UW Shoulder and Elbow Academy: Bone ingrowth humeral components - two reasons why we avoid them. What if Propionibacterium are on them?

In some circles there is an interest in humeral components that encourage the ingrowth of bone by having a surface that is porous coated, grit blasted, or trabecular metal. We avoid such humeral components for two reasons.

First, a smooth humeral stem can be securely fixed in the humeral canal without bone ingrowth using impaction grafting as shown in this link and here and here. The advantage of impaction grafting is that a special surfaced prosthesis is not required and the need to broach the endosteum for a tight fit is eliminated.

Second, all humeral components are at potential risk for revision, whether because of malposition, infection, or the need for revision to a different prosthesis. Recently, the revision burden was reported to be 4.7% for hemiarthroplasty and 15.0% for total shoulder arthroplasty. Another recent article reported that revision cases comprised 8.8% and 8.2% of all shoulder arthroplasties in 2009 and 2011. Yet another recent study found an 11% revision rate for total shoulders. These numbers suggest that at least 1 in 12 shoulder arthroplasties - over 6000/yr - are requiring revision!

Revision of an ingrowth humeral stem usually requires not only a split in the humeral shaft (see this link) but also circumferential osteotomization of the bone around the prosthesis with further fragmentation of the cortex and compromise of the fixation of the revision prosthesis.

Here is a recent example. After a prior surgery to remove a loose glenoid, intraoperative cultures were overwhelmingly positive for Propionibacterium in spite of the lack of serologic or intraoperative findings suggestive of infection. The ingrowth stem was not removed even though it had been placed in retroversion. The postoperative films are shown below.

However, the symptoms of infection - including pain in the arm - did not respond to a postoperative course of IV antibiotics and oral suppression. A subsequent revision was carried out to remove the humeral implant out of concern for its being a host to Propionibacterium in a biofilm. In addition to the issue of bone ingrowth into the proximal porous coated stem, close examination of the films pointed to 'spot welds' between the bone and the prosthesis (marked with lines on the x-ray below left) corresponding to the junction of the porous coated and the smooth section of the stem (below right).

As anticipated, a humeral split and multiple circumferential cuts with a curved osteotome were required before the prosthesis could be removed with a needle nosed vice-grip and mallet.

After a thorough washout, the humerus was reconstructed with a smooth stemmed component, Vancomycin soaked allograft, and bodice reconstruction of the humeral shaft.