INTRODUCTION

As might be expected, it is clear that the most important distinction to make when planning a revision procedure is the status of bone; that is, is the failure with an intact or compromised osseous structure? If osseous integrity is compromised, what is the extent and nature of this feature?

There are several expressions of osseous deficiency. These include (1) deficient length, resulting from absent distal humerus or proximal ulna; (2) compromised bone quality secondary to osteolysis characterized by ballooning of an intact cortex; and (3) periprosthetic fracture. We have classified the periprosthetic fracture as proximal (I) (metaphyseal), with stem involvement (II), or past the stem (III) (Fig. 66-1). From a revision standpoint, the type II fracture involving the stem or a type III fracture proximal or distal to the tip of the stem are of greatest significance. Because the rationale and technique of revision have been covered in detail in Chapter 65, herein we describe three techniques that are used for reimplanation revision following failed total elbow arthroplasty in the face of osseous deficiency. These techniques include impaction grafting, allograft strut grafting, and allograft prosthetic composite. Obviously, impaction grafting and strut grafting can be used in conjunction with these other techniques as well. Implant selection is an important issue. Some have employed unlinked devices in some cases of revision with osseous compromise.⁴ Others have employed custom designs for such problems.⁵ We believe that these problems require a linked implant with multiple sizing options.

FIGURE 66-1 Mayo Classification of periprosthetic fractures about the elbow. The types II and III are typically managed with strut grafting

IMPACTION GRAFTING

SURGICAL TECHNIQUE

The exposure is routine. As in all revisions, it is imperative that the ulnar and radial nerve be identified and protected.

Bone preparation follows the same basic steps:

1. All membranous tissue associated with the osteolytic expansion is removed.

Note: The radial nerve is vulnerable in those with osteolysis extending to the spiral groove and may be injured when removing the membrane.¹⁴

2. The normal medullary canal is identified.

3. A revision system (Zimmer, Warsaw, IN) allows execution of the impaction technique. The “tube with the tube” system is employed (Fig. 66-2).

4. The inner tube is inserted through the outer tube. The smaller tube should extend at least two diameters into normal host bone. The outer tube should extend the length of the osteolytic portion.

5. Allograft or autograft that has been through a bone mill or morcellized bone with fragments measuring 2 to 4 mm are tamped or impacted tightly around the outer tube. Serial tamps have been designed to allow this to be done effectively and efficiently.

6. The cement is mixed in the cartridge and the cartridge is screwed onto the adapter of the inner tube.

7. Cement is delivered into the host bone for the distance (D), which is equal to at least two diameters of the canal width.

8. The inner tube is withdrawn for a distance (D) so as to align with the outer tube and both the inner and outer tubes are then withdrawn while cement is being delivered to the canal formed by the vacated and larger tube.

9. A bone graft is placed behind the flange. This bone graft extends past the expanded osteolytic region if there is a fracture or distortion at the junction between the osteolytic and the host bone interface.

10. The implant is then inserted through the cement channel and into the host bone.

FIGURE 66-2 The technique of impacting grafting. A, A smaller diameter tube is inserted through a larger diameter tube and extends past the larger diameter tube for a distance (D) by approximately 2 cortical diameters into the normal host bone. The outer diameter tube is inserted to the depth of the osteolytic segment. B, Small chips of cancellous bone are tightly impacted around the outer diameter tube. C, The cement cartridge is attached to the inner diameter tube, which is withdrawn to the level of the larger outer tube while cement is being injected. D, At this point, both tubes are withdrawn together while cement continues to be delivered through the inner tube. E, The implant is inserted through the cement and impacted graft composite and into normal host bone.

Pearl.

Care is taken to tightly impact the cancellous bone to avoid collapse of the void left by removing the outer tube. Furthermore, the cement is injected while in a relatively viscus state to allow better intrusion into the cancellous bed and to allow more controlled insertion of the stem.

RESULTS

The value of impaction grafting was described originally for acetabular deficiencies by Schreurs et al.¹³ A 20-year experience with impaction grafting and cemented acetabular component revealed excellent long-term results as well as biopsy evidence of viability and incorporation of the impacted graft. This provides a rationale and a basis for the application at the elbow. We have recently reviewed our experience with 12 patients undergoing revision elbow arthroplasty using an impaction grafting technique between the years 1993 and 1997.⁹ The average surveillance period was 6 years, with a minimum 2-year follow-up. Recognizing that this represents a salvage-type revision, at the time of last follow-up, eight were intact after the index procedure (Fig. 66-3). In addition, two other patients underwent revision because of loosening and a third required revision due to a subsequent failure of the unrevised ulnar component. One patient developed an infection. Of the eight remaining patients, marked radiographic improvement was observed. At the final follow-up after the additional revision in three patients, five were excellent and four were good with three fair results. This suggests that impaction grafting is a viable option as a salvage for the difficult revision with poor bone quality due to osteolysis.

FIGURE 66-3 A, A patient with marked osteolysis of the distal humerus with gross loosening. B, Seven years following impaction, grafting shows a well-fixed implant with no evidence of loosening.

BONE STRUTS

We have adopted the excellent experience of strut reconstruction about the hip, as popularized by Gross,⁶ for application at the elbow. The results have been gratifying.