Tibial migration of a novel cementless total knee arthroplasty

Abstract

Aims

The search for enhanced fixation and longevity in total knee arthroplasty (TKA) continues. Radiostereometric analysis (RSA) allows early detection of implant migration predictive of future aseptic loosening and is the recommended method for evaluating new implants. This multicentre RSA study evaluated early fixation and migration stability of a novel cementless tibial baseplate used in primary TKA.

Methods

A total of 30 patients undergoing primary TKA were enrolled at two academic centres. All patients received a porous cementless tibial baseplate combined with either a cementless or cemented femoral component. The patella was selectively resurfaced based on surgeon preference and was resurfaced using either a cementless onlay or a cemented inset biconvex design. During surgery, RSA beads were inserted into the tibia, femur, and patella. RSA examinations were performed at six weeks (baseline), six months, 12 months, and 24 months postoperatively. The primary outcome was change in maximum total point motion (MTPM) of the tibial baseplate over 24 months. Secondary outcomes included component migration patterns and patient-reported outcome measures (PROMs).

Results

In total, 29 patients completed minimum 24-month follow-up. Mean tibial baseplate MTPM was 1.018 mm (SD 0.742) at six months and 1.210 mm (SD 0.571) at 24 months. The change in MTPM between six and 12 months was −0.096 mm (SD 0.551) and between 12 and 24 months was 0.095 mm (SD 0.603), remaining below accepted RSA thresholds associated with increased risk of aseptic loosening. Translational and rotational migration stabilized by six months. Patients reported substantial improvement in functional (Oxford Knee Score, Forgotten Joint Score) and health-related (EuroQol five-dimension questionnaire) PROMs and high satisfaction at 12 and 24 months.

Conclusion

This novel cementless tibial baseplate demonstrated early stabilization with no evidence of continuous migration over two years. These RSA findings suggest a low risk of future aseptic loosening, and support the early fixation performance of this implant design.

Cite this article: Bone Jt Open 2026;7(9):1239–1247.

Introduction

There continues to be a need for enhanced fixation and improved longevity in total knee arthroplasty (TKA). New porous surfaces and advances in 3D implant manufacturing aim to enhance biological fixation through improved osseointegration. Several new coatings and implant designs for cementless TKA have recently been introduced into the marketplace.^1-4 Unfortunately, the introduction of new implants or fixation methods carries a risk of higher revision rates, and it is not uncommon to see inferior results of newer implant designs.^5,6

Radiostereometric analysis (RSA) is a highly accurate method to detect early implant migration and has demonstrated predictive validity for late aseptic loosening.^7,8 It remains the recommended technique for evaluation of new materials and implant designs.^9,10 Acceptance thresholds for migration, particularly continuous migration, are well established and have been incorporated into stepwise algorithms for the safe introduction of new implant designs.^8,9,11-14

The purpose of this multicentre RSA study was to evaluate the early fixation of a novel porous cementless tibial baseplate used in primary TKA. We hypothesized that this implant would demonstrate acceptable early migration followed by secondary stabilization, with continuous migration remaining below accepted thresholds associated with increased risk of failure.

Methods

Study design and patients

This prospective multicentre single cohort study was conducted at two academic arthroplasty centres: London Health Sciences Centre, University Hospital (London, Canada), and Concordia Joint Replacement Centre (Winnipeg, Canada). Ethical approval was obtained at both sites, and the study was registered at ClinicalTrials.gov (NCT05226689).

A total of 30 patients were enrolled and received a fixed-bearing, cementless tibial baseplate (CONCELOC Porous Baseplate, Smith & Nephew, USA) in combination with a cruciate-retaining tibial insert, and either a cemented or cementless femoral component (LEGION, Smith & Nephew). The CONCELOC is a novel porous tibial baseplate that utilizes additive manufacturing to produce a highly porous titanium surface intended to replicate the structure of cancellous bone. The baseplate incorporates an anatomically positioned keel with antirotation press fit pegs to aid in achievement of initial stability. The tibial baseplate also consists of an anterior keel broach, not seen in previous designs, to improve fixation in dense bone and to mitigate fracture risk. The patella was selectively resurfaced using either a cemented inset biconvex button or metal-backed implant (CONCELOC Porous Patella, Smith & Nephew).

Patients were approached for enrolment in the study if they demonstrated symptomatic osteoarthritis of the knee, had a BMI of < 40 kg/m², were aged 21 years or older, able to provide written informed consent, had sufficient ligamentous function to warrant retention of the posterior cruciate ligament (PCL), and were willing to comply with the study procedures. Patients excluded from the study were those with active infection, medical contraindications for surgery, PCL deficiency, prior patellectomy, major (> 25°) coronal plane deformity, bilateral coronal plane deformity requiring simultaneous bilateral TKA, or were expected to receive contralateral TKA within one year, or had bone defects requiring revision implants.

Patient characteristics

Patient characteristics of the study cohort were typical for primary TKA patients, with a mean BMI of 30.6 kg/m² (SD 3.8) and mean age at surgery of 66 years (SD 6.0). A total of 16 patients were female (53.3%).

Surgical technique and implants

All surgeries in this study were performed by fellowship-trained arthroplasty surgeons (DDRN, BAL, EB, TRT) via a medial parapatellar approach to the knee. Neutral mechanical alignment was the target for all patients. The patella was selectively resurfaced depending on surgeon preference. Patellar resurfacing options were a cemented inset biconvex polyethylene button (GENESIS II Biconvex Patella, Smith & Nephew) or a metal-backed cementless implant (CONCELOC Porous Patella). Six to ten 1.0 mm tantalum RSA beads (Halifax Biomedical, Canada) were inserted into each of the proximal tibia and distal femur (if a cementless implant was used), and three to four beads into the patellar bone (if resurfaced with a metal-backed implant). Postoperative recovery for all patients followed standard-of-care guidelines at each institution.

Radiostereometric analysis

A baseline RSA examination was performed at six weeks following surgery with further examinations at follow-up intervals of six, 12, and 24 months. Patients were imaged in a supine position for RSA of the tibial and femoral components. For patients in whom a metal-backed patella was used, the patellar component was imaged with the patients lying on their side on the X-ray table with the operated knee slightly flexed (~15°) forward of the opposite knee, and the X-ray tubes positioned in a superoinferior orientation to avoid marker and implant occlusion by the anterior flange of the femoral component. Duplicate baseline examinations were obtained for both supine and lateral positions with repositioning of the patients between exposures to calculate intraobserver error.¹³ The RSA imaging suites at each centre have been previously described and validated for multicentre analyses.^15-17 Halifax Biomedical (Canada) performed all analyses using mbRSA (v. 4.2 RSAcore, the Netherlands) with tibial, femoral, and patellar implant models provided by the manufacturer.

Migration of the tibial baseplate with regard to the bone was assessed using maximum total point motion (MTPM) as well as overall subsidence (inferior movement). The six-week examination was used as the baseline against which all follow-up examinations were compared, including migration between 12 and 24 months. The primary endpoint was change in MTPM between 12 and 24 months. Continuous migration exceeding 0.2 mm between 12- and 24-month follow-up was considered potentially concerning based on established RSA criteria.

Patient-reported outcomes

Patient function was assessed through a variety of outcome scores including: Oxford-12 Knee Score (OKS),^18,19 EuroQoL five-dimension five-level questionnaire (EQ-5D-5L),²⁰ Forgotten Joint Score (FJS-12),²¹ Knee Injury and Osteoarthritis Outcome Score for Joint Replacement (KOOS JR),²² University of California at Los Angeles Activity scale (UCLA Activity),^23,24 as well as visual analogue scales (VAS) for knee pain and satisfaction with surgery. The VAS for pain was scored as 0 representing no pain (best outcome) and 100 representing the worst pain imaginable (worst outcome). Whereas the VAS for satisfaction was reversed with 0 representing complete dissatisfaction (worst outcome) and 100 representing complete satisfaction (best outcome). Perioperative data were also collected, including the length of surgery, tourniquet time, length of stay in hospital, blood loss, and type of anaesthesia used.

Statistical analysis

Based on migration thresholds established by Ryd et al,²⁵ this study was powered to detect a threshold of 0.2 mm of tibial baseplate migration (MTPM) between RSA examinations at 12 and 24 months postoperatively. While this threshold is based on cemented tibial baseplates, a recent study by Laende et al²⁶ demonstrated that while early postoperative migration of cemented and cementless baseplates differ, the one-year to two-year migration pattern is similar. Therefore, we applied the 0.2 mm threshold to the current study with an estimated SD of migration of 0.27 mm based on published data of 138 TKAs.²⁷ Using α equal to 0.05, a sample size of 26 patients was required to reject the null hypothesis with 95% power. Accounting for 15% loss to follow-up, a total of 30 patients were required. Independent-samples t-tests and 95% CIs were used to compare migration of the study groups with published thresholds of migration. Secondary outcomes of patient-reported outcome measures (PROMs) were compared with paired t-tests, with significance defined as p < 0.05.

Results

Surgeries occurred between May 2022 and January 2023 (Figure 1) and a total of 29 patients completed the study follow-up. One patient refused their 24-month follow-up, and another patient underwent revision due to persistent stiffness and pain. This patient had their patella resurfaced and the tibial insert exchanged, but the femoral and tibial components were retained; therefore, they remained in the study.

Fig. 1.

CONSORT patient flow diagram for the study. RSA, radiostereometric analysis.

Tabulated values for subsidence and maximum total point motion (mm) for the tibial baseplates, cementless femoral components, and metal-backed patellar components are listed in Table I.

Table I.

Tabulated values for subsidence (sub.) and maximum total point motion (MTPM) for the tibial baseplates, cementless femoral components, and metal-backed patellar components. Unit of measurement = mm.

Variable	6 wks (precision)		6 mths		12 mths		24 mths		6 to 12 mths		12 to 24 mths
	Sub.	MTPM	Sub.	MTPM	Sub.	MTPM	Sub.	MTPM	Sub.	MTPM	Sub.	MTPM
Tibial baseplate
Mean (SD)	-0.01 (0.098)	0.764 (0.427)	-0.053 (0.102)	1.018 (0.742)	-0.039 (0.157)	0.991 (0.457)	0.007 (0.181)	1.21 (0.571)	-0.003 (0.125)	-0.096 (0.551)	0.053 (0.17)	0.095 (0.603)
N	30	30	26	26	30	30	29	29	26	26	29	29
Femoral component
Mean (SD)	-0.013 (0.06)	0.408 (0.258)	0.166 (0.227)	0.907 (0.649)	0.124 (0.333)	1.077 (0.713)	0.225 (0.312)	1.057 (0.798)	-0.029 (0.224)	0.175 (0.415)	0.11 (0.191)	0.057 (0.486)
N	16	16	12	12	13	13	14	14	12	12	14	14
Patella
Mean (SD)		0.798 (0.251)		0.948 (0.481)		0.958 (0.352)		1.2 (0.579)		0.108 (0.335)		0.44 (0.683)
N		7		7		5		9		4		5

For the tibial baseplate, the change in MTPM between six and 12 months was -0.096 mm (SD 0.551) and between 12 and 24 months was 0.095 mm (SD 0.603), both of which were well below the published acceptance threshold of continuous migration. Cumulative MTPM at 24 months was 1.210 mm (SD 0.571). Cumulative mean subsidence of the tibial baseplate was 0.007 mm (SD 0.181) at 24 months with small changes between six and 12 months (-0.003 mm (SD 0.125)) and 12 and 24 months (0.053 mm (SD 0.170)).

Figure 2 depicts translational and rotational stability of the baseplate at six months. Small changes in sagittal and coronal tilt were observed from 12 to 24 months, but both values were within the precision threshold of the RSA system. The curves are started at time ‘zero’ because it is known that migration is zero at the time of surgery. The six-week values are the result of comparing double examinations. The result on the graph at six weeks represents the mean measurement error of the RSA system.

Fig. 2.

Means and SDs (error bars) of tibial baseplate translation (subsidence, maximum total point motion (MTPM), total translation) and rotation (sagittal tilt, coronal tilt, total rotation) over the follow-up period. The precision thresholds (dotted lines) represent the detection limit of the radiostereometric analysis system, based on the SD of double examinations performed at six weeks.

For the cementless femoral components, cumulative MPTM reached 1.057 mm (SD 0.798) at 24 months with a mean increase of 0.175 mm (SD 0.415) between six and 12 months, as well as 0.057 mm (SD 0.486) between 12 and 24 months.

Figure 3 depicts the MTPM and total translation of the cementless femoral components increasing up to 12 months before stabilizing, whereas rotational stability appears to be reached by six months.

Fig. 3.

Means and SDs (error bars) of cementless femoral component translation (subsidence, maximum total point motion (MTPM), total translation) and rotation (flexion tilt, varus tilt, total rotation) over the follow-up period. The precision thresholds (dotted lines) represent the detection limit of the radiostereometric analysis system, based on the SD of double examinations performed at six weeks.

The metal-backed patellar components displayed a cumulative MTPM of 1.200 mm (SD 0.579) at 24 months. Increases between six to 12 and 12 to 24 months were found, but the number of data points were very low (Figure 4).

Fig. 4.

Means and SDs (error bars) of metal-backed patella component translation (maximum total point motion (MTPM), total translation) over the follow-up period. The precision thresholds (dotted lines) represent the detection limit of the radiostereometric analysis system, based on the SD of double examinations performed at six weeks.

Precision of the RSA method was defined as the SD of measurement from double examinations obtained at six weeks for all 30 patients, and is shown in Figures 2 to 4 as dashed lines. Quality of RSA bead distribution in the bone (condition number, CN) was good for the majority of patients: mean CN = 49.9 (SD 55.3). One patient had CN at or higher than the recommended threshold of 120, but their data appeared consistent over the study period and they were therefore not excluded from the analysis. Repeatability of bead fitting between successive examinations (rigid body error, RBE) was below the recommended threshold of 0.35 for all patients and examinations.

Patient health and function scores improved substantially between preoperatively to 24 months postoperatively (Table II). Mean patient satisfaction was 89.4% (SD 8.6%) at 24 months. Statistical analyses on PROMs were not performed due to the small patient cohort, but postoperative improvement was as expected.

Table II.

Patient-reported outcome measures (PROMs) and improvement from preoperative to 24 months. Absolute numbers indicate the number of patients who completed the respective PROMs.

PROM	Preoperative	6 wks	6 mths	12 mths	24 mths	Improvement
OKS
Mean (SD)	24.3 (8.9)	28 (6.9)	38.2 (8.1)	39.9 (7.9)	40.7 (7.1)	16.4 (9)
N	30	30	30	30	30	30
EQ-5D-5L Index
Mean (SD)	0.665 (0.153)	0.748 (0.081)	0.813 (0.114)	0.851 (0.121)	0.84 (0.152)	0.175 (0.151)
N	30	30	30	30	30	30
UCLA
Mean (SD)	5.2 (2.3)	N/A	5.9 (1.7)	6.5 (1.6)	6.2 (2.0)	1.0 (2.3)
N	30	N/A	30	30	30	30
Mean (SD)	51.3 (13)	58.3 (10.3)	71.2 (14.1)	74.6 (17.4)	76.7 (15.3)	25.4 (16.0)
N	30	30	30	30	30	30
FJS
Mean (SD)	N/A	17.8 (18.8)	53 (28.6)	56.1 (30.6)	62.1 (29.0)	44.3* (28.6)
N	N/A	30	30	30	30	30
Satisfaction VAS
Mean (SD)	29.9 (17.9)	N/A	83.8 (18.5)	84.1 (19.6)	89.4 (8.6)	59.5 (18.5)
N	29	N/A	30	30	30	29

Data include one patient who was revised for stiffness at 16 months.

^*FJS improvement from 6 weeks to 24 months.

EQ-5D-5L, EuroQol five-dimension five-level questionnaire; FJS, Forgotten Joint Score; KOOS JR, Knee Injury and Osteoarthritis Outcome Score for Joint Replacement; N/A, not applicable; OKS, Oxford Knee Score; UCLA, University of California, Los Angeles activity scale; VAS, visual analogue scale.

Four adverse events were recorded during the study, including two intraoperative events: medial collateral ligament release requiring bracing, and a small crack in the posterior lateral condyle requiring cemented fixation of the porous femoral component. Two adverse events occurred during postoperative follow-up, including reported stiffness with no further treatment, and a revision procedure for persistent stiffness and arthrofibrosis 16 months postop resulting in a liner exchange and secondary resurfacing of the patella. This revised patient retained their tibial and femoral components and therefore was kept in the study for their 24-month visit.

Discussion

We hypothesized that this novel tibial component would meet acceptance thresholds for early migration followed by stable continuous MTPM between 12 and 24 months. We found that this implant followed the typical migration pattern for uncemented TKA.⁸ This implant had high early migration at six months (mean MTPM of 1.018 mm) which stabilized by 12 months (mean MTPM of 0.991 mm at 12 months). This reflects secondary fixation with osseointegration of the uncemented components. We found the change in MTPM for the tibia baseplate was 0.095 mm between 12 and 24 months, which is well below the 0.2 mm value for ‘at risk’ and ‘unacceptable’ groups described by Ryd et al.²⁵ This migration pattern is consistent with successful osseointegration and has been associated with a low risk of late aseptic loosening.

Comparative RSA studies of other porous cementless tibial components report similar early settling behaviour. Winther et al¹ analyzed the Regenerex (Zimmer Biomet, USA) porous titanium construct in 61 patients with RSA analysis of the tibial component. This surface exhibited a similarly high early migration pattern followed by stabilization through two years. These authors reported a mean MTPM of 1.60 mm at 12 months, and a mean MTPM of 0.06 mm between 12 and 24 months.

Linde et al² also analyzed the Regenerex porous titanium surface, investigating 27 cementless tibial components and 26 standard cemented tibial components (Vanguard, Zimmer Biomet). They demonstrated higher early migration in cementless tibial components but similar migration between one and two years, indicating long-term fixation. These authors reported a mean MTPM of 1.63 mm at 12 months, and acknowledged that this 12-month mean migration was higher than the suggested acceptance threshold in the literature.⁸ However, they observed stabilization of the implant between 12 and 24 months.

Sporer et al³ analyzed the migration of the 3D-printed Triathlon Tritanium TKA implant (Stryker, USA) in a cohort study with two years of follow-up of 29 patients with the use of RSA. In this investigation, mean tibial MTPM at 12 months was 0.688 mm, and the mean difference in maximum total point motion between 12 and 24 months was 0.021 mm for the tibial implant. These authors also analyzed the migration of a metal-backed patella and reported a change in mean MTPM of 0.089 mm between 12 and 24 months. By contrast, we identified a mean MTPM of 0.44 mm between 12 and 24 months for the cementless patellar design employed in this study.

Van der Lelij et al⁴ reported the five-year RSA results of the Triathlon Tritanium cementless 3D-printed tibial baseplate, showing no progression in migration beyond two years and comparable fixation to cemented counterparts. These authors reported a mean MTPN of 0.66 mm at five years and a mean MTPM of 0.02 mm between two and five years for the cementless group. However, they did observe one implant continuously migrating in the cementless group.

Puijk et al²⁸ recently published the results of the five-year migration of the uncemented Attune and Low Contact Stress (LCS) (DePuy, USA) tibial components. They analyzed 48 patients from their initial two-year RSA study at five-year follow-up, and reported a mean MTPM of 1.13 mm for 24 Attune components and 1.24 mm for 24 LCS components. Between two and five years, no significant migration (0.02 mm MTPM for the Attune and 0.13 mm MTPM for the LCS) was detected in any direction, with similar outcomes observed for both components.

Puijk et al²⁹ have since reaffirmed the 2012 RSA thresholds and proposed new fixation-specific six-month migration thresholds for early identification of unsafe designs. Curiously, they also reported that the 12- to 24-month mean continuous migration data were found to be less reliable for this purpose. For uncemented TKAs, they reported that a six-month mean MTPM was acceptable below 1.10 mm and unacceptable above 1.55 mm. The tibial implant migration in the present study was acceptable at both six months (mean MTPM of 1.018 mm) and at 12 months (mean MTPM of 0.991 mm). However, it is important to note that the baseline reference in the present study was at six weeks, and may have missed some early migration from the early weeks.

Clinically, the PROM improvements observed in the present study are consistent with contemporary TKA outcomes and comparable with those reported in RSA cohorts of both cemented and cementless implants. While PROMs are not sensitive to early fixation differences, the absence of clinical concerns, combined with favourable RSA findings, strengthens confidence in this implant’s early performance.

The major strength of this paper is that it is the first clinical RSA study on this novel cementless tibial baseplate design, and none of the authors had conflicts of interest with regard to implant design. These RSA findings suggest a low risk of future aseptic loosening and support the fixation performance of this implant design. Further, this study also characterized femoral and patella component migration, something relatively rare in RSA studies.

The limitations of this study are the modest sample size and relatively short follow-up; however, RSA is specifically designed to detect early failure signals through controlled exposure of a small number of patients. Notwithstanding this, it is possible that the study sample size could have missed a small proportion of implants that do not ingrow and continue to migrate, as observed by van der Lelij et al⁴ in their five-year follow-up study of the Triathlon Tritanium baseplate. Long-term revision risk still requires concomitant registry follow-up. Moreover, the reference assessment at six weeks may have missed some early implant migration between implantation and first assessment. These results restrict our ability to directly compare with the early (six-month) migration thresholds described by Puijk et al.¹¹ Finally, we acknowledge that this study is under-powered to detect differences in patient-reported outcome measures, and recognize that the femoral component migration and the patellofemoral joint can affect pain and function.

In conclusion, it has become evident through past implant design problems that a phased introduction of new technology in orthopaedics is needed. RSA has been shown to be an effective method for assessing micromotion, and remains the recommended technique for predicting late aseptic loosening and evaluating new materials and implant designs. This multicentre RSA study confirms stabilization of this novel cementless tibial baseplate used in primary TKA. Continuous tibial implant migration between one and two years remained below accepted thresholds associated with increased risk of aseptic loosening. These findings support the early fixation performance of this implant and justify continued longer-term surveillance.

Take home message

- As cementless fixation becomes more prevalent in total knee arthroplasty, it is imperative that the stability of new implants is examined.

- This study showed that a novel porous cementless tibial baseplate was stable using radiostereometric analysis, indicating a low risk for future aseptic loosening.

Data Availability

The datasets generated and analyzed in the current study are not publicly available due to data protection regulations. Access to data is limited to the researchers who have obtained permission for data processing. Further inquiries can be made to the corresponding author.