SEPTIMIU MURGU and HENRI COLT

Overview

Convex probe endobronchial ultrasound-guided transbronchial needle aspiration is used for diagnosing (and staging) intrathoracic lymphadenopathy in lung cancer,¹ other intrathoracic tumors,² and lymphadenopathy from extrathoracic malignancies.³ The aspiration yield for benign causes of intrathoracic lymphadenopathy is less than with cancer, mostly because of lack of tissue architecture in cytology specimens, but sarcoidosis, and bacterial, mycobacterial, and fungal infections can be diagnosed.^4–6 High-frequency endobronchial ultrasound (EBUS), using a radial scanning probe, is used to diagnose peripheral pulmonary nodules⁷ and to define airway wall structures in tracheal stenosis, tracheobronchomalacia, excessive dynamic airway collapse, and tumor invasion.^8–10

Extraluminal vascular structures (superior vena cava, azygos vein, main pulmonary artery and its branches, ascending and descending aorta, aortic arch, pulmonary veins, and left atrium) are visualized using convex probe EBUS (see Figure 25-5 in the imaging case box). Bronchoscopists can assess volume status, pericardial fluid, or the presence of pulmonary embolism.^11,12 Thoughtful resource use for intensive care unit (ICU) admissions¹³ warrants that EBUS should be considered in critically ill patients with suspected lung, mediastinal, cardiovascular, and hemodynamic disorders. This chapter illustrates potential applications of EBUS and EBUS–transbronchial needle aspiration (TBNA) in patients admitted to the ICU or undergoing general anesthesia. Moreover, we suggest algorithms built on the strengths of EBUS diagnostic abilities, its minimally invasive nature, and its proven safety profile for the noncritically ill patient setting (based on our own institutional experience).

Intrathoracic lymphadenopathy

Many patients admitted to the ICU demonstrate enlarged intrathoracic lymph nodes on computed tomography (CT) scans performed for a variety of reasons. Sampling lymph nodes could assist in management. For example, patients with hypoxemic respiratory failure and parenchymal or interstitial infiltrates may have associated lymphadenopathy of variable significance. Although a patient with pneumonia likely has reactive lymphadenopathy, needle aspiration could reveal the actual organism. Patients with sarcoidosis and anthracosis have characteristic findings on EBUS-TBNA specimens (granulomatous reaction and anthracotic histiocytes, respectively).^4,14

EBUS-TBNA can make the diagnosis of primary lung cancer or extrathoracic malignancy, metastatic to mediastinal, and hilar lymph nodes. It allows staging or restaging of patients with known or suspected malignancy.1–3 This information is used to redefine level of care, ICU priority scores, and guide treatment. A patient with cancer on mechanical ventilation, for example, may be at risk for transbronchial or transthoracic lung biopsies. EBUS-TBNA can make a diagnosis of progressive disease, infections, or granulomatoid reaction (Figure 25-1).

Extrinsic central airway obstruction, pulmonary artery compression, and superior vena cava syndrome

Extrinsic compression is often amenable to interventional procedures, such as airway stent insertion,¹⁵ so that patients with respiratory failure on mechanical ventilation may be weaned from ventilators.¹⁶ Bronchoscopy and EBUS-TBNA are performed for the dual purpose of diagnosing the central airway obstruction (CAO) (Figure 25-2) and making a cytohistologic diagnosis by sampling impinging masses. Doppler mode is useful to assess the vascularity of the mass and suggests tumor engorgement with potential worsening airway obstruction in the supine position or hypervolemic state. A rapid diagnosis allows prompt referral for stent insertion and systemic therapy. A patient with a large mediastinal cyst could even be drained¹⁷ rather than be referred for open surgery.

Main pulmonary arteries (PA) and the superior vena cava (SVC) can be compressed by mediastinal tumors. This may cause pulmonary hypertension and SVC syndrome, respectively. Hypoxemia and hypercarbia can lead to worsening of PA pressures and cause acute cor pulmonale. Fluid overload and the supine position lead to increased tumor volume and worsened SVC compression, potentially precipitating SVC syndrome and its consequences on hemodynamics and upper airway patency. Convex probe EBUS convex with Doppler mode allows tumor measurement and visualization of the PA, SVC compression, and intratumoral vascularity.

Expiratory central airway collapse, tracheal stenosis, and airway wall tumor invasion

Expiratory central airway collapse (ECAC) is a reported cause of hypercarbic respiratory failure as well as inability to wean from mechanical ventilation. Diagnosis can be made with white light bronchoscopy (WLB), but assessment of airway wall structures requires visualization in cross section (Figure 25-3). A high-frequency (20 MHz) radial probe, used through a bronchoscope with a working channel of 2.8 mm, allows identification of hypoechoic and hyperechoic layers of the airway wall and identifies cartilage abnormalities in diffuse malacia, excessive dynamic airway collapse (EDAC), and tracheal stenosis.^8–10 Cartilage integrity in ECAC, its fracture in posttracheostomy or postintubation stenosis, its invasion by extraluminal tumor, and precise identification of cross-sectional extent of hypertrophic stenotic tissues in tracheal stenosis can impact management. For instance, a patient with ECAC-related respiratory failure might have excessive dynamic airway collapse on WLB but no evidence of cartilage abnormality. This should prompt a search for causes of obstructive ventilatory impairment resulting from small airway disease/parenchymal changes.¹⁸ On the other hand, a patient with diffuse malacia, from relapsing polychondritis, for example,¹⁰ may warrant stent placement to restore airway patency (see Figure 25-3). A patient with cartilage invasion by primary lung tumor might warrant systemic chemoradiotherapy, whereas in tracheal stenosis, if the cartilage is destroyed, the risk of recurrence is high after laser-assisted or balloon dilation. EBUS findings might accelerate appropriate referrals in each of these instances (see Figure 25-3).

Peripheral pulmonary nodules

In critically ill patients, peripheral pulmonary nodules may be caused by primary or metastatic malignancy, or fungal (i.e., coccidioidomycosis, invasive pulmonary aspergillosis, cryptococcus, histoplasmosis), mycobacterial (i.e., tuberculosis), or bacterial septic emboli. They are also seen in patients admitted to the ICU with diagnoses of autoimmune disease, granulomatosis with polyangiitis, or rheumatoid arthritis. If tissue diagnosis is needed, patients on mechanical ventilation are at high risk for pneumothorax (PTX) from transbronchial lung biopsy or CT-guided fine-needle aspiration (FNA). Furthermore, there are dangers transporting critically ill patients outside the ICU. Radial probe EBUS has a diagnostic yield of 50% to 80%, depending on the clinician’s ability to visualize the nodule, the size of the nodule, presence of air bronchus sign on CT scan, the nodule’s distance from the hilum, and the position of the probe in relation to the nodule.¹⁹ Compared with CT-guided FNA, lower rates of PTX and bleeding are described.⁷

Pulmonary embolism

Central pulmonary embolism (PE) (defined as thrombus in the pulmonary trunk or a main or a lobar PA) is diagnosed using contrast-enhanced chest computed tomography (chest CT angiogram). Acute PE with hemodynamic instability, contrast allergy, or severe renal dysfunction may occur and often preclude intravenous contrast administration. Bedside transthoracic echocardiography (TTE) has low sensitivity for confirming PE, although it might reveal right ventricular dysfunction. Pulmonary angiograms or ventilation-perfusion scans require transport to radiology, which is risky in unstable patients. Because the majority of patients who succumb from PE die within 2 hours from the onset of symptoms (and those who do survive may succumb from recurrent PE), a firm diagnosis is warranted. Visualization of pulmonary arteries by using convex probe EBUS is possible because depth of penetration is approximately 5 cm. Vessels run adjacent to the central airways and there is no interposition of ventilated lung parenchyma between airways and vessels.¹² In a prospective, multicenter pilot study of 32 patients with CT angiogram-confirmed, hemodynamically stable PE, CT angiogram documented 101 PEs, of which 97 (96%) were also detected by EBUS.¹² EBUS missed only 4 emboli (1 in a middle lobe and 3 in a left upper lobe artery). There were no EBUS-related complications. The entire main pulmonary trunk, main PAs, the lower lobe arteries, and at least 10 mm of the upper lobe and middle lobe arteries can be examined in just a few minutes (Figure 25-4A and B, see Figure 25-1D). Whether EBUS might worsen hemodynamic instability because of increased intrathoracic pressures during bronchoscopy is unknown. A large endotracheal tube (No. 8.5-9) should probably be used to avoid expiratory flow limitation, which might lead to intrinsic peak end-expiratory pressure (PEEP) and worsening hypotension. Future research in hemodynamically unstable patients may be hampered by the fact that diagnostic bronchoscopy is generally contraindicated in patients with ongoing cardiac ischemia or severe refractory hypoxemia, both of which can occur in the setting of acute massive PE.

Pericardial effusion

EBUS allows visualization of the middle mediastinum and images pericardial effusion around the left atrium (see Figure 25-4). Although most pericardial disorders are diagnosed by TTE or transesophageal echocardiography (TEE), in case of acute hemodynamic instability during general anesthesia or in the ICU, EBUS could be useful because it allows assessment of pulmonary vessels, SVC, and portions of the pericardium. If TTE is suboptimal (obesity, chest deformity, hyperinflation while on positive-pressure ventilation, emphysema, subcutaneous edema or emphysema, lack of patient cooperation, inability to rotate the patient for optimal scanning, presence of surgical dressings) and TEE is not readily available, EBUS can be used to visualize posterior cardiac structures in great detail. A patient with hypotension in whom tamponade is suspected might thus warrant EBUS when TTE is nondiagnostic and/or not readily available. We have seen patients in whom pericardial effusion around the left atrium is clearly identified (see Figure 25-4C and D). In theory, when tamponade physiology affects the left atrium (LA), the respiratory variation in LA chamber size and diastolic collapse could be detected. The single scanning plane and limited chamber visualization, however, limit the benefit of EBUS in patients with pericardial disease. Relevant to patients in the operating room under general anesthesia or in the ICU on positive-pressure ventilation is that a reversed sequence of intrathoracic pressure occurs during the respiratory cycle. Intrathoracic pressure is higher during inspiration, and therefore venous return is smaller. Further experience is needed to better understand anatomic structures in scanning planes before EBUS can be recommended for cardiac assessment.

Volume status

The assessment of fluid responsiveness is relevant to managing patients with acute hypotension. If patients on mechanical ventilation have no significant respiratory variation in the inferior vena cava (IVC) or SVC diameter, it is unlikely that cardiac output will increase with volume expansion, which could actually be harmful. Whereas the diameter of the IVC at the junction with right atrium is measured using the subcostal view on TTE, the SVC diameter measurement is usually performed by TEE.²⁰ Measurement of IVC diameter is less predictive than SVC diameter because the former is unreliable in patients with high intraabdominal pressures, whereas the latter directly reflects the relationship between central blood volume and intrathoracic pressures. Maximal and minimal SVC diameter (measured at the top of the right atrium) are observed during respiration.²⁰ Studies using TEE demonstrate that the collapsibility index of the SVC, defined as [(maximum diameter on expiration − minimum diameter on inspiration)/(maximum diameter on expiration)], is highly predictive of fluid responsiveness. Results show that in septic patients on mechanical ventilation, an SVC collapsibility index above 36% predicted an increase in cardiac index after volume expansion of more than 11%, (specificity 100% and sensitivity 90%).²⁰ We observed that cyclic increase in intrathoracic pressure can induce partial or complete collapse of the vessel seen by EBUS, when the SVC is visualized with the scope in the lower trachea/proximal right main bronchus and oriented anterolaterally toward the right (see Figure 25-4E and F). Using the signal from the ventilator, airway pressures are displayed on the screen of the ultrasound monitor, allowing accurate timing of SVC cyclic changes during the respiratory cycle. Interpretation should be evaluated in the context of the patient’s clinical condition, radiographic findings, degree of hypoxemia, urine output, and fluid balance (see Figure 25-4G).

Conclusions

EBUS is an exciting imaging technique for visualizing mediastinal lymph nodes, major vessels, tumors, and airway structural changes. Its role in critically ill patients in the ICU or operating theater is promising.

Pearls and highlights

References