Cross-contamination between Tracheostomy and Sternotomy Following Percutaneous Dilation Tracheostomy and Its Influence on Sternal Wound Infections in Cardiac Surgery

Background: The impact of percutaneous dilation tracheostomy (PDT) on development of post-median sternotomy wound infection (SWI) is still debated. In a retrospective study, we aimed to investigate the frequency of cross-contamination and whether PDT changes the microbial spectrum in SWI. Methods: In this study, between January 2010 and May 2013, all patients with post-median sternotomy PDT were included (group A and B). Our control group C comprised patients with SWI without postoperative PDT. Pre-, peri-, and postoperative parameters, day and duration of the PDT, duration of mechanical ventilation and microbiological data from tracheal secretion, sternal wound, blood cultures, and clinical outcomes were collected. Results: From 265 patients who received PDT, 25 (9.43%) developed SWI after PDT (group A). In eight (32%) patients from group A, identical pathogens were isolated from the tracheostomy and sternal wound. The majority of patients with cross-contamination had polymicrobial infections. The common pathogens isolated from both sites were: 77% tracheal pathogens (Coagulase-negative staphylococci (n = 2), Candida spp. (n = 5), Klebsiella spp. (n = 3)) and 23% gram-negative bacteria (n = 3). In group A versus C, there was significantly more mediastinitis (28% (7/25) versus 12.2% (16/131), p = 0.0001). Seven (87.5%) patients with cross-contamination had CDC II-III infection. In our control group (C) of SWI without postoperative PDT, 73.2% gram positive species and only 1.5% (2/131; p = 0.0012) Candida spp. were detected. The overall mortality was high in group A versus C (48% (n = 12) versus 3.82% (n = 5), p = 0.0001). Mortality within the first 30 postoperative days was 12% (n = 3) versus 0 (p = 0.003). One patient who died in the first 30 POD had cross-contamination. Pearson-test did not demonstrate a correlation between the day of PDT and the development of SWI. Conclusion: Although risk of microbial cross-contamination from PDT to the sternal wound is high (32%) due to neighboring anatomical structures, its clinical relevance remains uncertain. Coincidence of PDT and SWI remains low. In our study, we could detect changes in the microbial spectrum: the incidence of candida spp. was significantly higher (20% versus 1.5% p = 0.0012) in patients with SWI after PDT than in SWI without PDT. At the same time, the incidence of S. aureus (8% versus 29% p = 0.026) was significantly lower. Large multicenter studies would be needed to further clarify the described interferences.