Impact of systemic antifungal therapy on the detection of Candida species in blood cultures in clinical cases of candidemia

The diagnosis and follow-up of candidemia still rely on blood cultures (BCs). In vitro studies show that antifungals can significantly modify the result of blood culture not containing adsorbing agents. We aimed to evaluate, under clinical conditions, the impact on BC yeast detection of systemic antifungal therapy (SAT). Patients (n = 125) experiencing candidemia at Grenoble University Hospital (France) were included in a 4-year retrospective study. The Plus Aerobic/F (Aerobic) and Plus Anaerobic/F (Anaerobic) bottles, which both contain adsorbing resins and the non-resin selective Mycosis IC/F (Mycosis) bottles, were compared using multivariate hierarchical models adjusted for clinical characteristics. The positivity rate (PR) is decreased in patients with SAT (p < 0.01), abdominal surgery (p = 0.01), and hemodialysis (p = 0.02). In all bottles, SAT reduces PR by a factor of 0.16 (95 % CI: [0.08; 0.32]) and increases the time to positivity (TTP) by a factor of 1.76 ([1.30; 2.40]; p < 0.01). In the presence of SAT, TTP is higher in non-resin bottles (Mycosis) than in resin bottles (RR = 1.76, [1.30; 2.40]); however, the TTP in nonresin and resin bottles remains comparable. Although discordant results are observed with and without SAT (37 and 58 % respectively), we showed that the presence of SAT decreases significantly the agreement rate by a factor of 0.29 (CI: [0.12; 0.68]). The combination of Anaerobic and Mycosis bottles allowed a 100 % positivity rate for C. glabrata. SAT significantly affects BC results. Because they provide additional and complementary results, this study supports the concomitant use of resin and selective bottles, especially in patients receiving SAT.