Augmenting Osteointegration of Proximal Femoral Endoprosthesis Using Porous Titanium Bridging Endosteal Collars: preliminary medium-term results

Aseptic loosening remains a major cause for failure of proximal femoral replacements (PFR). To reduce the incidence of aseptic loosening a PFR was developed with a selective laser sintered, porous titanium-bridging collar and endosteal component. This study reports the short-term outcomes of this implant in a consecutive series of patients.


Results
Twenty-three patients were available for follow-up with one death in the early postoperative period. One patient died 18 months post-surgery due to causes unrelated to surgery, two cases of super cial infection were managed with antibiotics and two cases of deep infection managed with debridement, antibiotics, and implant retention (DAIR). At the last follow up the mean Harris Hip Score was 78 (68-86) and the mean Musculoskeletal Tumour Society Score was 20 (16)(17)(18)(19)(20)(21)(22)(23)(24). The radiographic assessment did not reveal any evidence of loosening with the integration of the endosteal collar observed in all cases.

Conclusions
Our medium-term results demonstrate good functional outcomes with no evidence of radiographic loosening and osteointegration of the endosteal collar in a consecutive series of patients. Further studies are required to assess the mid to longterm outcomes seen with this implant.

Level of Evidence:
Level IV -Retrospective Case Series Background: Proximal femur replacement (PFR) is a viable option for various aetiologies including oncology surgery, revision arthroplasty, and non-unions following fracture xation [1][2][3], and for this reason, the use of this implant is increasing. Apart from the increased cost associated with the implants used, PFR surgery is also associated with an increased complication rate compared to conventional total hip arthroplasty [3,4]. With an implant survival rate of only 73% at 5 years, failure of PFR is demanding on the patient, surgeon, and healthcare payer often leading to more extensive surgery by way of total femur replacement (TFR) or amputation in some cases [3][4][5]. Due to the increasing utilization of PFR in clinical practice, and devastating consequences of failure, attention has turned to improve implant survival to achieve better patient outcomes and decreased strain on the healthcare system. Aseptic loosening remains a major cause for failure of PFR with loosening rates of up to 29% at 4-10 years [4,6] with aseptic loosening typically being associated with poor functional outcomes and increased incidence of revision surgery [7]. With the increase in the use of PFR worldwide, in part due to expanding indications for this type of implant, efforts have been aimed at developing implants with newer materials involving advanced technology which would increase prosthesis longevity [8][9][10].
Aseptic loosening has been shown, to begin with, osteolysis at the bone-prosthesis junction followed by extension along the stem, nally culminating in prosthesis loosening [8].
Histologic studies by Blunn and Wait have shown that aseptic loosening could be reduced by preserving load transfer at the prosthesis shoulder through osteointegration leading to reduced stresses around the cement mantle [11]. To reduce aseptic loosening and improve prosthesis longevity, implant designs aimed at achieving better osteointegration at the bone-prosthesis junction where they would form a 'purse string' that seals the bone interface thereby preventing migration of wear particles [8].
Finite element analysis studies have shown that better osteointegration at the bone-prosthesis junction would prevent aseptic loosening by increasing load transfer at the shoulder of the prosthesis reducing stresses at the stem /bone interface [12]. By enhancing osteointegration and thereby increasing prosthesis longevity, better functional outcomes could also be achieved with these major surgeries.
The Pantheon PFR (Adler Ortho S.p.A, Milan, Italy) was designed to improve the integration of the bone with the prosthesis. The modular system is made up of a bridging collar which is designed to sit on top of the femoral resection with an endosteal extension that is designed to achieve a press-t within the diaphyseal bone. The novel bridging collar is a substantially porous titanium alloy (Ti6Al4V) with a pore size of 1000 microns manufactured by 3D printing using selective laser sintering of alloy powder ( Fig. 1).
Due to the theoretical bene ts of the porous titanium bridging collars in improving the integration of the prosthesis at the boneprosthesis junction our unit has recently started using this PFE with the aim of this study to report the clinical and radiographic outcomes of a consecutive series of patients managed with this implant.

Materials And Methods:
Between October 2017 and August 2019, 24 patients with a mean age of 69.1 years (50 -84) were managed with the Pantheon PFR with an endosteal titanium porous bridging collar. The etiology of the cases was as follows: ten patients had resections due to oncological diagnosis, ten patients had revision joint replacement surgery following aseptic loosening, three patients had surgery for non-union of the proximal femur and one patient was a second stage revision for a previously infected hip replacement. Of the patients who had an oncological diagnosis, one patient received neoadjuvant and adjuvant chemotherapy before and after surgery. No patient received radiotherapy. The median pre-operative ASA grade was 2 (range 1 -3). This study was performed as an audit project and all patients had consented for their radiographs and clinical details to be used in this study.

Surgical technique:
All procedures were undertaken using the posterior approach. After adequate resection of the proximal femur and reaming of the canal, endosteal reaming was performed to allow for the bridging collar to sit tightly in the femur. Facing reamers were used to facilitate the ush tting of the prosthesis on the cut femoral surface. A trial reduction is performed, con rming the optimal implants required. For cemented implantation, the canal is washed and brushed and a cement restrictor is applied at the desired level. The femoral canal was then retrograde washed and dried with ribbon gauze. Cement was inserted using a cement gun with a third-generation technique. Attention to the cementing technique was of utmost importance to achieve good osteointegration in the bridging collar. To avoid cement being introduced into the porous section of the collar, cement was only introduced into the canal to within approximately 3 cm of the cut surface to allow the cement to rise under pressurization to the bottom of the collar. The collar has a non-porous end to divert any cement away from the porous surface. Twelve patients required an acetabular component. Five also required a dual mobility bearing. The decision to use a dual mobility bearing was made by the operating surgeon when they felt that there was a need for more stability due to the resection and associated soft tissue insu ciency. The remaining six patients who had good acetabular cartilage underwent bipolar-bearing replacement.
All patients had a standard physiotherapy protocol which involved either partial or full weight bearing, at the discretion of the surgeon, for 6 weeks following surgery followed by full weight-bearing thereafter for all patients. All patients received thromboprophylaxis for 5 weeks following surgery in the form of subcutaneous Dalteparin injection (5000 units s/c once a day) in line with NICE guidance. All patients were followed up at six weeks, twelve weeks, six months, and then 6 monthly intervals.
Antero-posterior and lateral radiographs of the entire femur were taken during follow-up visits. The bone-prosthesis junction involving the collar was assessed in all radiographs in both views to identify the presence of any radiolucent lines. The evaluation was done independently by three individuals: two orthopaedic fellows and a consultant. Any gap of more than 1mm observed was considered radiolucent. Any disagreement was settled by mutual agreement. Outcomes were assessed using the Musculoskeletal Tumor Society (MSTS) and Harris Hip Score (HHS) at the last follow-up.
Continuous data were analyzed with simple descriptive statistics. All patients consented to the study.

Results:
At the time of our analysis, 23 patients were available for follow-up (Table 1) (Fig. 2, Fig. 3 & Fig. 4). The endosteal collar showed solid osteointegration with the parent bone. No evidence of loosening was seen in any of the radiographs nor did any of the patients have thigh pain on weight-bearing or deep palpation. The main nding of our study was that Pantheon PFR prosthesis showed good clinical and functional outcomes in the short term with no case showing evidence of loosening and no cases requiring revision.
Over the years PFR has emerged as a useful option in both young and elderly patients for various aetiologies. Young patients, mostly after oncological resection, have high demand and an increased life expectancy and hence implants with increased longevity are necessary. Likewise in elderly patients, following PFR a further revision surgery may not be feasible due to increased morbidity and mortality associated with the procedure [4]. Hence, achieving osseointegration at the bone-prosthesis junction and thereby preventing the cascade of loosening is vital. The porous endosteal titanium bridging collar was designed at aiming to achieve ingrowth, prevent loosening, and thereby ensuring better outcomes. Our early results do not show evidence of loosening and this is re-assuring. Longer follow-ups of these patients are essential to monitoring for osteointegration.  [14]. Our results also indicate that modular PFE is associated with favorable results both with implant functional and clinical outcomes.
Numerous designs and advances in technology have been made to achieve better osseointegration of prosthesis to the bone.
Good integration at the bone-prosthesis junction reduces aseptic loosening by preserving load transfer at the shoulder of the implant [16] as shown by Chao and Sim. To facilitate good integration, the material used and the coating becomes important.
Porous coating implants have been shown to demonstrate accelerated bone ingrowth in animal models [17] and the same concept has been implemented in prosthesis designs for humans [18] however, studies on retrieved implants with these surface coatings often show limited or no osteointegration [19]. In primary distal femoral replacement, a hydroxyapatite-coated collar can lead to signi cantly reduced loosening of those implants that were identi ed to have osteointegration on radiographs [20] however only 60% of the implants were osseointegrated. To combat this low incidence of integration, laser sintered porous collars have been designed and in animal studies, osteointegration has been shown throughout the porous structure [9].
Titanium alloys have been shown to have excellent mechanical properties including high speci c strength and biocompatibility to mimic bone. The prosthesis we used was based on these principles and showed good promising early results with no evidence of radiolucent lines suggestive of loosening.
PFR surgeries are uncommon and only done in a few centers. Hence, there is a relative paucity of studies reporting outcomes for PFR. While HA-coated collars have been studied and reported, there are few studies reporting outcomes of porous titanium bridging collars in PFR. Our study analyzing the functional and radiological outcomes of this novel prosthesis is the central strength of the study. Our promising results would encourage surgeons to use this prosthesis in PFR surgery.

Limitations of our study
Our study is not without limitations. Short follow-up, retrospective nature of the study, and single-center results are important limitations of our study. Retrospective studies always have the inherent risk of variability in data collection but however, we carefully scrutinized the patient data and collected all details. These are results only from a single-center, however, surgeries involving PFR happen only in few tertiary centers in the hands of experienced surgeons thus, even if a multi-centric study analyzing the cases from all other centers was performed the number of cases per annum would still be expected to below. We intend to follow-up on these cases and analyze the long-term results of these patients. The rarity of PFR surgeries performed makes it di cult to establish cohorts and compare results among them due to the low numbers. We also intend to perform a multi-center study involving centers that have used this prosthesis and increased numbers would help us overcome this limitation and would enable outcome analysis cohorts based on etiology.
Our short-term results demonstrate good functional outcomes with the integration of the endosteal collar and no evidence of radiographic loosening in a consecutive series of patients. Further studies are required to assess the long-term outcomes seen with this implant.

Conclusion:
Aseptic loosening remains an important factor, detrimental to good outcomes, following PFR. Our short-term study shows the use of titanium porous endosteal bridging collars in PFR to be promising resulting in reliable osteointegration at the boneprosthesis junction and also allowing good clinical and functional outcomes. Further studies are required to assess the mid to long-term outcome seen with this implant. This Ethics committee of our hospital (OUH Trust) approved this study as an audit project the reference number of approval was 6647 All patients consented for their clinical details and radiographs to be used for this study Consent for publication: All patients gave their consent for their details to be used anonymously and gave their consent for publication Data Availability statement: No linked research data is available with this submission. For readers who want to access the data, they can write to the authors and then the data will be made available on request.
Competing Interests: The authors declare the following competing interests