Most tumours involving the proximal humerus should be treated by a musculoskeletal tumour specialist. Most of these specialists will have excellent broad orthopaedic skills and will be experienced arthroplasty surgeons. However, many will also be less familiar with operations around the shoulder and perhaps less experienced with the use of proximal humeral locking plates. Therefore, the management of fractures or incipient fractures in the proximal humerus in relation to a tumour deposit brings an opportunity to combine the knowledge and experience of shoulder trauma surgeons, who understand the complexities of applying a locking plate to a fractured shoulder, and tumour surgeons who recognise the variety of surgical options for periarticular pathology and have a specialist knowledge in the management of bone defects.
This paper shares the wisdom of a group of tumour surgeons who have combined these skills to preserve the proximal humerus, to stabilise the fracture, and to augment a large bone defect.
Probably the most common surgical solution for a proximal humeral tumour is a hemiarthroplasty or endoprosthesis. If the rotator cuff and tuberosities can be preserved then the expectation will be for good long-term function. However, if the cuff or tuberosities need to be sacrificed the outcome in terms of shoulder function, after a hemiarthroplasty or endoprosthesis, is likely to be much poorer.1 Excision arthoplasty of the shoulder severely compromises function. The use of reverse geometry prostheses may have a role in these circumstances,2 but that role has not yet been clearly identified and there remains concern about loosening, infection and survivorship with these implants. Osteoarticular allograft reconstruction has been attempted but has a high complication rate.3 Therefore the best anticipated outcome would be expected from a procedure which preserves the native articular surface, the tuberosities and the rotator cuff. Using an intramedullary device would necessarily violate the rotator cuff and could potentially seed the tumour within the humerus or precipitate new metastases. Blade plate insertion is technically demanding and non-locking plates struggle to provide adequate support in osteoporotic bone. The final solution would seem to be a proximal humeral locking plate which is designed to provide stable structured support against valgus collapse of the head. However even a locking plate cannot provide support in an empty shell. The use of injectable cement to fill defects after evacuation of tumours is a tried and tested formula for orthopaedic tumour surgeons. By combining a locking plate and cement the authors have managed to achieve their aims of excising the tumour, filling the defect, stabilising the fracture, and restoring function quickly.
But is this a technique which can be easily transferred to the management of osteoporotic proximal humeral fractures? Many of the same management issues apply such as poor quality or absent bone in the head, the desire to preserve the articular surface and to restore early and useful function. Experience of trying to reduce and hold a proximal humeral fracture in osteoporotic bone shows that the bone is often fragmented and anatomical reduction is seldom possible. The application of a locking plate is usually achievable and the tightening of the screw heads in the plate is a seductive sensation giving false reassurance as to the solidity of the reconstruction. Early motion is achievable and adequate function is often restored. However late collapse of the head with subsequent penetration of the screw tips into the glenohumeral joint is a well recognised complication.4
Support of the subchondral bone by inserting a substance which can be ‘poured’ into all the defects and which then solidifies to bolster the locking plate and which also prevents late collapse of the head would seem an ideal solution. The use of cement in these circumstances, while initially appealing, does raise some concerns. In order to ‘pour’ cement into all the necessary defects it would also tend to extravasate through fracture lines and potentially into the glenohumeral joint or even near important neurovascular structures. Cement in and around fracture lines would impede and could prevent fracture healing. If the fracture is so supported that it cannot ‘settle’ or unite other complications will ensue such as periprosthetic fracture and hardware failure. Large tracts of cement in the head will also impede revision surgery both if the position of the plate has to be altered or if conversion to an arthroplasty is required.
A better alternative to cement would be a material with the same handling characteristics but which does not impede, but rather aids fracture healing, and which would not unduly interfere with revision surgery. Some newer bone substitutes are coming close to the mark but they remain expensive and their use in the management of proximal humeral fractures has not yet been established. A recent study examining the evidence for the use of bone substitutes to augment surgical stabilisation of fractures of the femoral neck was disappointing.5 There are issues concerning both the biological and biomechanical properties of these substitutes.
Therefore, this paper has much to be commended. It recommends, to other orthopaedic tumour surgeons, a useful alternative for the treatment of fractures through tumours in the proximal humerus. It also provides a surgical option for an experienced shoulder surgeon to treat such a fracture when the tumour has been established as a metastasis or multiple myeloma and which does not need to be treated specifically at a tumour centre. But perhaps of broadest value is the concept that osteoporotic fractures of the proximal humerus can be treated by a combination of a locking plate with a solid bone substitute. Locking plates have revolutionised the management of proximal humeral fractures but there are still many associated complications and perhaps this combined approach, gleaned from the experience of trauma and tumour surgeons, will provide the next major milestone for the surgical treatment of these difficult fractures.
1. Cannon CP, Paraliticci GU, Lin PP, Lewis VO, Yasko AW. Functional outcome following endoprosthetic reconstruction of the proximal humerus. J Shoulder Elbow Surg 2009;18:705-10.
2. De Wilde LF, Plasschaert FS, Audenaert EA, Verdonk RC. Functional recovery after a reverse prosthesis for reconstruction of the proximal humerus in tumor surgery. Clin Orthop 2005;430:156-62.
3. Potter BK, Adams SC, Pitcher JD Jr, Malinin TI, Temple HT. Proximal humerus reconstructions for tumors. Clin Orthop 2009;467:1035-41.
4. Owsley KC, Gorczyca JT. Fracture displacement and screw cutout after open reduction and locked plate fixation of proximal humeral fractures [corrected]. J Bone Joint Surg [Am] 2008;90-A:233-40.
5. Lindner T, Kanakaris NK, Marx B, et al. Fractures of the hip and osteoporosis: the role of bone substitutes. J Bone Joint Surg [Br] 2009;91-B:294-303.
Brownlow H, Consultant Orthopaedic Surgeon
Royal Berkshire NHS Foundation Trust, Reading, United Kingdom