Analysis of the clinical efficacy of robot-assisted left total knee arthroplasty performed by surgeons with varying levels of experience

Abstract

This study sought to compare the clinical outcomes of left total knee arthroplasty performed under robotic assistance by right-handed surgeons of differing experience levels, and to assess how such technology may improve operative precision and mitigate experience-related variations in results. A retrospective review was performed for 120 patients suffering from left knee osteoarthritis, all of whom received total knee arthroplasty from December 2020 to June 2025. Patients were divided into Group A (senior surgeons, > 10 years of experience) and Group B (junior surgeons, < 2 years of experience), each comprising 60 patients. Each group was further subdivided into robot-assisted total knee arthroplasty (RTKA) and conventional total knee arthroplasty (CTKA) subgroups, with 30 patients in each. Recorded parameters included operative time, incision length, intraoperative blood loss, postoperative hip-knee-ankle (HKA) angle, sagittal and coronal tibial and femoral component angles (LTC, FTC, LFC, FFC), as well as Knee Society Score (KSS), visual analogue scale (VAS) for pain, range of motion (ROM), and Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) scores on postoperative days 3 and 90. Complication rates were also compared. The results showed that while senior surgeons had shorter operative times and better early radiographic outcomes in conventional TKA, robotic assistance significantly improved component alignment and early clinical scores for junior surgeons, resulting in no statistically significant inter-group differences. in both radiographic and clinical outcomes. Notably, in the RTKA subgroups, there were no statistically significant disparities noted in operative outcomes when comparing procedures performed by senior and junior surgeons in any intraoperative, radiographic, or clinical parameters. Additionally, junior surgeons performing conventional left TKA had a higher incidence of femoral anterior cortical notching (NOTCH). In conclusion, robot-assisted left total knee arthroplasty enhances procedural consistency and reduces the performance gap between surgeons of different experience levels, particularly aiding junior right-handed surgeons in overcoming technical challenges associated with non-dominant side surgery, thereby showing potential to promote greater consistency in surgical outcomes and patient safety.

Introduction

Total knee arthroplasty (TKA) represents the standard surgical treatment for advanced knee osteoarthritis [1]. Its long-term success, however, depends critically on precise implant positioning and correct restoration of lower limb mechanical alignment. Achieving these technical objectives is inherently subject to the operating surgeon’s proficiency and accumulated experience. Studies have shown that a surgeon’s level of experience is significantly correlated with the precision of component positioning, the incidence of complications, and patient functional recovery following TKA. Surgeons with greater experience typically possess advantages in alignment control and soft tissue balancing due to their accumulated expertise [2]. Beyond experience, another potential technical variable affecting operative precision is the congruence between the surgeon’s hand dominance and the side of the operated limb [3]. Currently, the majority of surgeons are right-handed. When performing left-sided TKA, they often encounter the so-called “non-dominant side” challenge due to differences in the surgical viewing angle, the tactile feedback during instrument manipulation, and habitual motor patterns. The literature indicates that during left-sided TKA, right-handed surgeons may exhibit a greater tendency for malpositioning of femoral or tibial components in either the coronal or sagittal plane, which could subsequently influence postoperative knee biomechanics and clinical function scores [4]. This technical disparity arising from side adaptation may be more pronounced among less experienced junior surgeons, potentially further widening the gap in TKA outcomes between them and their senior counterparts.

In recent years, with the increasing adoption and widespread application of robotics in orthopaedic surgery, the technology for robot-assisted TKA has matured significantly. Robot-assisted TKA utilises personalised preoperative planning and intraoperative real-time tracking and navigation to aid surgeons in achieving precise osteotomy and prosthesis implantation, thereby reducing the absolute reliance on the surgeon’s personal experience and spatial judgement [5]. Consequently, in theory, robot-assisted TKA has the potential to mitigate the variability in surgical outcomes attributable to surgeon hand dominance and differences in experience level. Current reports suggest that robot-assisted TKA can shorten the learning curve for junior surgeons, lower the technical barrier for performing TKA, and promote the homogenisation of surgical outcomes [6, 7]. However, in clinical practice, whether robot-assisted TKA can achieve homogeneity in surgical outcomes among surgeons with different experience levels and reduce the disparity in clinical efficacy when right-handed surgeons of varying seniority perform left-sided TKA remains inadequately investigated, with a lack of targeted research reports. Therefore, this study retrospectively analyses the clinical efficacy of robot-assisted left-sided TKA performed by two right-handed surgeons with different levels of experience in our department. The findings are reported as follows.

Materials and methods

Case selection

In this study, the senior surgeon group (Group A) consisted of three attending orthopedic surgeons, each with over 10 years of independent TKA experience. The junior surgeon group (Group B) consisted of three orthopedic fellows in their final year of training, each with less than 2 years of experience in performing TKA as the primary operator under supervision. All participating surgeons were self-reported right-handed individuals who exclusively use their right hand for writing and precision tool manipulation. All six surgeons underwent a standardized, simulator-based training program on the robotic system prior to performing any clinical RTKA cases included in this study.

Inclusion Criteria: (1) End-stage knee osteoarthritis (Kellgren-Lawrence grade IV) with unsatisfactory outcomes following standard conservative treatment. (2) All patients underwent primary left-sided TKA performed by a right-handed surgeon. (3) Complete clinical data and the ability to comply with follow-up. Exclusion Criteria: (1) Patients who underwent unicompartmental knee arthroplasty or right-sided TKA. (2) Patients diagnosed with non-osteoarthritic conditions such as rheumatoid arthritis or post-traumatic arthritis. (3) Patients who underwent contralateral knee arthroplasty during the study period. (4) Patients who were lost to follow-up or did not adhere to the follow-up protocol. (5) Severe flexion contracture (> 15°), external malformations or substantial bone defects requiring augmentation were exclusion criteria for this study.

Clinical data

A cohort of 120 patients diagnosed with left-sided knee osteoarthritis and undergoing total knee arthroplasty between December 2020 and June 2025 constituted the study population. Based on the surgeon’s experience, patients were divided into Group A (senior surgeon group: >10 years of experience) and Group B (junior surgeon group: < 2 years of experience), with 60 patients in each group. Within each group, patients were further subdivided based on the use of robotic assistance into a robot-assisted total knee arthroplasty (RTKA) subgroup and a conventional total knee arthroplasty (CTKA) subgroup, each comprising 30 patients. All patients underwent preoperative standing long-leg radiographs and lateral knee radiographs. The preoperative Hip-Knee-Ankle (HKA) angle was measured to quantify coronal plane deformity. The mean preoperative HKA angle was comparable between Group A and Group B (Varus deformity: 172.5° ± 5.2° vs 172.8° ± 4.8°), as well as between the RTKA and CTKA subgroups within each experience group (P > 0.05). The two groups demonstrated comparable baseline characteristics, with no statistically significant differences observed (P > 0.05) (Table 1). All patients in this study were informed and voluntarily participated, having provided signed informed consent. A post-hoc power analysis based on the primary outcome of postoperative HKA angle (SD ≈ 2.0°, clinically meaningful difference = 1.5°) indicated that a sample size of 30 per subgroup provided > 80% power at α = 0.05. While this may be underpowered for detecting differences in low-frequency complications, it is adequate for comparing alignment and early functional scores.

Table 1.

Comparison of general data between the two patient groups

Parameter	A Group (n = 60)	B Group (n = 60)	t/χ2	P
Age (years, x̅ ± s)	67.8 ± 6.9	66.5 ± 7.3	1.059	0.291
RTKA Group (n = 30)	68.1 ± 7.0	66.0 ± 7.5	1.121	0.267
CTKA Group (n = 30)	67.5 ± 6.9	67.0 ± 7.2	0.273	0.786
P	0.752	0.588	-	-
Sex (Male/Female)	22/38	25/35	0.301	0.583
RTKA Group (n = 30)	11/19	12/18	0.067	0.796
CTKA Group (n = 30)	11/19	13/17	0.139	0.709
P	1	0.796	-	-
BMI (kg/m², x̅ ± s)	26.4 ± 3.1	27.0 ± 3.5	−1.020	0.310
RTKA Group (n = 30)	26.2 ± 3.0	26.8 ± 3.4	−0.735	0.466
CTKA Group (n = 30)	26.6 ± 3.2	27.2 ± 3.6	−0.695	0.490
P	0.618	0.653	-	-
KL Grade (Grade III/IV)	18/42	15/45	0.417	0.518
RTKA Group (n = 30)	9/21	7/23	0.287	0.592
CTKA Group (n = 30)	9/21	8/22	0.071	0.789
P	1	0.771	-	-
Preoperative KSS (points, x̅ ± s)	48.3 ± 10.2	46.8 ± 11.5	0.780	0.437
RTKA Group (n = 30)	48.8 ± 10.5	47.5 ± 11.8	0.461	0.647
CTKA Group (n = 30)	47.8 ± 10.0	46.1 ± 11.3	0.610	0.544
P	0.695	0.628	-	-
Preoperative VAS (points, x̅ ± s)	6.5 ± 1.3	6.7 ± 1.5	−0.805	0.422
RTKA Group (n = 30)	6.4 ± 1.4	6.6 ± 1.6	−0.536	0.594
CTKA Group (n = 30)	6.6 ± 1.3	6.8 ± 1.4	−0.575	0.567
P	0.552	0.602	-	-
Preoperative ROM (degrees, x̅ ± s)	102.5 ± 14.8	99.8 ± 16.2	0.971	0.333
RTKA Group (n = 30)	103.2 ± 15.1	100.5 ± 16.5	0.671	0.505
CTKA Group (n = 30)	101.8 ± 14.6	99.1 ± 16.0	0.670	0.506
P	0.706	0.731	-	-
Preoperative WOMAC (points, x̅ ± s)	58.6 ± 9.4	60.1 ± 10.7	−0.836	0.405
RTKA Group (n = 30)	58.0 ± 9.8	59.4 ± 10.5	−0.542	0.590
CTKA Group (n = 30)	59.2 ± 9.1	60.8 ± 11.0	−0.594	0.555
P	0.607	0.581	-	-

Surgical technique

Standardized Operative Setting: To control for ergonomic variables, a standardized operating room setup was employed for all left TKA procedures. The patient was positioned supine with a tourniquet on the proximal thigh. A lateral post and a foot roll were used for stabilization. The operating table height was adjusted to a standard level for all surgeons. For both CTKA and RTKA, the primary surgeon consistently stood on the right side of the patient (ipsilateral to their dominant hand) to approach the left knee. Standard manufacturer-provided cutting guides were used in the CTKA procedures. The degree of knee flexion for femoral and tibial preparation followed the standard steps recommended by the prosthesis manufacturer and the robotic system protocol.

To minimize chronological bias, both RTKA and CTKA procedures were performed concurrently during the study period. The decision to use robotic assistance was primarily based on device availability and patient consent after explaining both techniques, rather than a temporal shift in standard practice. Furthermore, the senior and junior surgeons performed both types of procedures throughout the study duration, ensuring that the learning curve for RTKA was distributed across the timeline and not confined to its introduction phase. The anaesthetic method employed for all patients was general anaesthesia. A pneumatic tourniquet was applied during the procedure. Joint access was achieved through a standard midline skin incision followed by a medial parapatellar arthrotomy. Patellar denervation was performed in all cases. Posterior-stabilised (PS) knee prostheses were implanted in all patients. RTKA Group: Following joint exposure, femoral and tibial trackers were installed. Registration of the various reference points was performed sequentially according to the robotic system’s prompts. Following registration, the surgical plan was verified and adjusted if necessary. After confirmation of the plan, robotic-assisted bone resection was carried out. Trial components were then inserted to assess joint stability. Upon achieving satisfactory stability, the joint cavity was irrigated prior to the implantation of the final prosthetic components. CTKA Group: Following joint exposure, the surgical procedure was performed sequentially, including distal femoral osteotomy, multi-plane femoral osteotomy using appropriately sized four-in-one cutting guides, intercondylar notch osteotomy, and tibial plateau osteotomy. After completion of the osteotomies, trial components were inserted to assess knee stability and flexion-extension gaps. Upon satisfactory assessment, the joint cavity was irrigated, and the definitive knee prosthesis components were implanted.

Postoperative management postoperative

Care followed a uniform protocol including a sequential anti-coagulation regimen with tranexamic acid for thrombosis prophylaxis, alongside anti-inflammatory and analgesic medications as part of perioperative management. On postoperative day one, standard anteroposterior and lateral knee radiographs, as well as standing long-leg alignment radiographs, were obtained. Relevant angles were measured and recorded from these images. Functional rehabilitation of the knee joint was initiated under the guidance of a specialised physiotherapist. Upon discharge, patients were instructed to return for regular follow-up appointments.

Observation indicators

Parameters recorded for each group included operative time, incision length, and intraoperative blood loss. Radiographic measurements comprised the postoperative hip-knee-ankle (HKA) angle, sagittal tibial component angle (LTC), coronal tibial component angle (FTC), sagittal femoral component angle (LFC), and coronal femoral component angle (FFC). Clinical outcomes assessed at postoperative day 3 and day 90 included the Knee Society Score (KSS), visual analogue scale (VAS) for pain, range of motion (ROM) of the knee, and the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) score. Complications were also recorded for each group. Radiographic Measurement Protocol and Reliability: All postoperative radiographic measurements (HKA, LFC, FFC, LTC, FTC) were performed independently by two orthopedic surgeons who were not involved in the surgical procedures and were blinded to the surgeon’s identity (experience group) and the surgical technique (RTKA or CTKA). Measurements were conducted using the hospital’s Picture Archiving and Communication System (PACS) with standard digital measurement tools. To assess inter-observer reliability, the intraclass correlation coefficient (ICC) was calculated for each parameter based on the measurements from both evaluators. The ICCs ranged from 0.85 to 0.93, indicating excellent agreement. Any measurement discrepancy greater than 2° was reviewed jointly, and a consensus value was reached and used for final analysis.

Statistical analysis

Statistical analysis was performed using SPSS 26.0 and R 4.3.1 software. Measurement data were verified to follow a normal distribution by Shapiro-Wilk test and expressed as mean ± standard deviation (x̄ ± s). Between-group comparisons were conducted using the independent samples t-test, with results reported as mean difference (Mean Difference, MD) and its 95% confidence interval (95% CI). Categorical data were presented as case counts (percentage), and between-group comparisons were performed using the χ² test or Fisher’s exact probability method. Given that this study involved multi-level, multi-index intergroup comparisons (e.g., Group A vs. Group B, intra-group comparisons of RTKA vs. CTKA, inter-group comparisons of CTKA, and inter-group comparisons of RTKA), to control for the risk of type I error (false positive) inflation due to multiple comparisons, the P-values for the primary study endpoints (imaging and clinical functional indicators) were corrected for false discovery rate (FDR) using the Benjamini-Hochberg procedure. The corrected P-values, denoted as Padj, were used in the main text and tables, with Padj < 0.05 considered statistically significant.

Results

Comparison of intraoperative indicators

The total operative time and incision length in Group A were significantly shorter than those in Group B (P < 0.05). Nevertheless, intraoperative blood loss did not differ significantly between the two groups (P > 0.05). Within-group comparisons revealed that the operative time and incision length in the RTKA subgroups were significantly longer, while the intraoperative blood loss was significantly less, compared to the respective CTKA subgroups (P < 0.05). Nevertheless, when analyzing the robot-assisted TKA subgroups separately, operative duration, incision length, and intraoperative blood loss were comparable between patients whose surgery was performed by Group A and Group B surgeons (P > 0.05). In the conventional TKA subgroups, the mean operative time was significantly shorter for patients operated on by Group A surgeons compared to those in Group B (P < 0.05, Table 2).

Table 2.

Comparison of operative Time, incision Length, and intraoperative blood loss between the two patient groups

Parameter	A Group (n = 60)	B Group (n = 60)	t/χ2	P	MD (95% CI)	Padj
Operative Time (min, x̅ ± s)	105.6 ± 24.1	120.8 ± 26.3	−3.275	< 0.001	−15.2 (−23.8, −6.6)*	0.002*
RTKA Group (n = 30)	130.5 ± 11.2	142.5 ± 12.6	−1.924	0.060	−12.0 (−24.8, 0.8)	0.068
CTKA Group (n = 30)	80.7 ± 8.5	99.1 ± 10.8	−7.135	< 0.001	−18.4 (−23.1, −13.7)**	< 0.001**
P	0.001	0.001	-	-	-	-
Incision Length (cm, x̅ ± s)	13.9 ± 1.1	14.8 ± 1.0	−4.729	< 0.001	−0.9 (−1.3,−0.5)**	< 0.001**
RTKA Group (n = 30)	15.0 ± 0.7	15.5 ± 0.6	−1.667	0.102	−0.5 (−1.1, 0.1)	0.104
CTKA Group (n = 30)	12.8 ± 0.8	14.1 ± 1.0	−1.946	0.057	−1.3 (−2.6, 0.0)	0.051
P	0.001	0.001	-	-	-	-
Intraoperative Blood Loss (ml, x̅ ± s)	94.5 ± 34.8	100.2 ± 39.1	−0.834	0.406	−5.7 (−19.3, 7.9)	0.413
RTKA Group (n = 30)	68.2 ± 18.5	72.5 ± 21.0	−0.858	0.394	−4.3 (−14.2, 5.6)	0.394
CTKA Group (n = 30)	120.8 ± 24.3	127.9 ± 27.5	−1.098	0.276	−7.1 (−20.0, 5.8)	0.279
P	0.001	0.001	-	-	-	-

MD: mean difference; CI: confidence interval; Padj: P-value corrected by the Benjamini-Hochberg method. **Significance markers: *Padj < 0.05, **Padj < 0.01, *Padj < 0.001

Postoperative radiographic indicators

The LFC, LTC, and HKA angles in Group A patients were superior to those in Group B patients (P < 0.05). No statistically significant differences were detected, however, in either the FFC or LFC alignment angles (P > 0.05). Within-group comparisons showed that the HKA, LFC, FFC, LTC, and FTC angles in the RTKA subgroups were superior to those in the respective CTKA subgroups (P < 0.05). Between-surgeon comparisons within the CTKA subgroups revealed that the LFC, LTC, and HKA angles for Group A surgeons were superior to those for Group B surgeons (P < 0.05), while no significant differences were observed in the FFC and LFC angles (P > 0.05). Conversely, among patients who underwent robot-assisted TKA, all measured alignment angles showed no statistically significant differences between the procedures performed by Group A and Group B surgeons (P > 0.05, Table 3).

Table 3.

Comparison of postoperative LFC, FFC, LTC, FTC, and HKA angles between the two patient groups

Parameter	A Group (n = 60)	B Group (n = 60)	t/χ2	P	MD (95% CI)	Padj
HKA (°, x̅ ± s)	178.5 ± 1.8	177.2 ± 2.3	3.512	< 0.001	1.3 (0.6, 2.0)**	0.001**
RTKA Group (n = 30)	179.2 ± 1.2	178.9 ± 1.4	0.923	0.361	0.3 (−0.3, 0.9)	0.361
CTKA Group (n = 30)	177.8 ± 1.9	175.5 ± 2.1	4.512	< 0.001	2.3 (1.3, 3.3)***	< 0.001***
P	0.001	0.001	-	-	-	-
FFC (°, x̅ ± s)	88.7 ± 1.7	88.4 ± 2.0	0.889	0.376	0.3 (−0.4, 1.0)	0.376
RTKA Group (n = 30)	89.2 ± 1.3	88.9 ± 1.5	0.802	0.426	0.3 (−0.4, 1.0)	0.426
CTKA Group (n = 30)	88.2 ± 1.9	87.9 ± 2.2	0.595	0.554	0.3 (−0.7, 1.3)	0.554
P	0.021	0.046	-	-	-	-
LFC (°, x̅ ± s)	90.5 ± 1.6	89.6 ± 1.9	2.813	0.006	0.9 (0.3, 1.5)**	0.008**
RTKA Group (n = 30)	91.0 ± 1.1	90.8 ± 1.3	0.667	0.508	0.2 (−0.4, 0.8)	0.508
CTKA Group (n = 30)	90.0 ± 1.8	88.4 ± 1.7	3.902	< 0.001	1.6 (0.8, 2.4)**	< 0.001**
P	0.009	0.001	-	-	-	-
LTC (°, x̅ ± s)	89.2 ± 1.5	88.3 ± 1.9	2.934	0.004	0.9 (0.3, 1.5)**	0.005**
RTKA Group (n = 30)	89.8 ± 0.9	89.6 ± 1.1	0.802	0.426	0.2 (−0.3, 0.7)	0.426
CTKA Group (n = 30)	88.6 ± 1.6	87.0 ± 1.8	3.750	< 0.001	1.6 (0.7, 2.5)**	< 0.001**
P	0.001	0.001	-	-	-	-
FTC (°, x̅ ± s)	89.8 ± 1.4	89.5 ± 1.7	1.125	0.263	0.3 (−0.2, 0.8)	0.263
RTKA Group (n = 30)	90.1 ± 1.0	89.9 ± 1.2	0.667	0.508	0.2 (−0.3, 0.7)	0.508
CTKA Group (n = 30)	89.5 ± 1.6	89.1 ± 1.9	0.952	0.345	0.4 (−0.4, 1.2)	0.345
P	0.049	0.047	-	-	-	-

MD: mean difference; CI: confidence interval; Padj: P-value corrected by the Benjamini-Hochberg method. **Significance markers: *Padj < 0.05, **Padj < 0.01, *Padj < 0.001

Postoperative clinical indicators

At postoperative day 3, Group A patients exhibited significantly lower VAS and WOMAC scores, and significantly higher KSS scores and ROM, compared to Group B patients (P < 0.05). However, it’s not by postoperative day 90 (P > 0.05). Within-group comparisons showed that at both postoperative day 3 and day 90, patients in the RTKA subgroups had significantly better VAS scores, KSS scores, WOMAC scores, and ROM than those in the corresponding CTKA subgroups (P < 0.05). Between-surgeon comparisons within the CTKA subgroups revealed that, at postoperative day 3, patients operated on by Group A surgeons had significantly lower VAS and WOMAC scores, and significantly higher KSS scores and ROM, compared to those operated on by Group B surgeons (P < 0.05). It’s not by postoperative day 90 (P > 0.05). By contrast, in the robot-assisted TKA subgroups, clinical scores at both postoperative day 3 and 90 did not differ significantly between the patients treated by surgeons from Group A and those from Group B (P > 0.05; Table 4).

Table 4.

Comparison of VAS Score, KSS, WOMAC Score, and ROM at postoperative day 3 and day 90 between the two patient groups

Parameter	Group A (n = 60)	Group B (n = 60)	t/χ2	P	MD (95% CI)	Padj
VAS Score at POD 3 (points, x̅ ± s)	3.2 ± 0.8	3.8 ± 0.9	−4.000	0.001	−0.6 (−0.9,−0.3)	< 0.001
RTKA Group (n = 30)	2.8 ± 0.7	3.0 ± 0.8	−1.000	0.321	−0.2 (−0.6, 0.2)	0.321
CTKA Group (n = 30)	3.6 ± 0.7	4.6 ± 0.7	−5.657	0.001	−1.0 (−1.4,−0.6)	< 0.001
P	0.001	0.001	-	-	-	-
VAS Score at POD 90 (points, x̅ ± s)	1.5 ± 0.6	1.6 ± 0.7	−0.877	0.382	−0.1 (−0.3, 0.1)	0.382
RTKA Group (n = 30)	1.3 ± 0.5	1.4 ± 0.6	−0.707	0.482	−0.1 (−0.4, 0.2)	0.482
CTKA Group (n = 30)	1.7 ± 0.6	1.8 ± 0.7	−0.599	0.551	−0.1 (−0.4, 0.2)	0.551
P	0.006	0.018	-	-	-	-
KSS at POD 3 (points, x̅ ± s)	68.5 ± 6.2	64.0 ± 7.1	3.784	0.001	4.5 (2.2, 6.8)	< 0.001
RTKA Group (n = 30)	72.0 ± 5.5	70.5 ± 5.8	1.034	0.306	1.5 (−1.4, 4.4)	0.306
CTKA Group (n = 30)	65.0 ± 5.2	57.5 ± 5.0	5.657	0.001	7.5 (5.0, 10.0)	< 0.001
P	0.001	0.001	-	-	-	-
KSS at POD 90 (points, x̅ ± s)	85.5 ± 4.8	84.5 ± 5.5	1.042	0.300	1.0 (−0.9, 2.9)	0.300
RTKA Group (n = 30)	87.0 ± 4.0	86.2 ± 4.2	0.745	0.460	0.8 (−1.3, 2.9)	0.460
CTKA Group (n = 30)	84.0 ± 4.9	82.8 ± 5.8	0.924	0.359	1.2 (−1.3, 3.7)	0.359
P	0.012	0.012	-	-	-	-
WOMAC Score at POD 3 (points, x̅ ± s)	35.2 ± 6.0	40.1 ± 7.2	−4.142	0.001	−4.9 (−7.3,−2.5)	< 0.001
RTKA Group (n = 30)	32.0 ± 5.5	34.5 ± 6.2	−1.639	0.107	−2.5 (−5.6, 0.6)	0.107
CTKA Group (n = 30)	38.4 ± 4.8	45.7 ± 5.5	−5.292	0.001	−7.3 (−10.0,−4.6)	< 0.001
P	0.001	0.001	-	-	-	-
WOMAC Score at Postoperative Day 90 (points, x̅ ± s)	20.5 ± 5.1	21.8 ± 5.8	−1.300	0.196	−1.3 (−3.3, 0.7)	0.196
RTKA Group (n = 30)	18.8 ± 4.5	20.0 ± 5.0	−0.943	0.350	−1.2 (−3.8, 1.4)	0.350
CTKA Group (n = 30)	22.2 ± 5.0	23.6 ± 6.0	−0.998	0.322	−1.4 (−4.2, 1.4)	0.322
P	0.007	0.018	-	-	-	-
ROM at Postoperative Day 3 (°, x̅ ± s)	95.2 ± 8.5	90.1 ± 9.8	3.902	0.001	5.1 (2.0, 8.2)	0.002
RTKA Group (n = 30)	98.5 ± 7.0	96.8 ± 7.5	3.125	0.002	1.7 (−1.9, 5.3)	0.360
CTKA Group (n = 30)	91.9 ± 8.2	83.4 ± 8.5	0.923	0.360	8.5 (4.4, 12.6)	< 0.001
P	0.001	0.001	-	-	-	-
ROM at Postoperative Day 90 (°, x̅ ± s)	115.8 ± 6.0	114.5 ± 7.2	1.099	0.274	1.3 (−1.0, 3.6)	0.274
RTKA Group (n = 30)	117.5 ± 5.0	116.8 ± 5.5	0.525	0.601	0.7 (−2.0, 3.4)	0.601
CTKA Group (n = 30)	114.1 ± 6.2	112.2 ± 7.5	1.089	0.280	1.9 (−1.6, 5.4)	0.280
P	0.021	0.007	-	-	-	-

MD: mean difference; CI: confidence interval; Padj: P-value corrected by the Benjamini-Hochberg method. **Significance markers: *Padj < 0.05, **Padj < 0.01, *Padj < 0.001

Complications in both patient groups

No severe complications such as acute infection or neurovascular injury occurred in either group. Postoperative complications primarily included femoral anterior cortical notching (NOTCH), local wound exudate, and muscular calf vein thrombosis of the operated limb. In both groups, the RTKA subgroups each had one case of the aforementioned complications. In the CTKA subgroup of Group A, there were 2 cases of NOTCH, 1 case of local wound exudate, and 1 case of muscular calf vein thrombosis. In the CTKA subgroup of Group B, there were 4 cases of NOTCH, 1 case of local wound exudate, and 1 case of muscular calf vein thrombosis. The overall incidence of complications did not differ significantly between the two groups (P > 0.05). However, patients in the CTKA subgroup of Group B exhibited a higher incidence of femoral anterior cortical notching.

Discussion

The impact of surgeon experience level on TKA outcomes

It is well established that a surgeon’s experience and technical skill significantly influence surgical outcomes, with experience being closely linked to the surgeon’s level of seniority [8]. In TKA, due to the multitude of factors involved—including the placement angles of the femoral and tibial components, osteotomy angles and depths, and soft tissue balancing—achieving favourable surgical outcomes under conventional techniques is highly dependent on the surgeon’s experience [9]. Numerous studies have confirmed that surgical experience is a crucial variable affecting TKA results. Surgeons with greater seniority or higher surgical volumes typically achieve better restoration of lower limb alignment, lower complication rates, and superior early functional outcomes [10, 11]. In this study, within the conventional TKA group, the senior surgeon achieved superior HKA, LFC, and LTC angles, along with better VAS and KSS scores at postoperative day 3. This highlights the importance of the senior surgeon’s experience in achieving optimal prosthesis positioning and restoring lower limb alignment, and further demonstrates their superior control over soft tissue balancing in the knee joint [12], as evidenced by the advantageous VAS and KSS scores at postoperative day 3. However, in this study, the advantage conferred by experience was no longer significant by postoperative day 90. This may be attributed to factors such as ongoing patient rehabilitation and self-adaptation over time, which gradually diminish the early impact of technical disparities; this observation aligns with findings from some long-term follow-up studies [13]. Nonetheless, it is important to note that our study identified a direct risk associated with inexperience manifesting in specific complications, namely a significantly higher incidence of NOTCH in the conventional surgery subgroup performed by the junior surgeon compared to the senior surgeon group. This is likely related to the junior surgeon’s relative lack of experience in selecting osteotomy angles and prosthesis sizes [14, 15]. These findings support the view of Sassoon et al. [16] that osteotomy errors are a common type of mistake during the learning curve, underscoring the importance of initial training and supervision.

The impact of surgeon hand dominance on surgery performed on the non-dominant side

Numerous studies have reported that right-handed surgeons can experience disadvantages when operating on their non-dominant side (the left side), a phenomenon particularly pronounced in orthopaedic surgery where high-precision spatial orientation and instrument manipulation are crucial [17]. Research indicates that during arthroscopic surgery, a surgeon’s precision and efficiency in instrument handling decrease when performing non-dominant side procedures, resulting in a steeper learning curve [18]. A multi-centre study by Jaglarz et al. [19] found that when right-handed surgeons performed left-sided TKA while standing on the patient’s right side, the deviation in the hip-knee-ankle angle was significantly greater compared to right-sided procedures, with corresponding increases in the rates of abnormal mechanical axis valgus and varus angles. Some scholars have reported [20] that when right-handed surgeons perform left-sided TKA, the placement angle of the femoral component in the sagittal plane (LFC) may differ from that in right-sided surgeries, a discrepancy potentially attributable to changes in the surgeon’s standing position and instrument grip habits. Other studies suggest that non-dominant side surgery may affect the surgeon’s accuracy in judging tibial rotational alignment [21, 22]. Furthermore, research suggests that the biomechanical habits and anatomical perspective differences stemming from hand dominance constitute a potential technical variable affecting joint replacement surgery, particularly TKA, which demands a high degree of symmetrical operation [23]. In this study, the observed difference in the incidence of NOTCH between the junior and senior surgeon groups may be attributed to the following: when performing this procedure on the left side, the visual field, the angle for holding the cutting guide, and the trajectory of the oscillating saw differ from a right-handed surgeon’s ingrained muscle memory and habitual patterns. Junior surgeons, having not yet developed a stable and adaptable operative system, find it more difficult to compensate for the discomfort associated with this “mirrored operation.” Consequently, they are more prone to osteotomy errors such as excessive depth or anterior translation, leading to NOTCH. This aligns with the findings of Bäthis et al. [24] in spinal surgery, who observed that hand dominance affects instrument trajectory, with a more pronounced impact on junior surgeons. Therefore, our study not only demonstrates the challenge posed by hand dominance in TKA but also reveals a significant interaction between this challenge and surgeon experience level—namely, that insufficient experience amplifies the technical risks associated with non-dominant side surgery.

Robot-assisted surgery contributing to a reduction in outcome variability of surgical outcomes among surgeons of different experience levels

With the continuous development of artificial intelligence, the application of robot-assisted technology in biology [25] and surgery has gradually matured. Multiple studies have confirmed that, compared to conventional TKA, RTKA offers significant advantages in achieving predetermined alignment targets and reduces outliers [26–28]. In this study, while differences in postoperative radiographic parameters were observed between patients operated on by senior and junior surgeons using conventional techniques, these differences became non-significant with the use of robotic assistance. This finding robustly supports the notion of robotic systems acting as “experience equalisers.” The mechanism lies in the robotic system placing the critical steps of surgery—from 3D planning to osteotomy execution—within an objective, data-driven framework. Preoperative planning determines a personalised osteotomy plan, while intraoperative real-time navigation and boundary control ensure the accuracy of its execution. This significantly reduces the absolute reliance on the surgeon’s intraoperative spatial judgement and the stability of freehand techniques. Consequently, as demonstrated in this study, it effectively assists junior surgeons in overcoming the barrier of limited experience and helps them consistently navigate the challenges associated with “non-dominant side” operations. Ultimately, this enables surgeons of different experience levels to achieve highly consistent and precise surgical outcomes for their patients [29–32]. This provides a practical and feasible technological pathway towards the standardisation and homogenisation of surgical care quality.

Limitations of this study

Firstly, this study is a single-centre retrospective analysis with a limited sample size. Future validation through multicentre, prospective, large-sample studies is warranted. Although three surgeons were included per group to mitigate the operator effect, residual inter-individual skill variance within an experience level cannot be entirely ruled out. For low-frequency complications such as the ‘anterior femoral cortical notch (NOTCH)’, there may be issues with insufficient statistical power, and the observed trends require confirmation through studies with larger sample sizes. Secondly, the follow-up duration is relatively short; the impact of robotic assistance on the long-term survivorship of the prostheses requires investigation through longer-term follow-up. However, we acknowledge that more detailed assessment of sagittal and rotational deformities was not routinely quantified. Thirdly, all surgeons in this study were from a single surgical centre. The relationship between seniority and level of experience may vary across different centres. Therefore, the generalisability of our findings needs to be further confirmed by multicentre studies. Fourth, this study did not evaluate economic factors. As our data confirm, RTKA required longer operative times, which has implications for operating room throughput and direct costs. The additional expenses associated with robotic system acquisition, maintenance, and consumables were also not analyzed. A comprehensive cost-effectiveness analysis, balancing the improved precision and potential reduction in revision surgery against these increased upfront costs, is crucial for widespread adoption and was beyond the scope of this clinical efficacy study. Future research incorporating long-term survivorship data and detailed health economic models is warranted to fully assess the value proposition of robotic assistance in this specific context.

In summary, while confirming the challenges posed by surgeon experience disparity and “non-dominant side” surgery in conventional TKA, this study is the first, within a comparative framework, to systematically demonstrate the superior efficacy of robot-assisted TKA in mitigating these disparities. The findings indicate that robot-assisted TKA not only universally enhances surgical precision and safety but also significantly reduces outcome variability stemming from differences in surgeon experience and habits. This holds significant importance for promoting the homogenisation of surgical technique and ensuring patient safety.

Abbreviations

CTKA

Conventional Total Knee Arthroplasty

HKA

Hip-Knee-Ankle

LTC

Lateral Tibial Component

FTC

Frontal Tibial Component

LFC

Lateral Femoral Component

FFC

Frontal Femoral Component

KSS

Knee Society Score

VAS

Visual Analogue Scale

ROM

Range of Motion

WOMAC

Western Ontario and McMaster Universities Osteoarthritis Index

BMI

Body Mass Index

NOTCH

Femoral Anterior Cortical Notching

PS

Posterior-Stabilised (knee prosthesis)

CONSORT

Consolidated Standards of Reporting Trials

Author contributions

HPW and MYW managed perioperative care, conducted follow-up, and authored the initial manuscript. YPL, HX, SJL, HPW, XZS and MYW oversaw all preoperative planning and surgical procedures, and assisted in revising the article. Patient follow-up was managed by XQY, XZS, and GCM. All authors read and approved the final manuscript.

Funding

This research was funded by Sichuan Provincial Primary Health Development Research Center in 2024, North Sichuan Medical College (No. SWFZ24-Q-86), Sichuan Provincial Primary Health Development Research Center in 2025, North Sichuan Medical College (No. SWFZ25-Y-35), Chengdu High-tech Medical Association “2023 annual Flurbiprofen gel paste treatment of osteoarthritis special research fund” project (2024013), Sichuan Provincial Rehabilitation Medical Association’s 2025 Annual Scientific Research Projects (No.SCKFKY20250210), 2025 Annual Science and Technology Bureau of Panzhihua City project (2025ZD-S-2) and 2024 Annual Science and Technology Bureau of Panzhihua City project (2024ZD-S-87) who is not involved in study design, data collection, analysis and interpretation or manuscript preparation.

Data availability

The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.

Ethics approval and consent to participate

This study was approved by The Institutional Review Board of Panzhihua Central Hospital (Pankelun Trial No. [2024-08]) and written informed consent to participate was obtained from all of the individual participants included in the study.

Consent for publication

Written informed consent was obtained from the patient for publication of this paper.