Recommendation 1: For patients with multiple comorbidities or critically ill patients at high risk of bronchoscopy, and in whom the etiology of pulmonary inflammation remains unclear - requiring differentiation among causes such as infection, drug-induced lung injury, immune-related pneumonitis, and tumor progression - a multidisciplinary discussion is recommended prior to performing bronchoscopy. (Consensus level 90%)

An MDT is fundamental to ensuring the safe performance of bronchoscopy. This team should include physicians from relevant specialties such as Respiratory and Critical Care Medicine, Medical Oncology, Infectious Diseases, Radiology, Rheumatology and Immunology, Microbiology, and Pharmacology, and Anesthesiology. If the patient has specific comorbidities, corresponding departments should be involved in the decision-making process. For example, patients with coronary heart disease (especially those with a history of coronary stent implantation) require evaluation by a cardiologist; those with cerebrovascular disease necessitate input from a neurologist. The MDT should conduct a comprehensive assessment focusing on both the “necessity” and “feasibility” of bronchoscopy.

Recommendation 1: The MDT discussion covers topics such as the patient’s baseline condition and organ function, the status of tumor treatment, the severity of pneumonia, the necessity of bronchoscopy, and the specific content of the examination. (Consensus level 98%)

(1) Patient’s baseline condition and organ function risk.

a. Performance status: This assesses the patient's ability to perform daily activities and their overall health condition. An ECOG score of 4 indicates a high-risk factor for procedures, necessitating careful evaluation of the procedure’s necessity and associated risks^[32].

b. Control of comorbidities: This includes management of acute or subacute conditions such as hypertensive crises, diabetic ketoacidosis or hyperosmolar states, acute cardiovascular events, aortic dissection, pulmonary embolism, and severe electrolyte imbalances. Any uncontrolled acute conditions should be prioritized for correction. If stabilization proves difficult, a thorough risk-benefit analysis for the procedure should be conducted.

c. Respiratory and circulatory system function: This refers to patients with Type I or Type II respiratory failure (especially those with PaO₂/FiO₂ ≤ 200 mmHg, or decompensated respiratory acidosis), severe airway stenosis or obstruction, decompensated heart failure, malignant arrhythmias, or circulatory instability such as shock^[33,34]. These patients are at high risk for bronchoscopy and should be approached with extreme caution.

(2) Disease-related factors evaluation.

a. Tumor-related factors: This includes the type and stage of the tumor, as well as an estimation of the patient’s survival based on tumor biology and clinical progression. For patients with a short-expected survival (e.g., < 3 months), the benefit of invasive procedures should be carefully reconsidered.

b. Treatment-related factors: A comprehensive inquiry into the patient’s recent history of anti-tumor treatments and the treatment phase is required (whether they are in the post-chemotherapy myelosuppression period). It is essential to clarify their use of chemotherapy, targeted therapy, radiotherapy, and immunotherapy. Additionally, the potential risk of treatment-related pulmonary damage should be carefully considered, such as targeted therapy-related pneumonia, CIP, or radiation-induced pneumonitis.

c. Pneumonia severity, necessity of bronchoscopy, and examination content: Tools like the CURB-65 score should be used to objectively assess pneumonia severity. A score of ≥ 2 indicates a higher risk for bronchoscopy, requiring careful evaluation of the procedure's benefits and risks. Imaging assessments should be performed to evaluate pulmonary involvement and lesion characteristics. The necessity of bronchoscopy in identifying the etiology of pneumonia and guiding treatment. Furthermore, the specific type of bronchoscopy and required tests for the patient should be discussed.

Recommendation 1: For patients suspected of having malignancies involving the central airways, major vessels, or large cavitary lesions, a contrast-enhanced chest CT or bronchial artery CTA is recommended prior to bronchoscopy to assess bleeding risk. If the imaging suggests a high risk and bronchoscopy is still indicated, full preparedness for managing massive hemorrhage must be established. (Consensus level 100%)

Recommendation 2: For patients undergoing anti-angiogenesis therapy, if only BAL or PSB is planned, and the patient’s general condition is clinically stable with no coagulopathy, it is considered unnecessary to discontinue the medication in advance. However, if invasive procedures such as biopsies or interventional treatments are planned, it is recommended to stop the medication based on its half-life, unless the procedure is urgent. (Consensus level 100%)

Recommendation 3: In the absence of underlying coagulation disorders or related comorbidities, routine blood and coagulation tests performed within one week prior to bronchoscopy are considered acceptable. For patients with known hematologic diseases or coagulation abnormalities, a more stringent timeframe of 24 h is recommended for obtaining these laboratory indicators to ensure an accurate pre-procedural bleeding risk assessment. (Consensus level 97%)

Post-chemotherapy patients are prone to thrombocytopenia, coagulation abnormalities, or malignant tumors invading the trachea, bronchi, pulmonary arteries, or bronchial arteries, which can lead to an increased risk of airway hemorrhage. Therefore, a comprehensive assessment of the risk of airway hemorrhage is required before the procedure. This includes a detailed inquiry into the patient’s history of bleeding, as well as their medication history, particularly the use of antiplatelet agents, anticoagulants, and anti-vascular-targeted therapies. There is currently no universally established standard for the optimal timeframe for pre-bronchoscopy coagulation tests. Our expert panel reached the following consensus based on clinical practice and patient risk stratification: For patients without underlying coagulation disorders or related comorbidities, routine blood tests and coagulation function tests obtained within one week prior to the procedure are considered acceptable. For patients with known hematologic diseases or coagulation abnormalities, we recommend obtaining these laboratory indicators within 24 h before bronchoscopy to ensure accurate assessment of bleeding risk. Abnormal blood counts or coagulation functions should be corrected proactively. Antiplatelet or anticoagulant medications should be discontinued according to their half-life^{[34,39,40-44]}.

In patients with suspected tumor invasion of the central airways, major vessels, or large cavitary lesions, pre-procedural imaging with contrast-enhanced chest CT or bronchial artery CTA is strongly recommended. This evaluation is critical for identifying high-risk features, including pulmonary artery involvement, pseudoaneurysms, or signs of abnormal vasculature such as enlarged/tortuous bronchial arteries or bronchial artery-pulmonary artery shunts. For these high-risk cases, a multi-disciplinary consultation involving specialists from pulmonology, critical care, radiology, thoracic surgery, and anesthesiology is essential to weigh the necessity of bronchoscopy and formulate a comprehensive management plan. Furthermore, emergency protocols for massive hemorrhage must be established, including readiness for endobronchial balloon tamponade, bronchial or pulmonary artery embolization, and emergency thoracotomy^[39,45,46].

Anti-angiogenic drugs (e.g., bevacizumab, anlotinib, recombinant human endostatin) inhibit neovascularization and can increase vascular fragility, thereby elevating the risk of bleeding during invasive procedures^[47-49]. The decision to withhold these drugs before bronchoscopy should be stratified by procedural risk: (1) Low-risk procedures (e.g., BAL, PSB, airway inspection without biopsy): Drug discontinuation is generally unnecessary in clinically stable patients without coagulopathy. (2) High-risk procedures (e.g., biopsy, interventional therapy): It is recommended to withhold the drug prior to the procedure based on its half-life, unless the situation is urgent. The recommended discontinuation times, based on the pharmacokinetics of these agents, are as follows: bevacizumab for 4-6 weeks, anlotinib for 1 week, and recombinant human endostatin for 24 h. These are general guidelines, and the final decisions must be individualized, incorporating the patient’s complete medication history, comorbidities, and hepatic/renal function.

Recommendation 1: Most patients receiving standard chemotherapy for solid tumors are at low risk and do not require routine prophylactic antibiotics before bronchoscopy. Prophylactic antibiotics are recommended for high-risk patients. (Consensus level 90%)

The probability of secondary bacterial infection following bronchoscopy is approximately 6%-8%, with common pathogens including coagulase-positive or negative staphylococci, non-hemolytic or β-hemolytic streptococci, and Klebsiella species^[50,51]. The prophylactic use of antibiotics for bronchoscopy remains controversial in clinical practice, particularly in patients with tumor-related immunosuppression. Multiple studies have clearly shown that routine prophylactic antibiotics after diagnostic bronchoscopy have not been demonstrated to reduce the incidence of post-procedure fever, pneumonia, or endocarditis^[52,53]. Therefore, guidelines from the British Thoracic Society (BTS) and others generally do not recommend routine use^[54]. However, recent large-scale evidence suggests potential benefits in specific high-risk populations. For example, a nationwide Japanese database study (involving more than 68,000 patients undergoing diagnostic bronchoscopy for noninfectious diseases) demonstrated that oral prophylactic antibiotics (such as aminopenicillins or fluoroquinolones) were significantly associated with a reduction in post-procedure infections (OR 0.60; 95% CI 0.45-0.80), particularly in elderly patients (> 70 years), those with malignancy, or those undergoing biopsy or bronchoalveolar lavage^[55]. Nevertheless, this benefit must be weighed against risks such as antimicrobial resistance and the lack of clear improvement in mortality.

Neutropenia (granulocytopenia) and lymphopenia have different mechanisms of immunodeficiency, leading to distinct infection risks and evidence bases. For patients with malignancies who develop neutropenia following chemotherapy, they are classified into low-risk and high-risk groups based on the severity of potential complications. Low-risk patients are those expected to have neutropenia (absolute neutrophil count [ANC] < 0.5 × 10⁹/L) lasting ≤ 7 days, with no or few comorbid conditions; high-risk patients are those with neutropenia lasting > 7 days and/or significant medical comorbidities^[56]. Currently, most patients receiving standard chemotherapy for solid tumors are classified as low-risk and do not require prophylactic antibiotics before bronchoscopy. However, for high-risk patients, prophylactic antibiotics are recommended.

Lymphopenia is common in patients receiving anti-tumor therapy or those with hematologic diseases. Radiotherapy, chemotherapy (e.g., cyclophosphamide, cisplatin, methotrexate, and taxanes), targeted therapy (e.g., mTOR inhibitors, tyrosine kinase inhibitors targeting VEGFR or PDGFR), and immune checkpoint inhibitors can induce lymphopenia^[57-60]. In a prospective Danish population-based study of 98,344 individuals, lymphopenia (defined as a lymphocyte count below the population’s 2.5th percentile, i.e., < 1.1 × 10⁹/L) was linked to a higher risk of both infection-related hospitalization and mortality^[61]. Furthermore, early lymphopenia has also been identified as a risk factor for chemotherapy-induced febrile neutropenia^[62] and early death following chemotherapy^[63]. Given that evidence regarding antibiotic use during bronchoscopy in lymphopenic patients is limited, the expert panel recommends the prophylactic administration of antibiotics for those with lymphopenia (defined as a lymphocyte count below the population’s 2.5th percentile) undergoing this procedure.

Bronchoscopy inherently carries certain respiratory-related risks, and such risks are significantly heightened in cancer patients who have undergone anti-tumor therapy - characterized by potential multi-organ dysfunction and complicated by pneumonitis. Therefore, it is of crucial clinical importance to optimize preoperative risk stratification assessment, formulate a comprehensive respiratory support plan, and select an appropriate procedure site.

Recommendation 1: Pre-procedural risk stratification should be conducted based on the patient’s baseline cardiopulmonary function, type and severity of comorbidities, clinical symptoms, coagulation function parameters, and the complexity of the intended procedure. The level of respiratory support must then be tailored accordingly, commensurate with the assessed risk. (Consensus level 97%)

(1) Patients classified as low-risk (basically normal pulmonary and cardiac function; resting respiratory rate of 12-20 breaths/min; oxygenation index (PaO₂/FiO₂) ≥ 300 mmHg; platelet count ≥ 100 × 10⁹/L; normal coagulation function; requirement for diagnostic procedures only; and absence of conditions such as airway stenosis or a history of massive hemoptysis) may undergo the procedure in a standard bronchoscopy room, accompanied by oxygen inhalation via nasal cannula and ECG monitoring.

(2) Patients classified as moderate-risk (moderate pulmonary dysfunction; mild cardiac insufficiency (NYHA Class Ⅰ-Ⅱ); resting respiratory rate of 21-25 breaths/min; PaO₂/FiO₂ of 200-300 mmHg; platelet count of 60-100 × 10⁹/L or mild coagulation abnormalities; requirement for small-scale therapeutic procedures; or presence of mild airway stenosis) should undergo the procedure in a bronchoscopy room equipped with high-flow nasal cannula oxygen therapy (HFNC) or non-invasive positive pressure ventilation (NIPPV). Local epinephrine infusion and a hemostatic balloon should be prepared and readily available.

(3) Patients classified as high-risk (severe pulmonary dysfunction; severe cardiac insufficiency (NYHA Class Ⅲ-Ⅳ); complicated by serious underlying diseases such as uremia, liver cirrhosis, or cerebrovascular disorders; resting respiratory rate > 25 breaths/min; PaO₂/FiO₂ < 200 mmHg; platelet count < 60 × 10⁹/L; INR > 1.5; requirement for large-scale therapeutic procedures; or presence of severe airway stenosis or a recent history of massive hemoptysis) should undergo the procedure in an ICU or a hybrid operating room, equipped with comprehensive respiratory and circulatory support, such as Cone-Beam Computed Tomography (CBCT). Pre-procedural preparations should include HFNC and NIPPV, with intubation and mechanical ventilation conducted in advance. If necessary, intravenous access for rapid blood transfusion, a complete array of emergency medications, and a hemostatic balloon should be immediately available.

BAL is a procedure in which saline is instilled into the bronchial alveoli via a bronchoscope and then aspirated for the collection of alveolar surface liquid and/or the clearance of material occupying the alveoli. For patients without respiratory failure, a total lavage volume of 100-300 mL is recommended, divided into three to five aliquots. The minimal total volume retrieved should be ≥ 5% of the instilled volume; an optimal return is ≥ 30%. A minimal volume of 5 mL of the pooled BAL sample is required for cellular analysis (optimal 10-20 mL), striking a balance between minimizing lavage volume and ensuring an adequate specimen in patients with respiratory failure^[64]. BAL is the preferred method for identifying pathogens in infectious pneumonia and for differentiating between infectious and non-infectious pneumonias. Laboratory analysis of the fluid should be strategic and directed at the relevant differential diagnoses. Tests employed may include ay include cell differential and count, microbiological smears and cultures, fungal antigen testing, pathogen nucleic acid testing, metagenomic next-generation sequencing (mNGS) for lower respiratory tract pathogens, special stains, and cytopathological examination^[65-67].

PSB is a sampling technique that employs a sheathed brush. This brush is advanced through the bronchoscope to the bronchial segment exhibiting the most significant radiographic infiltration, most rapid progression, or endoscopic evidence of purulent secretions to collect lower respiratory tract specimens. Typically performed alongside BAL, PSB is primarily used for pathogen identification in infectious pneumonia and for cytopathological examination. While it samples a smaller area and offers a narrower spectrum of detectable analyses than BAL, its principal advantage lies in obtaining uncontaminated lower respiratory tract specimens via the protected sheath, thereby facilitating reliable microbiological culture and cytopathological studies^[68-70].

Bronchial mucosal biopsy involves obtaining tissue samples from endobronchial lesions. It is indicated for establishing an etiological diagnosis in cases of bronchoscopically identified mucosal abnormalities, such as nodules, erosions, swelling, or congestion. Notably, antineoplastic agents like immune checkpoint inhibitors and RAF/MEK inhibitors can induce sarcoid-like reactions^[71,72]. Hence, this procedure can yield diagnostic value in patients with suspected new-onset sarcoidosis, even in the absence of bronchoscopically visible mucosal changes. In confirmed cases of infectious pneumonia, bronchial mucosal biopsy is generally not necessary. However, it may be considered if infections such as mycobacterium tuberculosis or fungi are suspected, particularly when BAL or PSB results are negative and significant mucosal abnormalities are present^[73,74].

Autofluorescence bronchoscopy, narrow-band imaging bronchoscopy, and confocal laser endomicroscopy (CLE) share the primary design objective of detecting airway tumors and/or pre-cancerous lesions. Each technique, however, provides a distinct view of mucosal pathology - through differences in fluorescence signals, vascular patterns, and microscopic cellular architecture, respectively. Consequently, they offer an irreplaceable advantage over traditional white-light bronchoscopy in the accurate differentiation among inflammation, interstitial lung disease, and neoplasia^[75-77].

TBLB obtains small lung tissue specimens, providing crucial evidence for pathological diagnosis. It is primarily indicated for lesions located distal to the segmental bronchi that are not visible by conventional bronchoscopy. The TBLB procedure is generally considered to be safe. However, compared to techniques like BAL, PSB, and bronchial mucosal biopsy, TBLB carries higher risks of bleeding and pneumothorax^[78]. Not all patients require TBLB, it should be actively considered when the risk assessment is acceptable in the following scenarios: (1) For undiagnosed conditions: When the etiology remains unclear despite clinical presentation, laboratory tests, imaging features, and BAL results; (2) For complex or multifactorial etiologies: Such as suspected drug-associated pneumonia with concurrent infection; (3) When pathological typing is needed to guide treatment decisions; (4) Upon initial treatment failure: For example, when clinically suspected immune drug-related pneumonia shows poor response to empirical corticosteroid therapy; (5) For atypical or focal radiographic presentations: TBLB demonstrates high success rates and diagnostic value for lesions located within the reach of endobronchial ultrasound.

Technically, TBLB includes traditional fluoroscopy-guided approaches and modern guided bronchoscopy techniques. The latter significantly improves the diagnostic yield for peripheral pulmonary lesions, with an average diagnostic sensitivity of approximately 70%, superior to traditional TBLB techniques (approximately 55%)^[79,80]. Commonly used guided bronchoscopy techniques currently include radial endobronchial ultrasound (R-EBUS), guide sheath (GS), virtual/electromagnetic navigation, multimodal augmented reality navigation bronchoscopy, cone-beam CT, and robotic bronchoscopy^[81]. Among these, navigation bronchoscopy combined with EBUS-GS is the most frequently used in clinical practice.

Cryobiopsy (CB) is a sampling technique that utilizes cryoadhesion to obtain tissue specimens. It comprises two distinct procedures: Endobronchial cryobiopsy (EBCB) for visible lesions and TBCB for peripheral lesions beyond endoscopic visualization. Compared to forceps biopsy (FB), CB is characterized by larger specimen size, superior architectural preservation, and a higher diagnostic yield. For endobronchial lesions, EBCB carries an increased risk of mild-to-moderate bleeding compared to FB^[82,83]; it is therefore recommended when FB sampling is inadequate or when a specific pathology necessitates a larger sample. Similarly, TBCB is not recommended for routine use. Its primary indication arises when the etiology remains elusive despite comprehensive evaluation-including clinical, laboratory, radiological, BAL, and conventional TBLB findings-and when there is a high clinical suspicion of a diffuse interstitial lung disease or peripheral pulmonary nodule requiring histological confirmation.

EBUS-TBNA is a minimally invasive technique that enables real-time ultrasound-guided precise puncture and biopsy of mediastinal and hilar lesions adjacent to the trachea or bronchi, characterized by accurate positioning, operational safety, and high diagnostic efficiency^[84,85]. While not routinely recommended, it should be actively utilized in specific scenarios: (1) Etiological differentiation, when imaging reveals extrapulmonary lesions adjacent to airways requiring distinction between treatment-related pneumonia, tumor recurrence, or lymph node metastasis; (2) Diagnosis of specific pathogen infections, such as obtaining lymph node tissue for pathogenic testing in suspected tuberculosis; (3) Evaluation of systemic diseases, particularly when granulomatous diseases like immunotherapy-induced sarcoidosis are suspected, where demonstrating non-caseating granulomas in lymph nodes provides crucial diagnostic evidence.

The rigid bronchoscope provides a safe, reliable, and powerful operative channel, serving as a key technical platform-rather than a routine diagnostic tool-for managing high-risk and complex scenarios^[86,87]. Its primary indications include: (1) performing tissue biopsy in high-risk patients when pathological confirmation is essential yet unattainable by conventional diagnostics, and when factors such as high bleeding risk or severe hypoxemia requiring ventilatory support necessitate a secured approach to manage potential complications like major hemorrhage; (2) managing severe treatment-related airway complications, such as those following radiotherapy or immunotherapy, which may involve central airway obstruction from necrotic material or pseudomembranes, or rapidly progressive stenosis and granulomatous proliferation, all requiring urgent intervention under rigid bronchoscopy-including debridement, resection, balloon dilation, ablation, or stent placement.

(1) Microbiological examination

a. Bacterial Smear and Culture: Used for diagnosing bacterial pneumonia, covering both aerobic and anaerobic bacteria.

b. Fungal smear and culture: Aids in detecting fungal infections, including yeasts, molds, and dimorphic fungi.

c. Tuberculosis (TB) smear and culture: Employs acid-fast staining and mycobacterial culture to identify Mycobacterium tuberculosis and non-tuberculous mycobacterial infections. Weakly acid-fast positivity is a characteristic feature of Nocardia infection.

d. Polymerase chain reaction (PCR) based molecular testing: Enables confirmation of infections such as viruses, Legionella, and Pneumocystis jirovecii, and also detects Mycobacterium tuberculosis as well as rifampin resistance through rpoB gene mutation analysis.

e. Antigen testing: Includes assays for Cryptococcus neoformans capsular antigen and galactomannan (GM), which are critical for diagnosing cryptococcosis and aspergillosis, respectively.

f. mNGS: mNGS is theoretically capable of identifying any pathogen within its reference database and encompasses both DNA and RNA sequencing. However, due to its high cost, technical complexity, and challenging interpretation, it should not replace conventional diagnostic methods. Instead, its use is particularly valuable for immunocompromised hosts, cases with negative conventional test results, critically ill patients, or those with rapid disease progression^[88]. Furthermore, when an RNA viral infection is suspected, an mNGS approach that includes RNA sequencing is warranted.

g. tNGS: This method uses pre-capture and enrichment of pathogen-specific nucleic acids prior to high-throughput sequencing and bioinformatics analysis, thereby increasing pathogen read counts and improving detection sensitivity. Compared to mNGS, tNGS offers a shorter turnaround time, reduced sequencing data volume, and lower cost. However, mNGS is recommended when rare or novel pathogens not covered by tNGS panels are suspected^[89,90].

(2) Cytological and immunological analysis

a. Cell Differential Count: Determines the proportional distribution of cell types, including neutrophils, lymphocytes, eosinophils, basophils, and macrophages.

b. Special stains: Utilizes histochemical techniques for specific diagnoses: Periodic acid-Schiff (PAS) stain for secondary pulmonary alveolar proteinosis; India ink stain for Cryptococcus infection; and Grocott’s methenamine silver (GMS) stain for Pneumocystis jirovecii and other fungi.

c. Flow cytometry: Provides immunophenotypic data, such as the CD4⁺/CD8⁺ lymphocyte ratio, to support the diagnosis of conditions like sarcoidosis and CIP^[91,92].

d. Rapid on-site evaluation (ROSE): ROSE is a technique that involves the immediate preparation, staining, and microscopic assessment of specimens obtained during endoscopic procedures via aspiration, biopsy, or brushing. Typically using Diff-Quik staining, ROSE enables rapid differentiation between benign and malignant lesions. Beyond this binary distinction, ROSE provides critical intraoperative guidance through the following functions: (1) Rapid recognition of characteristic inflammatory patterns such as suppurative inflammation, granulomatous inflammation, organizing pneumonia, eosinophilic pneumonia, and diffuse alveolar damage^[93]; (2) Immediate detection of suspected pathogens^[93,94]; (3) Real-time assessment of specimen adequacy, determining whether sufficient diagnostic material has been obtained^[93].

e. Cytopathological examination: Aims to identify malignant cells, aiding in the diagnosis and assessment of neoplastic processes.

(1) Routine pathology

a. Hematoxylin and eosin staining (H&E): The cornerstone histological method for evaluating tissue architecture, characterizing inflammatory patterns (e.g., suppurative inflammation, diffuse alveolar damage, granulomas), distinguishing infectious from non-infectious etiologies, and detecting malignant cells.

b. Special stains: The selection of special stains is guided by initial findings on H&E-stained sections. For instance, an acid-fast stain is used to detect Mycobacterium tuberculosis; a GMS stain aids in identifying fungi such as Pneumocystis jirovecii; an India ink preparation can reveal the capsule of Cryptococcus spp.; and Masson’s trichrome stain is employed to visualize fibrotic changes.

(2) Molecular testing: Molecular testing includes PCR-based assays and mNGS. PCR-based assays are used, for example, to detect Mycobacterium tuberculosis and rifampin resistance-associated rpoB gene mutations. Regarding mNGS separately, testing performed on TBLB or lung biopsy specimens demonstrates sensitivity and specificity comparable to BALF. Nonetheless, due to the greater procedural complexity and risks involved, tissue samples should not be the first-choice specimen. They can, however, serve as a valuable alternative or supplementary option when BALF is inconclusive or unavailable^[95,96].

(3) Immunohistochemistry: Immunostaining with markers such as CD68 and CD3 aids in characterizing inflammatory cell populations. It is also widely used to determine tumor lineage and subtype using markers including TTF-1, Napsin A, P40, estrogen receptor (ER), progesterone receptor (PR), etc.

Diagnosing infection requires integrating visual cues, cytological patterns, and microbiological evidence.

Bronchoscopic findings: Under white light, infectious inflammation typically presents with diffuse mucosal hyperemia, edema, and often intraluminal purulent secretions. Autofluorescence imaging displays large, poorly demarcated areas of pink or dark red abnormal fluorescence, while narrow-band imaging reveals uniformly thickened and tortuous mucosal capillaries in a “bushy” pattern. CLE further demonstrates alveolar exudates, thickened alveolar fibers, and disorganization of the elastin network. In specific cases, CLE allows for direct visualization of filamentous branching fluorescent structures, possibly indicating Aspergillus hyphae^[97,98].

Cytological and histopathological findings

BALF cytology: An elevated neutrophil count (> 3%) is commonly seen in acute bacterial or fungal infections. An elevated lymphocyte count (> 15%) may suggest viral pneumonia or tuberculosis. Eosinophilia (> 1%) is frequently associated with parasitic infections^[64].

Histopathology: The observation of pathogens (such as fungal hyphae, spores, or acid-fast bacilli) or characteristic infection-related pathological changes (e.g., caseating necrotizing granulomas, suppurative inflammation) in tissue specimens provides direct diagnostic evidence for an infectious disease.

(3) Microbiological evidence:

a. Pathogenic microorganisms: detection of organisms such as Yersinia pestis, Bordetella pertussis, Mycobacterium tuberculosis, Burkholderia mallei, and Talaromyces marneffei via culture, PCR, or mNGS strongly suggests their role as the causative agent^[88].

b. Opportunistic pathogens: for organisms including Pseudomonas aeruginosa, Staphylococcus aureus, Streptococcus pneumoniae, nontuberculous mycobacteria, Cryptococcus neoformans, Pneumocystis jirovecii, Cytomegalovirus (CMV), Nocardia species, and Aspergillus species, determining clinical significance requires comprehensive assessment of host factors, medication history, imaging, and treatment response.

c. Colonizing organisms: oropharyngeal flora (e.g., anaerobic bacteria, actinomycetes, Corynebacterium pseudodiphtheriticum) generally have a low likelihood of causing infection but may become pathogenic in mixed infections such as aspiration pneumonia or lung abscess. Common CAP and HAP pathogens are listed in Table 2.

Non-infectious etiologies, primarily treatment-related pneumonitis and malignancy, are often diagnoses of exclusion characterized by specific interstitial patterns.

(1) Bronchoscopic findings

a. Interstitial Patterns on CLE: CLE findings for CIP, targeted drug-induced pneumonitis, or radiation pneumonitis correspond to various interstitial pneumonia patterns: an organizing pneumonia (OP) pattern features increased density and focal consolidation with slender, heterogenous but non-distorted alveoli; a non-specific interstitial pneumonia (NSIP) pattern shows marked increased density with architecturally preserved but thickened alveolar walls and a characteristic “glossy” crystalline coating; a chronic hypersensitivity pneumonitis (CHP) pattern involves consolidation with mild structural distortion; and a fibrotic pattern exhibits severe architectural distortion and destruction, appearing as a “hair-curler” or “corkscrew” pattern in severe cases^[99,100].

b. Malignancy: the hallmark of malignant tumors on CLE is the presence of clusters of enlarged, dark polygonal cells accompanied by the distinctive phenomenon of synchronous cell streaming^[101].

(2) Cytological and histopathological findings

a. Cytology: a BALF lymphocytosis (> 15%) is a common feature in CIP and sarcoidosis. Eosinophilia (> 1%) is frequently associated with drug-induced lung injury^[64].

b. CIP pathology: mechanisms are not fully understood. Histopathological manifestations are diverse, including OP, acute lung injury, fibrosis, non-caseating granulomas, and diffuse alveolar damage^[25,102,103]. Given the lack of specific features, routine lung biopsy is not generally recommended; diagnosis relies on clinical assessment and exclusion of infection. Biopsy aids diagnosis in challenging cases.

c. Radiation pneumonitis pathology: changes range from acute-phase findings (inflammatory infiltration, endothelial swelling, fibrin thrombus) to chronic-phase sequelae (fibrosis, vascular distortion)^[104,105]. Like CIP, routine biopsy is not recommended due to non-specific features; diagnosis is based on radiotherapy history and imaging within the radiation field.

d. TKI-induced pneumonitis pathology: a spectrum of histopathological patterns has been documented, such as organizing pneumonia, hypersensitivity pneumonitis, diffuse alveolar damage, pulmonary eosinophilia, and nonspecific interstitial pneumonia^[106,107]. Like CIP and radiation pneumonitis, TKI-induced lung injury lacks distinctive pathological features; thus, routine lung biopsy is not indicated for diagnosis. Diagnosis should instead rely on a documented history of TKI exposure, supported by consistent clinical and imaging manifestations, after comprehensive exclusion of alternative etiologies including infection and heart failure. In diagnostically challenging situations, lung biopsy may be utilized as an adjunct to clarify the diagnosis.

Regardless of etiology, accurate interpretation depends on sample quality and a structured diagnostic approach.

(1) Quality control

A qualified BALF sample is defined by cellular criteria: squamous epithelial cells < 1% and columnar epithelial cells < 5%. A squamous proportion > 5% signifies upper respiratory contamination. If criteria are not met, re-collection should be pursued; otherwise, results must be interpreted with caution.

(2) Interpretation of negative or discordant results

a. Negative results: negative smear/culture/mNGS should first raise consideration for non-infectious etiologies. If suspicion for infection remains high, a comprehensive reassessment (history review, repeat sampling, multi-site mNGS) is warranted.

b. Discordant results: clinicians should avoid relying on a single test. Assessment must weigh method sensitivity (e.g., mNGS vs. culture difficulties for fungi/TB), microbial load, prior antibiotics, and specimen quality. The final distinction between true infection, colonization, and contamination requires integrating immune status, radiographic features, and treatment response.