Surgical accuracy and clinical outcomes of image‐free robotic‐assisted total knee arthroplasty

The development of total knee arthroplasty (TKA) for knee osteoarthritis (OA) has a good reputation for its effectiveness in reducing joint pain and improving range of motion. We aimed to review our early results using the image‐free robotic‐assisted technology in knee arthroplasty.

Both of them are imageless semi-active handheld robotic TKA systems. CORI ® robotic surgical system is an upgraded version of Navio ® , it equips with new camera technology and cutting technology which provide faster robotic surgical procedure compared to the Navio ® Surgical System. Previous studies have demonstrated the efficacy of Navio ® /CORI ® in maintaining accuracy of component placement. 6,7 In this study, we evaluated the component position in the coronal and sagittal plane, the relationship with functional outcome and the complication comparing robotic-assisted TKA and conventional TKA.
This study aimed to compare the surgical accuracy and clinical outcome between robotic-assisted techniques and manual conventional techniques. CRE2021.324). All surgeries were done in a tertiary hospital by two senior surgeons in the adult joint replacement team who practice on this field for more than 10 years using a medial para-patellar approach and posterior stabilised implant (Journey II) assisted by Navio/CORI robotic surgical system. An age and sex-matched control group with same sample size was recruited who undergo conventional TKA for comparison. All patients received identical wound closure technique and postoperative recovery protocol (Enhanced recovery after surgery adult joint reconstruction per-operative analgesic & antiemesis protocol adopted in the hospital). All patients followed the standard physiotherapy adult joint reconstruction rehabilitation protocol. Patients were discharged from hospital once their mobility allows outpatient care. Clinical data collected included the patient's demographic data, operation record, X-rays of the knee, standing lower limb scannogram and functional score both pre-operatively and post-operatively.

| MATERIALS AND METHODS
Clinical assessments were performed pre-operatively, at 6 and 12 months post-operatively. Clinical data collected in the assessment included the degree of knee deformity, range of movement, Knee Society Score (KSS), Knee Function Score (KFS) and complications. All patients were assessed by the operative team 1 month before the operation for functional scores (KSS and KFS). Post-operative assessments were performed at 6 months and 1 year. The minimal clinically important differences of KSS and KFS were defined as 9 and 10 points respectively. 8 The X-ray dated closest to the operation date was used for measurement of pre-operative radiological parameters including Hip knee ankle angle (HKA), mechanical lateral distal femoral angle (mLDFA) and mechanical medial proximal tibial angle (MPTA). The first X-ray dated after the index operation was used for measurement of post-operative radiological parameters including HKA, femoral component coronal alignment, tibial component coronal alignment and posterior tibial slope ( Figure 1). Posterior tibial slope was measured against fibular shaft axis. [9][10][11] Two independent orthopaedic surgeons performed all radiological measurements to reduce observation bias. The interobserver reliability was evaluated by comparing the radiological measurements on the same set of radiographs between two independent orthopaedic surgeons. The outcome was defined as acceptable when the values were within 3°and as outliers when deviated from the optimum angle by more than 3°.
All total knee arthroplasties were performed using measured resection surgical technique. The robotic-assisted TKA group, surgeries were performed using either NAVIO or CORI robotic-assisted arm system. They have a navigation system for planning and a handheld burr system for the execution of the planned cuts. Two 4 mm bone pins were placed four fingers breadth superior to the patella in the centre of the femur through arthrotomy incision and the other were two placed four fingers breadth inferior to the tibial tubercle on the medial side of the tibial crest through separated incision. Tracker arrays were fixed on the pins and their positions were confirmed using the infrared camera. Various points were verified using the point Probe ( Figure 2). The hip centre and ankle centre were defined by pivoting movement. The preoperative knee motion was collected, the knee axis was defined and surface mapping was done ( Figure 3). Patient-specific implant planning was done by the surgeon using the virtual model provided by the system. Bone cutting was done with the combined use of burrs and saws ( Figure 4). Real-time feedback for balance was done with the aid of the robotic system. Further soft tissue balancing was performed accordingly, and implants were inserted manually with cement. The conventional TKA group was performed with intramedullary femoral guide (3°of external rotation with respect to posterior condyle axis) and extrameduallary tibial guide. The tibial cut was aim for perpendicular to tibial shaft in coronal plane with a 3°posterior slope in sagittal plane. In both group, patella was replaced in all patient except one case in conventional group which was due to thin patella.

| DISCUSSION
The primary aim of TKA is to re-establish a normal mechanical axis using a well-fixed stable prosthesis. Jeffery et al. 3 reported that accurate coronal alignment to restore the mechanical axis is important in preventing loosening. He reported a 24% loosening rate in 8 years     Certain limitations of the present study should be mentioned.
First, the follow-up period was too short to assess long-term functional outcomes and complications such as loosening. The short follow up period was limited by the availability of robotic system in local hospital. Second, patients with complicated OA of the knee like bone stock deficiency, and severe flexion contracture were excluded.
However, these patients may be the ones who benefit most from the robotic TKA as the anatomical landmark became unreliable and precise bone cutting became difficult.

| CONCLUSION
In this study, both conventional and robotic TKA achieved similar short-term clinical outcomes. In summary, we concluded that semiactive robotic TKA systems NAVIO/CORI (Smith & Nephew) for TKA achieved significant better radiological alignment and shorter hospital stay than conventional TKA. T A B L E 4 Inter-rater analysis of measurements between trainee (1) and orthopaedic specialist (2) using Navio/ CORI.