Influence of Cartilage on Femoral Component Rotation in Robotic‐Assisted Total Knee Arthroplasty: A Model‐Based Quantitative Analysis and Clinical Data Evaluation

ABSTRACT

Objectives

Femoral component rotation affects knee function and component survival in total knee arthroplasty (TKA). However, the presence of posterior femoral condylar cartilage leads to discrepancies in the femoral posterior condylar axis (PCA) between robotic‐assisted TKA and manual TKA. The purpose of this study was to investigate the relationship between the discrepancy in thickness of the medial and lateral posterior femoral condylar cartilage and the discrepancy between robotic‐assisted and manual rotation of the femoral component.

Methods

In the computed tomography (CT) modeling simulation section, we retrospectively reviewed a total of 18 preoperative knee CT scans of patients who underwent robotic‐assisted TKA with different femoral prosthesis sizes between January 2022 and January 2023 to measure the mean posterior femoral condylar distance between femurs of different sizes. In the prospective clinical study section, we prospectively measured the cartilage thickness of the medial and lateral posterior condyles in 60 patients who underwent Mako‐assisted TKA between October 2023 and December 2024.

Results

According to our mathematical model of the difference between robotic and manual femoral component rotation in the presence of different femoral sizes and differences in medial and lateral posterior condyle cartilage thicknesses, the maximum value of angular discrepancy of PCA was 4.02° and the minimum value was 1.13°. The average cartilage thickness difference between the medial and lateral posterior femoral condyles was 0.29 ± 0.97 mm (−2.00 to 2.10 mm). The mean difference in femoral component rotation between robotic and manual TKA was 0.35° ± 1.21° (−2.61° to 2.82°).

Conclusion

For most patients with posterior femoral condylar cartilage, the PCA determined by robotic‐assisted surgery was greater than that determined manually. Therefore, when surgeons perform TKA with robotic assistance, it is important to be aware of this discrepancy in femoral component rotation to avoid complications such as poor component survival due to inadequate rotation of the femoral component.

The presence of posterior femoral condylar cartilage leads to discrepancies in the femoral posterior condylar axis (PCA) between robotic‐assisted TKA and manual TKA. For most patients with posterior femoral condylar cartilage, the PCA determined by robotic‐assisted surgery was greater than that determined manually.

1. Introduction

In total knee arthroplasty (TKA), rotation of the femoral component affects patellofemoral function, knee stability, and tibial polyethylene insert wear, which have major impacts on patient prognosis [1, 2, 3, 4, 5]. A previous study revealed that the external rotations of robotic‐assisted TKA are generally larger than those of manual TKA [6]. One possible reason for this finding is that robotic‐assisted TKA uses a gap‐balancing workflow, which generally results in larger external rotations [7, 8], and another possible reason is that the posterior condylar cartilage thickness affects femoral component rotation [9, 10, 11].

Most robotic‐assisted TKAs construct 3D models with the help of preoperative computed tomography (CT) scans. However, because the CT model usually does not include the cartilage of the posterior femoral condyle, the posterior condylar axis (PCA) defined by the Mako TKA system is the line tangential to the posterior condylar cortical surface [12]. Manual TKA with the posterior reference technique usually uses a femoral sizing guide attached to the posterior condylar surface, which in turn determines the degree of femoral component rotation based on PCA. If cartilage wear occurs, there will be a difference in cartilage thickness between the medial and lateral posterior condyles, resulting in a difference between robotic and manual PCA [9, 10, 11].

Previous studies have confirmed that the mean difference in the degree of external rotation with and without cartilage is between 0.1° and 1.8°, and in extreme cases, it can even exceed 4° [9, 11, 13, 14, 15]. However, in addition to varying degrees of cartilage wear, the condylar distance could also influence the determination of femoral component rotation via PCA. Therefore, for individuals, it is more meaningful to evaluate individual differences than to understand only changes in the overall mean. To our knowledge, no research has provided a simple and accurate method for assessing the effects of external rotation on femoral components.

The aim of this study was to: (1) establish a mathematical model to precisely quantify the influence of differences in the thicknesses of the medial and lateral posterior condyle cartilage on the external rotation of the femur during robotic‐assisted TKA; (2) describe the distribution and frequency of such differences in clinical practice cases to provide a quantitative reference for the evaluation of external rotation in robotic‐assisted TKA.

2. Materials and Methods

The study was divided into two parts: a CT modeling simulation and a prospective clinical trial. The study was approved by our institutional ethics committee (2023KY130‐HS001) and registered in the Chinese Clinical Trial Registry (ChiCTR2400085440).

2.1. CT Modeling Simulation

In this section, we reviewed patients who underwent TKA surgery using a Triathlon PS (Stryker) knee prosthesis assisted by the Mako TKA system (Stryker, Mahwah, NJ, USA) at our institution from January 2022 to January 2023 and evaluated the size of the femoral component used. We categorized these patients into six groups according to their femoral component size, ranging from 1 to 6. Three patients were randomly selected from each group, for a total of 18 patients included in the study. The demographic information of the patients is shown in Table 1. After informed consent was obtained from the patients themselves, the preoperative CT scans of the 18 patients were modeled and analyzed via Mimics, version 25.0 (Materialize, Leuven, Belgium).

TABLE 1.

Patient demographic information of the CT modeling simulation.

Patient demographics	Value
Age (years)	66.17 ± 11.33 (37–80)
Height (m)	1.63 ± 0.08 (1.50–1.80)
BMI (kg/m2)	25.67 ± 2.68 (19.03–29.05)
Sex
Male	7 (38.9%)
Female	11 (61.1%)
Side
Left knee	9 (50.0%)
Right knee	9 (50.0%)
HKA (°)	171.25 ± 3.54 (165.00–176.50)
FTA (°)	176.15 ± 3.02 (169.70–179.60)
LDFA (°)	88.71 ± 3.34 (82.70–96.50)
MPTA (°)	86.04 ± 3.53 (78.70–90.50)
K‐L scale	3.50 ± 0.62 (2–4)

Note: Quantitative data are given as the mean ± SD, with the range in brackets; categorical data are given as the number, with the corresponding percentage in brackets.

Abbreviations: FTA, femoro‐tibial angle; HKA, hip‐knee‐ankle angle; K‐L scale, Kellgren–Lawrence scale for radiographic classification of osteoarthritis; LDFA, mechanical lateral distal femoral angle; MPTA, mechanical medial proximal tibial angle.

First, the intramedullary axis of the distal femur was constructed by connecting two midpoints of the medullary cavity. In the view perpendicular to the distal femoral intramedullary axis, the line tangential to the posterior condylar cortical surface was defined as the robotic PCA. The distance between the two points where the robotic PCA is tangent to the posterior condyle was measured and recorded as the condylar distance. (Figure 1) In previous studies, the average difference in the thickness of the medial and lateral posterior condyle cartilages was 1.4–1.7 mm [11, 15]. To simulate real cases, we selected three different thickness differences of 1, 2, and 3 mm between the medial and lateral posterior condyle cartilages. For the CT scan of each patient, the presence of 1 mm, 2 mm, and 3 mm of cartilage on the surface of the lateral posterior femoral condyle was simulated. The line tangential to the medial posterior condylar cortical surface and the lateral posterior condylar cartilage surface was defined as the manual PCA in the view perpendicular to the distal femoral intramedullary axis. The angular difference between manual and robotic PCA was subsequently measured. This measurement step was carried out by two measurers separately, and the final results were averaged.

FIGURE 1.

Measurement of the posterior femoral condylar distance. The posterior femoral condylar distance was measured in the plane perpendicular to the distal femoral intramedullary axis. We defined the line tangential to the posterior condylar cortical surface as the robotic PCA. The distance between the two points where the robotic PCA is tangent to the posterior condyle was measured and recorded as the condylar distance.

Based on the above data, we developed a mathematical model of the difference between robotic and manual femoral component rotation in the presence of different femoral sizes and differences in medial and lateral posterior condyle cartilage thicknesses. The thickness of the medial and lateral posterior condyle cartilage can be measured in the bone registration step of the Mako TKA system, while the condyle distance can be found in Table 3 according to the selected femoral component size in the intra‐operative planning.

$$\text{Angle}=\arctan \frac{\text{difference in cartilage thickness}}{\text{condylar distance}}$$

TABLE 3.

Angular discrepancy of PCA in relation to cartilage thickness and component size.

Component size of Triathlon PS (Stryker)	Mean posterior intercondylar distance (mm)	Lateral minus medial value of cartilage thickness
		1 mm (°)	2 mm (°)	3 mm (°)
1	42.73 ± 1.05	1.34	2.68	4.02
2	43.91 ± 1.01	1.30	2.61	3.91
3	45.42 ± 0.73	1.26	2.52	3.78
4	46.71 ± 1.40	1.23	2.45	3.68
5	48.73 ± 0.84	1.18	2.35	3.53
6	50.51 ± 0.99	1.13	2.27	3.40

Note: Quantitative data are given as the mean ± SD.

2.2. Prospective Clinical Trial

A total of 60 patients who underwent primary TKA using a Triathlon PS (Stryker) knee prosthesis assisted by the Mako TKA system were prospectively enrolled between October 2023 and December 2024. All the surgeries were performed by the same senior orthopedic surgeon. The inclusion criteria for this study were as follows: (1) age less than 80 years; (2) knee osteoarthritis; and (3) proposed treatment by unilateral robotic‐assisted TKA. The exclusion criteria for this study were as follows: (1) severe valgus knee (> 15°); (2) severe flexion contracture deformity (flexion contracture > 15°); and (3) severe varus knee (> 15°). The demographic information of the patients is shown in Table 2.

TABLE 2.

Patient demographic information of the prospective clinical trial.

Patient demographics	Value
Age (years)	67.65 ± 6.23 (53–80)
Height (m)	1.62 ± 0.07 (1.45–1.75)
BMI (kg/m2)	26.28 ± 2.53 (20.80–31.30)
Sex
Male	17 (28.3%)
Female	43 (71.7%)
Side
Left knee	38 (63.3%)
Right knee	22 (36.7%)
Varus/valgus
Varus	57 (95.0%)
Valgus	3 (5.0%)
HKA (°)	171.16 ± 4.86 (156.00–179.60)
FTA (°)	174.62 ± 3.88 (164.80–179.90)
LDFA (°)	89.78 ± 4.41 (80.70–105.30)
MPTA (°)	86.30 ± 4.65 (77.30–97.30)
K‐L scale	3.52 ± 0.50 (3–4)

Note: Quantitative data are given as the mean ± SD, with the range in brackets; categorical data are given as the number, with the corresponding percentage in brackets.

Abbreviations: FTA, femoro‐tibial angle; HKA, hip‐knee‐ankle angle; K‐L scale, Kellgren–Lawrence scale for radiographic classification of osteoarthritis; LDFA, mechanical lateral distal femoral angle; MPTA, mechanical medial proximal tibial angle.

The surgery was performed according to the Mako TKA Surgical Guide. After completing the osteotomy, the medial and lateral posterior condylar osteotomy blocks were removed. A syringe was used to penetrate the cartilage at the most convex part of the block, and then the depth of penetration was marked on the syringe. After pulling out the syringe, a Vernier caliper was used to measure the distance from the tip of the syringe to the marker. (Figure 2) This procedure was repeated twice, and the final result was averaged.

FIGURE 2.

Measurement of cartilage thickness of the posterior femoral condyles. (a) Using a syringe to penetrate the cartilage of the posterior femoral condyles. (b) Using a Vernier caliper to measure the depth of syringe penetration.

2.3. Statistical Analysis

The demographic data and measurement results of the study are reported as the means ± standard deviations. The intraclass correlation coefficient (ICC) was calculated to evaluate the angular discrepancy between the two observers' measurements. p < 0.05 was considered to indicate statistical significance. All the statistical analyses were performed via SPSS, version 24.0 (SPSS, Chicago, IL, USA), and the graphs were drawn via GraphPad Prism, version 8.30 for Windows (GraphPad Software, San Diego, CA, USA).

3. Results

3.1. CT Modeling Simulation

The ICC between the two measurements was 0.83 (95% CI [0.61–0.93], p < 0.001), which indicated that the measurements were in good agreement. The average preoperative distance between the posterior femoral condyles of patients with femoral components of sizes 1–6 ranged from 42.73 mm to 50.51 mm. Thus, the angle discrepancy between manual and robotic PCA can be calculated according to the trigonometric function when there is a thickness difference between the cartilage of the medial and lateral posterior femoral condyles. For a certain difference in cartilage thickness, the smaller the femoral component is, the larger the angle difference, as shown in Table 3.

3.2. Prospective Clinical Trial

The mean cartilage thickness of the medial posterior condyle was 1.34 ± 0.89 mm (0–3.40 mm). The mean cartilage thickness of the lateral posterior condyle was 1.63 ± 0.77 mm (0–3.00 mm). The average cartilage thickness difference was 0.29 ± 0.97 mm (lateral minus medial, −2.00 to 2.10 mm), as shown in Figure 3. In three cases of valgus knees, the thickness of the posterior condylar cartilage was as follows: 1.2 mm medial, 2.3 mm lateral; 1.3 mm medial, 1.5 mm lateral; and 2.1 mm medial, 0.4 mm lateral. Substituting the data on cartilage thickness and prosthesis type from the 60 cases in the clinical trial into the formula yielded the expected difference between manual and robotic PCA, which averaged 0.35° ± 1.21° (−2.61° to 2.82°), as shown in Figure 4.

FIGURE 3.

The average difference in cartilage thickness between medial and lateral posterior femoral condyles.

FIGURE 4.

The actual differences in PCA caused by cartilage thickness in real cases. We substituted the cartilage thickness and prosthesis type of the 60 cases into the formula to derive the expected difference between manual and robotic PCA. Three cases of valgus knees were highlighted with red dots.

4. Discussion

This study provides, for the first time, a simple and accurate method for assessing the effects of different sizes of femoral components and cartilage thicknesses on the external rotation of femoral components. Surgeons can estimate the size of the prosthesis and cartilage thickness between the medial and lateral posterior femoral condyles to understand the discrepancy between the PCA defined by the TKA‐assisted robot and the PCA determined by traditional manual instrumentation. This could help surgeons obtain a more accurate understanding of the real external rotation of femoral components.

4.1. The Influence of Posterior Condylar Cartilage and Condylar Distance on Femoral Component Rotation

In CT image‐based robotic‐assisted TKA for defining the PCA, a line that is tangential to the bone surface of the internal and external posterior femoral condyle, which does not consider the thickness of the cartilage, is usually selected. In contrast, traditional manual TKA requires the surgeon to hold the reference guide close to the posterior condylar surface, whereas in most patients, there is residual cartilage [13, 16]. According to our formula, the greater the difference in cartilage thickness between the lateral and medial posterior condyles is, the greater the influence on external rotation. This difference is amplified in smaller prostheses (those with smaller condylar distances). According to our results, when the difference in cartilage thickness between the medial and lateral posterior femoral condyles reached 3 mm with a femoral component of size 1, the difference in the angular rotation of the femoral component reached 4.02°, which is not negligible in TKA (Table 3).

4.2. The Distribution and Frequency of Femoral Component Rotation Differences in Clinical Practice

With respect to the real difference in cartilage thickness among the real arthritis population, we found that most of the cartilage of the lateral posterior femoral condyle was thicker than that of the medial posterior femoral condyle. The average cartilage thickness difference was 0.29 ± 0.97 mm (−2.00 to 2.10 mm). These findings are similar to those of several previous studies [9, 14, 17]. When combined with the actual implant size, our data showed that the external rotation of the femoral component for robotic‐assisted TKA is 0.35° ± 1.21° (−2.61° to 2.82°) greater than that for manual TKA. In some cases, we could find when the external rotation of the femoral component for robotic‐assisted TKA is set at 5°, it is actually equivalent to 3° for manual TKA.

In the present study, 3 valgus knees were also included. In theory, the lateral posterior condyle of the valgus knee should be worn more; accordingly, the cartilage thickness of the medial posterior condyle should be greater than that of the lateral posterior condyle. However, according to our results, medial cartilage wear was worse in 2 patients, and only 1 patient had worse lateral cartilage wear. Since the sample size of the valgus knee was not large enough, more data are needed for further evaluation in the future.

This angular discrepancy may lead surgeons, especially those newly performing robotic‐assisted TKA, to set the femoral component external rotation angle at a smaller value than appropriate. An insufficient femoral component external rotation angle may affect patellar tracking, ligament balance during knee flexion, and tibiofemoral congruity; this ultimately leads to a poorer prognosis for the patient [18, 19, 20]. For cases in which the external rotation of the femoral component varies significantly from expectations, we recommend that the surgeon measure the thickness of the medial and lateral cartilage with a probe intraoperatively to allow a more accurate estimation of the true external rotation of the femoral component.

4.3. Strengths and Limitations

Our research comprehensively considered the influence of posterior condylar cartilage thickness and condylar distance on the femoral component rotation via PCA, and for the first time proposed an easy‐to‐use mathematical model to provide guidance for clinical practice.

The limitations of our study are as follows. First, measurement error may exist in measuring cartilage thickness by puncturing the cartilage with a needle. However, this method is an easy way for doctors to obtain cartilage thickness during surgery. Our data could provide accurate judgment for doctors under the same conditions. Second, this study only applies to surgeons who adjust the external rotation of the femoral component according to the PCA, and it is not applicable to surgeons who prefer other axes. Third, the sample size of this study, especially for the valgus knee section, was not large enough. Further studies with larger sample sizes are still needed. Fourth, our research focused on robotic‐assisted TKA using the Triathlon PS (Stryker) knee prosthesis, and the results may not be generalizable to all robotic‐assisted TKA surgeries.

5. Conclusions

Overall, there was an angular discrepancy between the robotic PCA and manual PCA when there was a difference in the cartilage on the posterior femoral condyle. In most patients, thicker cartilage on the lateral posterior condyle results in greater external rotation of the femoral component when robotic‐assisted TKA is performed. Surgeons, especially those newly performing robotic‐assisted TKA, should be aware of this difference to avoid inappropriate femoral component rotation.

Author Contributions

This study was conducted under the guidance of R.L. and W.C. The article was written by H.‐M.A. and Q.‐D.W. The statistical analysis was performed by W.G. The formal analysis was done by G.W. All authors read and approved the final manuscript.

Ethics Statement

This study was approved by our Institutional Ethics Committee (No. 2023KY130‐HS001) and registered in the Chinese Clinical Trial Registry (ChiCTR2400085440).

Conflicts of Interest

The authors declare no conflicts of interest.

An H.‐M., Wei Q.‐D., Gu W., Wang G., Chai W., and Li R., “Influence of Cartilage on Femoral Component Rotation in Robotic‐Assisted Total Knee Arthroplasty: A Model‐Based Quantitative Analysis and Clinical Data Evaluation,” Orthopaedic Surgery 17, no. 12 (2025): 3406–3411, 10.1111/os.70196.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.