Mid-term outcomes of a novel liner design in kinematically-designed cruciate-retaining total knee arthroplasty

Abstract

BACKGROUND

Medial dished (MD) liner designs for cruciate-retaining (CR) total knee arthroplasty (TKA) are a relatively novel development. MD tibial inserts have a more constraining medial side, which allows for more similar kinematics and function to a native knee.

AIM

To evaluate the clinical results and patient-reported outcomes after CR TKA procedures utilizing a kinematically designed medial dish system.

METHODS

A multicenter, retrospective cohort review of 139 primary elective TKAs utilizing a kinematically designed CR Knee System (JOURNEY™ II CR MD; Smith and Nephew, Memphis, TN, United States) at three different institutions with a minimum of two years of follow-up. Demographic information, clinical outcomes, and patient-reported outcome measures were collected and analyzed.

RESULTS

With up to 3.7 years from surgery, overall implant survivorship was 98.6%. There were significant postoperative increases in the average Knee Injury and Osteoarthritis Outcome Score for Joint Replacement scores (17.4 at 6 months, 26.1 points at two years or more, P < 0.001).

CONCLUSION

The combination of high implant survivorship and substantial improvements in patient-reported outcome measures suggests that the medial dish tibial insert represents a safe and effective option within TKA. Additional investigation is necessary to evaluate the long-term survivorship of this design.

Key Words: Total knee arthroplasty; Medial dished insert; Cruciate-retaining; Native kinematics; Kinematically-designed; Medial dished liner; Constrained liner

Core Tip: This multicenter retrospective review evaluates outcomes following cruciate-retaining total knee arthroplasty using a novel, kinematically designed medial dished tibial insert. With an implant survivorship of 98.6% and significant improvements in subjective outcome measures such as the Knee Injury and Osteoarthritis Outcome Score for Joint Replacement, the results of this study suggest this implant design effectively replicates native knee kinematics. These findings support the use of the medial dished tibial insert as a promising option in cruciate-retaining total knee arthroplasty, warranting further long-term investigation.

INTRODUCTION

Projections estimate that total knee arthroplasty (TKA), one of the most frequent operations in the United States, will reach 719000 annual procedures by 2040[1]. Despite its prevalence, patient satisfaction remains a significant challenge; only 81% of patients report being satisfied postoperatively, leaving 19% as very dissatisfied, dissatisfied, or neutral. Furthermore, satisfaction specifically with pain relief is inconsistent, ranging from 72% to 85% based on the activity being performed[2,3]. These statistics underscore a clear need for advancements in TKA to more effectively alleviate pain and restore knee function.

The pursuit of improved long-term outcomes, joint function, and patient satisfaction has spurred the recent development of new implant models and techniques. Although cruciate-retaining (CR) and posterior-stabilizing (PS) implants are currently the most prevalent, designs that specifically guide the motion of the medial compartment are gaining increased traction. These implants are designed to mimic the knee joint’s physiological kinematics by employing the medial pivot motion of the medial compartment while simultaneously reducing lateral congruence[4]. The recently developed medial dished (MD) insert (JOURNEY™ II CR MD; Smith and Nephew, Memphis, TN, United States) represents a modification of these concepts. It utilizes an asymmetric design, characterized by a more conforming medial compartment, to encourage adequate posterior femoral rollback without fully relying on a medial pivot mechanism. The architecture is specifically engineered to approximate the natural knee’s anatomical structure and functional mechanisms, which is hypothesized to yield improved contact stress distribution and decreased polyethylene wear[5-7]. This study’s objective is to assess the clinical results and survivorship of the implant for a TKA procedure utilizing kinematic alignment that features a tibial MD liner, based on a minimum two-year follow-up. Our specific aims include examining the frequency of adverse events and re-operations, analyzing patient-reported outcome measures (PROMs), and determining the implant's overall survival rate.

MATERIALS AND METHODS

Study design

A retrospective, multicenter, cohort study of 139 primary TKAs from three institutions was performed to evaluate the clinical outcomes of a kinematically-aligned TKA with an MD tibial insert design. Institutional review board approval was obtained. Patients who underwent primary, elective TKA using the above system between February 2022 and July 2023 with at least two years of follow-up were included in the study. In an effort to decrease confounders, we included patients of one surgeon from each center in the study, all of whom were fellowship trained in joint arthroplasty. This resulted in our final cohort of 139 knees in 137 patients with follow-up times ranging from two to 3.7 years.

Data collection and measures

To evaluate this system, patient demographics, intraoperative data, implant information, short- and mid-term clinical outcomes, and PROMs were recorded. All information was obtained through retrospective chart review of patients’ electronic medical records (EMR) (Epic Systems, Verona, WI, United States). The demographic variables were sex, age, race, smoking status, insurance status, American Society of Anesthesiologists (ASA) score, body mass index, Charlson comorbidity index, and primary diagnosis. The intraoperative and implant variables collected included operative time, type of anesthesia, laterality, patellar resurfacing/fixation, and bearing surface. The clinical outcomes included length of stay, discharge disposition, 90-day emergency department visits, 90-day readmissions, and all-time incidence and indication of revisions. Knee Injury and Osteoarthritis Outcome Score for Joint Replacement (KOOS, JR) scores were evaluated preoperatively, six months postoperatively, and two years postoperatively.

Statistical analysis

Excel software (Microsoft Corporation, Richmond, WA, United States) and Statistical Package for the Social Sciences (SPSS) (Version 28; International Business Machines Corporation, Armonk, NY, United States) were utilized to analyze data. Statistical significance was reached at P-values below 0.05. The frequencies of nominal and ordinal variables were calculated as a percentage. The means, ranges, and standard deviations were calculated for ratio and interval variables. A Kaplan-Meier survivorship analysis was done to assess implant survivability up to three years following primary TKA.

RESULTS

A total of 139 knees in 137 patients with follow-up times ranging from two years to 3.7 years were included in this study. The primary diagnosis for TKAs was primary osteoarthritis (n = 137, 98.6%). Two patients were indicated for TKA due to post-traumatic osteoarthritis. Patient ages ranged from 39 to 85 years, with a mean age of 66.0 years. There were 86 female patients (61.9%) and 53 (38.1%) male patients. Of the 139 knees in 137 patients, 94 were White (67.6%), 21 were Black or African American (15.1%), three were Asian (2.2%), and 21 identified as “Other” (15.1%). The majority were never smokers (n = 80, 57.6%), followed by former smokers (n = 55, 39.6%), and current smokers (n = 4, 2.9%). The most common ASA scores were two and three (n = 99, 71.2%; n = 37, 26.6%, respectively). Two patients (1.4%) had an ASA one score, and one patient (0.7%) had an ASA four score. The average body mass index was 31.4 (20.5 to 51.2). The mean Charlson comorbidity index score was 3.4 (0 to 13). The average time to follow-up was 851.8 days (730 to 1359). The average operative time was 100.8 minutes (71 to 196). The most common form of anesthesia was spinal/regional/block (n = 125, 89.9%). General anesthesia was utilized in the remaining 14 patients (10.1%). There were 61 right knees (43.9%), 76 left knees (54.7%), and two cases of bilateral TKAs (1.4%) (Tables 1 and 2). Data on patellar resurfacing were also gathered. Among the cohort, cemented fixation was used in 123 patients (88.5%), while 12 patients (8.6%) received press-fit components. Patellar resurfacing was not performed in the remaining four patients (2.9%). Regarding the bearing surface, almost three-quarters of the implants (n = 100, 71.9%) were cobalt chrome-on-polyethylene. An oxidized zirconium-on-polyethylene bearing surface was used in the other 28.1% of cases (n = 39) (Table 2).

Table 1 Patient demographics, n (%).

	Knees (n = 139)
Sex
Male	53 (38.1)
Female	86 (61.9)
Age (years) (range)	66.0 (39 to 85)
Race
White	94 (67.6)
Black or African American	21 (15.1)
Asian	3 (2.2)
Other	21 (15.1)
Smoking status
Current	4 (2.9)
Former	55 (39.6)
Never	80 (57.6)
ASA score
1	2 (1.4)
2	99 (71.2)
3	37 (26.6)
4	1 (0.7)
BMI (kg/m2) (range)	31.4 (20.5 to 51.2)
CCI1	3.4 (0 to 13)
Primary diagnosis
Primary OA	137 (98.6)
Post-traumatic OA	2 (1.4)
Time to follow-up (days) (SD)	851.8 (135.9)

¹n = 110 due to differences in data availability between centers.

ASA: American Society of Anesthesiologists; BMI: Body mass index; CCI: Charlson comorbidity index; OA: Osteoarthritis; RA: Rheumatoid arthritis; SD: Standard deviation.

Table 2 Intraoperative and implant variables, n (%).

	Knees (n = 139)
Operative time (minutes) (range)1	100.8 (71 to 196)
Anesthesia
General	14 (10.1)
Spinal/regional/block	125 (89.9)
Laterality
Right	61 (43.9)
Left	76 (54.7)
Bilateral	2 (1.4)
Patellar resurfacing/fixation
Cemented	123 (88.5)
Porous (cementless)	12 (8.6)
Non-resurfaced	4 (2.9)
Bearing surface
Oxidized zirconium-on-polyethylene	39 (28.1)
Cobalt chrome-on-polyethylene	100 (71.9)

¹n = 91 due to differences in data availability between centers.

The average length of stay was 1.4 days (0.3 to 7.2). The majority of patients were discharged home (n = 132, 95.0%), and seven went to a skilled nursing facility (5.0%) (Tables 3 and 4). During the 90-day postoperative period, two patients (1.4%) presented to the emergency department, and six patients (4.3%) required hospital readmission. One patient was readmitted for wound dehiscence, one for deep vein thrombosis, and one for periprosthetic joint infection. The other three were for non-orthopedic reasons, including syncope, psychiatric issues, and gastrointestinal bleeding. As illustrated in the Kaplan-Meier graph, a high implant survival rate of 98.6% was observed over a three-year period, during which only two patients required revision surgery (Figure 1). In both cases, periprosthetic joint infection was the indication for the revision procedure. KOOS, JR scores were recorded pre- and postoperatively (Table 4). The preoperative KOOS, JR score was 42.8. At six months postoperatively (plus or minus three months), the average score was 60.6, and at two years or more postoperatively, the average was 70.8. The magnitude of change from the preoperative baseline was also determined. The cohort demonstrated a mean KOOS, JR score improvement of 17.4 points at six months, which further increased to an average of 26.1 points by the two-year mark (P < 0.001). This indicates that there was continued improvement in KOOS, JR scores in the three years following TKA.

Figure 1 Kaplan-Meier survivorship analysis and survival confidence intervals. A: Kaplan-Meier survivorship analysis; B: Survival confidence intervals. TKA: Total knee arthroplasty.

Table 3 Clinical outcomes, n (%).

	Knees (n = 139)
Mean LOS (days) (range)1	1.4 (0.3 to 7.2)
Discharge disposition
Home	132 (95.0)
SNF	7 (5.0)
90-day ED visits	2 (1.4)
90-day readmissions	6 (4.3)
Wound dehiscence	1
DVT	1
PJI	1
Non-orthopedic related	3
MUA	1 (0.7)
All-time revisions	2 (1.4)
Septic revision	2
Aseptic revision	0

¹n = 117 due to differences in data availability between centers.

LOS: Length of hospital stay; SNF: Skilled nursing facility; ED: Emergency department; DVT: Deep venous thrombosis; PJI: Peri-prosthetic joint infection; MUA: Manipulation under anesthesia.

Table 4 Patient-reported outcome measures¹.

		Mean KOOS, JR (SD)	P value
Preoperative	n = 63	42.8 (18.0)
Six months	n = 73	60.6 (14.4)
Two years	n = 57	70.8 (19.1)
Δ Preop to six months	n = 61	17.4 (18.9)	< 0.001
Δ Preop to two years	n = 46	26.1 (19.7)	< 0.001

¹Variations in n due to differences in patient follow-up and availability of patient-reported outcome measures.

KOOS, JR: Knee Injury and Osteoarthritis Outcome Score for Joint Replacement; SD: Standard deviation; Δ Preop: Difference between pre- and postoperative scores.

DISCUSSION

Within a follow-up period extending to 3.7 years, the MD tibial insert in CR TKA has shown both excellent implant survivorship and very good clinical outcomes. Clinically, patients reported significant improvements across multiple domains, including pain management, activity endurance, sports functionality, and overall quality of life. These positive results were substantiated by the statistically significant increase observed in postoperative KOOS, JR scores.

As the annual volume of TKA procedures continues to rise, there is a corresponding demand for advancements in implant design that lead to better patient satisfaction. It has been demonstrated by previous clinical, radiographic, and in vivo analyses that the natural knee joint does not operate as a simple four-bar mechanism[4,8-10]. Instead, the natural knee’s mechanics involve a relatively constrained medial side, while the less constrained lateral compartment accommodates both rotational and translational movements. This biomechanical concept can be replicated using specific insert designs. The MD insert, for example, utilizes an asymmetrical design with increased medial congruency to promote natural joint motion, and it has achieved an excellent survivorship of 98.6% at two to 3.7 years of follow-up.

Although CR and PS implants demonstrate impressive ten-year overall survivorship rates exceeding 90%, previous studies indicate that approximately 20% of patients remain dissatisfied with their surgical outcomes[11-13]. Certain demographic factors have been correlated with patient satisfaction outcomes in some studies[13,14]. The design of conventional CR and PS prostheses may also contribute to patient dissatisfaction and reduced functional outcomes, as these designs have been associated with kinematic abnormalities such as paradoxical motion, mid-flexion instability, and insufficient roll-back[15-17]. The medial stabilized (MS) concept emerged in the 1990s as a solution to these kinematic abnormalities. This design was developed to mimic the knee's natural asymmetric constraint by utilizing a highly conforming medial compartment paired with reduced lateral congruency[18]. A systematic review comparing the mid-term outcomes of MS prostheses against standard CR and PS articulations revealed a comparable five-year survivorship of approximately 96.5%. Furthermore, the review demonstrated improved stability and lower revision rates for MS designs (2.4%) vs the CR and PS cohorts (13% and 7.2%, respectively)[19-21]. Additionally, patients who received an MS prosthesis showed substantial enhancements across various functional scores. Notably, the KOOS, JR score exhibited an average increase of 39.3 points from the preoperative baseline to the two-year follow-up[19]. The authors ultimately concluded that MS TKA designs are associated with superior outcomes in high-end function[19]. In our own patient cohort, the investigated MD insert also achieved comparably high survivorship rates and significant improvements in functional outcome scores. It is noteworthy that while some cases did require revision surgery, none of these revisions were attributed to issues with implant longevity, specifically loosening or instability, which are common sources of early implant failure. Therefore, future studies are required to evaluate the long-term outcomes, including survivorship, PROMs, and complication rates, in order to more effectively compare this implant to its variants.

To contextualize the clinical performance of the MD liner, it is important to compare it with existing tibial insert designs, including medial congruent (MC) and PS inserts. MC inserts derive their stability from a highly congruent medial polyethylene insert. The design philosophy of the MC insert is analogous to that of the MD insert, as both aim to re-establish native knee kinematics in addition to stability. This is accomplished by featuring a moderately confined medial compartment, thereby shifting most rolling and translational movements to the lateral compartment[22]. Compared to the CR design, the MC bearing design achieved greater tibial external rotation during the extension phase of the swing, known as the “screw-home mechanism”. This enhanced rotation is attributed to the MC design’s increased anterior lip height, more posterior dwell point, and modified sagittal congruency[23]. A previous study reported that the MC design was associated with reduced pain and improved range of motion relative to the PS group. Furthermore, it demonstrated an increased Forgotten Joint Score-12 when compared against the CR group, even though outcomes did not vary drastically overall[23]. An additional study observed that patients with an MC insert design achieved a superior range of knee flexion, as assessed clinically, compared to those with a PS insert design[24]. Published studies show discrepancies regarding the relative performance of MC and CR inserts. A double-blinded, randomized, and controlled investigation comparing these two designs found that the MC design provided knee mechanics closer to the physiological norm and was linked to higher patient satisfaction. This satisfaction was particularly evident in patients’ reported confidence when performing everyday tasks[24]. However, not all studies concur, as some have reported no significant differences between the MC design and the standard CR concept. A randomized study involving 60 patients, for example, analyzed implant migration after they were provided with either a CR or an MC insert. The findings indicated no significant variation in implant survivorship or PROMs when comparing the two groups[25]. A study comparing the PS and MC designs noted that while the PS group showed significantly greater improvement at three months postoperatively, the outcomes for both groups were comparable by the one-year follow-up mark[26]. A major disadvantage associated with the PS design, however, is the accessory femoral box cut. This step carries the risk of compromising the femoral condylar strength and bone stock, both of which may be critical for subsequent revision procedures[27,28].

One of the most significant contributors to early implant failure after primary TKA is postoperative knee instability[29]. Consequently, the balance between adequate stability and an optimal range of motion requires meticulous management, as excessive constraint may elevate the risk of early implant loosening and failure. One study comparing PS and MC implant outcomes found that after achieving sufficient stability intraoperatively, using the lowest necessary level of joint constraint contributed to superior ROM when measured at the last clinic visit[27]. It is theorized that enhancing congruency could lead to improved stability that better replicates natural knee kinematics. Nonetheless, this presents a potential trade-off, as there is no consensus on whether increased articular conformity also elevates the likelihood of early aseptic loosening. In the future, in vivo studies are necessary to investigate this relationship by comparing how varying degrees of constraint, congruency, and stability impact clinical outcomes.

Limitations

Several limitations within this study must be recognized. First, as a retrospective analysis, it is susceptible to various biases, such as selection bias. It is also possible that confounding variables were missed in the data collection process. Additionally, the lack of a control group prevented the accurate control of potential confounders. Further, as a multicenter study, there is an increased risk of unmeasured confounders stemming from variations in pre- and postoperative standards of care as well as intraoperative protocols and techniques. Although the generalizability of these results may be constrained by the study’s relatively small sample size (n = 139), this study is the first to provide an analysis of this specific implant design, which has not been investigated in prior literature.

CONCLUSION

This MD tibial insert represents a promising and effective alternative when compared to conventional CR and PS liner designs, contributing to the ongoing effort to achieve more natural knee kinematics. To increase generalizability and statistical power, further, larger-scale studies including control groups are necessary to assess the long-term outcomes of medially stabilizing implant designs across several key domains: Functionality, joint stability, kinematics, patient satisfaction, and implant durability.