Persistent candidemia matters: risk factors and clinical outcomes in a retrospective matched cohort study.

OBJECTIVES

Persistent candidemia remains incompletely understood, with inconsistent definitions and conflicting evidence on outcomes. We aimed to identify factors associated with persistent candidemia and to assess its impact on complications and 30-day mortality in a single-centre cohort.

METHODS

We conducted a retrospective matched cohort study including patients aged ≥ 14 years with at least one positive peripheral blood culture for Candida species (September 2014-September 2024) at a 900-bed tertiary hospital in Valencia, Spain. Persistent candidemia was defined as isolation of the same Candida species from blood cultures obtained ≥ 5 days after initiation of appropriate antifungal therapy. A total of 105 patients with persistent candidemia were compared with 105 with non-persistent candidemia, matched by Charlson comorbidity index. Multivariable logistic regression identified factors independently associated with persistence and 30-day mortality.

RESULTS

Persistent candidemia was identified in 105/515 (20.4%) patients with follow-up cultures. Patients with persistent candidemia were younger (59.3 vs. 66.0 years, p = 0.02) and had higher Candida scores (2 [IQR 1-3] vs. 2 [1-2], p = 0.03). Persistent candidemia was associated with central venous catheter presence (80% vs. 65.7%, p = 0.03), parenteral nutrition (61.9% vs. 46.7%, p = 0.04), and ICU-acquired candidemia (55.2% vs. 37.1%, p = 0.01). Early catheter removal within 48 h was less frequent (27.2% vs. 56.8%, p < 0.01) and independently associated with a reduced risk of persistence (aOR 0.20; 95% CI 0.09-0.42). Thirty-day mortality was higher in patients with persistent candidemia (29.5% vs. 16.2%, p = 0.03). After adjustment, persistent candidemia (aOR 2.10; 95% CI 1.03-4.26) and ICU-acquired candidemia (aOR 3.38; 95% CI 1.57-7.29) remained independently associated with increased 30-day mortality.

CONCLUSIONS

Persistent candidemia occurred in one in five patients and was independently associated with increased 30-day mortality. Early catheter removal emerged as the only independent protective factor, underscoring the importance of prompt source control and systematic follow-up blood cultures in the management of candidemia.