Origin of propionibacterium in surgical wounds and evidence-based approach for culturing propionibacterium from surgical sites

Abstract

Background: To explore the origin of Propionibacterium in surgical wounds and to suggest an optimized strategy for culturing this organism at the time of revision surgery, we studied the presence of this organism on the skin and in the surgical wounds of patients who underwent revision arthroplasty for reasons other than apparent infection.

Methods: Specimens were cultured in broth and on aerobic and anaerobic media. The presence and degree of positivity of Propionibacterium cultures were correlated with sex. The results of dermal and deep cultures were correlated. Times to positivity and the yields of each media type and specimen source were investigated.

Results: Propionibacterium grew in twenty-three of thirty cultures of specimens obtained preoperatively from the unprepared epidermis over the area where a skin incision was going to be made for a shoulder arthroplasty; males had a greater average degree of positivity than females (p < 0.002). Twelve of twenty-one male subjects and zero of twenty female subjects who had cultures of dermal specimens obtained during revision shoulder arthroplasty had positive findings for Propionibacterium (p = 0.0001). Twelve of twenty male subjects and only one of twenty female subjects had positive deep cultures (p = 0.0004). The positivity of dermal cultures for Propionibacterium was significantly associated with the positivity of deep cultures for this organism (p = 0.0001). If Propionibacterium was present in deep tissues, it was likely that it would be recovered by culture if four different specimens were obtained and cultured for a minimum of seventeen days on three different media: aerobic, anaerobic, and broth.

Conclusions: Because the surgical incision of dermal sebaceous glands may be a source of Propionibacterium in deep wounds, strategies for minimizing the risk of Propionibacterium infections may need to be directed at minimizing the contamination of surgical wounds from these bacteria residing in rather than on the skin. Obtaining at least four specimens, observing them for seventeen days, and using three types of culture media optimize the recovery of Propionibacterium at the time of revision surgery.

Figures

Fig. 1

Results of culture of specimens from the surface of the unprepared epidermis of eighteen healthy male volunteers (square symbols) and twelve healthy female volunteers (round symbols). The degree of positivity (vertical axis) is a numerical value assigned to the quasi-quantitative result of the culture (no growth = 0, one colony = 0.1, broth only = 0.2, 1+ = 1, 2+ = 2, and 3+ = 3). The epidermis of the male subjects had a higher rate and degree of positivity than the female subjects.

Fig. 2

Culture results for twenty female patients (left of the vertical line) and twenty-one male patients (right of the vertical line) who underwent revision shoulder arthroplasty for issues other than clinically apparent infection (i.e., stiffness, pain, or component loosening). Triangles indicate the result of dermal cultures obtained immediately after the skin incision (Initial dermis). Squares indicate the result of dermal cultures obtained immediately prior to the administration of antibiotics (Final dermis). Solid dots indicate the results of the deep cultures. The vertical axis indicates the total degree of positivity for each of the specimens obtained with use of the system described for Figure 1. For the deep cultures, the Propionibacterium score indicates the sum of the degrees of positivity for all deep cultures of specimens obtained from the shoulder. In contrast to what was found for the male patients, the dermal and deep cultures for the female patients were consistently negative. Male patients with positive dermal cultures often had positive deep cultures as well.