Skip to main content
Springer logoLink to Springer
. 2026 Mar 12;20(1):346. doi: 10.1007/s11701-026-03162-6

Robot-assisted total knee arthroplasty: adverse events related to the robotic system

Xunzhou Song 1,#, Mingyou Wang 1,#, Hongping Wang 1, Zhuodong Tang 1, Yuping Lan 1,✉
PMCID: PMC12979380  PMID: 41813935

Abstract

Robot-assisted total knee arthroplasty (RA-TKA) aims to enhance surgical precision and reproducibility. However, technical malfunctions occasionally occur. This study analyses single-centre RA-TKA malfunction cases to explore causative factors and reflect on the relationship between surgeons and robotic assistance. A retrospective analysis was conducted on 188 consecutive RA-TKA procedures performed between January 2020 and June 2025 by the same senior arthroplasty consultant using five distinct robotic systems (Brainlab template-guided, HURWA automated alignment, Smith & Nephew CORI milling-drilling, BoneSage automated alignment, and Jianjia template-guided). Cases meeting predefined ‘malfunction’ criteria. Malfunction defined as: inability to complete critical steps according to the robot’s predetermined plan due to any cause, necessitating intraoperative abandonment of robotic assistance or major plan adjustments; or direct robot-related factors causing poor postoperative alignment (mechanical axis deviation > 3°) or abnormal soft tissue balance requiring early intervention. Systematic classification and statistical analysis of malfunction causes. Seventeen malfunction cases were categorised into three primary groups: (1) Robot malfunction (3 cases, 17.6%): intraoperative robotic system failure; (2) Technical malfunction (7 cases, 41.2%): loosening of positioning/reference frame (5 cases), misplacement of calibrator in osteotomy zone (2 cases); (3) Workflow and learning curve issues (2 cases, 11.8%): significant time prolongation and subsequent abandonment due to unfamiliarity with specific new system procedures (2 cases); (4) Conflict between surgeon decision-making and robotic planning (5 cases, 29.4%): surgeons performed unplanned mobilisation or osteotomy adjustments based on experience, ultimately resulting in failure to achieve anticipated balance/alignment. The success of RA-TKA relies on a reliable platform, standardised procedures, and human-machine collaboration. Clinical observations from 188 cases revealed: (1) 17 cases (9.0%) experienced robot-related malfunctions; (2) Different robotic systems exhibited distinct failure modes; (3) Malfunction rates were higher during the early learning curve phase; (4) All malfunctions were resolved through surgeon intervention without causing surgical failure. This indicates that technical malfunctions can be progressively mitigated through system optimisation and training, while intraoperative human-machine coordination and stepwise verification remain critical for surgical safety.

Keywords: Robot-assisted surgery, Total knee arthroplasty, Surgical malfunction, Complications, Learning curve, Intraoperative decision-making

Introduction

Total knee arthroplasty (TKA) is the gold standard procedure for treating end-stage knee osteoarthritis. Its long-term success rate is closely linked to the precise restoration of postoperative lower limb alignment, soft tissue balance, and accurate prosthesis positioning. Conventional TKA relies heavily on the surgeon’s experience and surgical instruments, with reported rates of postoperative alignment abnormalities (mechanical axis deviation > 3°) reaching 10–30%. This correlates with increased early revision rates [1] and reduced patient satisfaction.

To enhance surgical precision and reproducibility, robot-assisted surgical systems (RAS) have rapidly advanced in joint surgery. RA-TKA theoretically reduces human error through preoperative or intraoperative 3D planning, real-time dynamic optical tracking, and physical osteotomy guidance or spatial constraints, enabling precise, individualised osteotomy and soft tissue balance [2]. Multiple studies have confirmed RA-TKA’s superiority in improving prosthesis alignment accuracy [3, 4]. Nevertheless, its clinical implementation faces considerable challenges, including substantial costs, a distinct learning curve, potentially prolonged operative times [5], and distinct malfunction patterns compared to conventional techniques. Existing literature predominantly highlights RA-TKA’s advantages, with limited systematic analysis of real-world implementation difficulties or malfunction cases leading to unmet surgical objectives. Comparative studies examining differences in malfunction patterns between robotic systems employing distinct technical principles—such as active constraints versus passive guides, burrs versus oscillating saws, and CT-based modelling versus image-free modes—remain notably scarce.

This study aims to retrospectively analyse consecutive cases performed over the past five years by a senior arthroplasty surgeon using five distinct technical pathway RA-TKA systems. By thoroughly examining the characteristics, root causes, and intraoperative management strategies of malfunction cases, this study aims to reveal potential risks and pitfalls in the practical application of RA-TKA. It seeks to provide evidence-based insights for optimising surgical workflows, refining team training programmes, and ultimately achieving safer and more reliable human-machine collaboration.

Materials and methods

Study design and patient cohort

This study constitutes a single-centre retrospective case series analysis. Following approval by the hospital ethics committee, all RA-TKA cases performed between January 2020 and June 2025 by the corresponding author (a professor with over 25 years’ experience in joint surgery and an annual TKA caseload exceeding 200 procedures) were reviewed. Inclusion criteria: primary knee osteoarthritis or rheumatoid arthritis, undergoing initial RA-TKA. Exclusion criteria: prior history of periprosthetic osteotomy, severe bone defects requiring modular prostheses or augmentation blocks, or simultaneous multi-joint arthroplasty. Ultimately, 188 patients were included, of whom 17 were classified as malfunction cases.

Robotic systems

This study involved five distinct RA-TKA systems, listed chronologically by introduction:

System A (Guide-assisted + Reciprocating Saw Type, 2020): Brainlab system. ① Intraoperative optical modelling acquisition; ② Infrared tracking guided placement and fixation of handheld osteotomy guides; ③ Surgeon performing osteotomy via guide using conventional reciprocating saw. (Figures 1 and 2)

Fig. 1.

Fig. 1

Surgeon using a conventional oscillating saw to perform bone resection through a fixed guide block

Fig. 2.

Fig. 2

Osteotomy Plan and Anticipated Flexion-Extension Gap

System B (Automatic Alignment + Pendulum Saw Type, 2023): HURWA robotic system. ① Based on preoperative CT modelling; ② Robotic arm automatically navigates to osteotomy site with pendulum saw; ③ Surgeon performs osteotomy under spatial constraints of robotic arm while monitor displays real-time osteotomy position. (Figures 3 and 4)

Fig. 3.

Fig. 3

The surgeon awaits the robotic arm’s automated alignment and, upon successful alignment, manually initiates the oscillating saw to commence bone cutting

Fig. 4.

Fig. 4

Display showing real-time cutting position of the saw blade and the area already cut

System C (Automatic Alignment + Drill-and-Milling Type, 2025): Smith & Nephew CORI System. ① Modelling based on intraoperative optical data acquisition; ② Robotic arm provides active safety zone constraints while the surgeon controls drilling and milling for osteotomy; ③ High-speed (60,000 RPM) milling drill performs osteotomy, preventing cutting deviations caused by saw blade deflection and vibration. (Figures 5 and 6)

Fig. 5.

Fig. 5

Surgeon utilising a high-speed burr for bone cutting during surgery

Fig. 6.

Fig. 6

Osteotomy Plan

System D (Automatic Positioning + Pendulum Saw Type, 2025): Bone Saint Yuan Hua Robot. Operating on a principle similar to System B, the robotic arm automatically positions and assists the surgeon in performing osteotomy. (Figures 7 and 8)

Fig. 7.

Fig. 7

Following successful alignment, the surgeon manually activates the oscillating saw to perform bone cutting

Fig. 8.

Fig. 8

Osteotomy Planning Diagram

System E (Automatic Alignment + Guide Plate + Oscillating Saw Type, 2025): Jianjia Robotics. The principle is similar to System A, but the guide plate is held by a robotic arm, with the surgeon performing the osteotomy using a conventional oscillating saw. (Figures 9 and 10)

Fig. 9.

Fig. 9

Robot-guided automatic alignment of osteotomy guide plates under infrared guidance

Fig. 10.

Fig. 10

Osteotomy Planning Diagram

Surgical technique and data collection

All procedures employed the standard medial parapatellar approach. Registration, alignment, planning, and operation of the robotic system strictly adhered to each manufacturer’s recommended protocols. Intraoperative pneumatic tourniquets were utilised throughout. Prosthesis selection was based on patient anatomy and surgeon judgement, with all cases utilising posterior-stable cemented prostheses.

Definition of malfunction cases

This study defines ‘RA-TKA malfunction’ specifically as instances where the intraoperative robot-assisted process failed to proceed as planned, or where although the robotic surgery was completed, robot-related factors directly led to poor measurement outcomes. This includes:

① Intraoperative technical abandonment: Complete abandonment of the RA-TKA robot due to registration failure, loss of tracking, equipment malfunction, etc., resulting in conversion to manual TKA.

② Significant deviation from plan: Completion of RA-TKA osteotomy, but subsequent extensive unplanned additional osteotomy or soft tissue release performed by the surgeon due to dissatisfaction with balance or alignment, with postoperative imaging confirming mFTA > 3° deviation from neutral position.

③ Robot-directly related complications: Such as coordinate system abnormalities caused by loosening of tracers or reference frames, requiring immediate intraoperative correction.

Statistical analysis

Statistical analysis was performed using SPSS 26.0. Quantitative data are presented as mean ± standard deviation. Intergroup comparisons employed either t-tests or Mann-Whitney U tests based on data normality. Qualitative data are expressed as frequencies and percentages. A P value < 0.05 was considered statistically significant.

Results

Among 188 RA-TKA procedures, 17 cases (9.04%) met the study’s definition of ‘malfunction’. The mean patient age was 68.4 ± 7.2 years, with no significant gender distribution. While no absolute concentration of malfunction cases was observed across the five robotic systems, their occurrence correlated with the early implementation phase of each system (Group A, which commenced earliest, exhibited a 35.3% malfunction rate). All malfunction cases were managed through immediate intraoperative intervention by the surgeon—whether by converting to conventional technique, adjusting osteotomy, or performing corrective balancing—ensuring safe prosthesis implantation without catastrophic outcomes. Short-term follow-up revealed no statistically significant differences in functional outcomes between the malfunction and successful groups.

A summary analysis of causes for the 17 malfunction cases is presented in the table below.

Malfunction Cause Number of cases Percentage Systems involved Management and outcome
Robot malfunction Intraoperative system software freeze/error 3 17.6% A (1), B (1), D (1) System reboot attempted; prolonged surgery time (+25 min). One case was converted to conventional surgery due to recurrent failure.
Technical malfunction Intraoperative loosening of femoral/tibial reference frames 5 29.4% A (3), B (2) Cuts stopped immediately; frames re-fixed and registration verified, leading to minor plan adjustments. One case resulted in a minor resection step-off (<1 mm).
Calibration device erroneously placed on a planned resection surface 2 11.8% B (2) Upon discovery, device was removed, re-positioned on intact bone, and registration repeated. Did not affect final outcome but prolonged surgical time.
Processes and Learning Curves Unfamiliarity with the workflow of a new system 2 11.8% C (2) Procedural hesitation; estimated skin-to-cut completion time exceeded 90 minutes. Converted to manual TKA to control operative duration.
The Practitioner-Robot Decision Conflict Surgeon’s experience-based release/adjustment vs. robotic plan 5 29.4% A (2), B (2), D (1) The robotic system indicated "balanced," but the surgeon’s tactile perception suggested tightness/laxity. Empirical adjustments were made, with postoperative imaging or clinical assessment confirming suboptimal balance or alignment deviation.

Discussion

Robotic-Assisted Total Knee Arthroplasty (RA-TKA) stands as a pivotal advancement in the intelligent digitalisation of joint surgery. Through three-dimensional planning and real-time constraints, it aims to surpass traditional surgical techniques in prosthesis alignment and soft tissue balance, delivering safer and more efficient outcomes for patients. Current research predominantly focuses on its advantages [6, 7], with a notable lack of literature analysing the causes of malfunction in RA-TKA during actual clinical practice, particularly regarding the distinct malfunction patterns across different technical platforms. Therefore, in-depth analysis of malfunction cases is crucial for refining technical workflows, safeguarding patient safety, and promoting the rational adoption of this technology.

This study analysed 17 malfunction cases from 188 RA-TKA procedures performed by a single surgeon using multiple robotic platforms at a single centre, revealing key challenges encountered in clinical practice. Reliably integrating robotic technology into standard TKA workflows requires overcoming several obstacles. The findings confirm the existence of three common issues: technical malfunctions, learning curves, and human-machine decision conflicts. Furthermore, by comparing malfunction patterns across different technical pathways, this study provides evidence-based insights for optimising RA-TKA practice.

Technical reliability and the ‘Fragile digital pathway’

RA-TKA establishes a visual pathway from virtual three-dimensional planning to physical skeletal manipulation. This study identified multiple recognised ‘vulnerable points’ within this pathway as significant causes of surgical interruption. This is most evident in the highest proportion of technical malfunctions (41.2%). Firstly, the stability of the tracking system is fundamental. In this study, four RA-TKA procedures experienced repeated blood contamination and loss of tracking with passive infrared reflector reference spheres. Tracking was rapidly restored after alcohol wiping, hence these cases were excluded from malfunction statistics. Previous research suggests active infrared emitter reference frames offer more reliable tracking, though their larger volume may interfere with manipulation. Secondly, rigid fixation of the reference frame to bone is critical. All reference frames in this study employed dual-screw fixation with 2 mm diameter screws inserted via spiral drilling. Two femoral frames loosened due to osteoporosis, one tibial frame loosened from unstable single-cortical fixation, and two cases experienced coordinate drift from loosening between the screw and frame clasp. In osteoporotic cases, micro-movements due to unstable fixation or accidental impact may compromise coordinate accuracy. Theoretically, enhancing retention through double-cortical screws or larger thread diameters is feasible, though this increases risks of bone tunnel trauma and stress fractures [7, 8]. Future tracker designs must achieve a better balance between interference resistance, miniaturisation, and precision.

Multidimensional learning curve

This study demonstrates that the RA-TKA learning curve represents a complex, multidimensional adaptation process. Research by Vermue et al. [5] indicates that the RA-TKA learning curve primarily manifests as prolonged operative time, with no significant learning curve effect on prosthesis alignment accuracy – findings consistent with our observations. Adaptation initially involves decision-making patterns. Even experienced conventional TKA surgeons require an adaptation period (approximately 5–10 cases) when first using the robot, particularly when confronting the system’s multi-plane osteotomy and soft tissue balancing integrated solutions displayed on the screen. In this study, surgeons using System A as their first RA-TKA system exhibited significantly higher average operating times and malfunction cases compared to other groups, primarily due to repeated deliberation and modification of osteotomy and joint space release plans. Secondly, there is adaptation to tool handling variations. System C (high-speed burr) presented the most representative cases. Both malfunctions stemmed from unfamiliarity with the new osteotomy tool—including the burr’s tactile feedback, cutting rate, and depth control—resulting in significantly prolonged osteotomy times. A positive finding regarding the burr was that, once mastered, its inherent elimination of lateral vibration enabled ‘zero’ malfunctions related to osteotomy surface precision in this study. In contrast, malfunctions with the traditional oscillating saw system were dispersed across multiple stages: tracking, guide plate positioning, and operation. For instance, two cases involved calibration pins being placed within the osteotomy zone, necessitating secondary pin placement, re-registration, re-planning, and other procedures that extended operative time. Finally, the entire surgical team required coordination. RA-TKA demands that assistants and nurses possess a clear understanding of novel instrument handover protocols, registration procedures, and equipment management workflows. The three cases in this study experiencing recurrent intraoperative robotic system malfunctions and the two cases converted to manual TKA due to unfamiliarity with the workflow underscore the critical importance of team coordination and adaptation to new technological processes under real-world pressures. Studies on robotic learning curves by Kayani et al. [2] in 2019 and Londhe et al. [9] in 2024 similarly indicate that as surgical teams become more attuned, team anxiety gradually diminishes and surgical efficiency progressively improves.

Tool evolution: the precision-efficiency paradox of grinding drills versus oscillating saws and future prospects

Our observations provide clinical data for the academic debate on ‘grinding drills versus oscillating saws’. Grinding drills (e.g., System C) theoretically offer superior osteotomy precision by avoiding the offset and tremor inherent in oscillating saw systems through high-speed milling, potentially providing greater control when handling sclerotic or deformed bone. However, this comes at the cost of extended bone processing time. In their study of the Honghu robot learning curve, Zhixin et al. [10] identified osteotomy operation time as a significant component of the learning curve. This reveals a clinical trade-off paradox between precision and efficiency. Furthermore, while studies suggest that the microstructure of bone surfaces produced by grinding drills may facilitate bone cement infiltration [11], this advantage has yet to be confirmed as beneficial for long-term prosthesis survival rates or improved functional scores. Future efforts should focus on optimising milling cutter geometry, cutting dynamics algorithms, and force feedback to resolve the precision-efficiency paradox.

Human-machine conflict and synergy: the evolution from ‘operator’ to ‘system administrator’

The five cases of malfunction attributed to ‘human-machine conflict’ in this study hold the greatest clinical significance. In these instances, the system algorithm indicated perfect gap balance, yet the surgeon, relying on experience, over-adjusted the plan, resulting in abnormal gaps. Ultimately, partial correction of the force line was sacrificed to restore equilibrium. This reveals a shift from ‘experience-driven intuition’ to a new decision-making paradigm of ‘data-driven planning with experience-based calibration.’ Blind adherence to either approach during surgery risks error: complete trust in the machine may overlook registration anomalies or under-simulated soft tissue dynamics; absolute reliance on experience may lead to excessive adjustments due to sensory misdirection or reference frame drift. These cases underscore the imperative for surgeons to evolve from surgical “operators” to ‘managers’ of the RA-TKA system. When conflicts arise, the initial response should not be empirical release or osteotomy, but rather the initiation of a systematic diagnostic process: first, verifying the integrity of the entire digital pathway (tracking ball clean? Reference frame stable? Registration points displaced?), followed by multi-angle, dynamic objective assessment using robotic gap pressure sensors or distance measurement devices. Research by Bardou-Jacquet et al. [12] confirmed that integrating pressure sensors with robotic systems optimises ligament balance precision. Yang Qianqian et al. [13] summarised application risks of the Tianji surgical robot, emphasising that only after ruling out technical malfunctions and confirming objective data accuracy should experience be cautiously applied for final fine-tuning. This ‘augmented intelligence’ model represents the ideal future of human-machine collaboration.

Remediability of malfunctions and the ultimate value of advanced expertise

All 17 malfunction cases ensured safe prosthesis implantation through the surgeon’s immediate intraoperative intervention. Short-term follow-up revealed no statistically significant difference in functional outcomes between the malfunction group and the successful group. This demonstrates that the ‘malfunctions’ defined in this study were predominantly procedural and technical in nature, rather than irreversible clinical failures. It further confirms that while robotic systems enhance precision and predictability, the surgeon’s individual competence remains the ultimate and most critical safeguard for surgical safety when confronting complex intraoperative contingencies, technical malfunctions, or unplanned events. This does not diminish the value of robotic technology but rather clarifies its current developmental stage: robots serve as exceptional ‘assistants’ and ‘quantitative tools’, whilst surgeons remain the ‘primary operators’ and ‘ultimate decision-makers’ responsible for the overall procedure [14, 15].

Limitations of the present study

This single-centre retrospective analysis, conducted entirely by a highly experienced specialist, may have partially masked risks of more severe consequences from certain malfunctions due to the surgeon’s exceptional corrective abilities. Consequently, conclusions should be applied cautiously when extrapolating to surgeons of varying experience levels. The inclusion of multiple robotic systems limited the absolute number of malfunction cases, constraining deeper multivariate risk factor analysis. Furthermore, surgeons undergo an adaptation process with different robotic systems. It warrants consideration whether prolonged use of a single robotic system could significantly reduce malfunction rates. Future multicentre, prospective studies involving surgeons of varying technical proficiency are required to comprehensively quantify the incidence of various malfunction patterns, their risk factors, and their impact on medium-to-long-term outcomes. This would provide more robust evidence for the standardised application and education of RA-TKA techniques [16].

Conclusions

The success of robot-assisted total knee arthroplasty (TKA) hinges upon three core prerequisites: a reliable technical platform, standardised surgical procedures and training protocols, and the surgeon’s comprehension of robotic logic coupled with human-machine collaborative decision-making capabilities. Based on clinical observations from 188 cases, key findings include: (1) 17 cases (9.0%) experienced robot-related malfunctions; (2) distinct failure modes were exhibited across different robotic systems; (3) malfunction incidence was higher during the early learning curve phase; (4) All malfunctions were resolved through surgeon intervention without causing surgical failure. This indicates that technical malfunctions can be progressively mitigated through system optimisation and training, whilst intraoperative human-robot collaboration and stepwise verification remain critical for ensuring surgical safety.

Author contributions

Xunzhou Song and Mingyou Wang conceived and designed the study, and drafted the manuscript. Hongping Wang and Zhuodong Tang contributed to data collection and assisted in manuscript revision. Yuping Lan performed all robotic surgical procedures, participated in critical review of the manuscript, and conducted data verification.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Song Xunzhou and Wang Mingyou have contributed equally and are designated as joint first authors.

References

  • 1.Parratte S, Pagnano MW, Trousdale RT, Berry DJ (2010) Effect of postoperative mechanical axis alignment on the fifteen-year survival of modern, cemented total knee replacements. J Bone Joint Surg Am 92(12):2143–2149 [DOI] [PubMed] [Google Scholar]
  • 2.Kayani B, Konan S, Huq SS, Tahmassebi J, Haddad FS (2019) Robotic-arm assisted total knee arthroplasty has a learning curve of seven cases for integration into the surgical workflow but no learning curve effect for accuracy of implant positioning. Knee Surg Sports Traumatol Arthrosc 27(4):1132–1141 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Sires JD, Wilson CJ (2021) CT validation of intraoperative implant position and knee alignment as determined by the MAKO total knee arthroplasty system. J Knee Surg 34(10):1133–1137 [DOI] [PubMed] [Google Scholar]
  • 4.Li Z, Chen X, Wang X, Zhang B, Wang W, Fan Y, Yan J, Zhang X, Zhao Y, Lin Y, Liu J, Lin J (2022) HURWA robotic-assisted total knee arthroplasty improves component positioning and alignment - a prospective randomized and multicenter study. J Orthop Translat 33:31–40 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Vermue H, Luyckx T, Winnock de Grave P, Ryckaert A, Cools AS, Himpe N, Victor J (2022) Robot-assisted total knee arthroplasty is associated with a learning curve for surgical time but not for component alignment, limb alignment and gap balancing. Knee Surg Sports Traumatol Arthrosc 30(2):593–602 [DOI] [PubMed] [Google Scholar]
  • 6.Sicat CS, Chow JC, Kaper B, Mitra R, Xie J, Schwarzkopf R (2021) Component placement accuracy in two generations of handheld robotics-assisted knee arthroplasty. Arch Orthop Trauma Surg 141(12):2059–2067 [DOI] [PubMed] [Google Scholar]
  • 7.Liu Gang L, Guoqing W, Yang et al (2023) Early clinical efficacy of MAKO robotic-assisted total knee arthroplasty [J]. Chin J Orthop Joint Surg 16(03):232–238 [Google Scholar]
  • 8.Kayani B, Konan S, Tahmassebi J, Pietrzak JRT, Haddad FS (2018) Robotic-arm assisted total knee arthroplasty is associated with improved early functional recovery and reduced time to hospital discharge compared with conventional jig-based total knee arthroplasty: a prospective cohort study. Bone Joint J 100-B(7):930–937 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Londhe SB, Rudraraju RT, Shah RV, DeSouza C, Shetty V, Khan FS, Bajwa S (2024) Analysis of robot-specific operative time and surgical team anxiety level and its effect on alignment during robot-assisted TKA. J Robot Surg 18(1):86 [DOI] [PubMed] [Google Scholar]
  • 10.Zhi Xin P, Hangyu P, Wei et al (2023) Preliminary study on the learning curve of Robot-Assisted total knee arthroplasty using the Honghu system [J]. Chin J Orthop Joint Surg 16(04):340–346 [Google Scholar]
  • 11.Ong MT, Yu M, Cao M et al (2025) Robotic-assisted burr preparation significantly improves cement penetration but not immediate fixation strength in total knee arthroplasty. Knee Surg Sports Traumatol Arthrosc 33(11):3935–3943 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Bardou-Jacquet J, Murgier J, Laudet F, Fabre T (2022) Combining load sensor and robotic technologies for ligament balance in total knee arthroplasty. Orthop Traumatol Surg Res 108(5):102889 [DOI] [PubMed] [Google Scholar]
  • 13.Yang Qianqian D, Lixiang G, Dawei et al (2022) Clinical risk analysis and countermeasures for the Tianji orthopaedic surgical robot [J]. Biomedical Eng Clin Pract 26(04):497–501 [Google Scholar]
  • 14.Zhang S, Kong X, Chai W (2022) Review of clinical applications of arthroscopic surgical robots in 2021. Orthopedics 13(6):562–567 [Google Scholar]
  • 15.Liu J, Kong X, Chai W (2025) [Current Status and Prospects of Domestic Total Joint Replacement Surgical Robots in Clinical Practice]. J Clin Orthop Res 10(1):52–56 [Google Scholar]
  • 16.Kolodychuk N, Su E, Alexiades MM et al (2021) Can robotic technology mitigate the learning curve of total hip arthroplasty? Bone Jt Open 2(6):365–370 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

No datasets were generated or analysed during the current study.


Articles from Journal of Robotic Surgery are provided here courtesy of Springer

RESOURCES