1 Background

Total knee arthroplasty (TKA) is one of the most common surgical procedures performed in the United States with a prevalence of 1.52 % in the US population.¹ Currently, there are over 4.7 million people in the US with a TKA, with annual procedures expected to grow to 1.26 million annual procedures by 2030.^1,2 Although this procedure has been done for more than 50 years, patient dissatisfaction remains low at 15–20 %.^2–5 The exact reason for patient dissatisfaction is multifactorial. Common reasons for dissatisfaction are residual pain, limited function, poor surgical technique, malalignment of limbs, and postoperative complications.^6–8 Patient related comorbidities such as depression, smoking, and education can lead to poor outcomes but cannot entirely explain dissatisfaction. Poor flexion can limit the patient's ability to execute basic activities in their everyday lives, such as squatting, climbing stairs, and running.⁸ Poor range of motion is linked to several factors including malalignment and improper gap balance.⁹

Instability is one of the most common reasons for the failure of modern TKA, accounting for 7.5–29 % of failures.^10,11 Flexion instability is an increasingly recognized form leading to patient morbidity.¹² Factors leading to flexion instability include joint line elevation, excessive posterior condylar resection paired with undersized femoral components, increased posterior tibial slope, and over-release of the posterior cruciate ligament in the cruciate retaining knee. Revision surgery is often required when addressing patients with flexion instability.

When using conventional methods, the flexion gap can be difficult to fully assess. Improper resection of the knee and sizing of implants can lead to instability within the joint. Measured resection uses several bone landmarks to determine proper rotation of the femoral component and gap balance.¹³ This technique has been scrutinized by other studies for lack of accuracy among several steps within the procedure and difficulty to be reproduced between patients.^14–16 A gap balanced technique aims to put equal tension on the collateral ligaments to obtain a rectangular flexion gap.¹⁴ Gap balancing techniques can potentially lead to improved flexion stability, however this technique is reliant on precise proximal tibial resection.¹⁴ Robotics in TKA has been heralded with mixed results, with some studies showing improved accuracy and outcome while others show no difference in results.¹⁷

The primary purpose of this study was to assess femoral size difference between surgeons with different techniques to determine if there is a difference in femoral sizing or polyethylene constraint. This study also aimed to determine the impact of each technique on restoring joint line anatomy within the knee.

2 Material and Methods

This retrospective data analysis study included all patients who underwent unilateral primary TKA by one of three surgeons. The inclusion criteria for the study involved patients who were 18 years of age or older at the time of operation, with a minimum of 90 days of follow-up. This study was approved by an institutional review board.

Femoral component and polyethylene sizing as well as the rate of semi-constrained polyethylene usage was compared between three high volume surgeons within the same practice who use the same knee implant system, but differing techniques. Surgeon 1 uses a restricted kinematic alignment (RKA) technique with robotic arm assistance. Surgeon 2 uses a mechanical alignment (MA) technique with a measured resection approach using standard instruments. Surgeon 3 uses a mechanical alignments technique with a gap balanced (GB) approach with standard instruments. A total of 1508 patients were enrolled in this initial study. The data acquired from this experiment was analyzed using a chi-square statistical test to determine differences between component sizing and polyethylene usage of surgical techniques. Statistical significance was set at p < 0.05.

To determine the ability of each surgical technique to restore joint anatomy and function, 100 patients who were greater than 90 days postoperative were chosen at random from the initial pool of patients for each surgeon. Preoperative and postoperative radiographic films for each patient were acquired. The radiography machine was set to 70 kVp and 16 mAs prior to completing the imaging protocol. These films were used to determine the distance of the joint line from the medial epicondyle, the distance of the joint line from the lateral epicondyle, the distance of the joint line from the fibular head, as well as the posterior femoral offset for each patient. The medial proximal tibial angle (MPTA) and the lateral distal femur angle (LDFA) were also determined for each patient by analyzing these images. The difference between preoperative and postoperative data was determined for each category for each patient. Knee Injury and Osteoarthritis Outcome Score for Joint Replacement (KOOS, JR) values were recorded following the six-week postoperative visit for each patient and the adjusted scores were used for statistical analysis. Each measurement was analyzed using a single-factor ANOVA test to determine if there were significant differences in joint anatomy restoration following TKA between the three surgical techniques. Statistical significance was set at p < 0.05.

Demographic data were recorded including patient age, sex, height, weight, body mass index (BMI), and reported race for all patients involved in the study. This data was analyzed using a single-factor ANOVA statistical test to determine any differences in the demographics of the patients for each of the surgical techniques.

The demographic data of the patients included in the study is shown in Table 1. Of the 300 patients examined for the study relating to joint anatomy, 128 were men (42.7 %) and 172 were women (57.3 %). The mean age of the patients was 68.59 for surgeon 1 (±8.86), 67.49 for surgeon 2 (±9.14), and 69.91 for surgeon 3 (±9.05). The average height of all patients was 66.8 ± 4.25 inches and the average weight was 191.67 ± 42.8 pounds. The majority of patients involved in this study (89.67 %) identified as white.

3 Results

3.1

3.1 Component sizing

A total of 1508 patients were included in this study among the three surgeons. A significant increase in formal size was identified for surgeon 1 compared to surgeon 2 and surgeon 3 (p < 0.0001) (Fig. 1). Surgeon 1 most commonly used size 11 femoral components at a rate of 21.11 %, while surgeon 2 and surgeon 3 most commonly used size 7 femoral components at a rate of 21.47 and 21.64 %, respectively. A significant decrease in polyethylene size was identified for surgeon 1 compared to surgeon 2 and surgeon 3 (p < 0.0001) (Fig. 2). Surgeon 1 used 10 mm poly the most at a rate of 70.85 %; surgeon 2 used 11 mm poly the most at a rate of 29.21 %; surgeon 3 used 10 mm poly the most at a rate of 32.31 %.

Close

Fig. 1 Femoral component sizing for each surgeon by percentage.

Export to PPT

Close

Fig. 2 Polyethylene sizing for each surgeon by percentage.

Export to PPT

A significant decrease in tibial implant sizing was found for surgeon 3, but not between surgeon 1 and surgeon 2 (p < 0.0001) (Fig. 3). Surgeon 3 used size E tibial components the most at a rate of 28.46 %, while surgeon 1 used size E the most at 31.92 % and surgeon 2 most commonly used size D implants at 26.80 %.

Close

Fig. 3 Tibial component sizing for each surgeon by percentage.

Export to PPT

3.2

3.2 Joint anatomy

A significant decrease was observed in the change of distance from the medial epicondyle to the joint line (−0.1343 mm) (p < 0.05) and a significant increase was observed in the distance from the lateral epicondyle to the joint line (3.2043 mm) (p < 0.0001) when using the kinematic technique (surgeon 1) compared to the techniques used by surgeon 2 and surgeon 3 (Fig. 4). The mechanical alignment technique led to a significantly smaller distance from the fibular head to the joint line (9.384 mm) (p < 0.0001).

Close

Fig. 4 Change of distance from the medial epicondyle to the joint line, the lateral epicondyle to the joint line, the fibular head to the joint line and the change of posterior femoral offset. Results are presented in millimeters. An asterisk denotes statistical significance (p < 0.05).

Export to PPT

The RKA technique also led to significantly increased posterior femoral offset values (39.394 mm) (p < 0.001) (Table 2) and the largest increases in this value on average (+4.4412 mm) (p < 0.001) (Fig. 4). Although the RKA technique had the significantly smallest postoperative MPTA (86.7919 deg) (p < 0.0001) compared to the other approaches, it also has the significantly smallest change in MPTA (+0.5682 deg) (p < 0.0001) (Fig. 5). It was found that the MA technique led to the significantly largest postoperative LDFA on average (86.9009 deg) (p < 0.0001) as well as the largest increase in LDFA (+4.037 deg) (p < 0.001) among the three techniques evaluated (Fig. 5). No significant differences in average preoperative and six-week postoperative adjusted KOOS, JR values were observed between the surgeons, however the RKA technique led to the highest postoperative adjusted score. Average KOOS, JR values increased by the six-week postoperative visit for each of the three surgical techniques.

Table 2 Joint anatomy averages and standard deviations for 100 patients for each surgeon. Subjects were chosen at random. Bold P-values denote statistical significance (p < 0.05).

	Surgeon 1	Surgeon 2	Surgeon 3	P-value
Distance from the medial epicondyle to the joint line mean (mm), (SD)
Preoperative	34.76 (5.10)	36.68 (4.64)	38.80 (5.22)	>0.0001
Postoperative	24.62 (4.24)	35.26 (4.65)	36.50 (4.24)	0.0092
Change	−0.13 (5.29)	−2.29 (5.54)	−1.42 (5.07)	0.0159
Distance from the lateral epicondyle to the joint line mean (mm), (SD)
Preoperative	30.51 (5.69)	32.20 (5.41)	34.35 (4.97)	<0.0001
Postoperative	33.72 (4.55)	32.77 (4.34)	33.45 (4.04)	0.277
Change	3.20 (5.59)	0.58 (5.13)	−0.89 (5.49)	<0.0001
Distance from the fibular head to the joint line mean (mm), (SD)
Preoperative	16.37 (4.30)	15.28 (3.80)	16.09 (4.90)	0.186
Postoperative	13.67 (4.35)	9.38 (3.77)	13.86 (5.14)	<0.0001
Change	−2.70 (4.25)	−5.90 (3.67)	−2.24 (4.53)	<0.0001
Posterior femoral offset mean (mm), (SD)
Preoperative	34.95 (6.18)	35.04 (6.16)	36.02 (6.44)	0.408
Postoperative	39.39 (5.86)	36.55 (6.42)	36.85 (5.03)	0.00084
Change	4.44 (6.74)	1.51 (7.72)	0.82 (5.53)	0.00035
Medial proximal tibial angle mean (degrees), (SD)
Preoperative	86.22 (3.09)	86.47 (3.25)	86.16 (3.33)	0.773
Postoperative	86.79 (2.09)	88.91 (1.90)	88.81 (1.84)	<0.0001
Change	0.57 (3.44)	2.44 (3.54)	2.65 (3.68)	<0.0001
Lateral distal femur angle mean (degrees), (SD)
Preoperative	83.11 (2.62)	82.86 (2.84)	83.32 (2.42)	0.605
Postoperative	85.66 (1.99)	86.90 (2.15)	85.79 (2.07)	<0.0001
Change	2.55 (3.09)	4.04 (3.40)	2.56 (2.88)	0.00068

Close

Fig. 5 Change in medial proximal tibial angle (MPTA) and lateral distal femoral angle (LDFA). Results are presented in degrees. An asterisk denotes statistical significance (p < 0.05).

Export to PPT

4 Discussion

Total knee arthroplasty (TKA) is an extremely common surgical procedures performed in the United States with a prevalence of 1.52 % in the US population.¹ Overall patient function following TKA has improved, however this is sometimes overshadowed by highly variable patient satisfaction. The reasons for dissatisfaction include residual pain, limited function, poor surgical technique, malalignment, and postoperative complications.^6–8 Poor flexion limits the patient's ability to execute basic activities in their everyday lives, such as squatting, climbing stairs, and running.⁸ Poor range of motion (ROM) is linked to several risk factors and poor postoperative care but can also be caused by malalignment and improper gap balance leading to joint instability.⁹ Factors leading to flexion instability include insufficient distal femur resection in a preexisting flexion contracture, excessive posterior condylar resection paired with undersized femoral components, an increased posterior tibial slope, and over-release of the posterior cruciate ligament in the cruciate retaining knee.¹² Numerous concepts currently exist in TKA alignment and surgical methods, with different alignment goals.

Robotics have grown in popularity over the past decade, with more surgeons being trained in robotics and an increased utilization of robotic technology. Data regarding robotic assisted surgery is mixed, ranging from improved and more accurate alignment to no difference between manual knees.^18,19 Alternative alignment techniques moving away from standard mechanical alignment have also grown in popularity, and some studies have suggested improved outcomes and forgotten knee scores.²⁰ This study sought to determine differences in implant usage and alteration of the joint line between three different techniques.

The robotic kinematic alignment technique led to larger femoral components and smaller polyethylene sizes than both MA and GB techniques. Along with this, significantly increased posterior femoral offset was observed in the RKA group and closer restoration of the normal posterior femoral offset than the MA and GB techniques. While posterior femoral offset is only one marker for potential flexion instability, significantly decreased polyethylene sizes in the robotic RKA group was also found compared with both the MA and GB groups. There was also a significant decrease in variation of polyethylene sizes in this group as well, a surrogate for reproducibility.

Multiple significant differences were found in relation to joint line preservation and orientation. The robotic RKA group had a significant restoration of native MPTA and LDFA compared to both the MA and GB groups. This was significant and expected given the techniques utilized by each surgeon. This outcome is consistent with the goal of kinematic alignment, where native distal femoral anatomy is preserved leading to a slight varus alignment. Other studies have further discussed this difference and have detailed varus alignment as a functional advantage of kinematic alignment.^20,21 No significant difference has been found in the survivorship of kinematic and mechanical alignment knees.²² While some studies show no difference in patient satisfaction and function, more recent studies suggest improved patient satisfaction with kinematically aligned knees six months and two years following operation.^23,24

The robotic, kinematic alignment group had significantly better restoration of the native joint line compared to both the MA and GB group with addressing both the medial and epicondylar distance and the difference in distance to the fibular head. The large deviation was seen in fibular head height between the robotic RKA group and the MA, suggesting significant elevation of the native joint line in the MA group. This certainly is consistent with larger polyethylene sizes and smaller femoral components. As the joint line elevates, smaller femoral components are required in order not to overstuff the flexion gap. This alteration of the native joint line is believed to lead to a source of flexion instability.

Although there were differences in sizing, constraint usage, polyethylene variability, and joint line restoration and orientation no significant difference was observed in 90-day patient reported outcome measures (PROMs) recorded by KOOS, JR. There was also no difference in baseline demographics as would be expected since all surgeons were in the same orthopedic group. On average, patients reported scores over 70 points. All three surgeons examined within this study are considered to be high-volume joint replacement fellowship-trained physicians.

Limitations of this study including the retrospective nature make it difficult to establish causality. There is also minor variability in the timing of outcome measures recorded due to differences in patient follow-up along with normal patient attrition. To limit the impact of this variability, this study did not include reported outcomes recorded less than six weeks following surgery or greater than 16 weeks following surgery. This study may be underpowered to determine any differences in PROM's. This study is also limited by each surgeon independently performing the technique they were most comfortable with which may limit reproducibility of the study. Slight differences in surgical techniques, like ligament release, guidance systems, and surgeon performance, could have affected the results of this study.

5 Conclusion

Robotic RKA in TKA leads to significantly improved femoral offset, larger femoral sizes, and improved maintenance of native joint line anatomy compared to MA and GB techniques. The result of this is smaller polyethylene sizing and decreased usage of constrained polyethylene in TKA patients. While no significant difference in functional outcomes was observed this study may be underpowered to notice any significant difference in PROM's.

Patient-guardian consent

The study did not require consent to be obtained.

Credit author statement

John Dundon: Conceptualization, Methodology, Supervision, Project administration, Writing – Reviewing & editing, visualization, investigation. Nicholas Brown: Data curation, Writing – Original draft preparation, Writing – Review & Editing. Jen Escobar: Data Curation, Resources, Writing – Review & Editing. Paul Lombardi: Writing – Review & Editing, Project administration, Resources.

Ethical statement

The ethics committee, WCG, approved the study and consent on July 22nd, 2024. The committee's reference number is IRB00000533.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.