A Systematic Review and Meta-Analysis of Conventional Versus Robotic-Assisted Total Knee Arthroplasty

This study aims to compare the outcomes and advantages of total knee arthroplasty (TKA) performed using conventional surgical techniques with those conducted using robotic-assisted methods in terms of operation time, Oxford knee score, range of motion, tourniquet time, and Western Ontario and McMaster Universities Arthritis index. We performed a literature search through five databases, namely, PubMed, Cochrane Central, Scopus, Embase, and Web of Science, from inception until July 3, 2023. Randomized clinical trials (RCTs) and cohorts comparing conventional TKA with robotic-assisted TKA were included. The risk of bias of the included RCTs was determined using the Cochrane risk of bias tool and the National Institutes of Health tool for cohort studies. We conducted a meta-analysis using Review Manager 5.4. To analyze continuous data, we calculated the mean difference (MD) along with its corresponding 95% confidence interval (CI). By synthesizing data from a comprehensive analysis, the study unveiled noteworthy distinctions between robotic-assisted arthroplasty and conventional arthroplasty across critical parameters. First, a substantial alteration in the hip-knee-ankle (HKA) angle was observed, with the robotic-assisted approach demonstrating a significant difference (MD = 0.84, 95% CI = 0.25-1.43, p = 0.005). Second, in terms of operative time, a notable reduction in surgical duration was noted with conventional TKA (MD = 16.85, 95% CI = 8.08-25.63, p = 0.0002). The assessment of tourniquet time exhibited a significantly longer duration for robotic-assisted arthroplasty (MD = 35.70, 95% CI = 27.80-43.61, p < 0.001). Our findings indicate that conventional TKA outperforms robotic-assisted TKA, primarily due to its shorter operative and tourniquet times, along with a more favorable change in the HKA angle. However, it is worth noting that robotic-assisted TKA showed a slight advantage in pain outcomes, although this advantage was not statistically significant. To gain a more comprehensive understanding, we recommend conducting a large-scale randomized controlled trial that directly compares both TKA methods. This trial should evaluate costs and long-term outcomes while ensuring consistent follow-up durations among studies. Such an approach would greatly assist orthopedic decision-making and contribute to improved TKA outcomes.


Introduction And Background
Osteoarthritis is a common disability worldwide [1].Primary osteoarthritis is considered a degenerative condition, meaning that it results from the wear and tear of the joint over time.It is considered the most common joint disease or reason for disability in the United States [2].Among individuals aged 60 and above, knee osteoarthritis affects 13% of women and 10% of men [3].The increased rate in females may be related to pregnancy as it decreases calcium levels in the blood and bones [4].Knee osteoarthritis is a degenerative disease and develops progressively over a decade [5].Knee osteoarthritis can be classified into primary, which develops without a clear reason, and secondary, which develops due to a clear reason such as increased force on the joint, obesity, joint injuries, bone deformities, or certain metabolic diseases [3,4].
Total knee arthroplasty (TKA) is considered one of the most common procedures.In 2014, about 700,000 TKA procedures were performed in the United States [6].This procedure enhances the quality of life of patients and decreases pain associated with osteoarthritis [7].Although more than 20% of patients remain unsatisfied with the outcomes of this procedure, it is considered a cost-effective procedure that enhances the quality of life of a patient by saving about 1,000-12,000 euros for the patient [8].
TKA has traditionally been considered a therapeutic option for advanced stages of knee osteoarthritis [9].In the conventional approach, surgeons rely on their skills and experience for precise bone cuts and implant positioning.While they use guides and instruments, there is a potential for slight variations.Therefore, robotic-assisted approaches have been implemented to increase precision [10].On the other hand, the robotic-assisted approach is associated with increased operative time and cost as it requires more expensive equipment [10].There are three types of robotic systems, namely, autonomous, hands-on, and passive, which differ in terms of surgeon control of the operation [11].Robotics can assist in performing a minimally

Quality Assessment
The Cochrane risk of bias tool (version 1) was used to evaluate the included RCTs [14].This tool comprises the following categories: (1) identification of selection bias and other potential biases; (2) allocation of study groups; (3) blinding of participants and investigators; (4) evaluation of outcomes and the use of blinding in this assessment; and (5) randomization of the study population.The possibility of bias in judgment can be a high, low, or unclear risk of bias.

Data Extraction
Using Excel sheets, the extracted data contained the following items: (1) summary including study ID, site, study design, inclusion criteria, main outcomes/endpoint, and conclusion; (2) baseline data including study arms, sample size, age, follow-up, gender, operation side, underlying diagnosis, ROM, and preoperative knee score.Further, the following outcomes were used: (a) change in HKA angle, (b) HSS, (c) operative time, (d) Oxford knee score, (e) ROM, (f) tourniquet time, (g) WOMAC pain score.

Data Synthesis
We utilized RevMan version 5.4 for the statistical analysis in this study.We set the significance level at <0.05.For continuous data, we computed the mean difference (MD) and determined the 95% confidence interval (95% CI).Additionally, we assessed heterogeneity using both the I 2 and the chi-square test.Data were considered heterogeneous if the chi-square test yielded a p-value <0.1 and if the I 2 value exceeded 50%.Homogeneous data were analyzed using a fixed-effect model, while a random-effects model was applied for heterogeneous data.

Literature Search
We identified a total number of 1,762 results related to the topic.After removing duplicates, 1328 results remained.After manual screening, 53 articles were included for full content screening.After full content screening, 26 studies were included in the final analysis  (Figure 1).

Characteristics of the Included Studies and Patients
Our study encompassed a diverse range of research methodologies, comprising nine RCTs, one case-control study, three prospective cohorts, and 13 retrospective cohort studies.In total, our analysis involved 9,964 patients drawn from various countries, including the United States, China, South Korea, Belgium, the United Kingdom, Greece, Singapore, Germany, and Thailand.The study participants were predominantly elderly, aged over 60 years, with a higher proportion of females undergoing TKA compared to males.Furthermore, the primary indication for TKA across the majority of patients was osteoarthritis, albeit in varying stages (Table 1).2022 [22] RA TKA, 30

Quality Assessment Results
Of the nine included RCTs, seven had a high bias and two were high quality with low bias, as shown in Figure 2.For cohort studies, all studies were fair in quality, and only three studies had good quality.The case-control study exhibited fair quality (Tables 2, 3).

FIGURE 11: Forest plot of Western Ontario and McMaster Universities
Arthritis index.

Discussion
Patients typically seek a long-lasting TKA procedure that offers stability, effective pain relief, and enhanced functionality.So far, short-term follow-up studies have not demonstrated any superior clinical outcomes for robotic-assisted TKA when compared to conventional methods.Our comprehensive analysis unveiled several noteworthy findings in the comparative evaluation of conventional TKA and robotic-assisted TKA.Notably, the conventional TKA method exhibited superiority in multiple aspects.First, there was a significant difference in the change of the HKA angle, favoring conventional TKA, indicating that this approach resulted in a more favorable realignment of the lower limb compared to the robotic-assisted method.Second, operation time was notably shorter in conventional TKA, reflecting a more efficient surgical process.Moreover, tourniquet time favored the conventional approach, as robotic-assisted TKA required a longer duration of tourniquet application, temporarily restricting blood flow to the limb.This extended tourniquet time can have potential implications for patient outcomes and recovery.Conversely, when examining various outcome measures, robotic-assisted TKA did not demonstrate significant superiority over the conventional approach.Parameters such as postoperative ROM, the Oxford knee score (a measure of knee function and pain), and the WOMAC index (assessing pain associated with osteoarthritis) did not indicate significant advantages for the robotic-assisted TKA when compared to the conventional method.
The cost of TKA, whether performed conventionally or with the assistance of robotics, can vary significantly depending on several factors, including the location of the medical facility, the surgeon's fees, the type of implants used, the complexity of the procedure, and whether the patient has insurance coverage.In any possible scenario, the cost of robotic-assisted TKA would be significantly higher than conventional TKA [41][42][43].However, Cai et al. [16] reported that the total cost of rehabilitation had no significant difference between the two groups.Furthermore, Lonner et al. [44] found that robotic technology had the potential to play a cost-effective role due to its precision in surgical procedures and the relatively smooth learning curve associated with its use.
Xu et al. [38] noted that in robotic-assisted TKA, a significant portion of the additional surgical duration is allocated to tasks such as setup, femoral and tibial fixation, and alignment.This aspect appears to be a drawback of robotic-assisted TKA that should be improved to reduce the time allocated to non-surgical activities.The primary reasons for prolonged surgery time during the procedure were the intricate process of registering critical bone landmarks and the need for enhancements in the registration success rate.It is worth noting that as surgeons gain proficiency in the procedure, these challenges become less significant, as robotic-assisted TKA is associated with a high range of learning curve [45].
The extended duration of surgery in robotic-assisted TKA holds significant clinical importance, as studies have demonstrated that prolonged surgery times are associated with an elevated risk of peri-prosthetic joint infections [46].
Robotic-assisted TKA significantly reduced pain levels in the two studies by Song et al. [34] and Yang et al. [40].The overall outcome remains insignificantly in favor of robotic-assisted TKA, with no significant advantage observed.It is worth emphasizing that although robotic-assisted surgery may have the potential to alleviate pain in comparison to conventional methods, individual outcomes can differ.Factors such as the patient's overall health, the specific surgical procedure, and the surgeon's expertise are key determinants in assessing the extent of pain relief.Kayani et al. [47] reported that robotic-assisted TKA led to more substantial pain alleviation, enhanced early functional recuperation, and a shorter hospital stay.
Our paper has several notable strengths.We stand as pioneers in conducting a meta-analysis that directly compares conventional and robotic-assisted TKA.One of our significant advantages lies in our extensive sample size, which bolsters the robustness of our findings.Moreover, we meticulously included a wide range of studies, encompassing various types of evidence, to provide a comprehensive overview of the subject.Notably, our study unearthed a major and statistically significant difference between the two approaches, which holds great promise in guiding future orthopedic surgeons and practitioners in their decision-making processes.Our commitment to inclusivity is further underscored by our incorporation of all available RCTs related to this topic, bolstering the strength of our evidence base.To ensure the utmost rigor, we adhered to the guidelines for meta-analyses outlined by Cochrane, enhancing the credibility of our methodology.However, it is important to acknowledge several limitations in our study.High heterogeneity among the included papers posed a challenge, as did variations in study quality.Additionally, discrepancies in followup durations, diverse surgical teams, and a multitude of centers spanning different countries introduced significant individual variations.In light of these limitations, we strongly recommend a large-scale RCT directly comparing TKA using both approaches.Such a study should estimate the cost implications and consider long-term outcomes.Aligning follow-up durations among the studies would be essential to provide a more accurate estimation of long-term outcomes.This would facilitate a more comprehensive evaluation of the benefits and drawbacks associated with these surgical techniques.

Conclusions
Our findings support the superiority of the conventional TKA over the robotic-assisted TKA, as conventional TKA had lower operative time and tourniquet time.Additionally, the HKA angle change was superior in the conventional TKA.The superiority of the robotic-assisted TKA was in the pain outcome taking into consideration that the result was not significant.To provide clearer insights, we recommend a large-scale RCT comparing both TKA methods and assessing costs and long-term outcomes while aligning follow-up durations among studies.This would aid orthopedic decision-making and enhance TKA outcomes.

FIGURE 1 :
FIGURE 1: Preferred Reporting Items for Systematic Reviews and Meta-Analyses flow diagram.

FIGURE 2 :
FIGURE 2: Risk of bias graph summary for randomized controlled trials.

FIGURE 3 :
FIGURE 3: Forest plot of change in hip-knee-ankle angle.

FIGURE 4 :
FIGURE 4: Forest plot of hospital for special surgery.

FIGURE 5 :
FIGURE 5: Forest plot of hospital for special surgery after leaving one study out.

FIGURE 6 :
FIGURE 6: Forest plot of Change in Oxford knee score.

FIGURE 7 :
FIGURE 7: Forest plot of change in Oxford knee score after leaving one study out.

FIGURE 8 :
FIGURE 8: Forest plot of change in range of motion.