1 INTRODUCTION

Total Knee Arthroplasty (TKA) is the gold standard treatment for end stage knee osteoarthritis (OA). Many different designs exist and the medial pivot concept is gaining popularity in some regions. The SAIPH TKA (MatOrtho) was designed to replicate the natural asymmetry of the knee, in accordance with the kinematic principles described by Pinskerova et al. ¹ This implant is a second-generation medial pivot TKA, based on prior experience with the MRK design (MatOrtho). It has a highly congruent, ‘ball and socket’ medial polyethylene liner which offers more stability throughout range of motion than most other designs. However, there have been concerns that this may transmit more stress to the bone-implant interface predisposing to aseptic loosening, the leading cause of TKA revision ²

Our study therefore aimed to investigate this implant in accordance with the stepwise introduction of new technology as per Malchau ³ It has become evident through past implant design problems that a phased introduction of new technology in orthopaedics is important. Limiting exposure to untested implants during systematic assessment for early signs of failure is an important part of risk mitigation ^3–5 Radiostereometric Analysis (RSA) has been shown repeatedly as an effective method in assessing micromotion and predicting loosening in vivo years before symptoms and revision surgery ^6–8 enabling controlled exposure of a small number of patients to an untested implant with reliable results. Although micromotion patterns in TKA implants are well studied, a paucity of literature exists for the relatively new medial pivot designs. This is the first RSA trial on the SAIPH knee, carried out with the objective of investigating its early micromotion patterns and clinical performance.

2 Material and METHODS

Inclusion criteria for this study were symptomatic osteoarthritis eligible for primary arthroplasty surgery in ages >18 years. Exclusion criteria were planned or actual pregnancy, prior ipsilateral arthroplasty surgery, diagnosed osteoporosis or osteopenia and medications associated with it such as bisphosphonates, corticosteroids and Protos. Surgery was performed by a single experienced arthroplasty surgeon. Recruitment, operations, follow-up and radiological evaluations were all carried out at Fremantle Hospital, Western Australia.

The SAIPH Total Knee Arthroplasty is a cruciate-sacrificing, fully cemented implant system made from Cobalt-Chromium Alloy. SAIPH has a spherical medial femoral condyle and an Ultra High Molecular Weight Polyethylene (UHMWPE) liner which is highly congruent medially and implanted as a fixed bearing. The trochlea is lateralised in flexion to maximise the quadriceps lever arm about the centre of rotation, the medial condyle. All patients received patellar resurfacing with fully cemented implants.

The primary outcome was Maximum Total Point Motion (MTPM) of the tibial component versus the tibia at 12 and 24 months, measured using RSA. Migration and rotation of the tibial component about a virtual centroid were also calculated. RSA evaluation was carried out at 6 weeks, 12 and 24 months, hence capturing the micromotion experienced by the tibial component after initial settling-in, which is considered to bear little relevance to long term stability ⁸

Quantitative analysis and patient anthropometric data were collected in a centralised database for cataloguing, and analysis was conducted throughout the study course.

A sample size of 35 was set for this trial based on standard RSA procedure as it is generally considered that ∼25 patients are required to produce statistically valuable data and detect clinically significant migration exceeding experimental error ^9–11. We recruited an additional 10 patients to allow for attrition and technical issues. This implant is commercially available hence overpowering the study was deemed acceptable in ensuring strong results.

RSA investigations were carried out according to standardised procedure¹² using uniplanar supine imaging with RSA calibration cage 43, and analysed using UmRSA 6.0 software (RSA Biomedical, Umeå, Sweden). 1 mm tantalum RSA markers were implanted in the tibial metaphysis and the polyethylene liner during the arthroplasty surgery. These were subsequently used to determine relative 3D positioning of implants and track changes over time. Standardised digital radiographs were taken at a single radiological centre for all patients.

Rigid body centre translation and rotation of the polyethylene implant as well as Maximum Total Point Motion of the markers embedded in it were calculated relative to the tibial markers. The polyethylene is fixed to the tibial baseplate hence the two move as one. Migration was reported in millimetres and rotation in degrees on the X (medial-lateral), Y (caudal-cranial) and Z (anterior-posterior) axes. Patient positioning according to cardinal axes was set at first postoperative imaging. A cut-off for mean error of rigid body fitting was set at 0.35 mm. As the system must possess systematic and random error, accuracy was assessed by precision testing under repeatability conditions. These tests yielded very low error.

As a simple series, median values are presented along with mean, standard deviation and 95 % confidence interval in Tables 1 and 2. Data presented includes translation, rotation, MTPM and mean errors of rigid body fitting.

This study was funded by MatOrtho and supported by Fremantle Hospital.

3 RESULTS

35 patients of mean age 71 were recruited and operated over a period of 60 months. Fig. 1 outlines the patient flow, reasons for missing data and the data analysed at 12 and 24 months. No revisions were carried out during the study.

Close

Fig. 1 Patient flow.

Export to PPT

Migration and MTPM data are listed in Tables 1 and 2 Component subsidence was minor. At 12 months the tibial component median subsidence was 0.04 mm (−0.65-0.34 mm). At 24 months the median subsidence was 0.02 mm (−1.04-0.56 mm). Individual subsidence was quite consistent, as illustrated in Fig. 2. Translations on the other axes were minor (Table 2).

Close

Fig. 2 Individual patient implant subsidence at 12 and 24 months (mm).

Export to PPT

MTPM values are illustrated in Table 2 and Fig. 3. Median MTPM for the group was 0.44 mm at 12 months (0.10–2.71 mm), and 0.46 mm at 24 months (0.12–3.60).

Close

Fig. 3 Group Median MTPM at 12 months relative to risk categories as per Pijls et al. (mm).

Export to PPT

Rotational data for the tibial component are illustrated in Table 1. Signed rotations about all axes were minor. Absolute values for rotational movement were similarly low. Maximal rotation was 3.87° at 24 months and the highest median absolute rotations occurred on the x axis (flexion/extension).

4 Discussion

A recent meta-analysis by Pijls et al.¹³ stratifies implants into groups using mean MTPM at 12 months, accepting <5 % revision rate at 10 years. Based on this assessment MTPM <0.5 mm is classified as safe, 0.5–1.6 mm as at some risk and >1.6 mm as high risk. The micromotion pattern of this implant therefore falls well within safe parameters utilising median values ¹³ Mean values are less reliable considering the data spread however these are also reassuringly low. This migration and rotation data is consistent with stable fixation of the tibial baseplate in host bone.

There was no clear pattern of rotation as illustrated by the low but varied signed values, indicating minor rotation occurred in both directions on each axis in these patients. Measurement error can be assumed to contribute most significantly to the X axis rotations (flexion/extension) due to the plane of orientation of the implant markers. Z axis rotations (varus/valgus) remain low.

Limitations in this study are minimised by focusing on high quality measurements using the objective and precise method of RSA. All measurement and analysis was carried out by the author, with no conflicts of interest to declare. Findings at 24 month follow-up form the gold standard of RSA reporting. The cohort study protocol is not designed or able to directly compare this implant with others.

5 Conclusions

Benign micromotion patterns on accepted scales and time points, no revisions and high satisfaction provide ample support for this medial pivot total knee arthroplasty design.

CRediT authorship contribution statement

Nils Oscar Nivbrant: Formal analysis, Writing – original draft. Piers J. Yates: Conceptualization, surgical implantation, Writing – review & editing.

Ethics

Informed consent was obtained for all subjects in this study and ethics board approval was granted prior to commencement.

Disclosure

This study was funded by MatOrtho.

Consent

Informed consent was obtained for all subjects in this study and ethics board approval was granted prior to commencement.

Ethical statement

Study approved by local ethical review board prior to commencement and adhered to as per standard.

Funding sources

This study was funded by MatOrtho Ltd, Leatherhead, Surrey UK.