The Importance of Different Modularity of the Polyethylene Insert for Tibial Component Migration and Adaptive Bone Remodeling After Uncemented Total Knee Arthroplasty Using Trabecular Metal Technology (TMT) Zimmer NexGen®, (Monoblock vs. Modular Design)

Introduction In general a primary TKA is a durable and well-functioning solution with a 10-year implant survival of more than 90 % (Robertsson et al. 2001). Both implants to be tested in the present study have a trabecular metal coating of the tibial component. However, a different manufacturing technique of the baseplate with one having a monoblock polyethylene insert and the other a modular polyethylene insert have been used. Failure secondary to polyethylene wear is ranked as a frequent cause for revision. Many issues have been identified as contributory to polyethylene wear; the quality of the polyethylene, manufacturing process, nature of the tibial-femoral or patellofemoral articulation, area of contact, sterilization process, and issues pertaining to modularity (backside wear) (Lombardi et al. 2004, Lombardi et al. 2008). Long-term clinical studies have documented the success of direct compression-molded monoblock tibial components (Weber et al. 2002). However, the benefits of modularity are widely recognized and include the ability to fine tune soft-tissue balance after implantation of metallic devices.

New techniques are developed and the patients´ demands for increased durability of the implants are increasing. Failure because of aseptic loosening of the tibia component remains one of the leading courses of revision surgery in TKA (Sierra et al. 2004, Danish Knee Arthroplasty Register, Annual Report 2010). It is believed that both the implants, planned to be tested in the present study, with the trabecular surface coating will facilitate bone ingrowth and secure a better fixation to the host bone, thus increasing the implant survival and improving the function of the prosthesis.

One of the implants NexGen® , monoblock(TMT) has been used for routine clinical use abroad, but also in Denmark for some years, and so far the clinical results have been good (Bobyn et al. 2004). There have been made two randomized controlled studies using RSA mesurement of the implant showing promising results with good stable fixation at follow up between twelve and twenty-four month (Dunbar et al. 2009, Henricson et al. 2008).

The other type of prosthesis is identical but has a modular polyethylene design of the tibia component. Modular designs are used frequently in most TKA designs and have also shown good long-term results. However, there are potential problems concerning back-side wear and aseptic loosening because of polyethylene foreign body reaction (Lombardi et al. 2004, Lombardi et al. 2008, Weber et al. 2002).

There have been made previous randomized RSA studies evaluating uncemented modular TKA designs with good results showing stable fixation at twelve and twenty-four month (Dunbar et al. 2008, Hansson et al. 2005, Hansson 2008).

The TMT coating is a micro-casting of the normal trabecular bone performed using the element Tantalum (Ta-73). From previous experience Tantalum is very biocompatible, very strong and with a great resistance to corrosion (Black 1994). Trabecular metal has a porosious structure close to that of normal trabecular bone, and the biomechanical characteristics such as e.g. compressive strength and elastic modulus are very close to that of normal trabecular bone (Krygier et al. 1999, Zardiackas et al. 2001).

These characteristics are the reason why trabecular metal within the past decade have been used for production of different orthopedic implants, and the early clinical results so far have been very promising (Bobyn et al. 2004, Nasser \& Poggie 2004, Nehme et al. 2004, Nelson et al. 2003).

Quantitative measurements of the implant migration and adaptive bone remodeling of the bone close to the implants are generally accepted as suitable methods for an objective evaluation of the effect of joint replacement surgery and have been proposed as important examinations that shall bee performed before new implants is distributed for wide-spread use (Bobyn et al. 2004, Petersen \& Schrøder 2004, Østgaard \& Nielsen 2004).

Very small movements between an implant and the bone can be measured by special RSA X-rays after implantation of small (0.8-1.0 mm) Tantalum markers attached to the implant and in the bone (Selvik 1989), while changes in bone mineral or bone mineral density (BMD) of the bone in close relation to the implant can be measured by DEXA (Glüer et al. 1990, Mazess \& Barden 1988).

Micromovements with migration of the tibial component are seen after both cemented and uncemented TKA and with different polyethylene modalities during the first postoperative year. Subsequently most implants stabilize, while a minority migrate continuously beyond the first postoperative year (Ryd et al. 1995). The implants that continues to migrate are those with a high risk of later aseptic loosening. Several studies have evaluated the influence of different parameters such as mode of fixation (cemented/uncemented), surface coating e.g. hydroxyapatite, metal alloys, and design of the prostheses on the migration of the tibial component in primary TKA (Ryd 1986, Toksvig-Larsen 2000, Dunbar et al. 2008, Hansson et al. 2005, Hansson 2008 ), but whether the use of a modular design vs. a monoblock design on a trabecular metal (Tantalum) surface coatning will result in significant differences have still not been evaluated.

In studies using different bone densitometric techniques for measurements of changes in bone mineral of the proximal tibia after primary TKA a significant decrease in BMD is seen in most studies (Hvid et al. 1988, Karbowski et al. 1999, Levitz et al. 1995, Petersen et al. 1995, Regnér et al. 1999). In some of the studies the decrease in BMD reaches 20% - 36% (Hvid et al. 1988, Levitz et al. 1995, Petersen et al. 1995, Regnér et al. 1999), however, in other studies BMD remains unchanged (Bohr \& Lund 1987, Li \& Nilsson 2000, Spittlehouse et al. 1999) or shows a minor increase (Petersen et al. 2005).

Materials and methods Study design A prospective randomized study where the patients are randomized too receive a monoblock tibial component with trabecular metal coating (TMT Zimmer NexGen®) or a modular tibial component (TMT Zimmer NexGen®). Both prostheses are available and approved by the FDA and CE. The only difference between the implants is that of the polyethylene insert (monoblock vs. modular).

Place The operations were originally planned to be performed at the Knee Unit, Department of Orthopedic Surgery, Hørsholm Hospital. However, February 2011 this unit will be fused with the Department of Planned Elective Surgery Z, Gentofte Hospital, where more than 400 primary TKA will be performed each year. It is planned that the TKA operations shall be performed at the Gentofte Hospital, but a final decision can not be made until the above mentioned fusion have been fully implemented in the beginning of 2011.

RSA and DEXA facilities examinations will be performed at the Department of Orthopedic Surgery and Department of Radiology, Rigshospitalet, Copenhagen, Denmark.

Patients The participants of the study are patients that have been scheduled for a primary TKA using the usual indications for TKA at the Department of Planned Elective Surgery Z, Gentofte Hospital.

Criteria of inclusion:

1. Severe osteoarthrosis of the knee with indication for a TKA. 4. Age between 40 and 70 years.

Criteria of exclusion:

1. Diseases affecting the bone metabolism (osteoporosis, Pagets disease, hyperparathyreoidism etc.).
2. Patients estimated not to be able to understand the "Information to patients" papers or do not want to participate in the study.

Operation The operations are performed at the Department of Planned Elective Surgery Z, Gentofte Hospital by two experienced TKA surgeons that will be included in the research group. During surgery a minimum of 4 Tantalum markers (0.8 mm) is inserted in the bone of the proximal tibia and 6 markers are inserted in the polyethylene insert Intravenous antibiotics (1.5g cefuroxim) is administered three times during the first day and night of surgery and low molecular weight heparin is administered for prevention of deep venous thrombosis and pulmonary embolism. Local analgesics (Naropin ®, Toradol ®) are injected in the knee joint and in the soft tissue around the knee during surgery. Physiotherapy begins on the day of surgery and the patients will be mobilised with full weigth-bearing using crutches.

Implants Trabecular Metal Technology (TMT) Zimmer NexGen®, Monoblock Trabecular Metal Technology (TMT) Zimmer NexGen®, Modular Both implants are posterior cruciate ligament retaining implants. Apart from the different insert used (modular/monoblock) the designs of the two implants are identical.

The patella will be left unresurfaced.

Research plan

1. Clinical examination performed preoperatively and with follow-up after ¼, 1 and 2 years with evaluation of the knee function using the Knee Society's Knee Scoring System (Insall et al. 1989).
2. Conventional X-rays (standing and weight bearing) are obtained preoperatively, postoperatively before the patients are discharged and after 12 months with the aim of evaluating knee alignment, and to obtain a starting point for a possible later change in placement of the TKA components and development of osteolytic lesions (Ewald 1989).
3. Classification of the degree of arthrosis from preoperative X-rays estimating the Ahlbäck score (Ahlbäck 1968).
4. DEXA-scanning is performed using a Norland XR-46 scanner. Scans are performed with the aim of measuring BMD of the proximal part of the tibia in close relation to the tibial component, in the distale femur and the distal tibia bilaterally. The first scans are performed preoperatively (only proximal and distal tibia) not more than two weeks before surgery and postoperatively within the two weeks after surgery and with follow-up after 3, 6, 12, and 24 months. The precision of the BMD measurements of the distal and proximal tibia and distal femur will bee estimated from double measurements of patients (n=12) included in the study.
5. Special X-rays suitable for RSA measurements of migration of the tibial component (Selvik 1989) is performed within the first week after the operation and after 3, 6, 12, and 24 months. The precision of the RSA measurements is estimated from double examinations in 12 patients of the study.
6. Patient-reported outcome measurements. It is planned also to ask the patients preoperatively and after ¼, 1 and 2 years to do a patient-reported outcome measurement using the Oxford knee score (Dawson et al. 1998).
7. Long-term follow-up. If possible we intend to do also 5- and 10-year follow-up including all above-mentioned parameters.

Calculations of sample size DEXA (changes in BMD 0-2 years). The expected decrease in BMD in the proximal tibia after insertion of an uncemented TKA is 20%. In the study we would like to be able to measure a significant difference (minimal relevant difference = MIREDIF) in changes in BMD between the two groups of 7.5%. From a previously published study (Petersen et al. 2005) where BMD was measured in the proximal tibia in patients with primary uncemented TKA, we estimated a standard deviation (SD) of the changes in BMD that could be used for calculating sample size in our study. In the study a relatively large difference in SD (5.59%-12.40%) in the different ROI was seen, and thus our calculations are based on the average value form the different ROI.

Using Type I error = 5% Type II error = 15% and MIREDIF 7.5% SD = 7,53% Calculation resulted in a sample size of n = 18 in each group. RSA. There exist several published RSA studies, and we have decided to use the SD values from a study by Toksvig-Larsen et al.(2000). In that study 4 different uncemented tibial components (n=62) were examined and the specification of the SD was very clear. The SD used for calculation of sample size at 1 year of follow-up was very high because the SD in one of the 4 groups was much higher than in the other groups.

However, the calculation is performed using the parameter maximum total point motion (MTPM) two years postoperatively, which has shown to be the best predictor for later aseptic loosening of the a tibial component after primary TKA (Ryd et al. 1995).

Using:

Type I error = 5% Type II error = 15 % MIREDIF 0,6 mm SD = 0,7 mm Calculation resulted in a sample size of n = 24 in each group. In the study we have planned to include and randomize (block-randomization with 4 in each block) a total of 60 patients. This will secure a high statistical power even if some of the participants should drop out during the study.

Ethical considerations We do not expect that the patients participating in the study will experience any special side effects or complications related directly to the use of the two orthopedic implants. The difference between the patients of the 2 groups is expected only to be measurable using very precise techniques such as DEXA and RSA. Radiation dose to the patients from both RSA and DEXA scanning is low compared with many conventional X-ray examinations. The total accumulated radiation dose to a patient participating in all examinations during the study is estimated to be approximately 0,1 mSv. Tantalum has been used for more than 50 years and is a very biocompatible material without any tendency to corrosion (Black 1994). Tantalum beads for RSA have been used especially in Sweden in several clinical studies for more than 20 years. More than 20.000 beads have been implanted in more than 2.000 patients and no side effects from Tantalum beads have been observed (Kärrholm 1989).

The research protocol will be submitted for approval by The Danish Data Protection Agency, and The Scientific Ethical Committee of Københavns og Frederiksberg Kommuner. All patients will receive both oral and written information before informed consent to participate is obtained.

Statistical analysis The migration and the changes in BMD over time will be analyzed using repeated Student t test for paired data and for parameters (MTPM) that is per definition not normally distributed a non-parametric test will be used. The difference between migration and changes in BMD between the 2 groups will be evaluated by Student t test for upaired data.

Time frame Inclusion of the patients (n=60) is expected to take 1 year. We have planned to start the study during the spring 2011. Two years after inclusion of the last patient we would be able to analyze the data and start the process with the aim of publishing the data.

Publications We expect that the results of this protocol (when 2 years of follow-up have been completed) together with a clinical and radiological evaluation of the patients that have been operated on with insertion of an uncemented TKA using Trabecular Metal Technology (TMT) Zimmer NexGen®, Modular tibial component at the Department og Orthopaedic Surgery, Rigshospitalet will result in a phd-thesis and at least 2-3 articles in international orthopaedic journals.

The results of this protocol will be published in all circumstances.