**Background:** Candida airway colonisation is common in ICU patients, particularly those on mechanical ventilation, with reported rates of 30% after 48 hours of ventilation and up to 50% in ventilator-associated pneumonia. Known risk factors include broad-spectrum antibiotics, corticosteroids, immunosuppressants, mechanical ventilation, neutropenia, and diabetes. During the COVID-19 pandemic, respiratory Candida colonisation was reported in up to 21% of COVID-19 patients. However, data on whether SARS-CoV-2 infection independently increases the risk of respiratory tract Candida colonisation were scarce. This study aimed to assess the association of risk factors, particularly SARS-CoV-2 infection, with Candida respiratory tract colonisation, and to evaluate the effect of systemic antibiotic treatments.
**Methods:** The study had two components. First, a longitudinal observational analysis using the MIDAS database examined yeast prevalence in respiratory samples from 21 hospital units at AP-HM Marseille from 1 January 2018 to 31 March 2022 (17,408 respiratory samples total). Second, a retrospective case-control study (April–November 2021) included 300 patients: 100 cases (Candida airway colonisation defined by ≥3 respiratory samples growing yeasts on separate days), 100 positive controls (BactC: positive bacterial culture, no yeast), and 100 negative controls (NegC: no bacteria or yeast cultured). Clinical data were collected from electronic medical records. Multivariate unconditional logistic regression was performed comparing cases to all controls, BactC only, and NegC only.
**Key Results:** Between 2018 and 2022, yeast was isolated in 3,593 of 12,881 positive cultures (28%), with C. albicans predominating (74% of yeast isolates). Yeast prevalence increased from February 2020, coinciding with the COVID-19 pandemic onset, reaching 23% of positive cultures in April 2020 and 24% in May 2021. In the case-control study, 53% of cases had SARS-CoV-2 infection versus 29% (BactC) and 17% (NegC) (p<10⁻⁴). In multivariate analysis comparing cases to all controls, independent risk factors for Candida colonisation included: mechanical ventilation (OR 3.48, 95% CI 1.79–6.74, p=0.0002), diabetes (OR 2.12, 95% CI 1.10–4.07, p=0.0248), length of stay (OR 1.02, 95% CI 1.01–1.03, p<10⁻⁴), carbapenems (OR 3.00, 95% CI 1.58–5.68, p=0.0007), cefepime (OR 2.83, 95% CI 1.42–1.64, p=0.0030), piperacillin/tazobactam (OR 1.99, 95% CI 1.14–3.45, p=0.0148), and fungal coinfection (OR 6.71, 95% CI 1.67–26.97, p=0.0073). SARS-CoV-2 infection was not independently significant. Invasive fungal disease occurred in 16% of cases (5 candidaemia, 10 pulmonary aspergillosis, 1 C. albicans pleurisy) versus 3% in BactC and 0% in NegC (p<10⁻⁴). Mortality was 29% in cases, 23% in BactC, and 10% in NegC (p=0.0032).
**Clinical Implications:** This study demonstrates that while Candida airway colonisation increased during the COVID-19 pandemic, SARS-CoV-2 infection is a confounder rather than an independent risk factor. The key modifiable risk factors are mechanical ventilation, prolonged hospitalisation, diabetes, and broad-spectrum antibiotic use—particularly carbapenems and cefepime. Candida airway colonisation was associated with a sevenfold increased risk of invasive fungal disease, including candidaemia (5% of colonised patients). These findings suggest that antimicrobial stewardship and strategies to reduce duration of mechanical ventilation may help mitigate Candida colonisation risk in critically ill patients. The study's monocentric design and retrospective nature limit generalisability, but the use of three positive samples for case definition and two distinct control groups strengthens the findings.