patients. J Bone Joint Surg Am 2002;84-A Suppl 2:90-98
12 Merloz P. Concepts de base pour la chirurgie orthopédique assistée par ordinateur// Merloz P, Chirurgie orthopédique assistée par ordinateur. Cahiers d'Enseignement de la SOFCOT, N°80. Paris, Elsevier 2002:1-10
13 Jenny JY, Boeri C. Computer-assisted implantation of a total knee arthroplasty: a case-controlled study in comparison with classical instrumentation. Rev Chir Orthop Reparatrice Appar Mot 2001;87(7): 645-652
14 Wu H,Qiu WJ, Van Driessche Stéphane, et al. Role of computer-assisted relative digital image data in optimizing artificial total knee arthroplasty. Zhongguo Linchuang Kangfu 2005;9(22):1-3
15 Krackow KA. Revision total knee replacement ligament balancing for deformity. Clin Orthop Relat Res 2002;(404):152-157
16 Hofmann S, Romero J, Roth-Schiffl E, et al. Rotational malalignment of the components may cause chronic pain or early failure in total knee arthroplasty. Orthopade 2003;32(6):469-476
17 Kaper BP, Woolfrey M, Bourne RB. The effect of built-in external femoral rotation on patellofemoral tracking in the genesis II total knee arthroplasty. J Arthroplasty 2000;15(8):964-969
18 Saragaglia D, Picard F, Chaussard C, et al. Computer-assisted knee arthroplasty: comparison with a conventional procedure. Results of 50 cases in a prospective randomized study. Rev Chir Orthop Reparatrice Appar Mot 2001;87(1):18-28
19 Chauhan SK, Scott RG, Breidahl W, et al. Computer-assisted knee arthroplasty versus a conventional jig-based technique: A randomized, prospective trial. J Bone Joint Surg Br 2004;86(3):372-377
20 Matsumoto T, Tsumura N, Kurosaka M, et al. Prosthetic alignment and sizing in computer-assisted total knee arthroplasty. Int Orthop 2004;28(5):282-285
21 Nizard RS, Porcher R, Ravaud P, et al. Use of the Cusum technique for evaluation of a CT-based navigation system for total knee replacement. Clin Orthop Relat Res 2004;(425):180-188

三维骨建模系统在人工全膝关节置换中的优化作用:导航与常规手术早期
疗效比较○

吴 昊1，Stéphane Van Driessche2，Daniel Goutallier2
1广西壮族自治区人民医院骨科，广西壮族自治区南宁市 530021； 2法国亨利蒙多医院矫形与创伤外科，法国 94010
吴 昊，男，1963年生，广西壮族自治区南宁市人，汉族，1985年广西医科大学毕业，主任医师，主要从事人工关节置换与计算机辅助外科研究。
摘要
背景：人工全膝关节置换手术的重点是准确的假体三维定位，重建良好的下肢力线，维持膝关节韧带的平衡，避免髌股关节并发症，从而取得一个无痛稳定、功能良好和持久耐用的关节。
目的：通过对比导航和常规全膝关节置换的近期效果，讨论三维骨建模系统对人工全膝关节置换手术中精确重建下肢力线、旋转对位和韧带平衡的优化作用。
设计：分组对比观察。
单位：法国亨利蒙多医院矫形与创伤外科。
对象：选择2002-11/2003-06在法国亨利蒙多医院矫形与创伤外科进行的计算机辅助人工膝关节置换21例为导航手术组，年龄64~79岁，14例膝内翻，7例膝外翻；常规手术人工膝关节置换20例传统手术组，年龄65~83岁，15例膝内翻，5例膝外翻；纳入标准均为的骨性关节炎患者，排除复治的人工全膝关节翻修术患者，患者均知情同意。
方法：均采用后稳定型人工表面全膝关节（Hermes(r)法国Ceraver），两组的手术基本操作相同，导航手术组在Ceravision (r)无需影像资料的三维骨建模（Bone-Morphing）计算机辅助系统监控下进行。
主要观察指标：对比术前、术中和术后的相关影像资料，分析下肢力线重建和韧带平衡的结果；检查术后3个月手术膝关节的活动度、额面松弛度和髌骨稳定性。
结果：41例患者均进入结果分析。①所有患者都获得人工膝关节胫、股骨假体的满意对位植入和韧带平衡。②Ceravision系统对下肢力线的测量及在膝内外翻应力下的测量均比X射线片更精确；两组病例的下肢力线都在内外翻3°的范围内（P > 0.05）。③两组术后3个月的关节活动度比较差异不显著（P =0.06）；导航手术组膝关节额面松弛度优于常规手术组（P =0.03）。④两组均无髌骨失稳和脱位等并发症。
结论：应用以三维骨建模为基础的计算机辅助手术系统，优化了人工全膝关节置换，精确实现了截骨和三维对位，获得良好的膝关节屈伸位下关节等距间隙，保证良好的膝关节韧带张力与平衡稳定，取得比常规手术更合理的额面松弛度，术后的韧带平衡稳定更好，早期疗效满意。
关键词：膝关节置换；韧带平衡；计算机辅助系统；三维骨建模；医学植入体；生物力学
中图分类号: R318 文献标识码: A 文章编号: 1673-8225(2008)13-02564-05
吴昊，Stéphane Van Driessche，Daniel Goutallier.三维骨建模系统在人工全膝关节置换中的优化作用:导航与常规手术早期疗效比较[J].中国组织工程研究与临床康复，2008，12(13):2564-2568
[www.zglckf.com/zglckf/ejournal/upfiles/08-13/13k-2564(ps).pdf]
（Edited by Lu Y/Song LP/Wang L）