UW Shoulder and Elbow Academy: Propionibacterium in shoulder surgery

Propionibacterium acnes Infections in Shoulder Surgery

The abstract to this paper reads, "Perioperative shoulder infections involving Propionibacterium acnes can be difficult to identify in a patient who presents with little more than pain and stiffness in the postoperative period. Although indolent in its growth and presentation, infection of the shoulder with P acnes can have devastating effects, including failure of the surgical intervention."

The authors point out, as we've previously emphasized, that Propionibacterium are commonly found in failed shoulder surgeries, including arthroplasty, cuff repair and fracture fixation. These failures may present as pain, stiffness or mechanical loosening - only rarely with redness, swelling, tenderness, wound drainage, fever, elevated white blood cell counts, elevated C reactive protein, or elevated sedimentation rate. Our approach to the finding Propionibacterium in surgical wounds is explained here. In contrast to the statement of these authors, the data indicate that cultures for Propionibacterium should be on three different media and held for 17 days.

The authors suggest that Propionibacterium contamination from the skin of the patient can be prevented by "proper preparation of the surgical field" but as they point out this is difficult. Recent evidence has shown that the organisms lie in not on the skin, so that they are not susceptible to normal skin preparation. A great article on the ecology of skin bacteria is shown here. These two figures come from that article. Note that Propionibacterium are common in the oily areas (face, chest, back) and not the moist areas (axilla and groin). From the second figure one can see that Propionibacterium were NOT present in the axilla, contrary to popular belief.

They point out that presurgical preparation of the skin is ineffective in eliminating Propionibacterium from the hair follicles and sebaceous glands. The authors advocate changing to a new clean knife blade after the skin incision - a practice we have adopted as well. The bottom line is that there is no effective method of preventing Propionibacterium contamination of the surgical field.

We suggest that the risk of infection may also be reduced by copious irrigation with antibiotic-containing fluid and avoiding contact between the implants being inserted and the skin edge at the time of insertion.

The authors refer to our prior study demonstrating that patients having revision shoulder arthroplasty had a high risk of positive Propionibacterium cultures if they were male with osteolysis and humeral and/or glenoid component loosening. This study suggests that the effect of Propionibacterium is to create bone resorption rather than acute inflammation.