gms | German Medical Science

Background: In January 2014 an internal audit was performed in our department for short and long term outcome of prosthetic shuntgraft implantations. We found a 12.4% rate of shunt explantations due to graft infection. The majority of the cases was associated with Staphylococcus aureus. Graft infection in general as well as in our department is a serious, life threatening condition with high mortality, long time intensive care and in hospital treatment. The high prevalence of S. aureus rose serious concern about a potentially nosocomial source. The aim of our study was to clarify whether we have a nosocomial problem or not by using genomics techniques.

Materials and methods: Between December 1998 and December 2014 490 prosthetic shuntgrafts were implanted. After exclusion of 54 cases due to several reasons (extension of another shunt, interposition of another shunt, no proper shunt use…) 436 shunt remained for statistical analysis. Clinical data were taken retrospectively from the hospital data system, genetic analysis was defined prospectively and acquired from new cases in 2014.

Protein A is a 42 kDa surface protein originally found in the cell wall of Staphylococcus aureus. It is encoded by the spA gene. So far there are 14860 known different spA gene types (Ridom SpaServer) registered by 57 Ridom SpaServer usercountries with data from 115 countries. In 2014 we had three PTFE shuntgraft infections with involvement of S. aureus. Resistance testing as well as spA genetic typing was performed in all cases.

Results: 14.0% (61/436) prostheses had to be explanted, 12.4% (54/436) due to infection. In 77.8% (42/54) bacteria were found in blood and/or wound culture. One single germ was found in 78.6% (33/42), two germs in 16.7% (7/42) and four in 4.8% (2/42). S. aureus was present in 76.2% (32/42) and in all cases sensitive to methicillin. Staphylococcus epidemidis was found in 16.7% (7/42) of explantations.

Three different S. aureus spA-types were found: t015, t359, t6265. Having three unequal spA-types means, that these patients have different S. aureus sources, so a nosocomial problem could be excluded.

Conclusion: SpA genetic typing is a suitable technique to distinguish between nosocomial and individual sources for prosthetic shuntgraft infections. Now, after exclusion of a nosocomial problem we are going to implement chemoprophylaxis with intranasal mupirocin in hemodialysis patients who are nasal carriers of S. aureus. We know from literature, that this procedure is cost-effective.