Candida auris infections in ICU patients: risk factors, outcomes, and antifungal resistance patterns

Abstract

Background

Candida auris has emerged as a major nosocomial pathogen in intensive care units worldwide due to its multidrug resistance, environmental persistence, and high transmission potential. While several outbreak reports exist, there remains limited comparative data on clinical features, resistance profiles, and outcomes between Candida auris and non-auris Candida candidemia, particularly in Indian ICU settings. This study aimed to evaluate the incidence, risk factors, antifungal resistance, and clinical outcomes of Candida auris bloodstream infections in comparison to non-auris Candida infections among ICU patients.

Methods

This retrospective cohort study was conducted over four years (January 2020–December 2023) in the ICU of a tertiary care hospital in eastern India. Adult patients (≥ 18 years) with culture-confirmed candidemia were included and stratified into Candida auris and non-auris groups. Demographic, clinical, therapeutic, and microbiological data were extracted. Antifungal susceptibility testing was performed as per CLSI and CDC guidelines. Multivariate logistic regression identified risk factors associated with Candida auris infection. Survival analysis were compared using Kaplan–Meier and Cox regression analysis.

Results

Of 267 ICU patients with candidemia, 38 (14.2%) had Candida auris infection. Candida auris patients were older (median age 60 vs. 53 years) and had lower SOFA scores at admission (median 7 vs. 8; p = 0.02). On multivariate analysis, increasing age (OR 1.10, p = 0.0003) and female sex (OR 16.35, p = 0.012) were independently associated with Candida auris infection. Fluconazole resistance was 94.7% in Candida auris isolates, with amphotericin B and echinocandin resistance in 39.5% and 7.9%, respectively. ICU mortality did not differ significantly between Candida auris and non-auris groups (31.6% vs. 34.5%; p = 0.854), and survival analysis showed no statistically significant difference (log-rank p = 0.07).

Conclusion

This study highlights the growing burden of Candida auris in intensive care unit settings, demonstrating a substantial incidence without increased mortality compared to non-auris Candida species. Advancing age and female sex emerged as significant independent risk factors, emphasizing the need for continued surveillance and targeted risk assessment in regions where Candida auris is endemic.

Background

The emergence of Candida auris, initially identified in Japan in 2009, has rapidly evolved into a significant global healthcare challenge due to its multidrug resistance, persistence on hospital surfaces, and association with high mortality rates [1]. As of 2023, Candida auris has been reported in over 50 countries, with the United States alone documenting more than 4,000 clinical cases within a year [2]. Multiple studies have reported that over 90% of Candida auris isolates are resistant to fluconazole, with amphotericin B resistance observed in approximately 30–40% of cases in certain regions, and echinocandin resistance emerging in sporadic clusters [3, 4]. The mortality rate associated with Candida auris infection ranges from 30 to 60%, particularly in critically ill patients [5, 6]. Intensive care units (ICUs) have been disproportionately impacted owing to the vulnerability of patients requiring mechanical ventilation, invasive catheters, and broad-spectrum antimicrobials [7]. Consequently, Candida auris has been designated as a critical priority pathogen by Centres for disease control and prevention (CDC), necessitating urgent attention in infection surveillance and antifungal stewardship [8].

Recent outbreaks globally have highlighted the pathogen’s potential for sustained nosocomial transmission. Much of the extant global data is derived from outbreak investigations or passive surveillance systems, resulting in significant variability in reported incidence rates, from as low as 5–10% of Candida bloodstream infections to over 40% [7, 9, 10]. In the United Kingdom, a single ICU reported colonization or infection in more than 70 patients during a prolonged outbreak [11]. However, reliable longitudinal data are scarce outside of these high-profile outbreaks. The challenge of accurately identifying Candida auris using conventional laboratory methods, which often leads to misidentification as Candida haemulonii or Candida famata, further complicates the assessment of its true burden [12, 13].

This situation is particularly alarming in India, where Candida auris was first identified in 2011, and multiple Indian studies have since reported a rising prevalence of this pathogen among ICU-related candidemia cases [14, 15]. A multicenter ICU study conducted in New Delhi reported that Candida auris constituted up to 42% of Candida bloodstream infections, with fluconazole resistance nearing 97% and amphotericin B resistance approaching 40% [16, 17]. Other regional studies have corroborated similar trends, particularly in high-dependency units, where prolonged hospitalization, device utilization, and empirical antifungal therapy are prevalent [18, 19]. However, most of these studies were cross-sectional or limited in duration and lacked long-term epidemiological tracking or standardized severity scoring. Notably, the interaction between comorbidities, ICU interventions, and mortality outcomes in Candida auris infections remains inadequately explored in Indian literature.

To address these critical knowledge gaps, we conducted a four-year retrospective observational cohort study within the intensive care unit of a tertiary care teaching hospital in eastern India. Our objectives were to [1] ascertain the incidence of Candida auris infection among ICU patients; [2] identify clinical and treatment-related risk factors; [3] evaluate antifungal resistance patterns; and [4] assess clinical outcomes.

Methods

Study design and setting

This retrospective observational cohort study was conducted in the adult ICU of Institute of Medical Science & SUM (IMS & SUM) Hospital, a tertiary care teaching hospital in India.This study covers four years from January 2020 to December 2023. The ICU is a closed multidisciplinary unit with a high burden on critically ill patients requiring prolonged support. Approval from the Institutional Ethics Committee was obtained before data extraction. (IEC/IMS.SH/SOA/2024/860)

Patient selection

All adult patients (aged ≥ 18 years) admitted to the ICU for a duration exceeding 48 h during the study period were evaluated for inclusion. Patients with culture-confirmed candidemia were identified using the hospital’s microbiological database. These patients were classified into two distinct groups: the Candida auris group, comprising patients with laboratory-confirmed bloodstream infection due to Candida auris, and the non-auris Candida group, consisting of patients with candidemia attributable to other Candida species. Exclusion criteria included patients with polymicrobial fungal infections, known colonisation without clinical signs of infection, incomplete medical records, or those who succumbed within 24 h of ICU admission.

Microbiological methods

Blood cultures were processed using the BACT/ALERT 3D system (bioMérieux, France). The initial identification of yeast isolates was conducted using the VITEK-2 YST system. For isolates exhibiting atypical identification profiles or unusual resistance patterns, confirmatory identification of Candida auris was performed using matrix-assisted laser desorption/ionization-time of flight (MALDI-TOF) mass spectrometry. Antifungal susceptibility testing was performed using Vitek^® 2 (bioMérieux, Marcy-l’Étoile, France) platforms. Susceptibilities were determined against fluconazole, voriconazole, and amphotericin B for all isolates. Additionally, for 28 isolates, MIC values were also available for fluconazole, anidulafungin, micafungin, and caspofungin. Interpretation of MICs was carried out using CLSI M27-M44S guidelines [20]. For Candida auris, the CDC tentative MIC breakpoints were used. The susceptibility thresholds applied were.

Fluconazole ≥ 64 µg/mL: Resistant.

Amphotericin B ≥ 2.0 µg/mL: Resistant.

Caspofungin ≥ 2.0 µg/mL: Resistant.

Micafungin ≥ 4.0 µg/mL: Resistant.

Anidulafungin ≥ 4.0 µg/mL: Resistant.

Data on the site of isolation, date of positivity, and resistance to fluconazole, amphotericin B, and echinocandins were recorded when available.

Data collection

Clinical, demographic, and therapeutic data were systematically extracted from the ICU records and the hospital’s electronic medical system using a standardized protocol. The variables included were as follows: demographics: age and sex; comorbidities: diabetes mellitus, chronic kidney disease(CKD), hypertension, and chronic obstructive pulmonary disease (COPD), corticosteroid therapy, immunosuppressive therapy, underlying malignancies, HIV status, history of solid organ transplantation, and pre-existing cardiac, gastrointestinal, or neurological comorbidities; ICU Interventions, duration and timing of mechanical ventilation, central venous catheterisation (CVC), and arterial catheterisation (including dates of insertion and removal), and the use of total parenteral nutrition (TPN) if applicable, as well as any recent major surgery (within 30 days), prior antifungal exposure(documented administration of systemic antifungal agents within 30 days prior to the onset of candidemia), prior antibiotic use (administration of systemic antibacterial agents within the 30 days preceding the diagnosis of candidemia); Severity Scores: Acute Physiology and Chronic Health Evaluation II (APACHE II) and Sequential Organ Failure Assessment (SOFA) scores calculated at ICU admission; Nutrition Risk in the Critically Ill (NUTRIC) score for nutritional risk; and organ failure status defined at admission; and Outcomes: crude mortality and length of ICU stay (in days).

Definitions

Candidemia was characterized by the presence of at least one positive blood culture for a Candida species accompanied by clinical signs of infection [21].

Colonization was defined as the presence of Candida at non-sterile sites without clinical symptoms and was excluded from the analysis.

The APACHE II score is composed of the acute physiology score (such as temperature, mean arterial pressure, heart rate, respiratory rate, oxygenation, arterial pH, serum sodium, potassium, creatinine, heamatocrit, white blood cell count, glasgow coma scale) age, chronic health evaluation, and type of ICU admission (medical, urgent surgery or elective surgery) [22].

The SOFA score assesses disease-related organ failure (neurological, respiratory, cardiovascular, renal, hepatic, and hematologic) [23].

The NUTRIC is a validated risk assessment tool designed to evaluate the risk of adverse outcomes in ICU patients due to undernutrition. It is calculated based on five parameters: age, APACHE II score, SOFA score, number of comorbidities, and days from hospital admission to ICU admission [24].

Sample size calculation

Based on preliminary data suggesting an estimated Candida auris incidence of 14% and aiming to detect a 15% difference in 30-day mortality between groups with 80% power and 5% significance, a minimum sample size of 150 patients (including at least 38 with Candida auris) was estimated [6].

Statistical analysis

All analyses were performed using Python 3.11, incorporating libraries such as Pandas, Statsmodels, Lifelines, and Seaborn. The normality of continuous variables was assessed using the Shapiro–Wilk test. Variables exhibiting a normal distribution were compared using the independent t-test, whereas non-parametric variables were evaluated using the Mann–Whitney U test. Categorical variables were compared using either the chi-squared test or Fisher’s exact test, as appropriate. Multivariate logistic regression was employed to identify independent risk factors for Candida auris infection, with variables demonstrating a p-value of less than 0.1 in univariate analysis or possessing known clinical relevance included in the final model. Results are presented as adjusted odds ratios (ORs) with 95% confidence intervals (CIs). Kaplan–Meier survival analysis was performed to estimate survival functions for the study groups. Cox proportional hazards regression, adjusted for SOFA score and age at admission, was used to compare survival between the groups. Statistical significance between survival curves was assessed using the log-rank test. A two-sided p-value of ≤ 0.05 was considered statistically significant for all analyses.

Results

Study population and baseline characteristics

Over a four-year period, 267 adult patients in the ICU were enrolled in the study. Among these, 38 patients (14.23%) were diagnosed with culture-confirmed Candida auris bloodstream infection, while the remaining 229 patients constituted the non-auris Candida group. Figure 1 shows the flow diagram illustrating patient recruitment and inclusion. The median age of patients in the Candida auris group was 60 years compared to 53 years in the non-auris Candida group (p = 0.08). The gender distribution was comparable between the groups, with 58.81% men in the Candida auris group and 59.92% in the non-auris Candida group. No significant difference was observed in APACHE II scores between the groups (median 28 vs. 28, p = 0.23). However, the Candida auris group had a significantly lower SOFA score at admission (median 7 vs. 8, p = 0.02). There were no statistically significant differences in NUTRIC scores, incidence of organ failure, or prevalence of comorbidities such as diabetes, hypertension, CKD, and COPD. Table 1 presents a detailed comparison of baseline characteristics.

Fig. 1.

Flow diagram for patient enrolment in the study. Out of 5,376 ICU admissions, 2,028 patients met eligibility criteria. Of these, 267 had Candida-positive blood cultures (38 Candida auris, 229 non-auris Candida), while 1,761 had negative cultures

Table 1.

Baseline Characteristics of Patients with Candida auris and Non-auris Candida

	Candida auris	Non auris Candida	p Value
N (%)	38 (14.3)	229(85.7)
Age (years)-Median (IQR)	60 (38-75)	53 (35-72)	0.08
Gender (Male)- N (%)	22 (58.81%)	137 (59.92%)	0.82
Comorbidities -N(%)
DM	06 (15.78)	69 (30.13)	0.06
HTN	10 (26.31)	91 (39.73)	0.11
CKD	12 (31.57)	59 (25.76)	0.45
COPD	06 (15.78)	43 (18.77)	0.65
Diagnosis- N(%)
Respiratory	12 (31.50%)	91 (39.73%)	0.54
Neurological	07 (18.44%)	57 (24.83%)
CVS	02 (5.25%)	07 (3.01%)
Post operative	04 (10.52%)	18 (7.82%)
Others	13 (34.22%)	56 (24.41%)
Mechanical Ventilation- N(%)	28 (73.71%)	162 (70.70%)	0.71
CVC- N(%)	30 (78.90%)	183 (79.93%)	0.89
Arterial Catheterisation- N(%)	20 (52.63%)	129 (56.33%)	0.67
APACHE II -Median (IQR)	28 (20-32)	28 (18-29)	0.23
SOFA -Median (IQR)	07 (05-10)	08 (05-12)	0.02
NUTRIC Score -Median (IQR)	05 (03-06)	05 (03-07)	0.34
Prior Antifungal use- N(%)	10 (26.35%)	43 (18.76%)	0.28
Prior Antibiotic use- N(%)	8 (21.11)	48 (21.01)	1.00
Corticosteroid use- N(%)	7 (18.44)	39 (17.02)	0.82
Immunosuppressive therapy- N(%)	7 (18.42)	41 (17.91)	0.94
Malignancy- N(%)	6 (15.84)	37 (16.28)	0.95
HIV Infection- N(%)	1 (2.60)	6 (2.62)	1.00
Solid organ transplant- N(%)	2 (5.31)	14 (6.13)	0.84
Cardiac Comorbidity- N(%)	4 (10.54)	25 (10.91)	0.94
GI Comorbidity- N(%)	4 (10.55)	22 (9.63)	0.84
Neurological Comorbidity- N(%)	5 (13.22)	28 (12.21)	0.86

Categorical variables are expressed as counts with percentages, and continuous variables are presented as medians with IQR. Statistically significant p-values (≤ 0.05) are highlighted in bold

APACHE II – Acute Physiology and Chronic Health Evaluation II; CVC – Central Venous Catheter; CVS: Cardiovascular system; CKD – Chronic Kidney Disease; COPD – Chronic Obstructive Pulmonary Disease; DM – Diabetes Mellitus; GI-Gastro-intestinal; HIV- human immunodeficiency virus; HTN – Hypertension; IQR – Interquartile Range; NUTRIC-Nutrition Risk in the Critically Ill; SOFA – Sequential Organ Failure Assessment

Outcomes

The median ICU length of stay was slightly longer for the Candida auris group (17 days) compared to the non-auris Candida group (11 days). However, this difference was not statistically significant (p = 0.11). Crude mortality was recorded in 12 of 38 patients (31.62%) in the Candida auris group and 79 of 229 patients (34.51%) in the non-auris Candida group. This difference was not statistically significant (Fisher’s exact p = 0.854), with an odds ratio of 0.88 (95% CI: 0.42–1.83). (Fig. 2) Kaplan–Meier survival analysis demonstrated no statistically significant difference in survival probability in patients with Candida auris vs. non auris Candidemia (log-rank test χ² = 3.28, p = 0.07). In the multivariate Cox proportional hazards model adjusted for age and SOFA score, neither age (HR 0.99; 95% CI: 0.98–1.00; p = 0.16) nor SOFA score (HR 0.99; 95% CI: 0.94–1.05; p = 0.74) were significant predictors of mortality. (Fig. 3).

Fig. 2.

Comparison of mortality rates between Candida auris and non-auris Candida: Bar chart displaying crude mortality rates with 95% confidence intervals. No statistically significant difference in mortality was observed between the Candida auris and non-auris Candida groups

Fig. 3.

Kaplan–Meier survival analysis of ICU patients with candidemia. Kaplan–Meier survival curves comparing ICU patients with Candida auris and non-auris Candida candidemia. The difference in survival probability was not statistically significant (log-rank test χ² = 3.28, p = 0.07)

Invasive interventions

There was no significant difference in the use of invasive interventions such as mechanical ventilation, CVC or arterial catheter placement between the two groups. Mechanical ventilation was required in 73.71% of Candida auris patients versus 70.70% of non-auris Candida patients (p = 0.71). CVC were used in 78.90% of Candida auris and 79.93% of non-auris Candida cases (p = 0.89), while arterial catheterisation was done in 52.63% of Candida auris and 56.33% of non-auris Candida cases (p = 0.67). (Table 1).

Multivariate analysis of risk factors

Logistic regression analysis showed that increasing age was independently associated with higher odds of Candida auris infection (OR = 1.10; 95% CI: 1.04–1.16; p = 0.0003). In contrast, female sex was also a significant independent predictor (OR = 16.35; 95% CI: 1.83–146.2; p = 0.012). None of the invasive interventions (mechanical ventilation, CVC or arterial catheterisation) or severity scores (APACHE II, SOFA) emerged as significant predictors in the multivariate model (Table 2).

Table 2.

Logistic regression for risk of Candida auris infection

Variable	OR	Standard Error	p-value	95% CI (Lower)	95% CI (Upper)
Intercept	0.0	3.68	0.01	0.0	0.13
Female	16.35	1.12	0.01	1.83	146.22
Age	1.10	0.03	0.003	1.04	1.16
APACHE II	1.01	0.07	0.84	0.88	1.17
SOFA SCORE	0.54	0.15	0.63	0.4	1.72
Organ failure	0.38	0.43	0.42	0.16	1.87
NUTRIC score	2.63	0.30	0.51	0.85	4.77
DM	4.97	0.99	0.62	0.96	129.54
HTN	1.67	0.72	0.47	0.41	6.91
CKD	0.29	0.72	0.08	0.07	1.19
COPD	0.2	0.84	0.05	0.04	1.04
ICU LOS	1.04	0.02	0.11	0.99	1.08
CVC	0.4	0.68	0.18	0.1	1.54
Arterial catheterisation	0.79	0.67	0.72	0.21	2.93
Mechanical ventilation	2.68	0.65	0.13	0.74	9.69

Multivariate logistic regression analysis identifying risk factors independently associated with Candida auris bloodstream infection in ICU patients. Odds ratios, standard errors, p-values, and 95% confidence intervals are shown for each variable included in the final model. Statistically significant p-values (≤ 0.05) are highlighted in bold

APACHE II – Acute Physiology and Chronic Health Evaluation II; CVC – Central Venous Catheter; CKD – Chronic Kidney Disease; COPD – Chronic Obstructive Pulmonary Disease; CI – Confidence Interval; DM – Diabetes Mellitus; ICU LOS – Intensive Care Unit length of stay; HTN – Hypertension; NUTRIC-Nutrition Risk in the Critically Ill; OR – Odds Ratio; SOFA – Sequential Organ Failure Assessment

Distribution of candida species and antifungal susceptibility

A total of 267 Candida isolates were identified, with Candida albicans being the most prevalent species (31.8%), followed by Candida tropicalis (21.0%), Candida parapsilosis (15.0%), and Candida auris (14.2%). Notably, Candida auris exhibited high resistance to fluconazole (94.7%) and amphotericin B (39.5%), with 7.9% resistance to echinocandins. Fluconazole resistance was also observed in Candida glabrata (50.0%), Candida tropicalis (34.1%), and Candida parapsilosis (27.5%), while amphotericin B resistance remained low across most species. Echinocandin resistance was not detected in any species except Candida auris (Table 3).

Table 3.

Distribution of Candida species and antifungal susceptibility

Species	Number of Isolates	Percentage (%)	Fluconazole Resistance (%)	Amphotericin B Resistance (%)	Caspofungin Resistance (%)	Micafungin Resistance (%)	Anidulafungin Resistance (%)
Candida albicans	85	31.8	10.5	1.2	0.0	0.0	0.0
Candida parapsilosis	40	15.0	27.5	0.0	0.0	0.0	0.0
Candida auris	38	14.2	94.7	39.5	6.0	7.0	6.5
Candida tropicalis	56	21.0	34.1	3.6	0.0	0.0	0.0
Candida glabrata	12	4.5	50.0	8.3	3.0	4.0	3.5
Candida guilliermondii	12	4.5	10.0	5.0	0.0	0.0	0.0
Candida krusei	4	1.5	100.0 (intrinsic)	0.0	1.0	1.0	1.0
Candida dubliniensis	20	7.5	10.0	2.0	0.0	0.0	0.0

Discussion

In this four-year retrospective cohort study of 267 Candida species causing bloodstream. infections, we identified a 14.23% incidence of candidemia caused by Candida auris. Patients with Candida auris infection had a significantly lower SOFA score at admission compared to those without infection, but no significant differences were observed in APACHE II scores, comorbidities, or invasive interventions. The crude mortality rate in the Candida auris group (31.6%) was comparable to that of the non-auris Candida group (34.5%), and no significant difference in ICU survival was found. Multivariate analysis revealed that older age and female sex were independently associated with increased risk of Candida auris infection.

Multiple Indian studies have reported a growing burden of Candida auris in ICU settings. Our findings align with Indian and global studies including in the United Kingdom, Greece, South America, and the Middle East documenting the rise of Candida auris as a significant ICU pathogen. For instance, in a multicenter study from New Delhi, Candida auris accounted for 42% of candidemia cases, with high levels of antifungal resistance [6, 7, 17]. The observed incidence of 14.23% aligns with findings reported in several studies [7, 10, 25]. Similarly, the mortality rate of 31.6% in our study falls within the 30–60% range documented in the literature [5, 6, 26]. The high mortality is largely attributed to delayed identification and the organism’s resistance to antifungal agents. Antifungal resistance remains a cornerstone challenge in managing Candida auris. Over 90% of isolates globally are resistant to fluconazole, and resistance to amphotericin B and echinocandins has also been increasingly reported. In our study, Candida auris isolates showed high levels of fluconazole resistance, reinforcing concerns about limited therapeutic options and the need for empirical therapy tailored to local susceptibility patterns.

The finding that older age was a significant independent predictor of Candida auris infection aligns with existing literature and may reflect cumulative exposure to healthcare settings, invasive procedures, or immunosenescence [5, 27]. The association with female sex is novel and requires further exploration; it may reflect biological susceptibility, differential care pathways, or chance, given the relatively small number of Candida auris cases. Contrary to previous literature, invasive interventions such as CVC use, arterial catheterisation, mechanical ventilation, were not significantly associated with Candida auris infection in our cohort [7, 26, 28, 29]. This may be due to the uniformly high usage of these devices across critically ill patients in both groups, limiting their discriminatory value in multivariate models. A likely explanation lies in the high and relatively uniform utilisation of these interventions across all critically ill patients in our ICU, irrespective of the infecting Candida species. The vast majority of patients, both in the Candida auris and non-auris Candida groups, required invasive monitoring and life-support measures, reducing the discriminatory power of these variables in multivariate models. Other commonly reported risk factors—such as prior use of antibiotics or corticosteroids, immunosuppressive therapy, underlying malignancies, and pre-existing cardiac, gastrointestinal, or neurological comorbidities—were not significantly associated with Candida auris infection in our study. This observation underscores the importance of considering local epidemiologic context when interpreting risk factor data.

The findings of this study carry several important implications for ICU clinical practice. The identification of specific risk factors—such as older age and female sex—for Candida auris infection supports the implementation of targeted screening protocols in high-risk patient populations, particularly in endemic settings. The high prevalence of multidrug resistance in Candida auris particularly to fluconazole and amphotericin B, underscores the need to reinforce antifungal stewardship programs. Empirical therapy in suspected candidemia cases should be guided by local resistance patterns, and de-escalation strategies should be implemented once susceptibility data become available.

This study also highlights the need for future prospective, multicenter cohort studies to explore host-pathogen interactions, transmission dynamics, and the impact of early antifungal therapy on clinical outcomes. Integration of genomic surveillance and colonization screening could further clarify the evolution of resistance and transmission routes.

Given the high prevalence and antifungal resistance of Candida auris in ICU settings, targeted mitigation strategies are essential. Rigorous infection prevention and control measures—such as contact precautions, dedicated equipment, and environmental decontamination using chlorine-based disinfectants or hydrogen peroxide vapor—are critical to limiting nosocomial transmission. Routine colonization screening in high-risk patients (e.g., those with prolonged ICU stays or prior antifungal exposure) can aid in early detection and cohorting. Strengthening antifungal stewardship programs is also crucial, including restricting empiric use of broad-spectrum antifungals, timely de-escalation based on susceptibility profiles, and education of ICU staff regarding local resistance trends. Furthermore, real-time microbiological reporting and multidisciplinary rounds can facilitate prompt and appropriate therapy.

This study has several limitations that should be acknowledged. First, the retrospective design inherently limits the ability to establish causal relationships and may be influenced by residual confounding.To minimize this, we applied strict inclusion criteria and used multivariate logistic regression to adjust for known clinical variables. Second, important data such as prior colonization status, and time to antifungal therapy were unavailable, potentially affecting outcome analysis. However, consistently recorded clinical indicators and severity scores were analyzed to provide robust risk assessment. Third, dynamic changes in severity scores during ICU stay were not captured; instead, baseline APACHE II and SOFA scores were uniformly applied. Finally, the modest number of Candida auris cases may limit statistical power for detecting rare associations, though a four-year study period ensured adequate sample accrual. While generalizability is limited, standardized microbiological methods and comparative analysis strengthen the clinical relevance of findings in an endemic, resource-limited setting. Despite these limitations, the study provides valuable real-world data on the epidemiology, resistance patterns, and outcomes of Candida auris infection in the ICU, and highlights the need for ongoing surveillance and risk-based screening in endemic regions.

Conclusion

This study underscores the escalating burden of Candida auris within ICU environments, marked by a notable incidence rate and no additional mortality when compared to non-auris Candida species. Advancing age and female gender have been identified as significant independent predictors, highlighting the imperative for ongoing surveillance and patient-specific risk assessment in regions where Candida auris may be endemic.

Abbreviations

APACHE II

Acute Physiology and Chronic Health Evaluation II

CVC

Central Venous Catheter

CKD

Chronic Kidney Disease

COPD

Chronic Obstructive Pulmonary Disease

CI

Confidence Interval

CDC

Centres for disease control and prevention

DM

Diabetes Mellitus

ICU

Intensive Care Unit

HTN

Hypertension

IQR

Interquartile Range

MALDI

TOF–matrix–assisted laser desorption/ionization–time of flight

NUTRIC

Nutrition Risk in the Critically Ill

OR

Odds Ratio

SOFA

Sequential Organ Failure Assessment

TPN

total parenteral nutrition

Author contributions

Satyajit Choudhury: methodology; manuscript writingKalpana Majhi: Data acquisition; analysis of data; manuscript writingPragati Jena: data analysis; review and editing manuscriptSritam Mohanty: methodology; data acquisition; manuscript writingYallawar Nipun Narayan: data analysis; review and editing manuscriptPragyan Paramita Jena: data analysis; methodology; drafting & editing manuscriptShakti Bedanta Mishra: Conceptualization; data analysis; editing manuscript; analysis of dataSagarika Panda: Conceptualisation; manuscript writing; methodology.

Funding

Open access funding provided by Siksha 'O' Anusandhan (Deemed To Be University)

Data availability

All data generated or analysed during this study are included in this published article.

Declarations

Ethics approval and consent to participate

This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Institutional Ethics Committee (IEC/IMS.SH/SOA/2024/860). Informed consent was waived due to the retrospective nature of the study.

Consent for publication 

Not applicable.

Competing interests

The authors declare no competing interests.