1 Introduction

Unicompartmental knee arthroplasty (UKA) is a successful treatment for primarily single-compartment arthritis. The surgery boasts advantages including a bone-sparing technique, quick recovery, and excellent functional outcomes.^1–6 However, UKA survivorship has been demonstrated to be lower than TKA; one recent Finnish registry study demonstrated 69% UKA survivorship at 15 years, compared with 88% for total knee arthroplasty (TKA).⁷ The most common causes of revision following UKA include aseptic loosening (the main cause of early failures and in mobile-bearing implants) and progression of osteoarthritis (the main cause of late failures in fixed-bearing implants).⁸

Many surgeons advocate for treating patients who have isolated compartment osteoarthritis with a TKA, in part due to the lower survivorship. Furthermore, results following conversion from UKA to TKA are thought to be worse than a primary TKA.^9,10 A recent systematic review and meta-analysis found that eight of the 11 included studies demonstrated worse patient-reported outcomes and higher rates of reoperations following conversion of UKA to TKA when compared to primary TKA; the other three found statistically similar outcomes.¹¹ Even in the studies that found similar clinical outcomes, conversion from UKA to TKA is considered a more technically demanding surgery, requiring more operative time and the use of revision components like augments and bone grafting.^10–12

The advent of new technologies has led to renewed interest in evaluating whether outcomes following the conversion of UKA to TKA can be improved. One study demonstrated outcomes following computer-navigated conversion UKA to be satisfactory and similar to primary TKA at a minimum of 3 years postoperatively.¹³ A similar study investigating imageless robotic-arm-assisted revision UKA to TKA and primary TKA found comparable surgery time, alignment accuracy, and bone stock preservation between the two groups.¹⁴ Additionally, two studies have compared the outcomes of conventional and robotic conversion of UKA to TKA and found statistically similar functional outcomes and rates of complications.^15,16 Despite these encouraging early results, these studies typically have a small sample size and less than a 5-year follow-up.

Therefore, our purpose was to describe the surgical technique for converting UKAs to TKAs using robotic-arm-assistance and its evolution over time. In addition, we evaluated patient-reported outcomes, clinical complications, and survivorships following these procedures.

2 Methods

We conducted a retrospective review of consecutive patients who underwent robotic-arm-assisted UKA to TKA conversion (rcUKA). Surgery was performed by a single surgeon at a single institution between 2017 and 2022. A minimum of 1 year of follow-up was required for inclusion in the data analyses. Charts were reviewed to obtain the following data: patient demographics, indications for conversion from UKA to TKA, operative technique and implants used, postoperative complications, and patient-reported outcome measures (PROMs). Demographics included age, sex, body mass index (BMI), and comorbidities. Complications included prosthetic joint infections, superficial surgical site infections, aseptic loosenings, and revisions. PROMs included the Knee Injury and Osteoarthritis Outcome Score for Joint Replacement (KOOS JR) and the reduced Western Ontario and McMaster Universities Osteoarthritis Index (r-WOMAC).

A total of 44 robotic-arm-assisted UKAs in 41 patients were included in the study. The primary UKA was used to address the following compartments: medial in 37 patients, patellofemoral in 5 patients, and lateral in 2 patients. Patient demographics included a mean age of 72 years (range, 43 to 91), 15 of 41 patients were men (36.6%), and the average BMI was 32.63 (range, 24.3 to 46.7). The mean time elapsed from UKA to TKA was 6.8 years (range 1–16.6), and the mean postoperative follow-up was 1.8 years (range 1–6.6). The indication for UKA conversion to TKA was primarily osteoarthritis progression, occurring in 33 UKAs (75%). Other indications included 7 aseptic loosenings (16%), 2 unspecified pains (4.5%), 1 polyethylene wear (2.3%), and 1 prosthetic joint infection (2.3%).

Uncemented cruciate-retaining (CR) implants were used in 38 of the 44 robotic-arm-assisted TKAs (86.5%). The other six utilized cemented implants: four CR femurs (9.1%), six tibial baseplates (13.6%), four tibial stems (9.1%), and four medial tibial augments (9.1%). The mean polyethylene thickness was 10.1 mm (range, 9 to 16).

2.1

2.1 Robotic-arm-assisted UKA conversion to TKA technique

Robotic-arm-assisted UKA to TKA conversion surgeries were performed using the MAKO Total Knee SmartRobotics™ System (Stryker, Mahwah, New Jersey, USA). Therefore, every patient received an ipsilateral lower extremity CT per company protocol.

In the operating room, all patients were positioned in the supine position, using a tourniquet, with the operative leg prepped and draped in the standard fashion. A midline incision was made, and medial and lateral flaps were developed. A medial parapatellar arthrotomy was made from approximately 3 cm (cm) superior to the superior pole of the patella, extending to just medial to the tibial tubercle. At this time, a medial release and removal of any fat pad or other fibrotic tissue were removed from the knee joint.

Next, the equipment necessary for the robotic arm was set up. Femoral pins were placed intra-incisionally on the antero-medial surface of the meta-diaphyseal aspect of the femur. Tibial pins were placed 4–8 cm distal to the tibial tubercle along the medial border of the tibial shaft. All pins utilized bicortical fixation. The robotic arm was then registered prior to the removal of the tibial and femoral UKA implants. Registration was identical to that of a primary TKA, except on the medial aspect of the tibial and femoral condyles. There, the surface of the UKA implants was registered (after the removal of the polyethylene implant). Verification was also completed with the implants retained in bone (Fig. 1). Once the robotic arm was registered and verified, UKA implants were removed using osteotomes, micro-sagittal saws, and other standard implant removal techniques.

Close

Fig. 1 Verification screen using the MAKO Total Knee SmartRobotics™ System (Stryker, Mahwah, New Jersey, USA) demonstrating the verification probe adjacent to metal on the medial tibia.

Export to PPT

The bone cuts were determined using a gap balancing technique. The bone was cut in the following order: tibia, distal femur, anterior femur, anterior chamfer, distal femur, and posterior chamfer. Trial implants were inserted, and the knee was stressed at full extension and 90 degrees of flexion, both manually and using the robotic-arm technology, to assess for balanced gaps. If gap balancing was unsatisfactory, adjustments were made to the surgical plan, and the bone was recut as needed. Once gap balancing was acceptable using the trial implants, they were removed and final implants were inserted (using either the cemented or uncemented technique, determined on a case-by-case basis). The arthrotomy and incision were closed with sutures in the standard fashion.

The technique used evolved over time as the primary surgeon gained more experience. Originally, all patients were given cemented implants placed in mechanical alignment, commonly using revision components like tibial stems. By the end of the series, patients were given implants in functional alignment, more commonly using a cementless technique and less frequently using revision implants. Functional alignment is a TKA implant position technique that utilizes the precision of image-based robotics (and single radius of curvature implants) and aims to recreate the natural center of rotation of the knee, particularly the medial condyle.^17,18 In doing so, recreation of normal knee kinematics is thought to be accomplished. In our experience, functional outcomes at 1 year following robotic-arm-assisted TKA using functional alignment are better than those following conventional TKA using mechanical alignment. Fig. 2 shows preoperative and postoperative radiographs of two patients, representative of this evolution. Fig. 3 demonstrates the change in intraoperative plan between the two methodologies.

Close

Fig. 2 Representative images of the evolution of robotic-arm-assisted medial unicompartmental knee arthroplasty to total knee arthroplasty. Panel (A) was performed on a patient in 2017 using mechanical alignment, tibial stem, and cemented technique. Panel (B) was performed on a patient in 2022 using functional alignment and a press-fit primary tibial component.

Export to PPT

Close

Fig. 3 Robot-arm-assisted case planning screen demonstrates A) the mechanical alignment technique used originally in the case series, and B) the functional alignment technique used in the latter part of the series.

Export to PPT

3 Results

The PROMs significantly improved at 1-year follow-up for all patients included. The average KOOS JR score increased from 48.1 to 68.7 (P < 0.001), and the r-WOMAC score decreased from 25.7 to 10.6 (P = 0.003). Complications were rare, including two knees developing prosthetic joint infections (4.5%), one developing aseptic loosening of the femoral component (2.3%), and one developing a superficial surgical site infection requiring superficial irrigation and debridement (2.3%). Overall implant survivorship was 93.2% at an average of 1.8 years, and aseptic survivorship was 97.7%.

4 Discussion

In this study, we demonstrate that patients undergoing robotic-arm-assisted UKA to TKA conversion had improved patient-reported outcomes with low revision and complication rates. Improved preoperative planning and implant placement achieved with robotic-arm assistance give surgeons an alternative to manually revising UKA patients. As our robotic-assisted surgical technique evolved over time, we utilized primary uncemented implants in the majority of cases.

While conversions from UKA to TKA are successful, the literature suggests that there is tremendous room to grow. Lunebourg et al.⁹ evaluated whether clinical and functional outcomes following conversion of UKA to TKA were similar to primary TKA or revision TKA. In a propensity score-matched cohort of 144 patients who had an average of 7 years of follow-up, the authors found that functional outcomes, quality of life, complication profile, and implant survival were more like revision TKA, than primary TKA. In another study using the Norwegian Arthroplasty Registry, Leta et al.¹⁰ compared 768 TKA revisions to 578 UKA to TKA conversions. The authors demonstrated that, with few exceptions, the clinical and functional outcomes following UKA to TKA conversion were similar to TKA revision. Finally, a recent systematic review and meta-analysis found that UKA to TKA conversions had longer tourniquet time, worse Western Ontario and McMaster Universities Arthritis Index (WOMAC) pain scores and WOMAC functional scores, worse Oxford Knee Scores (OKS), and higher reoperation rates compared to primary TKA.¹¹ Robotics may be poised to close this gap in clinical and functional outcomes between UKA to TKA conversion and primary TKA.

While robotic-arm-assisted joint arthroplasty surgery is only approved by the United States Food and Drug Administration (FDA) for primary arthroplasty procedures, the technology has slowly begun to be used in revision surgery. Robotic-arm-assisted techniques have been used for conversions of UKAs to TKAs and have demonstrated similar results to conventional conversion surgery in terms of clinical and functional outcomes^15,16; furthermore, one study demonstrated similar implant position, bone stock preservation, and operative time to primary TKA, but did not compare clinical or functional outcomes.¹⁴ Additionally, robotics has been used successfully in revision TKA surgery in multiple case reports.^19–21 There is tremendous potential with this technology to improve our technique, decrease operative time, minimize bone loss, reduce the need for revision implants, and improve patient reported and clinical outcomes. This was demonstrated in our cohort; as the senior author became more experienced with the technique, primary cementless implants were able to be used in almost all cases.

Our study presents a detailed review of a novel, successful technique using robotic-assisted technology to perform UKA to TKA conversion surgery. However, there are some potential limitations. Our investigation is a retrospective case series of patients operated on by a single surgeon who had a small sample size. Future investigations should collaborate with other institutions to increase sample size and improve external validity. In addition, we present a case series and do not have a control group with whom we can compare our clinical and functional outcomes. Future studies should focus on comparing this procedure to conventional UKA to TKA conversion techniques as well as primary robotic TKA. Furthermore, we present only short-term outcomes, with an average of 1.8 years of postoperative follow-up. Longer-term follow-up will be necessary to evaluate in the future. Finally, the surgical technique was adjusted over time as the surgeon became more comfortable with the procedure. While this reflects the potential of this technology, it means that our patient sample is not homogenous.

5 Conclusion

In this single-surgeon case series, patients undergoing robotic-arm-assisted UKA to TKA conversion exhibited improved patient-reported outcomes and low revision and complication rates. Improved preoperative planning and implant placement achieved with robotic-arm-assistance gives surgeons an alternative to manually revising UKA patients and has the potential to improve the functional and clinical outcomes following these surgeries.

Patient consent

This study was a retrospective review of data from electronic medical records, all of which was standard of care. Therefore, no explicit patient consent was required to conduct or our study.

Ethical approval

This study received approval from the South County Hospital Institutional Review Board.

Funding

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

Statement of funding

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

CRediT authorship contribution statement

Matthew L. Magruder: Methodology, Writing – original draft, Preparation, Visualization, Writing – review & editing. Tanner McClure: Data curation, Formal analysis. Kevin Marchand: Data curation, Formal analysis. Michael A. Mont: Supervision, Conceptualization, Project administration, Writing – original draft, Preparation, Visualization, Writing – review & editing. Robert C. Marchand: Conceptualization, Resources, Writing – review & editing, Supervision.