Can tibia-first total knee arthroplasty using computer-assisted system improve anterior and posterior knee stability?

2.1Patient enrollment

Institutional Bioethics Committee for Medical Research approved the study and waived the requirement for informed consent from individual participants (Approval ID: A20-067). Patients with KOA with varus knee who met the indication for and underwent TKA from May 2019 to November 2020 were included, and MR-TKA was performed from May 2019 until February 2020. TF-TKA was performed from March 2020 to November 2020. All surgeries were performed by one senior board-certified orthopedic surgeon.

The indication of TKA was diagnosed on the basis of clinical findings, such as a decreased ROM, decline in activities of daily living, and radiological findings according to the Kellgren-Lawrence classification. The exclusion criteria were history of previous TKA who planned to undergo revision surgery, patients after high-tibial osteotomy, or patients with previous collateral ligament injury because of the potential effects of these conditions on postoperative AP knee stability.

2.2Surgical procedure for the MR technique

MR-TKAs were performed by using a cruciate-retaining, cemented, fixed-bearing implant (Attune; DepuySynthes, Inc., Warsaw, IN, USA) using a mid-vastus approach. Pneumatic tourniquets were applied to all patients. The anterior cruciate ligament was dissected and the posterior cruciate ligament (PCL) was retained in all cases. Patella resurfacing was not performed. Infrapatellar fat pad excision was only performed when the surgical visual field was insufficient and was minimized to avoid anterior knee pain [9]. We performed a mechanically aligned TKA to achieve neutral coronal mechanical limb alignment by making the femoral and tibial bone cuts such that the rectangular flexion and extension gaps were perpendicular to the mechanical axes. The surgery was performed according to the manufacturer’s instructions. The distal femoral cut was made by using an intramedullary alignment system. The posterior femoral cuts were made with a referencing guide to make the bony resection line parallel to the surgical epicondylar line and perpendicular to Whiteside’s line. The proximal tibial cut was made by using an extramedullary alignment system. Significant posterior osteophytes were removed. Trial components were inserted, and the ROM of the knee and other intraoperative parameters were checked as described above. Release of the collateral and retinacular ligaments to balance the flexion and extension gaps was not performed. The definitive components were introduced with cement fixation, and a final evaluation of the ROM was made.

2.3Surgical procedure of the TF technique using a CAS

TF-TKAs were performed by using the same implant and approach as used for the MR technique. First, anatomical landmarks and the individual biomechanical knee data were obtained; the ROM, varus-valgus angles of the leg axis, and gap sizes throughout the entire ROM were obtained at the beginning of surgery for registration of the navigation system [5]. A tibial osteotomy was first performed to ensure that the tibial bony resection was perpendicular to the anatomical tibial axis. Insertion of the PCL was protected by creating a bony island if necessary. After tibial bony resection, the necessary osteophyte removal was performed but medial collateral ligament (MCL) release, if required, was not performed. Following the tibial bony resection, the gap sizes were measured dynamically by performing varus-valgus stress tests for the medial and lateral gaps to balance the extension gap by using a tibial trial component. Those stress tests were performed from extension to full flexion. The amount of distal femoral osteotomy was determined parallel to the tibia shown on the screen of the navigation system. Then, the amount of the posterior femoral osteotomy and varus-valgus balance were determined on the screen of the navigation system to create an adequate gap. Based on the gaps obtained, the femoral component position was planned to reach total balance, meaning identical medial and lateral gaps throughout the entire ROM (Fig. 1). The distal femoral cut was made first followed by the dorsal cut. The same dynamic varus-valgus stress tests after femoral resections with all trial components placed were performed. The definitive components were introduced with cement fixation, and a final evaluation of the ROM was made.

Figure 1.

The right screen shows a graph of gaps from full knee extension to full flexion. The lateral gap is shown on the right side, and the medial gap is shown on the right side. The left screen shows the component alignments and estimated widths of bony resection.

2.4Intraoperative joint-gap measurement using a tensor

Following each bony resection, we fixed the tensor and femoral trial prostheses, and the patella was reduced. At this point, we measured the varus-valgus ligament balance and a joint tension at a joint center-gap setting equal to the tibial baseplate thickness, and appropriate polyethylene inserts of 0∘, 30∘, 45∘, 60∘, 90∘, and 120∘ of knee flexion were placed (Fig. 2). The varus-valgus ligament balances were defined as positive and negative, respectively.

Figure 2.

Measurements of the varus-valgus ligament balance and joint tension at a joint center-gap setting equal to the tibial baseplate and an appropriate polyethylene insert.

2.5Postoperative measurement of AP laxity by using a Kneelax 3 arthrometer

Immediately after surgery, anesthetized patients underwent knee AP laxity measurement by using a Kneelax 3 arthrometer with 30∘ and 90∘ of knee flexion, as described in a previous study by Shi et al. [10]. One surgeon performed the AP drawer test to apply a force of 132 N to the tibia relative to the femur anteriorly and posteriorly (Fig. 3). The AP laxity was calculated and recorded.

Figure 3.

Postoperative knee AP laxity measurements made by using a Kneelax 3 arthrometer at 30∘ and 90∘ of knee flexion.

2.6Statistical analysis

Data are presented as the mean and standard deviation. A priori power analysis was performed by using G* Power 3.1 (Franz Paul, Kiel, Germany). A sample size of 19 patients in the TF-TKA group and 38 patients in the MR-TKA group was calculated to achieve 80.2% power with an effect size of 0.8 to test the null hypothesis. There were 32 patients in the TF-TKA group and 41 patients in the MR-TKA group, and post hoc analysis showed 91.7% power with an effect size of 0.8. Values of P< 0.05 were considered to be indicative of statistical significance. All statistical analyses were performed by using EZR software.

3.1Patient characteristics

A total of 73 patients, 32 in the TF-TKA group and 41 in the MR-TKA group, were included, with 26 males and 47 females and an average age of 72.5 ± 7.4 (range: 43–87) years. The average preoperative hip-knee angle (HKA) was 12.2∘± 6.5∘ varus (range: 1∘–27.1∘) (Table 1). No significant differences in the male:female ratio, average age, and preoperative HKA were observed between the groups (Table 2).

3.2Comparisons of bony resection width

No significant differences in the width of bony resection in the distal femur and proximal tibia on the medial and lateral sides and in the dorsal femur on the lateral side between the TF-TKA and MR-TKA were observed. On the other hand, the width of bony resection in the dorsal femur on the medial side was significantly smaller in the TF-TKA group than in the MR-TKA group (9.0 ± 2.2 mm vs. 9.8 ± 1.3 mm, P= 0.039) (Table 3).

3.3Comparisons of intraoperative joint-gap measurements using a tensor

The horizontal gap balance was significantly closer in the TF-TKA group than in the MR-TKA group with 0∘, 30∘, 45∘, and 60∘ of knee flexion, but there were no significant differences between the groups for 90∘ and 120∘ of knee flexion (Table 4). No significant differences in joint-gap tensions among all knee flexion angles were observed (Table 5).

3.4Comparisons of intraoperative AP laxity measurements

Translation was significantly smaller in the TF-TKA group than in the MR-TKA group for AP laxity with 30∘ of knee flexion (8.8 ± 2.9 mm vs. 10.7 ± 3.1 mm, P= 0.0079). On the other hand, no significant AP laxity was observed for 90∘ of knee flexion (7.2 ± 2.8 mm vs. 7.2 ± 3.5 mm) (Table 6).

4.1Limitations

There were several limitations in this study. First, we did not assess the interobserver variation in instrumental measurements for the reason mentioned earlier. Second, all TKA surgeries were CR-TKAs performed by one consultant knee surgeon who used the mid-vastus approach. Therefore, these results cannot be generalized to TKAs using others approaches or to posterior-stabilized TKA. Third, a comparison between pre- and postoperative joint lines was not performed in this study but is a topic for future investigation. Fourth, it is uncertain if the lesser AP knee stability in both the mid- and deep-flexion angles and the closer to horizontal gap balances in extension-to-mid-flexion knee angles were related to the TF technique, use of the novel CAS system, or a combination of both. A future separate analysis would be required to answer this important clinical question.

Despite these limitations, this study is the first to show that TF-TKA using a novel CAS system achieved better AP knee stability in mid- and deep-flexion angles than those achieved by using the conventional MR-TKA, with closer to horizontal gap balances in extension-to-mid-flexion knee angles. Future comparative studies would be required to determine if these findings will result in better clinical outcomes and patient satisfaction.