Hemoptysis is one of the most common clinical problems for which bronchoscopy is indicated [7, 8]. Whether the patient complains of blood streaking or massive hemoptysis (expectoration of greater than 600 mL in 48 hours) [9], bronchoscopy should be considered to localize the site of bleeding and to diagnose the cause. Localization of the site of bleeding is crucial if definitive therapy such as surgery becomes necessary, and it is also useful to guide angiographic procedures. Bronchoscopy performed within 48 hours of the time that bleeding stops is more likely to localize the site of bleeding (34% to 91%) compared with delayed bronchoscopy (11% to 52%) [10]. Bronchoscopy is more likely to identify a bleeding source in patients with moderate or severe hemoptysis [11]. Clinical judgment must dictate whether and when bronchoscopy is indicated. For instance, it is not indicated in patients with obvious pulmonary embolism with infarction. Whenever patients have an endotracheal or tracheostomy tube in place, hemoptysis should always be evaluated, because it may indicate potentially life-threatening tracheal damage. Unless the bleeding is massive, a flexible bronchoscope, rather than a rigid bronchoscope, is the instrument of choice for evaluating hemoptysis.

In patients with diffuse pulmonary disease, the clinical setting influences the choice of procedure. When diffuse pulmonary infiltrates suggest sarcoidosis, carcinomatosis, or eosinophilic pneumonia, transbronchoscopic lung forceps biopsy should be considered initially because it has an extremely high yield in these situations. Transbronchial lung biopsy has a low yield for the definitive diagnosis of inorganic pneumoconiosis and pulmonary vasculitides [12]; when these disorders are suspected, surgical lung biopsy is the procedure of choice. In the case of pulmonary fibrosis and acute interstitial pneumonitis, transbronchial biopsy does not provide adequate tissue for a specific histologic diagnosis, although by excluding infection the procedure may provide sufficient information to guide therapy.

When an infectious process is suspected, the diagnostic yield depends on the organism and the immune status of the patient. In immunocompetent patients, bronchoalveolar lavage (BAL) has a sensitivity of 87% for detecting respiratory pathogens [13], and a negative BAL quantitative culture has a specificity of 96% in predicting sterile lung parenchyma. In bone marrow transplant recipients with pulmonary complications, BAL is diagnostic in 31% to 66% of patients [14]. BAL has a relatively poor sensitivity for detecting fungal infections in this population (40%) [15]. Transbronchial biopsy adds little additional information with respect to infectious processes, with a yield of less than 10% in bone marrow recipients [15,16 and 17]. Although noninfectious causes of pulmonary infiltrates may be identified by transbronchial biopsy, findings are nonspecific in the great majority of cases and must be weighed against a major complication rate of 7% [16]. In AIDS patients, the sensitivity of lavage or transbronchial lung biopsy for identifying all opportunistic organisms can be as high as 87% [18, 19]. Transbronchial biopsy adds
significantly to the diagnostic yield in AIDS patients and may be the sole means of making a diagnoses in up to 24% of patients, including diagnoses of Pneumocystis jiroveci, Cryptococcus neoformans, Mycobacterium tuberculosis, and nonspecific interstitial pneumonitis [20]. Lavage alone may have a sensitivity of up to 97% for the diagnosis of P. jiroveci pneumonia [21]. However, because induced sputum samples can also be positive for P. jiroveci in up to 79% of cases [21], induced expectorated sputum, when available, should be evaluated first for this organism before resorting to bronchoscopy.

In patients exposed to smoke inhalation, flexible nasopharyngoscopy, laryngoscopy, and bronchoscopy are indicated to identify the anatomic level and severity of injury [22, 23]. Prophylactic intubation should be considered if considerable upper airway mucosal injury is noted early; acute respiratory failure is more likely in patients with mucosal changes seen at segmental or lower levels [24]. Upper airway obstruction is a life-threatening problem that usually develops during the initial 24 hours after inhalation injury. It correlates significantly with increased size of cutaneous burns, burns of the face and neck, and rapid intravenous fluid administration [25].

Flexible bronchoscopy has a high yield in the evaluation of patients after major chest trauma. Patients may present with atelectasis, pulmonary contusion, hemothorax, pneumothorax, pneumomediastinum, or hemoptysis. Prompt bronchoscopic evaluation of such patients has a diagnostic yield of 53%; findings may include tracheal or bronchial laceration or transection (14%), aspirated material (6%), supraglottic tear with glottic obstruction (2%), mucus plugging (15%), and distal hemorrhage (13%) [26]. Many of these diagnoses may not be clinically evident and require surgical intervention.

Flexible bronchoscopy can identify a disrupted suture line causing bleeding and pneumothorax [27] following surgery and an exposed endobronchial suture causing cough [28].

When a nasotracheal or orotracheal tube of the proper size is in place, the balloon can be routinely deflated and the tube withdrawn over the bronchoscope to look for subglottic damage. The tube is withdrawn up through the vocal cords and over the flexible bronchoscope and glottic and supraglottic damage sought. This technique may by useful after reintubation for stridor, or when deflation of the endotracheal tube cuff does not produce a significant air leak, suggesting the potential for life threatening upper airway obstruction when extubation takes place. In patients on long-term ventilatory assistance with cuffed tracheostomy tubes, flexible bronchoscopy can help differentiate aspiration from tracheoesophageal fistula. With the bronchoscope in the distal trachea, the patient is asked to swallow a dilute solution of methylene blue. The absence of methylene blue in the trachea and its presence leaking around and out of the tracheostomy stoma provide accurate evidence of a swallowing abnormality and the absence of a tracheoesophageal fistula.

Clinicians’ ability to determine the probability of ventilator-associated pneumonia (VAP) is very limited, with a sensitivity of only 50% and a specificity of 58% [29]. Quantitative cultures obtained via bronchoscopy may thus play an important role in the diagnostic strategy. Quantitative cultures of BAL fluid and protected specimen brush (PSB), with thresholds of 10⁴ colony forming units (cfu)/mL and 10³ cfu/mL respectively, are most commonly employed prior to the initiation of antimicrobial therapy. Cultures of bronchial washings do not add to the diagnostic yield of quantitative BAL culture alone [30]. For a brief description of how to perform BAL and obtain PSB cultures, see the Procedure Section below.

For BAL, an evidence-based analysis of 23 prior investigations yields a sensitivity of 73% and a specificity of 82%, indicating that BAL cultures fail to diagnose VAP in almost one fourth of all cases [31]. A similar analysis of PSB cultures indicates a very wide range of results, with a sensitivity of 33% to greater than 95% and a median of 67%, and a specificity of 50% to 100% with a median of 95% [32, 33]. PSB is thus more specific than it is sensitive, and negative results may not be sufficient to exclude the presence of VAP [34]. Blind protected telescoping catheter specimens perform similarly to bronchoscopically directed PSB cultures [35, 36]. It is critical to note that colony counts change very quickly with antibiotic therapy. Within 12 hours of starting antibiotic therapy, 50% of all significant bacterial species initially identified in significant numbers had colony counts reduced to below the “pathogenic” threshold level. After 48 hours of therapy, only 14% of isolates are still present above threshold values [37]. It is therefore essential to obtain quantitative cultures before starting or changing antibiotics.

Despite the greater accuracy of quantitative bronchoscopic cultures, prospective randomized trials of early invasive diagnostic strategies employing bronchoscopy and quantitative lower respiratory tract cultures for VAP have not demonstrated significant advantages in mortality or other major clinical endpoints [38]. Recent evidence-based consensus guidelines for the management of VAP recommend obtaining quantitative or semiquantitative lower respiratory tract cultures when VAP is suspected, with subsequent management based on culture results and clinical response to therapy [39]. Further investigations will be required to clarify the role of quantitative cultures in management protocols.