**6. Patellar bone loss**

Compromised patellar bone stock poses signicant technical problems in primary and revision knee arthroplasty. In revision knee surgery bone deciency is normally secondary to loosening of the patellar button and osteolysis that affects severely the patella. In primary cases it is rare that the patella has been so eroded that resurfacing is not possible.

If there is functional loss of the medial or lateral collateral ligaments, soft tissue instability, inability to balance the flexion and extension spaces, or a severe valgus deformity, then a

**Treatment Options** *Clatworthy* 

<5 mm depth, <50% unicondylar/plateau involvement

5-10 mm depth, >50% unilateral condylar/plateau involvement

> 10-20 mm depth, unicondylar/plateau involvement

<20mm depth, bicondylar/plateau involvement

<20mm depth, bicondylar/plateau involvement

3 >20 mm depth

Compromised patellar bone stock poses signicant technical problems in primary and revision knee arthroplasty. In revision knee surgery bone deciency is normally secondary to loosening of the patellar button and osteolysis that affects severely the patella. In primary cases it is rare that the patella has been so eroded that resurfacing is

Prosthetic component translation, PMMA fill, morcelized allograft, autograft

Morcelized allograft or metal augments

Metal augments, metaphyseal sleeves, structural allografts

Metal augments (full block in tibia), metaphyseal sleeves, structural allografts

TM cones, structural allografts, custommade prostheses

Structural allografts, megaprostheses, custom-made prostheses, TM cones

*classification Rand classification* 

constrained condylar prosthesis is necessary.

*classification* 

Contained, undamaged metaphysis

Contained undamaged metaphysis

Contained undamaged metaphysis

Contained undamaged metaphysis

Uncontained damaged metaphysis

Uncontained damaged metaphysis

**6. Patellar bone loss** 

not possible.

**Defect Type** 

*Engh* 

1

2a

2a

2b

2b

Table 4. Surgical algorithm according to bone defect size.

Fig. 11. (a,b) Lateral radiographs of total knee prosthesis revision, with patella extremely dug out, long and thin on preoperative. (c) Note lateral patella subluxation, and reestablishment of correct balancing after TMT patella prosthesis implantation.

This happens in severe patello-femoral arthritis or inammatory arthropathy, when the patella may be thin and track laterally before and during arthroplasty. Treatment depends on the quality of the remaining bone stock and options include non-resurfacing, retention of the remaining thin patellar shell or total patellectomy (Pagnano et al., 1998). Nevertheless these solutions have been associated with lower functional results compared with resurfaced patella. A patellar bone grafting procedure has been described to provide patellar bone for possible future revision (Hassen, 2001). The "gull wing" patellar osteotomy (Kelly et al., 2002) has also been proposed in case of low demand patients, whereas in some cases it is possible to rebuild a damaged patella with K-wires in a reinforcing conguration to support the pegs of the patellar implant using the so called "rebar" technique (Tigani et al, 2009). Trabecular metal patella represents a viable therapeutic option for severe damaged patella; we experienced use with this technique (Tigani et al., 2009) in revision cases with more than 50% amount of residual bone, obtaining reliable bony xation despite the quality of residual bone (Fig. 11).

We therefore exclude TM patella in cases of previous patellectomy, where soft tissue have to be used for xation of the TM implant, because of reported migration and loosening of the implant in these difficult cases.

Management of Bone Loss in Primary and Revision Knee Replacement Surgery 405

The ''osteogenic'' potential of the graft corresponds to capacity of cells living within the donor graft to survive during transplantation, then proliferate and differentiate to osteblasts and eventually to osteocytes. ''Osteoinduction'' on the other hand is the stimulation and activation of host mesenchymal stem cells from the surrounding tissue, which differentiate into bone-forming osteoblasts. This process is mediated by a cascade of signals and the activations of several extra and intracellular receptors the most important of which belong to the TGF-beta family (Cypher & Grossman, 1996). "Osteoconduction" describes the facilitation and orientation of blood-vessel and the creation of the new Haversian systems into the bone scaffold. Finally these three properties together allow ''osteointegration''

The gold standard for regeneration of new bone is autologous bone graft, which contains a scaffold, osteoblasts, and the necessary signalling proteins and molecules. However, autograft is of limited availability and may be insufficient due to poor quality (eg, osteoporosis). Furthermore, it may fail in clinical practice as most of the cellular (osteogenic)

Thus alternatives to autograft bone have emerged. Perhaps the most common bone substitute is cancellous allograft, which is osteoconductive only, and rely on a viable vascularized bone bed for incorporation. Moreover bone graft provided by musculoskeletal tissue bank could provoke immune response and transmission of viral disease; the processing of allograft tissue lowers this risk but can signicantly weaken the biologic and

Bone substitutes could be used to replenish lost bone stock during total knee arthroplasty. A bone-graft substitute to be useful should be: osteoconductive, osteoinductive, biocompatible, bio-resorbable, structurally similar to bone, easy to use, and cost-effective

Recombinant growth factors such as bone morphogenetic proteins (BMPs) have osteoinductive capacity (Greenwald et al., 2001). Nevertheless because they are powerful in small amounts, and they are expensive, their indication in knee prosthetic surgery is still limited. Contrary platelet rich plasma (PRP) is largely available as it is prepared from centrifugation of autologous blood; it is an osteopromotive adjunct with the ability to enhance natural bone formation by stimulatory signals. Both BMPs and PRP need a scaffold

Demineralized bone matrix (DBM) corresponds to portion of bone without the mineral phases, extracted by strong acids (Peterson et al., 2004). The demineralization process leaves behind the growth factors, the noncollagenous proteins and collagen, and therefore DBM is osteoconductive and osteoinductive. However, few prospective, randomized clinical studies

Nowadays many synthetic substitutes are available, used either alone or in combination with other biologic adjuncts. Synthetic bone graft materials available include calcium sulphates, special glass ceramics (bioactive glasses) and calcium phosphates. Calcium phosphate-based implants have the most similar composition to human bone, in particular those made of hydroxyapatite (HA) (Paderni et al., 2009). HA-based substitutes provide an osteoconductive scaffold to which mesenchymal stem cells and osteoinductive growth factors can migrate and differentiate into functioning osteoblasts (Fujishiro et al., 2005). These materials can be sterilized and are moldable, but generally do not have sufficient

between the host bone and the grafting material surfaces (Giannoudis et al., 2005).

mechanical properties initially present in the bone tissue (Giannoudis et al., 2005).

elements do not survive transplantion (Sandhu et al., 1999).

(Giannoudis et al., 2005).

to support their bone regeneration properties.

delineate the efficacy of DBM, and the material can be costly.

mechanical properties to support full immediate weight bearing.
