Opportunistic pulmonary mycoses significantly increase morbidity and mortality among immunocompromised patients, resulting in over 1.6 million deaths annually from approximately 300 million cases. The global prevalence of this condition remains unclear (1). Factors, such as human immunodeficiency virus (HIV) and chronic diseases, including pulmonary tuberculosis (PTB), lung cancer and respiratory conditions related to smoking or pollution, have contributed to rising caseloads in low and middle income countries (2). In Indonesia, there is a lack of epidemiological data on pulmonary mycosis. Wahyuningsih et al. (1) estimated that it affects 2.89% of the population, equating to about 7.7 million people each year. Furthermore, Rozaliyani et al. (3) discovered fungal cultures in 68.9% of patients with multidrug-resistant tuberculosis (MDR-TB) and 44.5% of those with persistent asthma. Factors, such as chemotherapy, corticosteroids, HIV infection and prolonged antibiotic use, can lead to pulmonary mycosis (4, 5).
Pulmonary mycoses associated with thoracic cavity malignancies, including lung cancer, are often missed due to their similar clinical symptoms and non-specific radiological findings. About 56 of 3.430 lung cancer patients develop chronic pulmonary aspergillosis (CPA) and 2.6% develop invasive pulmonary aspergillosis (IPA) (6–8). Candida sp. and Aspergillus sp. were the most common fungal isolates in Indonesian lung cancer patients (3, 9). Sputum specimens are still used to diagnose pulmonary mycosis despite their low sensitivity (45%–56%) (10). Arifin Achmad General Hospital, Pekanbaru, found Aspergillus sp. in 6.8% and Candida sp. in 6.3% of 2.652 inpatients’ sputum (11).
Fungal cultures can be improved by using bronchoscopy-derived specimens, such as bronchial washing, brushing, bronchoalveolar lavage (BAL) and tissue, especially in thoracic malignancy. Studies indicate that tissue specimens detect fungal pathogens more accurately than sputum (12, 13). This examination method is rarely used in Indonesia, especially in Riau Province, despite its potential. With this background, this study aims to analyse the positivity rate of fungal cultures from sputum, bronchial washing and tissue specimens in cases of thoracic cavity malignancy in Riau Province.
This cross-sectional observational study assessed the positivity rate of fungal cultures obtained from sputum, bronchial washing and tissue samples in patients with thoracic cavity malignancies in Riau Province. The Microbiology Laboratory at the Faculty of Medicine, Riau University, analysed specimens collected during bronchoscopy procedures conducted at Arifin Achmad General Hospital in Riau Province. The study, scheduled to run from August 2024 to January 2025, received approval from the Ethics Committee of the Faculty of Medicine, Riau University, under Ethical Clearance Number B/III/UN19.5.1.1.8.UEPKK/2024.
This study involved patients with thoracic cavity malignancies who had clinical indications and were eligible for bronchoscopy. A total of 57 patients who met the inclusion and exclusion criteria were sampled consecutively, without stratification or randomisation. The study included adult patients who were clinically and radiologically diagnosed with thoracic cavity malignancy and had indications for bronchoscopy. Exclusions were made for patients with pulmonary mycosis, those who were taking antifungal medications and those who did not have sputum, bronchial washing, or tissue samples available.
Sputum specimens are collected prior to bronchoscopy, while bronchial washings and tissue are taken during the bronchoscopy procedure. Tissue specimens were collected by bronchial brushing or forceps. All specimens were grown on Sabouraud Dextrose Agar medium and incubated at 25°C for 1–5 days for macroscopic growth observation. The growth of yeast fungi was analysed by Gram staining, while that of mould fungi was examined by 10% potassium hydroxide (KOH). Yeast fungi findings, along with gram-negative/positive bacteria, after day 5, were considered non-pathogenic fungi. Fungal culture was positive if yeast/budding cells were found on the Gram stain and confirmed with VITEK® system (bioMérieux SA, Marcy-l’Étoile, France) for identification of fungal species. Microscopically intact hyphae on a 10% KOH examination are also considered a positive result.
The different fungal species identifications between the sputum, bronchial washings and tissues were followed by data analysis using IBM SPSS Statistics (IBM Corp., Armonk, NY, USA) software. Univariate analysis was conducted to determine the frequency distribution and was presented in the form of tables or diagrams. Furthermore, all variables were subjected to bivariate analysis using the chi-square test with an alternative Fisher test. A value of P < 0.05 is considered statistically significant and appropriate.
Between August 2024 and January 2025, 145 patients were hospitalised due to thoracic cavity malignancy. Out of these, 66 underwent bronchoscopy, and 57 patients who met the inclusion and exclusion criteria were included in the study (Figure 1).
Research sample selection.
Table 1 shows that most patients were over the age of 40 years old (91.23%), predominantly male (66.67%), and classified as underweight (49.12%). Additionally, a significant proportion were smokers with a high Brinkman Index (50.88%). A total of 50.88% of patients reported no history of pulmonary disease, and 66.67% had no comorbidities. Patients used bronchodilators, either with or without inhaled corticosteroids, 40.35% of the time. Chronic cough (84.21%) and moderate cancer pain (45.61%) were the primary symptoms observed. Most cases of thoracic cavity malignancy (91.23%) were diagnosed as lung cancer, with adenocarcinoma accounting for 36.94%. Fungal culture results were positive in 77.19% of the 57 patients with thoracic cavity malignancy.
Characteristics of research subjects
| Variables | N | % |
|---|---|---|
| Age (years) | ||
| • ≤40 | 5 | 8.77 |
| • >40 | 52 | 91.23 |
| Gender | ||
| • Male | 38 | 66.67 |
| • Female | 19 | 33.33 |
| BMI (kg/m2) | ||
| • Underweight | 28 | 49.12 |
| • Normoweight | 20 | 35.09 |
| • Overweight | 4 | 7.02 |
| • Obesity class I | 4 | 7.02 |
| • Obesity class II | 1 | 1.75 |
| Smoking status | ||
| • Non-smoker | 19 | 33.33 |
| • Mild brinkman index | 3 | 5.26 |
| • Medium brinkman index | 6 | 10.53 |
| • Severe brinkman index | 29 | 50.88 |
| History of pulmonary disease | ||
| • None | 29 | 50.88 |
| • Former TB | 5 | 8.77 |
| • COPD | 23 | 40.35 |
| Comorbidity | ||
| • None | 38 | 66.67 |
| • Hypertension | 12 | 21.05 |
| • Diabetes mellitus type 2 | 4 | 7.02 |
| • Heart failure | 4 | 7.02 |
| • Stroke | 1 | 1.75 |
| • Autoimmune disease | 1 | 1.75 |
| Drug history | ||
| • None | 17 | 29.82 |
| • Bronchodilators with/without inhaled | 23 | 40.35 |
| corticosteroids | ||
| • Anti-tuberculosis drugs | 5 | 8.77 |
| • Anti-hypertensives | 16 | 28.07 |
| • Anti-diabetics | 4 | 7.02 |
| • Oral corticosteroids | 1 | 1.75 |
| Clinical symptoms | ||
| • Shortness of breath | 23 | 40.35 |
| • Chest pain | 47 | 82.46 |
| • Chronic cough | 48 | 84.21 |
| • Haemoptysis | 5 | 8.77 |
| Cancer pain degree | ||
| • No pain | 10 | 17.54 |
| • Mild | 17 | 29.82 |
| • Moderate | 26 | 45.61 |
| • Severe | 4 | 7.02 |
| Types of thoracic cavity malignancy | ||
| • Lung cancer | 52 | 91.23 |
| • Mediastinal tumour | 2 | 3.51 |
| • Lung tumour metastasis | 3 | 5.26 |
| Pathology anatomy results | ||
| • Cancer type unknown | 16 | 28.07 |
| • Adenocarcinoma | 21 | 36.84 |
| • Squamous cell carcinoma | 14 | 24.56 |
| • Neuroendocrine carcinoma | 3 | 5.26 |
| • Tumour metastasis in lungs | 3 | 5.26 |
| Fungal culture results | ||
| • Positive culture | 44 | 77.19 |
| • Negative culture | 13 | 21.81 |
BMI, body mass index; COPD, chronic obstructive pulmonary disease; TB, tuberculosis.
A total of 44 out of 57 patients (77.19%) showed positive fungal culture results from at least one of the test specimens used in this study. As shown in Figure 2, the most identified fungi were Candida sp. (57.89%), Aspergillus sp. (31.58%), Cryptococcus sp. (3.51%) and Epicoccum sp. (1.75%).
Fungus positivity rates based on research subjects.
Figure 3 shows that sputum specimens had the highest positivity rate at 57.89%, followed by bronchial washing at 56.14% and tissue samples at 42.11%. The positivity rates of fungal cultures differed significantly across the types of specimens (Figure 4).
Fungus positivity rates based on research subjects.
Analysis of culture positivity rate and fungal species.
The positivity rate of Candida sp. differed significantly among the three specimen types (P = 0.002), with sputum showing the highest rate (49.12%). Similarly, the positivity rate of Aspergillus sp. also differed significantly (P = 0.013), with tissue samples demonstrating the highest distribution. No significant difference was observed for Cryptococcus sp. among the specimens (P = 1.000), while Epicoccum sp. was exclusively identified in tissue samples (Figure 4).
This study found that the positivity rates for sputum, bronchial washing and tissue specimens were similar. Tissue samples were utilised for all fungal culture identifications. However, the positivity rates for Candida sp. and Aspergillus sp. varied significantly across the three specimen types, reflecting their concordance patterns. Of the Candida sp. positivity cases, 21.22% were identified in all three specimen types, while 33.33% were detected only in sputum. Figure 5A illustrates that tissue specimens, considered the gold standard, had a lower positivity rate for Candida sp. compared to sputum and bronchial washing specimens. In contrast, Aspergillus sp. exhibited a concordance positivity distribution of 22.22% across all three specimen types, with a higher rate of 44.45% for tissue specimens. Figure 5B demonstrates that tissue specimens, as the gold standard, were more likely to test positive for Aspergillus sp. than either sputum or bronchial washing specimens.
Analysis of culture positivity rate and fungal species. (A) Candida sp. (B) Aspergillus sp.
Table 2 compares the diagnostic accuracy of sputum and bronchial washing specimens with that of tissue specimens, which are considered the gold standard. The accuracy measures include sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV). Sputum samples showed 70% sensitivity, 55.32% specificity and a 25% PPV for detecting Candida sp. In comparison, bronchial washing specimens exhibited 100% sensitivity, 74.47% specificity and a 45.46% PPV for Candida sp. detection, outperforming sputum specimens. For Aspergillus sp., sputum specimens had a sensitivity of 25% but demonstrated 100% specificity and PPV, while bronchial washing specimens revealed 50% sensitivity, 95.12% specificity and 80% PPV.
Accuracy of test specimens for fungal culture positivity
| Isolate | Sputum versus tissue | Bronchial washing versus tissue | ||||||
|---|---|---|---|---|---|---|---|---|
| Sn (%) | Sp (%) | PPV (%) | NPV (%) | Sn (%) | Sp (%) | PPV (%) | NPV (%) | |
| Candida sp. | 70.0 | 55.32 | 25.0 | 89.66 | 100.0 | 74.47 | 45.46 | 100.0 |
| Aspergillus sp. | 25.0 | 100.0 | 100.0 | 77.36 | 50.0 | 95.12 | 80.0 | 82.98 |
| Cryptococcus sp. | 100.0 | 98.21 | 50.0 | 100.0 | 100.0 | 100.0 | 100.0 | 100.0 |
| Epicoccum sp. | 0.0 | 100.0 | 0.0 | 98.25 | 0.0 | 100.0 | 0.0 | 98.25 |
NPV, negative predictive value; PPV, positive predictive value; Sn, sensitivity; Sp, specificity.
Both sputum and bronchial washing specimens detected Cryptococcus sp. with high sensitivity and specificity. Bronchial washing showed a PPV of 100%, exceeding that of sputum. Epicoccum sp. had 100% specificity and a 98.25% NPV but a sensitivity of 0%. Table 2 indicates that tissue specimens were more accurate than both sputum and bronchial washing specimens in identifying fungal cultures, especially for Aspergillus sp. and Epicoccum sp. In patients with thoracic cavity malignancies, bronchial washing may effectively detect Candida sp. and Cryptococcus sp.
Most respondents (91.23%) were over 40 years old, with 66.67% identifying as male. These findings align with the research of Pradnyaandara et al. (14) and Alfarisa et al. (15), which indicates that lung cancer is the most prevalent malignancy of the thoracic cavity among older men. In this study, a significant portion of patients (49.12%) were classified as underweight according to their body mass index (BMI). Malnutrition can potentially accelerate disease progression, increase fungal colonization, elevate mortality rates and diminish tolerance to treatment (16–18).
A severe Brinkman index (50.88%) among participants suggests that smoking is a significant risk factor for lung cancer (19, 20). Chronic inflammation, particularly among those with a history of chronic obstructive pulmonary disease (COPD) (40.35%), further raises the risk of lung cancer. Research indicates that 50–90% of lung cancer patients also have COPD. This condition leads to cellular mutations that foster cancer cell growth, driven by chronic inflammation, oxidative stress, and lung tissue remodelling. Additionally, the hypersecretion associated with COPD contributes to sputum and biofilm formation, creating an environment conducive to microorganism colonisation (19, 21, 22).
Comorbidities were identified in 66.67% of patients, with hypertension being the most prevalent condition associated with thoracic cavity malignancy, occurring in 21.05% of cases. These findings align with those of El-Badrawy et al. (23), who reported that 38% of lung cancer patients had no comorbidities. Additionally, Gupta et al. (24) found that 50% of lung cancer patients were diagnosed with hypertension. This study indicates that the use of bronchodilators, with or without inhaled corticosteroids (40.35%), is associated with an increased risk of fungal colonisation. This, in turn, may lead to a reduction in alveolar macrophage function and proinflammatory cytokines, which inhibits fungal colonisation in lung parenchyma (25, 26).
Chronic cough (84.21%) and chest pain (82.46%) emerged as the primary symptoms, potentially indicating pulmonary mycosis. Chronic coughing can result from impaired mucociliary clearance due to ongoing inflammation. The invasion of hyphae into lung epithelial tissue and blood vessel endothelium can lead to chest pain and lung infarction (27, 28). This study also identified lung adenocarcinoma (36.84%) as the most common type of thoracic malignancy. These findings are consistent with several studies indicating that lung adenocarcinoma is frequently associated with fungal colonisation (29, 30). Harada et al. (31) reported that 50% of adenocarcinomas were present in 85% of cases involving fungal colonisation in lung cancer patients.
El-Badrawy et al. (23) reported a fungal culture positivity rate of 68% among 100 lung cancer patients, which is comparable to the 77.19% observed in this study. The investigation identified the presence of Candida sp., Aspergillus sp., Cryptococcus sp. and Epicoccum sp. fungi in three test specimens. Specifically, Candida sp. constituted 59.65% and Aspergillus sp. 31.58% of the identified fungi. These findings align with those of Laroumagne et al. (30), who documented a positivity rate of 42.9% for Candida sp. and 6.2% for Aspergillus sp. across all lung cancer cases. Conversely, El-Badrawy et al. (23) found Aspergillus sp. in 36% and Candida sp. in 32% of the lung cancer patients.
Cryptococcus sp. (3.51%) and Epicoccum sp. (1.75%) are additional common findings in fungal cultures related to thoracic cavity malignancies. Immunosuppression is frequently associated with Cryptococcus sp. (32). Meanwhile, Epicoccum sp. is typically found in soil, dust and plants as a contaminant, with limited evidence supporting its role in pulmonary mycosis. Variations in fungal culture results may be attributed to the use of different test specimens, each exhibiting distinct sensitivities and specificities (33).
Fungal cultures showed positivity in 57.89% of 33 sputum, 56.14% of 32 bronchial washing and 42.11% of 24 tissue. Among 216 immunocompromised patients, the highest positivity rates were found in sputum (64.3%), bronchial washing (30.6%) and tissue (5.0%), as reported by Nugratama et al. (34). The diagnostic accuracy of each specimen type may influence positivity rates. Most positive fungal cultures were obtained from specimens of bronchial washing and tissue during bronchoscopy. In a study involving 98 lung cancer patients, Gupta et al. (24) reported that most fungal cultures were identified through bronchoscopy, with a positivity rate of 65.31% for BAL and 61.22% for bronchial washing. Ayçiçek et al. (29) found a fungal culture positivity rate of 16.4% from bronchoscopy compared to 15.3% from sputum.
Sputum cultures can be more positive due to contamination, specimen quality and fungal culture incubation time (24, 35). This study found low positivity in tissue samples, possibly due to damage to fungal cells or their components during collection, which could affect fungal cultures (36). The three sputum samples had different fungal culture positivity (P = 0.002). Sputum samples were most positive for Candida sp. (49.12%) and least positive for Aspergillus sp. (17.54%). According to Prakash et al. (37), immunocompromised patients had a 54% Candida sp. positivity rate with a P-value of 0.0001. Similarly, Sahara et al. (38) found Candida sp. in 80% of lung cancer patients. The humid upper airway and Candida sp. ability to colonise may explain the increased presence of these organisms in sputum. Positive results in fungal cultures are often considered contaminants (39).
The three tissue specimens showed significant variance in fungal culture prevalence, with Aspergillus sp. (28.07%) and Candida sp. (7.02%) having the highest and lowest prevalence (P = 0.013). In biopsy forceps specimens from lung cancer patients, Ayçiçek et al. (29) found 22.81% Aspergillus sp. positivity, significantly higher than bronchial washing (P < 0.001). Denning et al. (40) found Aspergillus sp. in 31.57% of biopsy tissues. Aspergillus sp., the most common pulmonary pathogenic fungus, penetrates epithelial and parenchymal tissue. Damage to the airway epithelium and lung parenchyma is the main cause of Aspergillus sp. colonisation and lung lesions. This confirms that tissue specimens are more often used to identify Aspergillus sp. than sputum or bronchial washing (41).
The results indicated no statistical significance (P = 0.000), although sputum samples revealed a higher percentage of Cryptococcus sp. (3.51%). Epicoccum sp. was identified with 100% accuracy in tissue samples. The low quantities of Cryptococcus sp. and Epicoccum sp. observed in this study led to statistically biased outcomes. Tissue specimens demonstrated greater reliability in detecting various fungi, particularly Aspergillus sp. and Epicoccum sp., while sputum specimens were mainly positive for Candida sp. and Cryptococcus sp. (32, 33).
This study compares sputum and bronchial washing specimens to tissue specimens, which are considered the gold standard for diagnosing malignancies in the thoracic cavity, in terms of fungal culture results. The positivity rate for Candida sp. in sputum (33.33%) was higher than that in tissue specimens (21.22%) (Figure 5A). Despite tissue specimens being the gold standard, they exhibit a lower positivity rate for Candida sp. The concordance analysis suggests that a high rate of Candida sp. positivity may result in false positives due to contamination or normal flora (33). Candida sp. can penetrate the airway epithelium by transitioning from yeast to hyphal form, which is attributed to their unique properties. Additionally, cancer necrosis facilitates the formation of biofilms by Candida sp., enabling them to invade adjacent organs. The positivity for Candida sp. in tissue specimens indicates high accuracy (29, 42).
Aspergillus sp. positivity was lower in sputum (0%) and bronchial washing (11.11%). All three test specimens had the same 21.22% Aspergillus sp. positivity rate. This suggests that sputum and bronchial washing positivity yield similar tissue results. If obtained exclusively from tissue specimens, Aspergillus sp. had 44.45% positivity (Figure 5B). Aspergillus sp. positivity is more common in tissue specimens, the gold standard, than in sputum and bronchial washing. Aspergillus sp. can invade and penetrate the lung epithelium and parenchyma, which increases its positivity in tissue specimens. Additionally, Aspergillus sp. can invade blood vessels and nearby organs (43, 44).
This study identified differences in the accuracy of diagnostic tests among the three specimens examined. Sputum and bronchial washing were found to be less accurate than tissue specimens in identifying all fungal species. Factors, such as the specimen collection technique, insufficient sample sizes and challenges in identifying fungal species, contributed to bias in the accuracy of the three test specimens (45, 46). Power et al. (47) defined a diagnostic test as useful if its sensitivity and specificity are both less than 150%. Other interpretations suggest that diagnostic tests with sensitivity and specificity values of £100% are likely ineffective. Conversely, diagnostic test results approaching 200% are considered optimal and highly recommended.
Many studies indicate that tissue specimens demonstrate greater sensitivity and specificity compared to sputum and bronchial washing for detecting fungal cultures, particularly Aspergillus sp. in lung cancer (43, 44). Tissue specimens yield more positive results for Aspergillus sp. because the fungus can invade the airway epithelium and lung parenchyma (45, 48). The results of diagnostic tests align with the analysis of specimen efficacy, indicating that tissue specimens produce the highest number of positive Aspergillus sp. culture results, whereas sputum and bronchial washing yield fewer positive results (44).
This study’s sputum specificity (55.32%) is lower than that of bronchial washing (74.47%), aligning with Kontoyiannis et al. (49), which reported a specificity of 60%. The discrepancy may be attributed to factors such as sputum sample contamination, poor quality and inadequate culture time. Candida sp. can transition from yeast to hyphal forms, facilitating their penetration of the airway epithelium. This characteristic may explain why bronchial washing demonstrates greater specificity than sputum samples. Additionally, necrosis induced by malignancy may contribute to the formation of biofilms by Candida sp., enabling them to invade lung parenchyma (24, 50).
Fungal culture positivity rates varied significantly among research subjects and test specimens, particularly sputum, bronchial washing and tissue samples. This study identified statistically significant differences in positivity rates for Candida sp. and Aspergillus sp. across the three specimen types, with sputum exhibiting the highest levels for Candida sp. and tissue specimens showing the highest levels for Aspergillus sp. Tissue specimens demonstrate higher diagnostic accuracy than sputum or bronchial washing for fungal culture identification, particularly for Aspergillus sp. and Epicoccum sp. Additionally, bronchial washing specimens may serve as a substitute for sputum in detecting Candida sp. and Cryptococcus sp. in patients with thoracic cavity malignancies.