Differences in Clinical Characteristics of AECOPD Patients with or without Candida Isolation from the Lower Respiratory Tract

Introduction

Chronic obstructive pulmonary disease (COPD) is a chronic inflammatory airway disease characterized by persistent airflow limitation. Acute exacerbation (AE) has always been regarded as a key indicator of COPD in clinical research. Bacteria and viruses are the most common infectious factors that induce AE, and bacterial colonization is an important cause of acute exacerbations of COPD (AECOPD), which is often overlooked.¹ In recent years, the effect of fungi derived from the lower respiratory tract (LRT) on COPD has attracted the attention of researchers. Studies have shown that abundant fungal biota exists in the induced sputum of COPD patients, with Aspergillus and Penicillium being associated with frequent AEs and high mortality.^2–4 Although the isolation rate of Candida from the LRT is much higher than that of other pathogenic fungi, it is often considered as contamination or simple colonization of Candida in the airway without clinical significance.^5,6 Therefore, there are few studies on the influence of positive Candida isolation from the LRT on COPD.

Candida is one of the most common opportunistic pathogenic fungi found in humans. It often exists on the surface of the LRT mucosa in the form of yeast; however, with changes in the environment, it can shift from a symbiotic to a pathogenic state. It can cause infection when mucosal or skin barrier function is disrupted, the number and function of neutrophils are impaired, cellular immunity is compromised, and metabolic dysfunction occurs. When the immune system functions properly, Candida typically coexists peacefully with the host as a colonizer, in contrast to yeast, mycelia have a stronger ability to invade and damage host cells.^7–9 At present, it is not clear whether the long-term presence of Candida in the LRT of COPD patients serves as a marker that potentially leads to AEs or affects disease severity. However, many studies have demonstrated that airway Candida colonization is associated with poor efficacy and prognosis in patients in intensive care units (ICUs).^10,11

A retrospective study of 16 hospitals in China over a 10-year period found that Candida pneumonia accounted for 11.4% of all fungal pneumonia cases, with COPD being the second most common comorbidity.¹² Generally, invasive candidiasis is secondary to colonization, and mucosal colonization is considered an independent risk factor for invasive candidiasis.¹³ Most patients experience varying durations and degrees of colonization before the onset of Candida infection, and multisite colonization is associated with a higher risk of Candida infections.¹³ It is believed that pathogen colonization in the LRT is an important factor in the pathogenesis of ventilator-associated pneumonia, and reducing bacterial colonization in the LRT is also an important prevention strategy for ventilator-associated pneumonia. Candida colonization serves as a precursor risk factor for Candida infection, which may be a risk factor and a predictive indicator for Candida infection in deep organs.^14–16

To further investigate the impact of positive Candida isolation in the LRT on AECOPD patients, we conducted a retrospective case-control study. This study compared the clinical characteristics of AECOPD patients with and without Candida isolation in the LRT and assessed the influence of Candida presence on AE events and long-term prognosis.

Materials and Methods

Study Design and Patient Population

This is a retrospective case-control study involving AECOPD patients hospitalized at the First Affiliated Hospital of Guangxi Medical University from January 1, 2018, to December 31, 2020. All patients or their immediate family members provided written informed consent. The study was approved by the Ethics Committee of the First Affiliated Hospital of Guangxi Medical University (no. 2022-KY-E-144, 2025-E1032). This study adheres to the Declaration of Helsinki.

The inclusion criteria were as follows: (1) Fulfillment of the COPD diagnostic criteria of the 2019 Global Initiative for Chronic Obstructive Pulmonary Disease (GOLD): patients exhibited symptoms such as recurrent cough, sputum production, shortness of breath, and pulmonary function tests indicating a forced expiratory volume in 1 s (FEV₁)/forced vital capacity (FVC) < 70% after inhalation of a bronchodilator, leading to a diagnosis of COPD. (2) Fulfillment of the diagnostic criteria for AE, including shortness of breath, cough, and/or sputum production with acute changes beyond normal daily variations and necessitating hospitalization.

The exclusion criteria were as follows: (1) age < 50 years; (2) hospitalization due to other systemic diseases; (3) no sputum production during hospitalization; (4) presence of oral candidiasis or oral leukoplakia; (5) isolation of fungi other than Candida from the LRT; (6) isolation of Candida from other specimens (such as blood, stool, or urine); (7) failure to obtain induced sputum or bronchoalveolar lavage fluid (BALF).

Population Groupings

In our study, LRT samples consist of induced sputum or BALF. According to the results of pathogenic detection in LRT samples from AECOPD patients during hospitalization, patients with at least one smear or culture positive for Candida were classified into the Candida-positive group; otherwise, they were classified into the Candida-negative group.

Data Collection and Follow-up

The data collected included demographic characteristics (e.g., age, sex, body mass index, duration of COPD, smoking status, COPD severity grading, complications and comorbidities) and clinical data (such as the frequency of AEs in the past year and laboratory examination results). Patients underwent follow-up assessments every 3 to 6 months. The follow-up period concluded on April 30, 2022, at which point the patients’ survival status was documented. The primary outcome of this study was AE events and the secondary outcome was death events.

Statistical Analysis

All data were statistically analyzed using SPSS 25 (IBM, Armonk, NY). For variables that were normally distributed, data were presented as mean ± standard deviation (SD), and comparisons between the Candida-positive and Candida-negative groups were performed using the independent sample t-test. For variables that did not follow a normal distribution, data were expressed as median with interquartile range (IQR), and between-group comparisons were conducted using the Mann–Whitney U- test. Categorical variables were expressed as percentages, and comparisons between groups were performed using the chi-squared test. The Kaplan-Meier method was used to generate survival curves for the two groups, and the Log rank test was used to compare survival rates between groups. Statistical significance was set at P < 0.05. Risk factors for AECOPD were identified using binary logistic regression analysis on the variables with statistically significant differences.

Results

Pathogen Isolation in the LRT

A flowchart illustrating the process is presented in Figure 1. Two hundred and twenty-five patients were enrolled in this study. According to the results of pathogen isolation from the LRT, the patients were divided into two groups: a Candida-positive group (n = 88) and a Candida-negative group (n = 137). The detected species included C. albicans (87.5%), C. glabrata (now called Nakaseomyces glabratus,¹⁷ 10.2%), C. tropicalis (6.8%), C. parapsilosis (2.3%), and C. krusei (now called Pichia kudriavzevii,¹⁷ 1.1%). Among patients in the Candida-positive group, 77 had bacterial isolates, including 69 cases of Gram-positive cocci, 1 case of Gram-negative cocci, 7 cases of Gram-positive bacilli, and 60 cases of Gram-negative bacilli. In the Candida-negative group, 114 had bacterial isolates, including 101 cases of Gram-positive cocci, 3 cases of Gram-negative cocci, 6 cases of Gram-positive bacilli, and 86 cases of Gram-negative bacilli. Although bacteria were detected in the LRT secretion smears of most patients, there was no significant difference in the positive rate of bacteria between the two groups (87.5% vs. 83.2%, P > 0.05). This suggests that the baseline characteristics of the two groups were comparable, including the positive detection rate of sputum microbial etiology. Viral testing for pathogens such as severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and influenza was negative in all included patients.

Figure 1 Candida detection chart of study population. AECOPD, acute exacerbation of chronic obstructive pulmonary disease. Candida glabrata has now been reclassified as Nakaseomyces glabrata, and Candida krusei as Pichia kudriavzevii.

Baseline Characteristics of Hospitalized AECOPD Patients

The mean age of the Candida-negative group was 69.18 ± 9.48 years, including 124 males and 13 females, and the mean age of the Candida-positive group was 71.40 ± 9.43 years, including 77 males and 11 females. There were no significant differences in sex, age, disease duration, body mass index, or complications between the two groups (P > 0.05). One hundred and ninety-five (86.7%) patients had a history of heavy smoking for a prolonged period, but the smoking index of the Candida-positive group was higher (54.35 ± 29.68 vs. 45.89 ± 28.93, P < 0.05). Furthermore, the Candida-positive group exhibited more severe symptoms, with higher COPD Assessment Test (CAT) scores and a higher proportion of patients with a modified Medical Research Council (mMRC) score of 4 and GOLD score of 4 (P < 0.05).

The percentage of FEV₁% (FEV₁% predicted, 41.94 ± 19.60 vs. 50.62 ± 22.88, P < 0.05) and the absolute FEV₁ values [1.18 (0.62, 1.61) L vs. 1.33 (0.73, 1.43) L, P < 0.05] in the Candida-positive group were significantly lower than those in the Candida-negative group. These findings suggest that Candida colonization may exacerbate pulmonary function impairment in COPD patients. We evaluated AEs in patients and found that more than half of the patients in both groups had experienced AEs in the past year, although this difference was not statistically significant. Notably, the proportion of patients in the Candida-positive group who required hospitalization following an AE (51/60, 85.0%) was higher than that in the Candida-negative group (51/82, 62.2%), suggesting that AEs were more severe in the Candida-positive group. Detailed data are presented in Table 1.

Table 1 Baseline Characteristics of Hospitalized AECOPD Patients with or without Candida Isolation

Laboratory Examination Results

An analysis of the initial routine blood examinations in the two groups, including white blood cell count (WBC), absolute value of neutrophils (NE), absolute value of eosinophils (EOS), and absolute value of lymphocytes (LYM), revealed that the LYM count in the Candida-positive group was significantly lower than that in the Candida-negative group (P < 0.05). Additionally, the peripheral blood CD8⁺ T-lymphocyte count in the Candida-positive group was lower (292.80 ± 185.70 cells/μL vs. 380.00 ± 236.10 cells/μL, P < 0.05). Although a similar trend was observed in the comparison of CD4⁺ T-lymphocyte counts between the two groups, the difference was not statistically significant. The data are presented in Table 2.

Table 2 Laboratory Examination Results in Hospitalized AECOPD Patients with or without Candida Isolation

Survival Analysis

Patients in both groups were followed up every 3 to 6 months. In the Candida-negative group, 19 patients (13.87%) died, and the survival rates at 12, 24, and 36 months were 93.4% (95% CI 89.28% - 97.52%), 91.0% (95% CI 86.1% - 95.9%), and 84.5% (95% CI 78.03% - 90.97%), respectively. Conversely, in the Candida-positive group, 33 patients (37.5%) died, and the corresponding survival rates at 12, 24, and 36 months were 80.7% (95% CI 72.47% - 88.93%), 70.1% (95% CI 60.49% −79.7%), and 59.1% (95% CI 47.93% - 70.27%), respectively. The survival rate in the Candida-positive group was significantly lower than that in the Candida-negative group (P < 0.001, HR 3.257, 95% CI 1.850–5.735; Figure 2).

Figure 2 Survival curve of hospitalized patients with acute exacerbation of chronic obstructive pulmonary disease with or without Candida isolation.

Positive Candida Isolation in the LRT and mMRC-4 as Independent Risk Factors for AEs and Hospitalization in COPD Patients in the Past year

AE has consistently been recognized as a key indicator of COPD in clinical research. Logistic regression analysis identified the presence of Candida in the LRT and mMRC 4 as independent risk factors for AE events in the past year [odds ratio (OR) = 1.949, P = 0.036; OR = 2.893, P = 0.012] and for COPD patients with AEs requiring hospitalization in the past year (OR = 1.931, P = 0.022; OR = 2.977, P = 0.001). These findings indicate that the positive Candida isolation in the LRT is not only associated with AEs but also serves as a risk factor for both AEs and hospitalization. The relevant data are shown in Table 3 and Table 4.

Table 3 Univariable Logistic Regression Analysis of Predictors for AE Events and Hospitalized AECOPD in the Past year

Table 4 Multivariate Logistic Regression for Predictors of AE Events and Hospitalized AECOPD in the Past year

Discussion

Previous clinical studies have demonstrated that more than half of hospitalized AECOPD patients were positive for Candida isolates from the LRT.^6,18 In the present study, 88 cases (39.1%) exhibited Candida presence in LRT specimens, with C. albicans accounting for the majority at 87.5%, a proportion significantly higher than that of other Candida species. The symptoms in the Candida-positive group were more severe than those in the Candida-negative group, and the CAT score and mMRC-4 grade ratio in the former were significantly higher than those in the latter. These findings are consistent with previous multicenter, large-sample case-control studies.¹⁸

The results also showed that airflow limitation was more severe in the Candida-positive group, as evidenced by a lower FEV₁% predicted level and a higher proportion of patients classified as GOLD-4 grade. Furthermore, follow-up data revealed that the survival rate in the Candida-positive group was significantly lower than that in the Candida-negative group. Consequently, the presence of Candida in the airway may be associated with exacerbation of symptoms, increased disease severity, and poor prognosis in patients with COPD. The anatomical characteristics of the LRT are different from those of the skin and mucous membranes. Colonization of the airway mucosa by Candida can lead to increased secretion and bronchial obstruction, resulting in the corresponding clinical manifestations or aggravation of the existing symptoms. Furthermore, C. albicans is a prevalent fungus responsible for allergic bronchopulmonary mycosis in addition to Aspergillus, and can cause allergic reactions or asthma-like symptoms in COPD.^19,20

AE constitute a critical event in the progression of COPD. We observed a significant trend of hospitalization after AEs in the Candida-positive group over the past year. Further analysis identified positive Candida isolation and mMRC-4 grades as two important independent risk factors for AEs and severe conditions requiring hospitalization after AEs in the past year. In addition to the previously mentioned factors contributing to aggravating symptoms and airflow limitation, interactions among mixed pathogens in the airway are also major causes that cannot be ignored. Although the role of Candida colonization in the progression of bacterial pneumonia remains controversial, the prevalence of mixed colonization or infection is notably high among patients with COPD. Our research showed a significant prevalence of concurrent detection of Candida and other bacterial pathogens in these patients. Candida, bacteria, and other pathogens exist in the airway for long period can influence each other through direct contact or indirect regulation, including enhanced physical contact or adhesion, mixed biofilm formation, the effects of quorum sensing molecules or metabolites, changes in virulence, alterations to the microenvironment, affecting host immunity, and even modification of disease outcomes.^21–24 For example, Staphylococcus aureus demonstrates significantly increased antimicrobial tolerance when exposed to exogenously supplemented farnesol or farnesol secreted by C. albicans in biofilms.²⁵ Moreover, C. albicans can augment the secretion of α and δ toxins by enhancing the accessory gene regulator quorum-sensing system of S. aureus, resulting in synergistic lethality during pathogen co-infection (whereas single microbial infection is not fatal).²⁶ Lactic acid produced by bacteria can inhibit the transition of C. albicans from yeast to mycelium and mask the key pathogen-related molecular pattern (β-glucan) on the surface of C. albicans cells, both of which are strategies to evade host immune detection.^27,28 Such interactions may contribute to the frequent AEs and disease progression in patients with airway Candida colonization. In our study, there were no significant differences between the two groups in either the isolation rate or the distribution of bacterial types, suggesting that bacterial co-detection was unlikely to be a major factor contributing to the differences in clinical outcomes between the two groups.

The use of antifungal drugs to eliminate Candida from airway colonization remains controversial, as the benefits of antifungal therapy for patients, including critically ill individuals, remain unclear.^29–31 Nonetheless, a previous study has demonstrated a significantly reduction in the incidence of ventilator-associated pneumonia caused by Pseudomonas aeruginosa after antifungal therapy in patients with airway C. albicans colonization.³² Furthermore, animal studies have shown that airway colonization by Candida in rats is associated with a marked increase in the incidence of bacterial pneumonia, which subsequently decreases following intervention with fluconazole or amphotericin B.^33,34 In COPD patients who have been using inhaled corticosteroids for a long time, C. albicans isolated from LRT specimens does not provide a definitive distinction between colonization and infection due to the absence of pathological evidence. Consequently, it remains unclear whether the elimination of Candida from the airway contributes to a reduction in the frequency of AEs, alleviating disease symptoms, or delaying the progression of COPD. This issue requires closer monitoring, comprehensive evaluation, and individualized assessment to determine the necessity of antifungal treatment.

The decline or disorder of immune function in COPD can impair respiratory defense mechanisms, which may lead to the colonization or infection by pathogens and accelerate disease progression. Studies have shown that the counts of CD4⁺ and CD8⁺ T-cells in the peripheral blood of AECOPD patients are lower than those in the stable phase.³⁵ In this study, we observed that the CD8⁺ T cell count in the Candida-positive group was significantly lower than that in the Candida-negative group, with a similar trend observed for CD4⁺ T-cell counts between the two groups. This observation raises the question of whether the presence of Candida in the airway exacerbates the degree of impaired cellular immunity, immune imbalance, or inflammatory responses. Interestingly, antifungal therapy in patients with hospital-acquired bacterial pneumonia who also exhibit Candida airway colonization showed decreased levels of inflammatory markers (IL-6, tumor necrosis factor-α, and interferon-γ) and an increased number of CD4⁺ T lymphocytes. This observation suggests that airway colonization by C. albicans may affect the immune response in COPD patients, which requires further investigation.³⁶

This study had several limitations. First, it was a retrospective study based on electronic medical records and was conducted at a single center, and subjective bias could not be completely avoided during the research process. Second, retrospective studies are limited in their ability to establish causality and control for confounding factors, which need to be further addressed by prospective studies. Finally, the study classified participants into different groups based on the results of Candida smear or culture of LRT specimens (mainly sputum); however, it was unable to clearly differentiate between colonization and infection, as pathological evidence was lacking.

Despite its limitations, this study reaches a convincing conclusion that airway Candida colonization should no longer be viewed as a simple bystander in COPD and may affect the degree of airflow limitation, the risk of AE, and long-term prognosis. Attention should be given to whether antifungal treatment could benefit AECOPD patients with Candida airway colonization.

Conclusions

Candida spp. were detected in the LRT secretions of more than one-third of hospitalized AECOPD patients. Candida isolation and an mMRC grade of 4 are independent risk factors for AECOPD. Candida colonization of the LRT may predict more severe clinical symptoms, greater airflow limitation, and a poorer prognosis in patients with AECOPD.