1 Introduction

Total knee arthroplasty (TKA) is widely recognized as an effective treatment option for patients with advanced knee osteoarthritis (KOA), offering pain relief and functional improvement. The growing demand for this procedure is driven by its robust clinical success.¹ Despite over 90 % implant survival at 10 years, data from joint replacement registries in several countries (e.g., UK, Canada, New Zealand) suggest that approximately a quarter of patients remain dissatisfied after surgery. The multifactorial nature of dissatisfaction is not yet fully understood, but component malpositioning may adversely affect joint kinematics through changes in joint line orientation and overall limb alignment.²

Computer-assisted surgical techniques have gained popularity since the mid-2000s, contributing to enhanced coronal and sagittal alignment precision and possibly lowering revision rates.³ These systems offer advantages such as improved mechanical axis accuracy, minimal bone resection, and better soft tissue preservation.⁴ More recently, innovations in navigation systems have enabled the intraoperative assessment of ligament behavior across the full range of motion (ROM).^5,6

It is worth noting that baseline lower limb alignment varies significantly between individuals. Some populations display a natural varus alignment (constitutional varus) without evident osteoarthritic changes on imaging.⁷ In such cases, using mechanically aligned TKA (MATKA) might necessitate aggressive bone resection and ligament releases, potentially leading to poorer outcomes and dissatisfaction. This has led to increased interest in restoring native joint anatomy, and the emergence of kinematically aligned TKA (KATKA) as an alternative approach.⁸

Despite these developments, there is limited evidence on the influence of preoperative factors—such as CPAK classification—on intraoperative kinematics, particularly TIR during flexion in knees treated with medial pivot implants.

We conducted intraoperative assessments using AR and CT-based navigation in patients undergoing KATKA or MATKA with an MP-CR prosthesis. Our hypothesis was that CPAK classification affects tibial internal rotation patterns, and the current study aims to evaluate this hypothesis.

2 Materials and methods

2.3

2.3 Surgical technique

All surgeries were performed by a senior board-certified orthopedic surgeon. The TKA procedure was conducted through a subvastus approach using a cemented, fixed-bearing MP-CR implant. The anterior cruciate ligament was excised, while the posterior cruciate ligament and deep medial collateral ligament were preserved. Patellar resurfacing was not performed in any case.

The NextAR system uses two sterile, disposable sensors—a tracker and an infrared camera—connected wirelessly to a control unit and smart glasses (Fig. 1). The system does not require an external camera or calibration and provides real-time spatial tracking with an error margin of ≤0.5°/0.5 mm. This configuration eliminates the need for bone pins by allowing direct attachment of sensors to the femur and tibia via the surgical incision (Fig. 2).^9,10

Close

Fig. 1 Overview of NextAR system.

Export to PPT

Close

Fig. 2 Overview of initial setup and the bone registration.

Export to PPT

For MATKA, femoral and tibial bone resections were made perpendicular to the mechanical axis, with the goal of achieving rectangular and symmetric flexion-extension gaps throughout the ROM. For KATKA, we followed Howell's unrestricted kinematic alignment principles. Bone resections were planned to match the implant thickness, preserving the native joint lines. The femoral component was resurfaced in the coronal plane, maintaining the patient's original LDFA (Fig. 3). In the axial plane, the posterior femoral cut was aligned at 0° to the posterior condylar axis, and sagittal adjustments were made to avoid notching (Fig. 4). Tibial resection was performed using an alignment guide (Fig. 5), and osteophytes on both tibial and femoral margins were removed. Posterior capsular release and removal of posterior osteophytes were performed in cases with flexion contracture.¹¹ The tibial baseplate was aligned to match the preoperative rotational reference (Fig. 6). Trial components were inserted, and the knee was assessed for ROM, ligament tension, implant rotation, posterior cruciate ligament tension, and patellar tracking. Final cementation was completed after confirming joint balance and removing excess cement.

Close

Fig. 3 Resurfacing of the femoral component on the coronal plane, while preserving the native LDFA.

Export to PPT

Close

Fig. 4 Adjusting the flexion/extension to the optimal position to prevent femoral notching in the sagittal plane.

Export to PPT

Close

Fig. 5 The tibial cut was made through the slot in the guide.

Export to PPT

Close

Fig. 6 The internal–external rotation of the tibial component was aligned parallel according to the preoperative registration.

Export to PPT

2.4

2.4 Intraoperative kinematic assessment

Intraoperative evaluation of tibial internal rotation (TIR) relative to the femur was conducted at flexion angles of 30°, 60°, 90°, and 120° using the NextAR navigation system (Fig. 7). The system generated a detailed report documenting planned vs. executed resections and ligament elongation throughout the procedure.⁶

Close

Fig. 7 Assessment of TIR relative to femur was performed using the NextAR system.

Export to PPT

3 Results

Table 1 presents the demographic and clinical characteristics of the two groups. TIR measurements differed significantly between CPAK type I and type II groups at 60° (2.1° ± 4.8° vs. −2.0° ± 4.2°, p = 0.009), 90° (7.0° ± 4.3° vs. 2.1° ± 5.6°, p = 0.002), and 120° (13.1° ± 5.2° vs. 7.5° ± 4.9°, p = 0.002) of knee flexion.

Multi-way ANOVA revealed a statistically significant interaction between sex and alignment technique (KATKA or MATKA) affecting CPAK classification at 90° of flexion (F = 6.4, p = 0.015). However, alignment method alone did not have a significant effect on TIR outcomes.

4 Discussion

This investigation yielded several key insights. Most notably, the tibial internal rotation (TIR) during knee flexion was significantly influenced by the preoperative CPAK classification, as measured intraoperatively using the AR-based CT navigation system. Additionally, a notable interaction between sex and alignment method (KATKA or MATKA) was observed at 90° of flexion. However, the alignment technique alone did not appear to significantly affect TIR.

To our knowledge, this is one of the few studies examining the relationship between CPAK classification and tibiofemoral kinematics during flexion using a medial pivot (MP) prosthesis. The MP design emulates the knee's native biomechanics by incorporating a “ball-and-socket” articulation in the medial compartment, which limits anteroposterior translation. This design promotes rollback of the femur on the lateral side while preserving stability medially.^14–16 Such behavior may be beneficial in preventing mid-flexion instability, which has been identified as a potential contributor to patient dissatisfaction after TKA.^17,18

Previous studies using fluoroscopic analysis have demonstrated that optimal medial stability combined with controlled lateral laxity in mid-range flexion correlates with favorable outcomes following MP-TKA.¹⁹ Compared to cruciate-retaining (CR) designs with gradually reducing radii, MP implants have been associated with reduced anterior-posterior laxity at mid-flexion angles.²⁰ Given that knee flexion function is closely linked to postoperative quality of life, particularly in Asian populations, optimizing mid-flexion stability is a critical factor in improving patient satisfaction.²¹

In recent years, combining kinematic alignment (KATKA) with MP designs has gained attention due to reported advantages in patient-reported outcomes and faster recovery compared to mechanical alignment with bi-cruciate substituting (BCS) implants.²² These benefits may stem from KATKA's ability to better replicate the patient's native joint line and kinematics, thus creating a more natural-feeling knee.

Although some randomized controlled trials (RCTs) and meta-analyses have suggested that KATKA offers improved postoperative ROM and clinical scores over MATKA, the present study adds a new perspective by indicating that preoperative CPAK classification—not alignment type—is the key factor influencing TIR outcomes.²³ Further investigations are warranted to clarify the optimal alignment strategy based on CPAK phenotype, as this may help guide more personalized approaches to TKA planning.

5 Limitations

This study has several limitations that should be acknowledged. First, we did not evaluate interobserver reliability for intraoperative measurements, which may introduce measurement bias. Second, the investigation was conducted by a single surgeon at one institution, potentially limiting the external validity and generalizability of the findings. Third, our patient cohort was composed exclusively of East Asian individuals, which may limit applicability to populations with different anthropometric characteristics or implant size distributions.

Additionally, selection bias may have influenced case inclusion, although surgical indications were consistently determined by a senior specialist. Despite these limitations, this is the first study to demonstrate that CPAK type I and II classifications are associated with distinct intraoperative tibial rotation patterns, irrespective of the alignment technique employed, as assessed using AR-based CT navigation. Further comparative studies are necessary to determine whether these intraoperative kinematic differences correlate with long-term functional outcomes and patient satisfaction.

6 Conclusion

Our findings suggest that preoperative CPAK classification (type I or type II) significantly affects tibial internal rotation during knee flexion at 60°, 90°, and 120°, as measured intraoperatively with augmented reality CT-based navigation. These effects were observed regardless of whether mechanically aligned or kinematically aligned TKA was performed. CPAK phenotype may serve as a valuable preoperative predictor of kinematic behavior in total knee arthroplasty.

CRediT authorship contribution statement

Eiichi Shiigi: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Supervision, Validation, Writing – review & editing. Tsuneari Takahashi: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing. Koji Yoshida: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Writing – review & editing. Masahiro Numa: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Writing – review & editing. Keiko Kamata: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Writing – review & editing. Katsushi Takeshita: Conceptualization, Project administration, Writing – review & editing. Takashi Sakai: Conceptualization, Project administration, Writing – review & editing.

Guardian/patient's consent

Individual patient's consent was waived due to retrospective nature of this study.

Ethical approval

This study was conducted in accordance with the principles of the Declaration of Helsinki. The Institute's Bioethics Committee for Medical Research at the Yamaguchi Prefectural General Medical Center approved the study (approval ID: 2020-J029) and waived the requirement for informed consent from individual participants, given the retrospective nature of the study. All patients received the current standard treatment.

Funding statement

Not applicable.