Background
Candida auris is an emerging multidrug-resistant fungus recognized as a global public health threat. Reference Du, Bing and Hu1 It has been implicated in multiple outbreaks worldwide, partly due to its high adaptability and intrinsic resistance to one or more classes of antifungal agents. Reference Du, Bing and Hu1,Reference Magnasco, Mikulska and Sepulcri2 Despite increasing colonization rates, the expansion of pilot screening programs, and high associated mortality, no standardized protocol exists in the United States for C. auris screening. Reference Du, Bing and Hu1,Reference Briano, Magnasco and Sepulcri3– Reference Rowlands, Dufort and Chaturvedi5 Notably, there is a paucity of studies evaluating the efficacy of C. auris screening in Southeastern or Appalachian states within the United States, highlighting a geographic gap in the current literature. Reference Rowlands, Dufort and Chaturvedi5–Reference Cheng, Brody and Ehni7 Additionally. the absence of standardized screening guidelines presents challenges for infection prevention, timely intervention, and containment. In February 2023, the University of Kentucky Healthcare (UKHC) implemented a targeted C. auris screening program. This study aims to (A) identify risk factors and common characteristics associated with C. auris colonization and infection using data from the screening program and (B) assess temporal trends in positivity rates.
Methods
This retrospective, observational study was conducted at UKHC, a 1,086-bed academic medical center, from July 1, 2021, to June 30, 2024. All adult patients (≥18 yr) tested for C. auris through either surveillance screening or clinical culture were included. The study was approved by the University of Kentucky Institutional Review Board. Before February 2023, C. auris screening was conducted exclusively during outbreak investigations. Following implementation, routine screening was expanded using axillary and groin swabs analyzed via polymerase chain reaction (PCR). Patients were screened if they met at least one of the following criteria: admission to an intensive care unit, admission from an outside facility with a wound or tracheostomy present on admission, or an active carbapenem-resistant organism infection. Surveillance screening was also performed in response to point prevalence testing when a patient was housed adjacent to an individual with newly identified C. auris who had not been under appropriate transmission-based precautions for more than 24 hours. Clinical cultures were defined as microbiological specimens, including respiratory, cerebrospinal fluid (CSF), blood, wound, or urine cultures, that yielded C. auris. For consistency in analysis, patients remained categorized according to their original classification, clinical or colonized, regardless of subsequent determinations by the care team. Cases were further classified as either community-onset, defined as sample collection before hospital day four, or hospital-onset, defined as sample collection on or after hospital day four, excluding emergency department visits. We compared patients with C. auris colonization to those with infection. χ2 and Fisher’s exact tests were used for categorical variables, and the Mann–Whitney U test was applied to continuous variables. We then conducted univariate logistic regression to identify factors associated with infection compared to colonization. Clinical variables found to be significant in univariate analysis were included in the multivariable analysis. A P value of <0.05 was considered statistically significant. The data analysis for this study was generated using SAS software, Version 9.4. Copyright © 2016, SAS Institute Inc.
Results
Between July 1, 2021, and June 30, 2024, 13,642 C. auris tests were performed, identifying 70 positive cases, including 13 preimplementation and 57 postimplementation (Figure 1). Among preimplementation cases, 7.69% were community-onset, whereas postimplementation community-onset cases accounted for 57.37% (Supplemental Table 1). Postimplementation monthly positivity rates ranged from 0% to 2.18%, with a mean of 0.96%. Among colonized patients, the primary reasons for testing were ICU admission (50.00%), point prevalence (20.00%), and wound (6.67%, Table 1). Among ten clinical cases, the most commonly identified clinical specimen types were bronchoalveolar lavage and urine (n = 5 each), with lower detection in invasive specimens (CSF = 1, blood = 2, Supplemental Table 2). Although C. auris was detected in clinical specimens, three cases did not receive anti-fungal treatment as the pathogen was deemed non-clinically significant. Thirty- and ninety-day mortality rates were comparable between clinical and colonized cases, with 30% versus 25% (P = 0.7377) and 40% versus 28.3% (P = 0.4561), respectively. No notable demographic differences were found between clinical and colonized patients. The mean age was 61.5 ± 15.2 years for clinical cases and 59.98 ± 14.7 years for colonized cases (P = 0.9557). Male sex predominated, accounting for 57.1% of the total cohort, with a higher, though statistically insignificant, proportion in clinical cases (80%) than in colonized cases (53.3%, P = 0.1147). The most common primary admission diagnoses were infectious diseases, neurology/neurosurgery, and pulmonary conditions, with no significant differences between clinical and colonized groups (P = 0.4035). In univariable analysis, significant risk factors for clinical cases included diabetes mellitus (P = 0.0143), osteomyelitis (P = 0.0273), tube feeding (P = 0.0293), tracheostomy or mechanical ventilation (P = 0.0404), and urinary diversion (P = 0.0121). These variables were included in the multivariable logistic regression model, in which none remained statistically significant.
Testing Volume and Positivity Rate Before and After Implementing Candida auris Admission Screening at the University of Kentucky (2021–2024).
Demographics, past medical history, and admission and testing information for Candida aurispositive patients
a Chronic obstructive pulmonary disease.
b Carbapenem-Resistant Organism.
c Substance use disorder / Intravenous drug use.
d Activities of daily living.
e Only includes those tested at UKHC; all outside tests (n = 12) excluded for this variable.
f Intensive care unit.
g Long-Term Acute Care Hospital.
h Skilled Nursing Facility.
I Multivariable logistic regression.
Discussion
In February 2023, UKHC implemented a targeted C. auris screening program for inpatients identified as high-risk based on internally defined criteria. The number of monthly tests ranged from 480 to 1,210, with a positivity rate of 0.96% on average. The proportion of community-onset cases increased from 8% to 54% with the implementation of C. auris screening. In univariable analyses, diabetes mellitus, osteomyelitis, tube feeding, tracheostomy, and urinary diversion were significantly associated with clinical infection compared to colonization. However, none of these variables remained statistically significant in multivariable logistic regression. Among colonized patients, the primary reasons for testing were ICU admission, point prevalence, and wound-related concerns.
Targeted screening facilitated early identification of colonized patients, enabling timely implementation of contact precautions and potentially preventing C. auris transmission to other hospitalized patients. Implementing targeted inpatient screening was feasible, with a relatively low positivity rate (∼1%). However, the absolute number of detected cases was 3.56 per month, demonstrating the utility of this strategy in identifying colonized patients and mitigating exposure risk. Similar rates have been reported in studies with comparable testing criteria. Reference Atrubin, Murphy and Silbert6–Reference Bolanos, Lie, Bauer, Herminia, Slade and Parada8 Following implementation, the proportion of community-onset cases increased from 8% to 54%, likely reflecting improved detection of colonized patients. While no outbreaks were observed before implementation, two occurred afterward, involving three and seven associated patients, respectively. Despite enhanced screening efforts, these outbreaks highlight the persistent risk of C. auris transmission, emphasizing the need for stringent infection control measures and continued surveillance.
Currently, there is no official guidance on screening opportunities from major public health entities; however, medical centers have implemented programs based on known risk factors. Reference Magnasco, Mikulska and Sepulcri2,Reference Leonhard, Chong and Foudraine4–10 Risk factors include being critically ill or immunocompromised, elderly age, diabetes mellitus, recent surgery, the presence of indwelling medical devices, the use of hemodialysis, a neutropenic state, chronic renal disease, mechanical ventilation, or the use of broad-spectrum antibiotic and/or antifungal drugs. Reference Du, Bing and Hu1,Reference Briano, Magnasco and Sepulcri3,Reference Rowlands, Dufort and Chaturvedi5,Reference Cheng, Brody and Ehni7,10 Screening protocols that have been developed comprise admission from other countries (either direct admission or 24 h duration) or the presence of an MDRO, Reference Leonhard, Chong and Foudraine4 admission from long-term acute care hospitals, outside hospitals, and skilled nursing facilities, and ICU admissions. Reference Rowlands, Dufort and Chaturvedi5–Reference Bolanos, Lie, Bauer, Herminia, Slade and Parada8
Of the 71 positive cases, 10 were clinical infections. Among these, five underwent screening within ±30 days of a positive clinical specimen, with four testing positive, yielding an estimated sensitivity of 80%. This finding suggests that a negative screening result may not reliably exclude the presence of C. auris, particularly in patients with intermittent or low-level colonization. Additionally, 2 of 7 patients with previously known colonization tested negative upon repeat C. auris screening, further underscoring the challenges of detecting colonization. Variability in colonization burden, sample collection techniques, or fluctuations in microbial load may contribute to these discrepancies.
This study has several limitations. The small sample size may have limited the ability to detect additional associations between clinical infection and colonization, along with generating concern about statistical power. Additionally, the study population reflects UKHC’s patient demographics and may not be generalizable to other U.S. healthcare settings. The cost per C. auris PCR test is approximately $50, which presents a potential financial burden. Although cost analysis was not within the scope of this study, a formal cost–benefit evaluation may be warranted in future work. Lastly, no demographic or clinical data were collected for negative patients, limiting comparative assessments.
In conclusion, the implementation of a targeted C. auris screening program at UKHC provided valuable insights into epidemiologic trends, patient demographics, and potential risk factors. Understanding these factors is critical for optimizing infection prevention strategies, refining screening protocols, and informing public health initiatives to mitigate C. auris transmission in healthcare settings.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/ash.2025.10151.
Acknowledgments
We would like to thank Goldrena Butler MSN, Sarah Cecil, Dacey Chitwood BPH, Megan Deaton, Courtney Riddell BPH, Glenda Roberts RN, Mindy Sekhon MPH, Kaira Shook, Dwayne Stafford, Tigre Suder RN, and Daniel Thompson, our colleagues in the Infection Prevention and Control Department for their assistance developing this screening program and support during the conduction of the present study.
Financial support
None.
Competing interests
None.
Open access
