In the intensive care unit (ICU), patients are at heightened risk for healthcare-associated infections due to factors such as invasive procedures, weakened immune systems, and prolonged hospital stays (Thomas-Rüddel et al., 2022). Candida species, including the emerging multidrug-resistant Candida auris, are prominent pathogens that contribute to these infections through skin colonization (Soriano et al., 2023; Lass-Flörl et al., 2024).

Despite the predominant feature of colonization, C. auris candidemia usually follows colonization and presents mortality rates up to 70%, resulting in this species, together with C. albicans, being listed on the World Health Organization (WHO) critical priority group of fungal pathogens (WHO, 2022). Candida species are the third leading cause of nosocomial bloodstream infections and reducing their transmission is an important aspect of ICU care (Poissy et al., 2022).

Given the critical importance of accurately screening and identifying Candida strains, systematic surveillance in high-risk settings, particularly among ICU patients, is essential (Ahmad and Asadzadeh, 2023). Global surveillance efforts have been implemented to monitor the spread and characteristics of C. auris across all continents (Worth et al., 2020; Contreras and Morgan, 2022; Taori et al., 2022; Piatti et al., 2022; de Melo et al., 2023; Rowlands et al., 2023; Vinayagamoorthy et al., 2022). However, the lack of C. auris surveillance in countries with low prevalence presents significant challenges to global public health, as undetected introductions and potential outbreaks could occur, leading to delayed responses and increased transmission. In this context, several studies highlight the importance of screening high-risk patients upon ICU admission, particularly those with a history of healthcare exposure in foreign hospitals (Worth et al., 2020; Briano et al., 2022; Ahmad and Asadzadeh, 2023). The early identification of potential community-acquired cases is crucial in preventing C. auris colonization and minimizing the risk of subsequent infections (Lass-Flörl et al., 2024).

Current methods such as polymerase chain reaction (PCR), matrix-assisted laser desorption/ionization-time of flight mass spectrometry (MALDI-TOF MS), and DNA sequencing are not universally accessible, require specialized expertise, and are time-intensive (Keighley et al., 2021; Maldonado et al., 2018). C. auris poses significant laboratory challenges due to frequent misidentification by standard microbiology assays and automated systems (Jones et al., 2024; de Cássia Orlandi Sardi et al., 2018). Real-time PCR and DNA sequencing are typically required for reliable identification, with the sequencing of specific DNA regions like the D1/D2 of 28S rDNA or internal transcribed spacer (ITS) helping to differentiate C. auris clades (Lockhart et al., 2022). Given these challenges, it is crucial to encourage and support the establishment of C. auris surveillance programs in a cost-effective manner. This involves providing affordable diagnostic tools to ensure that emerging cases are detected and managed swiftly. To address this gap, we aimed to assess the pattern of Candida colonization in ICU patients and unveil the prevalence of C. auris.

Our results highlight the variability of Candida species frequency depending on the local hospital epidemiology, and our study evidenced the importance of the ICU type (surgery vs. general), as already mentioned by other authors (Guinea et al., 2014). Findings from this work are in line with previous studies (carriage/infection) worldwide, showing that C. albicans still ranks as the main yeast colonizer on ICU patients but, in the last decade, a change in favor of NAC has been observed in the ICU (Sasso et al., 2017; Boattini et al., 2023).

Sequential sampling to assess the dynamics of Candida species over time revealed stable results, with a tendency toward the prevalence of C. albicans over NAC species. This stability in colonization patterns aligns with previous longitudinal studies that used respiratory tract samples (Willger et al., 2014; Krause et al., 2016). Our results demonstrate that C. albicans consistently maintained colonization, reinforcing the established understanding of its virulence factors. C. albicans is known to possess the most extensive range of virulence traits among Candida species, enabling it to persist and maintain long-term viability within the host (Staniszewska, 2020). In contrast, NAC species were more often associated with transient colonization, with species turnover observed between ICU admission and collection after 1 week of hospitalization (day 8).

The observed decline in Candida spp. prevalence between 2020 and 2022 could be attributed to several factors, including improvements in infection control practices, shifts in patient demographics, or the impact of external influences such as the COVID-19 pandemic, which altered hospital admission patterns and antimicrobial usage. However, the lack of statistical significance in this trend indicates that further data collection and extended observation periods are needed to determine whether this decline is truly meaningful or simply part of natural variability. Understanding these trends is essential for guiding future surveillance efforts, optimizing infection control strategies, and anticipating potential changes in Candida prevalence within hospital settings.

The Candida parapsilosis complex was prevalent (31.5%) among NAC species, in agreement with previous studies that showed C. parapsilosis as the second most common species after C. albicans found at Southern European hospitals (Portugal, Spain, Italy, and Greece) (Castanheira et al., 2020; Boattini et al., 2023). It is worth noting that C. parapsilosis ranked first at admission for one ICU (p=0.011). This cohort of patients was enrolled in our study before the COVID-19 pandemic, and we may point to the reinforcement of infection control measures taken during the pandemic years, as the hands of healthcare professionals are recognized as a major vector for C. parapsilosis nosocomial acquisition.

Of the 36 isolates identified as C. glabrata, all strains were confirmed to be C. glabrata stricto sensu. These findings align with other studies that have investigated the presence of cryptic species within the glabrata complex and found no evidence of their existence (Kaan et al., 2021; Nasri et al., 2023). Only cryptic species of the C. parapsilosis complex were isolated. These results point to a stability in the distribution and emergence of cryptic species in the ICU setting in Portugal (Faria-Ramos et al., 2014).

C. tropicalis is typically derived from the gastrointestinal tract, which may account for its low prevalence of only 4.2% among the Candida isolates. Additionally, C. lusitaniae was identified in the cohort at a frequency of 1.1%, with a significant association found in the FFH surgical ICU (p=0.012). This pattern reflects similar findings reported in Portugal, where C. lusitaniae was present at a prevalence of 2.6% (Pinto-Magalhães et al., 2019).

Our screening to unveil C. auris in ICU patients relied on qPCR given the critical need for rapid identification to implement public health measures promptly. Nonetheless, MSAA medium showed a specificity of 94.8%, albeit C. auris was not detected in this cohort. Commercial culture media used to differentiate C. auris from other Candida species also presented misidentification issues. Although CHROMagar Candida Plus C. auris can identify all C. auris strains, other species were misidentified as C. auris, such as C. parapsilosis complex species (Taverna et al., 2023). Additional selective media were described for the isolation of C. auris (Ibrahim et al., 2021; Das et al., 2021). As for MSAA, these selective media are based on the yeast growth conditions: thermo-resistance and halo-tolerance (Jeffery-Smith et al., 2018). MSAA has less inhibitors (as we used a 40°C incubation temperature instead of 42°C and 10% of sodium chloride instead of 12.5%) and stable conservation (4°C) after medium preparation compared with medium described by Das et al., which requires fresh preparation (Das et al., 2021). All C. auris strains representing different major clades were able to grow on MSAA plates. However, as for the study by Das et al., the C. auris clade II strain showed limited growth even after 48 h of incubation time, due to its longer doubling time at 42°C (Das et al., 2021). Over the SCA described by Ibrahim et al., MSAA has the advantage of reading at 48 h instead of 72 h (Ibrahim et al., 2021).

In our study, MALDI-TOF MS-based identifications showed total agreement with molecular procedures, reinforcing other authors who stated that MALDI-TOF MS systems can correctly identify C. auris in most cases (Mahmoudi et al., 2019). Overall, our results show that when qPCR is not available, the MSAA medium can be used for the routine screening of C. auris. The MSAA medium can be easily prepared, is straightforward to use as it does not require mycological or molecular expertise, and provides a simple and cost-effective tool for the detection and presumptive identification of C. auris in the clinical microbiology laboratory. From these findings, the proposed workflow to identify C. auris based on culture-based methodologies associated with the molecular identification of the isolates is a good option and allows antifungal susceptibility tests to be performed.

To the best of our knowledge, this prospective study is the first surveillance study for C. auris colonization in Portuguese tertiary hospital ICUs. One major strength is that it is specifically designed to screen for C. auris in a cohort of patients from two large hospital centers in the suburban Lisbon region. For over 2 years, our findings showed that C. auris is not prevalent in Portuguese hospitals. Our findings align with other prospective screening studies, namely from the UK (Sharp et al., 2021), Egypt (Khairat et al., 2021), and Germany (Heindel et al., 2022), which also reported no C. auris colonization among ICU patients. Additionally, our results are consistent with observations from Europe and North America, where C. auris colonization and infection in ICUs are typically sporadic and often associated with nosocomial outbreaks (Pacilli et al., 2020; Worth et al., 2020; Eckbo et al., 2021; Piatti et al., 2022). However, the situation is markedly different in India, where C. auris has become the leading Candida species isolated from the blood of critically ill patients, with a notable increase in cases during the COVID-19 pandemic (Shastri et al., 2020; Sharma and Chakrabarti, 2023).

In Europe, new C. auris cases were mainly diagnosed during the COVID-19 pandemic, particularly in Spain (n=591), Italy (n=291), and Greece (n=71) (Kohlenberg et al., 2022). Nevertheless, several studies carried out all over Europe in recent years reported only isolated cases of C. auris in hospital units, such as with the first Portuguese clinical case (Kohlenberg et al., 2022; Henriques et al., 2023). Furthermore, a multicentric retrospective study including all candidemia case isolates from six Southern European tertiary hospitals did not find C. auris (Boattini et al., 2023). In a 2022 prospective screening study by Contreras et al., C. auris isolates were recovered from critically ill patients with complicated clinical conditions who had a history of medical care not only in intensive care units but also in long-term care facilities (Contreras and Morgan, 2022). Similarly, the CDC’s pilot study on C. auris screening in several healthcare facilities across New York City between 2017 and 2019 found that colonizing isolates were predominantly obtained from patients in long-term care facilities and senior residences (Rowlands et al., 2023).

As the patients in these two hospitals did not remain in the ICU for extended periods (e.g., a median of 5 days), they may not have been exposed to the cumulative risks that typically lead to C. auris colonization. A shorter stay reduces the likelihood of encountering contaminated environments, decreases the time exposed to other colonized patients, and limits the use of invasive devices that could serve as entry points for C. auris. The absence of C. auris colonization in the two hospitals could indeed be due to the relatively short duration of hospital stays among the participants. This aligns with existing literature that indicates a higher risk of C. auris colonization after prolonged hospital or ICU stays, typically at approximately or after 20 days (Corcione et al., 2022). Das et al. (2018) highlighted a critical observation that patients in ICU settings typically acquired C. auris infections after a period of 15 to 20 days (Das et al., 2018).

Another potential limitation of our study was the collection of samples from only two sites for C. auris screening in the ICU setting. This approach may have reduced detection effectiveness, as Rowlands et al. (2023) emphasized the importance of including nasal swabs alongside bilateral axillary/inguinal mucosal sampling for comprehensive screening (Rowlands et al., 2023). Additionally, the study collected fewer swabs than anticipated (988 instead of 1,350) due to its premature termination. This early conclusion was influenced by several factors, including logistical challenges and external events, which resulted in fewer patients being swabbed than initially planned. Specifically, in the two ICUs at FFH, the study was halted prematurely in March 2020 due to the onset of the COVID-19 pandemic. The pandemic disrupted hospital research activities, leading to an early end to the study. At BAH, from October 2022 onwards, staff shortages created significant difficulties in continuing the study, particularly in the collection of swabs. These staffing limitations ultimately forced an early end to the research. Understanding the specific reasons behind these premature terminations is crucial for planning and designing future research studies.

Although C. auris colonization currently appears to be uncommon in Portuguese hospitals, Portugal has seen a significant increase in the number of immigrants from areas endemic for C. auris. This trend raises concerns about the possibility of the fungus being introduced and spread in the country and suggests that Portuguese hospitals should consider the protocol presented for an active screening.

Considering the above findings, this study, along with the various studies cited in the Discussion, underscores the need to implement C. auris screening alongside existing protocols for nosocomial microorganisms such as methicillin-resistant Staphylococcus aureus, vancomycin-resistant Enterococcus, and carbapenemase-producing Gram-negative bacteria. It is recommended that nasal and rectal swabs, in addition to axillary and inguinal swabs, be utilized for more comprehensive screening. Furthermore, screening should be systematically extended beyond the ICU to include palliative care units to enhance the early detection and management of C. auris.

In conclusion, by integrating molecular technologies with traditional culture assays, a screening protocol for C. auris was developed and implemented. This protocol involves initial screening using MSAA to identify potential C. auris isolates, followed by confirmation using traditional PCR or MALDI-TOF MS. Through our study, we also uncovered the colonization pattern of Candida in the ICU and contributed to the optimization of overall patient management.