Does percutaneous dilatational tracheostomy increase the incidence of sternal wound infection - a single center retrospective of 4100 cases

Lachmandath Tewarie, Rachad Zayat, Helga Haefner, Jan Spillner, Andreas Goetzenich, Rüdiger Autschbach, Ajay Moza

The impact of percutaneous dilatational tracheostomy (PDT) on the development of post-median sternotomy wound infection (SWI) and mediastinitis is still controversial. We aimed to investigate the frequency of cross-infection and incidence of SWI after PDT. In a retrospective design, out of a total of 4100 procedures, all patients who had undergone median sternotomy and postoperative PDT were included from January 2010 to May 2013. For comparison of the pathogens isolated from SWIs, data from all patients who developed an SWI without a PDT during the aforementioned period were also analyzed. Demographical, pre-, peri- and post-operative data were compared. Microbiologic analysis from cultures of sternal and tracheal wounds was performed. Day and duration of tracheostomy were correlated to SWI occurrence. Of the 265 patients who underwent a PDT, 25 (9.4 %) developed an SWI. In this cohort, identical pathogens were isolated from the tracheostomy and SWI in 36 % (9/25) of the patients. Of the pathogens isolated from the SWIs from the PDT + SWI group, 60 % were gram-positive bacteria, 20 % gram-negative bacteria and 20 % Candida spp. In the cross-infection group, the patients developed the following types of SWIs: 11.1 % CDC I, 55.6 % CDC II and 33.3 % mediastinitis (CDC III). The incidence of SWI in the group SWI + PDT was 9.4 % (9.4 % vs. 3.4 %, PDT + SWI and SWI w/oPDT , respectively, p = 0.0001). In group SWI w/oPDT , only 1.5 % (2/131 vs. 5/25; p = 0.001) Candida spp were isolated from SWI. The infection-related in-hospital mortality was high in groups PDT + SWI vs. SWI w/oPDT (20 % vs. 0 %, respectively; p = 0.0001). The statistical analysis did not demonstrate any correlation between time of performing PDT and occurrence of SWI. There was a high incidence of microbial cross-infection from the PDTs to the sternal wounds in our study. We did not detect any correlation between the time of performing PDT and occurrence of SWI. According to our data, PDT seems to increase the incidence of SWI, especially caused by Candida spp., after cardiac surgery, which results in a prolonged hospital stay. Therefore, early antifungal prophylaxis after a PDT might be reasonable in high-risk patients on long-term mechanical ventilation if there is an impending SWI.