Abstract
Background/Aim: Candidemia is a life-threatening fungal infection associated with high mortality and significant morbidity, particularly among older adults. Few studies have addressed mortality in this population. This study aimed to identify the risk factors for mortality in older patients with candidemia.
Patients and Methods: We conducted a retrospective study to identify the risk factors for mortality in older patients with candidemia using the Medical Information Mart for Intensive Care-IV database. The clinical characteristics and risk factors for mortality in patients younger or older than 65 years of age were determined. We also evaluated the effects of early antifungal therapy on outcomes in these patients.
Results: Among patients with candidemia, advanced age, congestive heart failure, cardiac arrhythmia, smoking habits, leukocytosis, and thrombocytopenia were significantly associated with in-hospital mortality. The early administration of antifungal therapy was related to improved survival in all age groups. Among older individuals who died during hospitalization, a higher prevalence of congestive heart failure, pulmonary vascular diseases, and smoking habit was observed. In contrast, the younger patients who died had a higher incidence of underlying metastatic cancer, coagulopathy, and cardiac arrhythmia.
Conclusion: Older patients with candidemia had higher in-hospital mortality rates than those of younger patients. Thrombocytopenia was found to be an independent risk factor, and early fungal therapy was associated with lower in-hospital mortality. Early identification of patients with candidemia and appropriate antifungal therapy are important for its treatment.
Introduction
Candida spp. are leading pathogens that cause infections in intensive care units (ICU). Candidemia is one of the most common nosocomial bloodstream infections worldwide. The crude incidence of candidemia in the US is estimated to be 8.7 per 100,000, accounting for substantial clinical and economic burdens (1). Candidemia results in increased mortality rates, extended lengths of hospital stay, and significantly increased healthcare expenditure (2, 3).
The mortality rate of candidemia is approximately 40%-75% in the general population (4-8). The incidence (3) and mortality rate are reportedly higher in older patients with candidemia (8-10). Age was identified as an independent predictor significantly associated with increased 28-day mortality among patients with candidemia in a recent study (11). Owing to the global shift toward aging populations, candidemia among older adults has become a critical issue. To date, few studies have elucidated the risk factors of mortality in older patients with candidemia.
We performed a retrospective study of adults with candidemia, utilizing data sourced from the Medical Information Mart for Intensive Care (MIMIC)-IV database. We conducted a comparative analysis of in-hospital mortality between cohorts based on age and receipt of antifungal therapy. The risk factors for in-hospital mortality in older patients were further evaluated.
Patients and Methods
Study population. This retrospective analysis used data collected from the latest version of the MIMIC-IV open-source clinical database (version 2.0, released on June 12, 2022) focused on patients diagnosed with candidemia. MIMIC-IV is a database comprising critical care patient data from the Beth Israel Deaconess Medical Center, including data from patients admitted to either the emergency department or ICU between 2008 and 2019 (12). The dataset comprises 431,231 hospital admission records and 73,181 ICU admission records, including 180,733 unique patients recorded for hospital admissions and 50,920 unique patients with ICU admission records (12).
The eligibility criteria for the study were as follows: 1) adults aged 18 years or older; and 2) patients with at least one positive blood culture for Candida species. Among the patients with two or more positive blood cultures, only the first was included. An episode of candidemia was defined as the isolation of Candida species from blood culture on one or more occasions. The onset of candidemia was defined as the day on which a blood culture first yielded Candida. We excluded participants with incomplete medical records and those under the age of 18. This approach ensured the specificity of the study cohort to individuals with a confirmed diagnosis of candidemia, based on standardized microbiological methods.
The study protocol was approved by the Institutional Review Board (IRB) of the Tri-Service General Hospital (approval number: TSGH: IRB A202405136).
Data collection. The following data were extracted: clinical demographics, comorbidities, use of indwelling catheters, laboratory results, microbial species, receipt of antifungal therapy, length of hospitalization, and outcomes. Comorbidities were identified based on the International Classification of Diseases, 10th Revision (ICD-10) codes, allowing for precise and standardized categorization of underlying health conditions. Patients aged > 65 years were classified as the older patient group, while those aged ≤ 65 years were considered the younger group. Clinical characteristics and laboratory data were extracted from the database and included for further analyses. Leukocytosis and leukopenia were defined as a white blood cell count higher than 12,000/μl and lower than 4,000/μl, respectively. Early antifungal treatment was defined as the administration of antifungal agents within 48 h of the collection of blood cultures that subsequently yielded Candida.
Statistical analyses. Data were analyzed using the PASW statistical package for Windows version 26 (SPSS, Chicago, IL, USA). The χ2 test with Fisher’s exact test was employed to assess categorical differences. Continuous variables were evaluated using the Student’s t-test, with results reported as median and interquartile range (IQR). Time to mortality, defined as the period from the onset of fungemia to death, was analyzed using Kaplan-Meier survival analysis, while the log-rank test was applied to compare univariate survival distributions among different patient cohorts. A logistic regression model was used to explore independent prognostic factors associated with in-hospital mortality. Univariate analyses were performed for each risk factor to determine odds ratios (OR) and 95% confidence intervals (CIs). All biologically plausible variables (p-value less than 0.10) in the univariate analysis were included in the multivariate logistic regression model using a backward selection approach. A p-value of less than 0.05 was considered statistically significant. We compared the characteristics between younger and older patients. The risk factors for in-hospital mortality in older patients were evaluated.
Results
In total, 402 patients with candidemia were included in this study, of whom 298 (74.1%) were admitted to the intensive care unit (ICU). Candida albicans (158, 39.3%) and Candida glabrata (130, 32.3%) were the two leading species isolated, affecting more than 70% of the patients in this study (Figure 1). A total of 240 patients (59.7%) received early antifungal therapy. The most frequently used antifungal agent was fluconazole, which was administered to 232 patients (57.7%). Of the 58 patients (14.4%) who did not receive antifungal treatment for candidemia, 49 died. Among these 49 patients, 45 deaths occurred before or on the same day that the positive blood culture results were reported.
Pie chart illustrating the relative distribution of various Candida species isolated from the patients enrolled in this study.
A total of 260 patients (64.7%) survived, and 142 patients (35.3%) died during hospitalization. The demographic and clinical characteristics of the two groups are presented in Table I. Patients who died were older (67 vs. 60, p=0.004) and more likely to have congestive heart failure (34.5 vs. 18.5, p=0.001), cardiac arrhythmia (48.6 vs. 36.9, p=0.026), coagulopathy (47.2 vs. 29.2, p<0.001), a smoking habit (35.9 vs. 23.8, p=0.011), dialysis (16.9 vs. 8.5, p=0.014), leukocytosis (63.4 vs. 44.2, p<0.001), and thrombocytopenia (50.7 vs. 13.5, p<0.001). Conversely, survivors were more likely to receive early antifungal therapy (73.8 vs. 33.8, p<0.001) and were more often treated with fluconazole (71.2 vs. 33.1, p<0.001) and micafungin (78.8 vs. 61.3, p<0.001). As listed in Table II, multivariate analysis showed that age (OR=1.022; 95%CI=1.004-1.040; p=0.016), congestive heart failure (OR=2.103; 95%CI=1.146-3.859; p=0.016), a smoking habit (OR=1.861; 95%CI=1.053-3.289; p=0.032), leukocytosis (OR=2.494; 95%CI=1.427-4.358; p=0.001), and thrombocytopenia (OR=7.145; 95%CI=3.867-13.205; p<0.001) were independent risk factor for in-hospital mortality. Receiving early antifungal therapy (OR=0.144; 95%CI=0.082-0.253; p<0.001) was the only independent protective factor against in-hospital mortality.
Clinical characteristics of patients with candidemia who survived or died in hospital.
Logistic regression analysis of predictors for in-hospital mortality among patients with candidemia.
Although age was an independent risk factor for in-hospital mortality, we further divided the cohort into two groups according to age. We defined patients under 65 years as the younger group and those aged 65 and above as the older group. As shown in Table III, the younger and older groups comprised 227 (56.5%) and 175 (43.5%) patients, respectively. In the older group, patients were more likely to have comorbidities such as congestive heart failure, renal failure, cardiac arrhythmias, chronic pulmonary disease, type 2 diabetes mellitus, hypertension, hypothyroidism, and leukocytosis. In contrast, the younger group was more likely to experience alcohol abuse, liver disease, pancreatitis, obesity, drug abuse, depression, and leukopenia. Older patients had a significantly higher in-hospital mortality rate than the younger group patients (42.9 vs. 29.5, p=0.006).
Clinical characteristics and outcomes of patients with candidemia stratified by age.
To elucidate the association between age and mortality risk among patients with candidemia, we stratified the cohort into younger (≤65 years) and older (>65 years) age groups. Supplementary Table S1 presents the analysis of risk factors for in-hospital mortality among patients with candidemia aged 65 years or younger. As shown in Supplementary Table S2, multivariate analysis revealed that those who died during the hospitalization were more likely to have cardiac arrhythmia (OR=2.312; 95%CI=1.013-5.279; p=0.047), coagulopathy (OR=2.859; 95%CI=1.258-6.499; p=0.012), leukocytosis (OR=4.611; 95%CI=1.975-10.766; p<0.001), and thrombocytopenia (OR=10.704; 95%CI=4.431-25.858; p<0.001). In contrast, early antifungal therapy was a protective factor against in-hospital mortality (OR=0.134; 95%CI=0.060-0.368; p<0.001). A total of 49 patients died before receiving any antifungal treatment indicating that they had no opportunity for early intervention. We performed further analysis after excluding these 49 patients. The results showed that early antifungal therapy was still associated with improved survival among patients with candidemia (Supplementary Tables S3-S4).
Table IV presents the 175 patients in the older group. Of these, 100 patients (45%) survived, while 75 patients (55%) died during hospitalization. As shown in Table V, multivariate analysis showed that those who died during hospitalization were more likely to have congestive heart failure (OR=2.259; 95%CI=1.096-4.654; p=0.027), pulmonary vascular disease (OR=2.711; 95%CI=1.053-6.985; p=0.039), smoking habits (OR=2.284; 95%CI=1.036-5.032; p=0.041), and thrombocytopenia (OR=3.725; 95%CI=1.632-8.499; p=0.002). In contrast, early antifungal therapy was a protective factor against mortality during hospitalization (OR=0.176; 95%CI=0.084-0.368; p<0.001).
Comparison of hospital outcomes for patients aged >65 years.
Logistic regression analysis of predictors for in-hospital mortality among patients with candidemia aged >65 years.
We stratified the patients into four cohorts based on age and whether they received early antifungal therapy. Kaplan-Meier survival analysis and log-rank test were conducted (Figure 2). Early antifungal therapy was associated with lower in-hospital mortality in the older and younger patient cohorts. The older cohort had a higher 30-day mortality rate than the younger cohort, even in those who received early antifungal therapy. In patients who did not receive early antifungal therapy, the older cohort still had a borderline higher 30-day mortality (p=0.06, Figure 2 cohorts B vs. cohorts D).
Kaplan-Meier analysis of in-hospital 30-day mortality of patients with candidemia. The patients were stratified by age and the timing of antifungal therapy. Early antifungal therapy is associated with significantly lower mortality rates in both younger and older cohorts. Specifically, patients under 65 years old who received antifungal therapy have notably lower mortality rates. Cohort A: Age ≤65, Early antifungal therapy; Cohort B: Age ≤65, Late antifungal therapy; Cohort C: Age >65, Early antifungal therapy; Cohort D: Age >65, Late antifungal therapy; **p<0.01; ****p<0.0001.
Discussion
The results of this study revealed significant differences in the clinical characteristics and in-hospital mortality between patients with candidemia who were younger than 65 years and those older than 65 years. We found that the risk factors of in-hospital mortality differed between younger and older groups. Both groups shared thrombocytopenia as a risk factor, while early antifungal therapy was a protective factor against in-hospital mortality in both groups.
As shown in Figure 1, C. albicans accounted for 39.3% of all species isolated in this study. This finding is consistent with previous studies identifying C. albicans as the leading cause of candidemia; however, non-albicans Candida accounts for approximately two-thirds of all cases (13, 14). As the antimicrobial susceptibility of Candida species could not be obtained from the database, we could not determine the resistance patterns of the isolates in this study.
Several studies have investigated the risk factors for mortality in patients with candidemia. These studies have shown varying results owing to differences in study populations and patient care. However, both our study and several previous studies consistently identify old age as an independent risk factor for mortality in patients with candidemia (3, 10, 15-17). In this study, we found that older patients had significantly more comorbid conditions than younger patients (Table I), which may contribute to their higher mortality. This finding highlights the importance of considering both physiological and clinical considerations in geriatric patients.
A recent study has found that a higher Charlson comorbidity index value emerged as independent predictors of increased mortality (17). In this study, we identified several factors associated with higher in-hospital mortality in older patients with candidemia, including congestive heart failure, pulmonary vascular disorders, smoking habits, and thrombocytopenia. In contrast, early antifungal therapy was the only independent protective factor against in-hospital mortality in this population (Table V). This is consistent with previous study (18), highlighting the importance of early antifungal therapy. Patients with congestive heart failure who develop candidemia face significant clinical challenges as both conditions independently contribute to high mortality and morbidity (3). The presence of congestive heart failure in patients with candidemia poses significant challenges, with clinical outcomes influenced by various factors, such as the use of devices like left ventricular assist devices and the timing of the infection. A recent study also found that cardiovascular disease was an independent risk factors for mortality in adult patients with candidemia (19). These findings highlighted the importance of evaluation of cardiovascular function in those with candidemia. Managing candidemia in these patients requires a multidisciplinary approach to address the complex interplay between conditions and enhance patient outcomes. Further research is essential to develop targeted interventions that reduce the risks associated with these coexisting conditions.
Pulmonary vascular diseases encompass a broad and diverse spectrum of underlying pathologies including venous thromboembolism, congenital malformations, inflammatory vasculitis, and pulmonary artery hypertension (20). One study reported that patients with pulmonary artery hypertension and sepsis have higher mortality rates than those without pulmonary artery hypertension (21). Although the direct connection between candidemia and pulmonary vascular diseases is not well defined, the interaction between infections, metabolic disorders, and environmental factors offers a basis for understanding potential risk factors. These factors should be carefully considered in the clinical management of patients with candidemia and underlying pulmonary vascular diseases.
We found that the in-hospital mortality rate was higher in older individuals with a smoking habit (Table V). One study showed that smoke condensates increase the virulence of C. albicans (22). Another study pointed out that smoke condensate significantly enhances the secretion of candidal histolytic enzymes and increases adherence to denture surfaces, thereby potentially predisposing patients to oral candidiasis (23). These findings suggest that smoking habits may be a predisposing factor for Candida infection; however, the correlation between cigarette smoking and the risk of mortality in patients with candidemia remains unclear. Further research is required to clarify this issue.
Our results indicated that thrombocytopenia and lack of early antifungal therapy were risk factors for mortality in patients with candidemia regardless of their age. Thrombocytopenia has previously been found to be a risk factor for mortality in patients with candidemia (24-26). Thrombocytopenia, as an indicator of disseminated intravascular coagulation (27), is associated with poor outcomes in infection cases (28). This may explain why it is a risk factor for mortality in patients with candidemia.
Several studies have documented that the lack of adequate antifungal therapy is a risk factor for mortality in older patients with candidemia (15, 29-31). Both these studies and our findings indicate that inadequate antifungal therapy is the universal risk factor for mortality in older patients with candidemia. This finding highlights the importance of providing adequate antifungal therapy for these patients.
To elucidate the influence of age and early antifungal therapy on the survival of patients with candidemia, we stratified patients with candidemia into four cohorts according to age and receipt of early antifungal therapy. A significantly lower in-hospital mortality was observed in both older and younger patients who received early antifungal therapy. Among those receiving early antifungal therapy, older patients had higher mortality rates than younger patients. These findings emphasize that old age is an important risk factor for mortality, even in patients receiving early antifungal therapy for candidemia.
This retrospective study has some limitations, including variations in patient care and inconsistencies in patient backgrounds. Another key limitation was the unavailability of antimicrobial susceptibility tests for Candida species. Because of this limitation, we could not evaluate the appropriateness of antifungal therapy as well as whether the treatment was definitive or empirical. To address this limitation, early antifungal therapy was included in the analysis. Additionally, we did not include biomarkers and Candida score as an adjunct for diagnosis. Despite these limitations, a major strength of this study was the large number of patients with candidemia. Furthermore, we used multivariate analysis to delineate the clinical demographics that were independently associated with in-hospital mortality in patients with candidemia.
In conclusion, the risk factors for in-hospital mortality differed between patients with candidemia who are younger than or older than 65 years. Thrombocytopenia was an independent risk factor of in-hospital mortality in patients with candidemia. Early fungal therapy was associated with lower in-hospital mortality rates. Clinicians should be particularly vigilant with older patients who have candidemia owing to their increased tendency for high mortality. Early recognition of patients with candidemia and prompt antifungal therapy are important for disease management.
Acknowledgements
Not applicable.
Footnotes
Authors’ Contributions
All Authors contributed to the study conception and design. All the Authors have approved the final manuscript. Wei-Ping Chen: Writing of the original draft, Conceptualization, Data curation, and validation; Tsung-Ta Chiang: Formal analysis, Supervision, Resources; Bing-Hong Liu: Software, Investigation, Methodology, Data curation, Formal analysis; Hao-Wen Chang: Writing – review & editing; Shih-Ta Shang: Funding acquisition, Validation, Visualization, Project administration, Writing – review & editing; Yung-Chih Wang: Conceptualization, Funding acquisition, Writing – review & editing.
Supplementary Material
The data that support the findings of this study are openly available in figshare at https://figshare.com/s/530beee1d6573b5f3250
Conflicts of Interest
The Authors have no relevant financial or non-financial interests to disclose.
Funding
This work was supported by grants from the Tri-Service General Hospital (TSGH-E-114278), National Science and Technology Council (114-2410-H-016-005-MY2), and the Armed Forces Taoyuan General Hospital (TYAF-GH-E-113059). The funders had no role in the study design, data collection and analysis, decision to publish, or manuscript preparation.
Artificial Intelligence (AI) Disclosure
During the preparation of this manuscript, a large language model (ChatGPT, GPT-5; OpenAI, San Francisco, CA, USA) was used solely for language editing and stylistic improvements in select paragraphs. No sections involving the generation, analysis, or interpretation of research data were produced by generative AI. All scientific content was created and verified by the authors. Furthermore, no figures or visual data were generated or modified using generative AI or machine learning-based image enhancement tools.
- Received October 24, 2025.
- Revision received November 14, 2025.
- Accepted November 17, 2025.
- Copyright © 2026 The Author(s). Published by the International Institute of Anticancer Research.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
References
Jump to section
Related Articles
Cited By...
- No citing articles found.