1 Introduction

For many years total knee replacement (TKR) surgery has demonstrated clinical success and cost-effectiveness in alleviating pain and enhancing the long-term function of patients who suffer from knee arthritis.¹ The incidence of surgical complications is low, yet up to 10–20% of patients are not satisfied with their outcome.^1–5 Dissatisfaction in high-demand, often younger patients, may be attributable to functional limitations of the implant, especially when walking long distances, running, squatting, kneeling or twisting. The majority of conventional TKR implants to date do not replicate the natural kinematics of the knee, which is now becoming recognised as an important element influencing the clinical outcome.

During flexion in the native knee, the tibial articulating surfaces slide and roll along the femoral condyles.⁶ Several studies have also confirmed that a natural initial rollback of the femur with respect to the tibia occurs mainly in the initial 30 degrees of flexion, followed by a sliding motion which is mainly during the motion arc.^7,8

This initial rollback movement allows correct positioning of the patella tendon to allow force transmission of the quadriceps through the joint. It is considered to reduce friction and abrasive wear to prostheses that exhibit this property (Fig. 1).¹⁰,¹¹

Close

Fig. 1 Comparison of the effects on the patella in dependence of the initial roll-back.

Export to PPT

The Physica KR implant is a Posterior cruciate ligament (PCL) retaining modular knee prosthesis with a Kinematic Retaining (KR) mechanism. It has been designed following the principles of standard tibiofemoral joint anatomy and the natural movements patterns of the native knee. The rationale of our study is to assess the clinical, radiological and survivorship outcomes in a KR TKR. This study builds on the published two-year follow-up data, which demonstrated early functional recovery with significant pain relief, when implanted a KR implant.¹² We hereby present the results of this five-year patient cohort study.

2 Methods

The investigation was structured as a multi-centre, non–blinded, prospective cohort research study focusing on the CE-certified Physica KR device (LimaCorporate, Italy), including a baseline control.¹²

An initial power analysis was conducted using Power and Precision v4 software (Biostat, NJ, USA).¹² According to standard procedures, the software projected a treatment difference of 44.0 points on the Knee Society Score (KSS), with a standard deviation of 20.0 and a significance level (alpha) of 0.05. This analysis indicated that a minimum sample size of 130 patients was necessary to achieve a statistical power of 0.80 for rejecting the null hypothesis.

With an expected dropout rate of approximately 20%, the target enrolment was set at 156 patients.¹²

After obtaining ethical approvals, 156 patients (87 men and 69 women, with an average age of 69 years) underwent Total Knee Replacement (TKR) utilising the Physica KR knee implant at three different European sites.¹² Patients selected for the study suffered from primary knee osteoarthritis, which had previously failed to respond to conservative management. Only once the decision for TKR had been made in an outpatient clinic by one of the senior authors, were the patients recruited to the trial.

All surgeries were performed by the individual senior authors at each European study site. Anaesthesia was either spinal or general, and a standardised surgical technique was performed. This was a medial parapatellar approach, with bone resections referenced using proprietary jigs aligned intramedullary for the femur and extramedullary for the tibia. Antithrombotic and antibiotic prophylaxis were administered as per each institution's standard protocol.

Patient evaluations were conducted pre-operatively (T0) and post-operatively at 6 weeks (T1), 6 months (T2), 1 year (T3), 2 years (T4), 3 years (T5), and 5 years (T6) using clinical and functional performance assessments.¹²

The primary endpoint was the Knee Society Score (KSS), while secondary endpoints included patient-reported outcome measures such as the Knee injury and Osteoarthritis Outcome Score (KOOS), Oxford Knee Score (OKS), Visual Analogue Scale (VAS) satisfaction, Forgotten Joint Score (FJS-12), and records of serious adverse events.¹²

Radiographic assessment included weight bearing long-leg films to assess overall alignment, as well as standard anteroposterior (AP), lateral and skyline projections. The severity of osteoarthritis was classified pre-operatively according to the Kellgren and Lawrence grading scale⁹ from grade 1 to 4. Post-operative radiographic assessment was carried out after 6 weeks, 6 months, 1 year, 2, 3 and 5 years.¹² Weight-bearing AP and lateral views of the knee, long-leg and skyline patella views were all required and evaluated according to the Knee Society radiographic assessment and some of its modifications.^10,13 Long-leg standing x-rays were acquired only at 6-week follow-up to analyse the accuracy of the components position with respect to the anatomical references. The pre-operative long leg standing x-rays were used as a baseline. Post-operative assessments included a modified evaluation of presence, location and thickness of periprosthetic radiolucent lines. Radiological evaluation was conducted by an independent qualified examiner.

Analysis of implant survivorship was calculated using the Kaplan Meier method¹⁴ with a confidence interval of 95%. The analysis was performed with the following end points: Revision of the components due to aseptic loosening, infection, osteolysis, or instability.

Data was analysed using SAS software (SAS Analytics, NC, USA). Paired t-tests were employed to compare pre-operative baseline measurements with follow-up results, evaluating functional and clinical outcomes, satisfaction, and safety profiles. Outcome measures were also compared across all the post-operative clinic visits using the Wilcoxon signed-rank test and further categorized by age and gender using the Wilcoxon rank-sum test at each review. P-values were adjusted for multiple comparisons using the false discovery rate (FDR) method, with values less than 0.05 considered statistically significant.

3 Results

At five years post-surgery, 134 patients (86.5% of our initial, total patient cohort) were retained for follow-up.

3.1

3.1 Functional clinical outcomes and patient satisfaction

3.1.1

3.1.1 Range of motion

Pre-surgery range of motion (116 ± 14) was generally recovered by six months (113 ± 14.6) and maintained at five years (117 ± 12.0).

3.1.2

3.1.2 KSS (Knee Society Score)

3.1.2.1

3.1.2.1 KSS scores clinical

Overall, 87.1% of patients showed an excellent to good clinical KSS score from 1 year to 76.9% at 5 years after surgery (see Table 1 and Table 2) (see Fig. 2).

Table 1 KSS Clinical Scores through five post operative years (n (%)).

	Preoperative	At six weeks	At six months	At 1 year	At 2 years	At 3 years	At 5 years
Description
Excellent	2 (1.3%)	33 (21.2%)	79 (50.6%)	106 (67.9%)	113 (72.4%)	106 (67.9%)	101 (64.7%)
Good	4 (2.6%)	43 (27.6%)	45 (28.8%)	30 (19.2%)	14 (9%)	20 (12.8%)	19 (12.2%)
Fair	28 (17.9%)	23 (14.7%)	7 (4.5%)	3 (1.9%)	3 (1.9%)	6 (3.8%)	7 (4.5%)
Poor	122 (78.2%)	53 (34%)	19 (12.2%)	10 (6.4%)	8 (5.1%)	8 (5.1%)	4 (2.6%)

Table 2 KSS Functional Scores through five post operative years (n (%)).

	Preoperative	At six weeks	At six months	At 1 year	At 2 years	At 3 years	At 5 years
Description
Excellent	22 (14.1%)	52 (33.3%)	120 (76.9%)	130 (83.3%)	121 (77.6%)	129 (82.7%)	116 (74.4%)
Good	18 (11.5%)	29 (18.6%)	14 (9%)	9 (5.8%)	8 (5.1%)	3 (1.9%)	8 (5.1%)
Fair	28 (17.9%)	20 (12.8%)	4 (2.6%)	2 (1.3%)	2 (1.3%)	5 (3.2%)	2 (1.3%)
Poor	88 (56.4%)	51 (32.7%)	12 (7.7%)	8 (5.1%)	7 (4.5%)	3 (1.9%)	5 (3.2%)

Close

Fig. 2 Mean Clinical KSS scores from preoperative period to five years.

Export to PPT

KSS functional scores at 6 months demonstrated excellent scores in 77% of patients. This result was maintained at the subsequent reviews up to 5 years.

3.1.3

3.1.3 KOOS (knee injury and Osteoarthritis Outcome Score)

All five KOOS dimensions highly improved from before surgery to five-years, and the changes were statistically significant (Table 3).

Table 3 KOOS sub-scores by timepoint (mean (+_SD)).

Visit	Symptoms or Stiffness Partial Score	Pain Partial Score	Function, Daily living Partial Score	Function, Sports Partial Score	Quality of Life Partial Score
before surgery (T0)	45.3 (18.0)	41.2 (13.3)	46.4 (16.9)	17.9 (17.5)	22.8 (15.0)
at six weeks (T1)	64.1 (16.0)	67.5 (16.1)	75.8 (14.9)	36.3 (27.0)	52.4 (18.9)
at six months (T2)	79.4 (14.8)	83.1 (16.1)	85.3 (14.1)	63.5 (24.6)	69.6 (22.3)
at one year (T3)	83.9 (14.8)	88.8 (14.7)	89.4 (13.8)	69.3 (23.8)	77.8 (20.9)
at two years (T4)	88.5 (13.4)	92.7 (11.9)	93.1 (11.5)	74.5 (22.3)	82.3 (20.1)
at three years (T5)	90.5 (10.6)	93.4 (12.4)	92.9 (11.5)	76.1 (22.9)	83.5 (19.6)
at five years (T6)	89.9 (11.2)	92.0 (11.9)	92.9 (12.3)	74.0 (24.7)	84.2 (20.6)

3.1.4

3.1.4 Oxford Knee Score (OKS)

80.1% of patients reported good or excellent score at five years, with a statistically significant and progressive improvement from 6 weeks of surgery (p < 0.01) (Fig. 3 and Table 4).

Close

Fig. 3 Mean Oxford Knee Score from preoperative to five years post-operative.

Export to PPT

Table 4 OKS at time points up to five years post-operative.

Visit:	Preoperative	At six weeks	At six months	At 1 year	At 2 years	At 3 years	At 5 years
Description
Excellent	0 (0.0%)	13 (8.3%)	74 (47.4%)	102 (65.4%)	111 (71.2%)	113 (72.4%)	106 (67.9%)
Good	4 (2.6%)	42 (26.9%)	48 (30.8%)	37 (23.7%)	18 (11.5%)	19 (12.2%)	19 (12.2%)
Fair	30 (19.2%)	52 (33.3%)	18 (11.5%)	6 (3.8%)	4 (2.6%)	4 (2.6%)	4 (2.6%)
Poor	122 (78.2%)	45 (28.8%)	7 (4.5%)	3 (1.9%)	2 (1.3%)	4 (2.6%)	4 (2.6%)

3.1.5

3.1.5 VAS satisfaction (Visual analogue patient satisfaction scale)

The change of VAS was already positive at 6 weeks after the surgery, with an average value of almost 74 mm. From 1 year to 5 years, the average VAS was over 85 mm.

3.1.6

3.1.6 FJS – 12 (The Forgotten Joint Score)

The FJS-12 increased from 64.5 at 1 year to 75.7 at 2 years (p < 0.01) and then remained constant throughout the subsequent follow-ups with a mean score of 79.2 at 5 years (Table 5).

Table 5 Overview of Patient Retention Rates, Clinical and functional Scores over Five Years Post-Surgery.

Score	Patients	KSS Clinical	KSS Function	OKS	FJS	VAS Satisfaction
Pre-Operative (T0)	156 (100%)	48.6 (14.6)	55.6 (18.54)	20.2 (6.8)	/	/
6 Weeks (T1)	153 (98.1%)	66.2 (16.3)	65.5 (22.0)	30.5 (7.8)	/	73.8 (19.9)
6 Months (T2)	150 (96.2%)	78.1 (14.6)	85.3 (18.9)	39.5 (6.9)	/	81.0 (19.6)
1 Year (T3)	151 (96.8%)	82.4 (13.5)	90.0 (17.1)	42.3 (5.9)	64.5 (27.1)	87.1 (14.7)
2 Years (T4)	138 (88.5%)	83.2 (12.4)	89.7 (19.7)	44.1 (5.0)	75.7 (26.5)	88.4 (16.6)
3 Years (T5)	141 (90.4%)	83.0 (11.6)	93.3 (13.1)	43.8 (5.9)	79.8 (24.2)	88.3 (16.1)
5 Years (T6)	134 (86.5%)	81.7 (9.3)	92.1 (12.6)	43.7 (5.6)	79.2 (26.4)	84.0 (19.4)

This table presents the retention rates of patients (n (%)) and the mean clinical and functional scorews ((±SD) over a period of five years following surgery. Additionally, VAS Satisfaction and FJS were evaluated starting from six weeks and one-year post-surgery, respectively.

This figure (Fig. 4) shows the AP, lateral and long leg alignment radiographic films of a patient's left knee before and after surgery with the Physica KR total knee replacement.

Close

Fig. 4 Radiographic evaluation.

Export to PPT

3.1.7

3.1.7 Radiographic outcomes

Post-operatively and throughout the entire study period, positioning of the TKR components was maintained, as well as limb alignment.¹²

Immediate post operative films showed 1 mm radiolucent lines in two patients (tibia - zone 4, femur - zone 1).¹² By 2 years post surgery, there were two femoral lucencies greater than 2 mm wide (in zone 1) and one tibial lucency over 2 mm in Zone 4.¹²

At 5 years, there were 1 mm femoral lucencies in 14 patients, in Zone 1, representing 9% of the study population. In the tibial component, there were 5 patients with 1 mm lucencies in Zone 4 and 6.

The majority of these radiographic lucencies were initially recorded at 6 months, and were non-progressive throughout the study.¹²

Fewer than <1.5% of patellar components demonstrated radiolucent lines at any time point.

3.1.8

3.1.8 Complications

122 potential safety events were recorded during the study period of 5 years, however, only four were considered related, or possibly related to device: 1 patient suffered loosening of the tibial implant, 1 patient with a post-operative haematoma, and 2 patients with a post-operative infection. Two patients (aseptic loosening, infection) underwent revision surgery.

3.1.9

3.1.9 Implant survivorship

The Overall implant survivorship at 5-years was 99.4%, with a confidence interval of 95.5–99.9 (Table 6).

Table 6 Parameters for Kaplan-Meier estimation.

Parameter	Procedure to 3 years	3–5 yearse
No. at Riska	156	154
No. of Events	0	1
No. Censoredb	2	47
Overall implant Survivorship (%)c	100	99.4
95% CId	100.0, 100.0	95.5, 99.9

a Number of subjects at risk at the beginning of interval.

b Subjects are censored because their last follow up has not reached the end of the time interval or because they are lost to follow-up.

c K-M Estimate made at the end of the time interval.

d Log-log confidence interval.

e One event occurred after 5 years.

4 Discussion

The Physica KR implant is designed to mimic native knee kinematic motion. Conventional TKR designs rely on a constant radius of both medial and lateral condylar entities.¹⁵ This symmetry maintains a constant rotational axis point on the tibia, which can lead to paradoxical anterior sliding of the knee.¹⁵ The articulating surface geometry of the Physica KR components is asymmetrical with a graduated radius of femoral condyle curvature, designed to recreate the rolling-gliding mechanism and constant ligamentous tension during the flexion arc, typical of a native knee.

This study employed several clinical outcome measures during the five-year follow up period. Traditional methods of evaluating ‘success’ following TKRs focus on implant longevity and radiological parameters. However, given that 10–20% of patients post TKR remain discontented with their outcomes^2,3, greater emphasis on Patient Reported Outcome Measures (PROMs) may be appropriate. This is reflected by the English National Joint Registry's aim to collect PROMs to assess and inform on pain and functional outcomes.¹⁶

KSS scoring showed statistically significant improvement from pre and post op scoring across all five domains at five years. The mean pre-operative OKS was 20.2, with a mean post-operative improvement of 23.6 (22.3–25.0 p < 0.01). More than eighty percent of the patients had a good or excellent OKS at five years. This improvement was statistically significant at 6 weeks postop (p < 0.01) and remained constant throughout the 5 years follow-up. At three years 113 patients out of 141 patients had an excellent OKS, however at five years due to the loss of 7 patients from the cohort this number falls to 106 despite the other parameter (good/fair/poor) patient numbers staying the same. The excellent rate at 5 years is therefore likely artificially lower than expected due to this, and closer to 72.4% rather than 67.9% (Table 4). Additionally, the mean FJS-12 increased from 64.5 at 1 year to 75.7 at 2 years (p < 0.01) and remained constant throughout subsequent follow up.

The clinical, radiological and survival data for the Attune, Triathlon, and Persona are summarised in (Table 7). The Physica KR implant showed comparable improvements in PROMs, with a small improvement in OKS at five years compared to the Attune, Triathlon and Persona. Additionally, a higher post op FJS score at five years which was statistically significant was reported with the Physica KR implant.

Progressive radiological lucencies can represent an early sign of subsequent failure in cemented TKRs, with greater than 9 mm considered a failure.¹⁷ 70% of knees in this study showed no radiological lucencies at either the femoral or tibial components. No patients had greater than 2 mm femoral lucencies, and only 4 patients had greater than 2 mm tibial lucencies. Although comparison studies with other implant designs (see Table 7) employed differing methods of assessing radiological outcomes, the frequency of lucencies observed is comparable.

The present study demonstrated excellent survivorship at five years of 99.4% (CI 95.5–9.9), with only two revisions in this period. This is comparable to other contemporary systems such as the Attune (97.8%),¹⁸ the Triathlon (97.6% (CI 95.6–99.6),¹⁹ and the Persona (98% at five years).²⁰

5 Conclusion

We acknowledge the two main limitations of this study; firstly, that there was no direct control group, and secondly the Covid-19 pandemic delaying a small proportion of our patient's follow up. Despite this, our study demonstrates that when using this novel TKR design – one can expect good-excellent functional outcomes, whilst achieving significant pain relief even in a high demand patient cohort. The Physica KR Implant has a remarkable survivorship of 99.4% at 3–5 years. This study adds further weight to the argument that implants with physiological kinematics should be in one's armoury when performing joint arthroplasty in high demand patient cohorts.

Ethical approval

Approval was obtained from the Regional Ethics Committee of the principal centres. The trial was registered through ClinicalTrials.gov (NCT02118272). For UK sites: IRAS project ID is 137766/REC reference is 13/EE/0438. For Italian site: EC reference is SPE_14.080.

The implant holds a European conformance CE Mark and has been registered with ODEP (Orthopaedic Data Evaluation Panel) and the Beyond Compliance Advisory Group for ongoing clinical assurance.

Funding

This research received funding from Lima Corporate Spa, Italy.

Informed consent

All patients underwent appropriate informed consent for inclusion in this study and for measurement, storage and analysis of their data.

CRediT authorship contribution statement

Michael D. Wilson: drafted the manuscript, All the authors participated in patient recruitment, the surgical procedures and post-operative care and follow-up, All authors read, amended and approved the final manuscript. Suraj Kohli: drafted the manuscript, All the authors participated in patient recruitment, the surgical procedures and post-operative care and follow-up, All authors read, amended and approved the final manuscript. James O. Smith: drafted the manuscript, All the authors participated in patient recruitment, the surgical procedures and post-operative care and follow-up, All authors read, amended and approved the final manuscript. Ian W. Barlow: designed and coordinated the study, All the authors participated in patient recruitment, the surgical procedures and post-operative care and follow-up, All authors read, amended and approved the final manuscript. Christian Carulli: All the authors participated in patient recruitment, the surgical procedures and post-operative care and follow-up, All authors read, amended and approved the final manuscript. Nicola Monteleone: All the authors participated in patient recruitment, the surgical procedures and post-operative care and follow-up, All authors read, amended and approved the final manuscript. Massimo Innocenti: designed and coordinated the study, All the authors participated in patient recruitment, the surgical procedures and post-operative care and follow-up, All authors read, amended and approved the final manuscript. Adrian R. Harvey: designed and coordinated the study, drafted the manuscript, All the authors participated in patient recruitment, the surgical procedures and post-operative care and follow-up, All authors read, amended and approved the final manuscript.