Persistence, Source Control, and Secondary Bacterial Bloodstream Infection in Catheter-Related Candidemia: A...

In catheter-related candidemia, a true catheter-source infection is often difficult to distinguish from a catheter that merely marks critical illness. Our primary aim was to identify predictors of 30-day mortality; secondary aims were to characterize species-specific patterns of persistent candidemia, its prognostic meaning within the host immune-inflammatory context, and the relationship between echinocandin therapy and secondary bacterial bloodstream infection (BSI).

We retrospectively reviewed 89 consecutive episodes of catheter-related candidemia at Bursa City Hospital (2021-2025). Catheter association required Candida spp. growth in concurrent catheter and peripheral blood cultures. The primary endpoint was 30-day mortality; persistent candidemia (≥72 h) and secondary bacterial BSI on days 3-30 were evaluated as exploratory endpoints. Raw MIC values were retrospectively reinterpreted according to EUCAST v12.1 breakpoints (effective 10 April 2026).

Thirty-day mortality was 53.9% (48/89). The multivariable model was significant (LR χ² = 12.12, df = 4, p = 0.017; Nagelkerke R² = 0.170); NLR was not significantly associated with mortality (OR = 1.061, 95% CI 1.00-1.13, p = 0.064). Catheter removal, initial echinocandin therapy, and persistent candidemia were not independent predictors, though catheter retention was associated with higher mortality on univariable analysis (83.3%; p = 0.033). Persistent candidemia occurred in 51.7% of episodes, most notably with C. parapsilosis and C. auris. The NLR-persistence interaction was not formally significant, but mortality was higher when both co-occurred (66.7% vs. 36.4%; OR = 3.50, p = 0.075). Secondary BSI occurred in 28.4% of episodes and was more frequent in the echinocandin group (39% vs. 20%; p = 0.062); sensitivity analyses did not change the main findings.

Mortality could not be explained by antifungal choice alone; these findings are exploratory. Host response, source control, species distribution, survival time, and device exposure are possible contributing factors warranting evaluation in prospective studies.