Physical Examination, Complementary Tests, and Functional Status Assessment

This patient had symptomatic idiopathic subglottis stenosis of moderate severity with a stenotic index* of 60% based on computed tomography (CT) measurements. The severity, extent, and possibly the morphology (shape) of the airway narrowing determine the magnitude of ventilatory impairment and the likelihood of response to a particular treatment strategy.¹ Cutoff values have been proposed to qualify the degree of narrowing as mild (<50%), moderate (51% to 70%), or severe (>71%).¹ Rather than relying on a subjective assessment of severity made at the time of white light flexible bronchoscopy, CT and morphometric bronchoscopy have been used to obtain objective measures that might assist in classifying stricture severity. With this latter technique, still images are captured with the bronchoscope in the center of the airway lumen proximal, distal, and directly at the level of the target abnormality while the tip of the scope is kept at a constant distance from the target area.¹ Images are then saved on CD-ROM or USB device. Cross-sectional area (CSA) of the airway is calculated by using image processing software such as the ImageJ analysis program (National Institutes of Health, Bethesda, Md; available free of charge at http://rsb.info.nih.gov/ij/) (Figure 7-2). As of this writing, however, it is unclear whether morphometric bronchoscopy measurements correlate with subjective estimations of airway narrowing, and whether objective measurement assessments of airway lumen impact decision making and outcomes. Relying on morphometric bronchoscopic measurements also has its disadvantages. For example, in our patient, the method could not be applied because normal tracheal lumen beyond the stricture could not be visualized. Passing the bronchoscope (outside diameter, 5.2 mm) through the stenotic lesion could have caused mucosal edema, worsened respiratory distress, and potentially precipitated respiratory failure. In our case, therefore, the degree of airway narrowing was calculated on the basis of CT images obtained at end inspiration.

Figure 7-2 Manual method of morphometric bronchoscopy using the ImageJ analysis program. The region of interest (ROI) manager function can be used to measure the area that is selected using the “polygon selection” function (the operator selects the area to be measured). The airway lumen cross-sectional area is measured in pixels.

White light bronchoscopy (WLB) is very useful in the evaluation of subglottic stenosis because it can detect inflammation and identify hypertrophic stenotic tissues. The airways, however, cannot be studied in cross-section; therefore WLB cannot be used to accurately assess the depth of hypertrophic tissues that make up the circumferential stricture, or to judge the integrity of cartilaginous airway structures. For this purpose, high-frequency endobronchial ultrasonography (EBUS) is useful because the radial EBUS (20 MHz) identifies hypoechoic and hyperechoic layers that correlate with laminar histologic structures of the central airways.² In fact, structural central airway wall abnormalities have been identified in patients with Wegener’s granulomatosis, tuberculosis, relapsing polychondritis, lung cancer,* or compression by vascular rings, and in patients with excessive dynamic airway collapse.^2–4 Endobronchial ultrasound examination using a 20 MHz radial probe can be performed at the time of initial diagnosis or at the time of treatment during flexible or rigid bronchoscopy to visualize airway wall structures at the level of stenosis and potentially to guide treatment decisions.⁵ For instance, for patients who are not surgical candidates, or if tracheal surgical expertise is not available, a simple stricture characterized solely by hypertrophic fibrotic tissue can be successfully dilated (with or without laser assistance) and will not require a stent; on the other hand, for a complex stenosis, in which the cartilage is destroyed, dilation alone will not be long-lasting, making stent insertion almost obligatory to maintain airway rigidity and patency.⁶

In the absence of obvious cartilaginous collapse, however, it is impossible to assess the integrity of the cartilage by WLB alone if prominent hypertrophic stenotic tissue is present (Figure 7-3). High-frequency endobronchial ultrasound, with its high resolution, allows visualization of stenotic tissue and cartilaginous structures (see Figure 7-3). In idiopathic tracheal stenosis, histologic studies of resected specimens showed integrity of the cartilage; in complex stenoses such as those after tracheostomy or intubation, histologic evidence revealed partial or total destruction of cartilage.⁷ Abnormal cartilaginous structures may be identified by EBUS (see Figure 7-3). Furthermore, results from EBUS can help determine the extent to which a lesion might be dilated or ablated by helping avoid damage to normal cartilage and damage to the peribronchial blood vessels^5,8 during dilation or bronchoscopic resection. EBUS can also be used to measure the diameters of both the normal airway and the stenosis, thereby potentially facilitating accurate assessment of the degree of narrowing and stent size selection.*⁸ EBUS is a minimally invasive procedure performed under moderate sedation or general anesthesia that adds extra time to a diagnostic bronchoscopic procedure.

Figure 7-3 White light bronchoscopy and high-frequency endobronchial ultrasound (EBUS) images in tracheal stenosis. A, Post intubation tracheal stenosis showing a circumferential stricture. B, The corresponding EBUS image shows hypertrophic tissues and the disrupted calcified cartilage. C, Idiopathic subglottic stenosis seen on white light bronchoscopy. D, The corresponding EBUS image reveals prominent hypertrophic tissues and the integrity of the cartilage.

Experimental studies of nonpulmonary tissues have shown that a novel optical technology, optical coherence tomography (OCT), may detect laser-induced tissue changes.⁹ Because laser-induced thermal injury disrupts the normal optical properties of tissues, OCT is capable of visualizing architectural features of the airway wall and may be a useful tool to define therapeutic target volumes in situ and to monitor tissue coagulation, cutting, and ablation intraoperatively. This may result in subsequent reduction of the iatrogenic collateral airway wall injury well described in experimental studies. The physical principles of OCT are analogous to ultrasound: When a beam of sound or light is directed onto a tissue, it is back-reflected (backscattered) differently from structures that have different acoustic or optical properties. The principal difference between ultrasound and OCT is that the speed of light (3 × 10⁸ m/s) is many orders of magnitude faster than that of sound (1500 m/s). With OCT, light is emitted from the source, directed into tissues, and reflected off internal structures. The longer the distance traveled, the longer the delay in returning to a detector. The delay in returning light from deeper structures compared with shallow structures is used to reconstruct images. With time domain OCT systems (currently commercially available for clinical use), in-depth profiling is performed by measuring echo time delay and the intensity of backscattered or reflected light.* The system we intended to use in this patient measured OCT echo time delays by comparing the backscattered or back-reflected light signal versus a controlled reference signal. The resolution of OCT is much higher than that of ultrasonography, and in studying central airway wall microstructures, it is possible to visualize upper airway wall layers (mucosa and submucosa) but not the entire human cartilage, because the depth of penetration in tissues is approximately 1.7 mm.¹⁰ OCT systems are also used during bronchoscopy (flexible or rigid) to prolong diagnostic and therapeutic procedures. In our patient, we considered that OCT could become part of our minimally invasive armamentarium for evaluating patients with tracheal stenosis. We therefore planned to use this technology as part of an internal review board–approved research protocol.

Another imaging technique, which is noninvasive, is vibration response imaging (VRI).^† This technology requires minimal patient cooperation and can be repeated as often as necessary.¹¹ VRI has been used in the evaluation of patients with asthma, chronic obstructive pulmonary disease (COPD), foreign body aspiration, and central airway obstruction.¹¹ Results from experimental studies suggest that different sound frequencies are generated by airways of different sizes, such that differential analysis of VRI might allow precise localization of pathologic processes in different compartments of the lung.^11,12 This is important in patients with central and concurrent peripheral airway obstruction, such as those with asthma/COPD and tracheal or subglottic stenosis. For instance, when a patient with COPD and post intubation tracheal stenosis develops nonspecific respiratory symptoms (cough, dyspnea, inability to raise secretions), VRI might be able to noninvasively localize the pathogenic process to the peripheral airway (e.g., COPD) or the central airway (e.g., tracheal stenosis), thus potentially avoiding the need for diagnostic computed tomography or bronchoscopy. This technology, however, was not available to us at the time of this patient’s evaluation.

With regard to physiologic studies, in this patient, the flow-volume loop was characteristic of fixed upper/central airway obstruction, revealing a “square” pattern caused by limitation of both inspiratory and expiratory flow. Published findings demonstrate absence of correlation between severity of obstruction as determined by the flow-volume loop and symptoms,¹³ or between spirometry-derived indices and radiologic assessment of airway obstruction.¹⁴ Furthermore, spirometry and flow-volume loops do not localize the anatomic level of the obstruction. Impulse oscillometry (IOS) can be used to localize the region of flow limitation in patients with concurrent central and peripheral airway obstruction. IOS is an effort-independent test during which brief pressure pulses of 5 to 35 Hz are applied during tidal breathing. Pressure-flow oscillations are superimposed on the subject’s tidal breaths, and real-time recordings are used to provide an estimate of total respiratory system impedance, including measurements of resistance* (R) and reactance^† (X) at different frequencies that might differentiate between central and peripheral components of airway obstruction.¹⁵ Resistance of lung tissue, caused by its viscoelastic properties, decreases with increasing oscillation frequency and becomes essentially negligible at 5 Hz. This means that resistance at 5 Hz relates mainly to flow resistance in the tracheobronchial tree. Increased R at a low oscillation frequency (5 Hz) reflects an increase in total respiratory resistance suggestive of airway obstruction. This is noted, for example, in patients with COPD. In these patients, no flow dependence of resistance is observed in the absence of upper or central airway obstruction.¹⁶

By contrast, the occurrence of upper/central airway stenosis is expected to generate increases in flow dependence of resistance,^‡ both in normal subjects and in COPD patients with peripheral airway obstruction.¹⁷ With an airway of decreasing size, a gradual increase in resistance is noted, as well as a gradual increase in the flow dependence of resistance (ΔR/ΔV). Furthermore, an increase in resistance at a higher frequency (20 Hz) reflects specifically increased central airway resistance.¹⁸ The IOS maneuver, which provides the advantages of requiring only passive cooperation during tidal breathing and does not cause respiratory fatigue,¹⁵ may in fact be more sensitive than spirometry for detecting upper/central airway stenosis. In one study, for eight patients who could be assessed after bronchoscopic intervention, R values were lower than before the intervention, but only one patient showed a post intervention R value within normal limits. By contrast, in 6 of 8 patients, post intervention ΔR/ΔV (flow dependence of resistance) values fell within normal limits, suggesting that the flow dependence of R (ΔR/ΔV), which is more specific of upper/central airway obstruction, should also be considered during examination of patients with tracheal stenosis and/or for whom R is expected to be increased owing to peripheral airway obstruction.¹⁶ This is probably the case for patients with concurrent stenosis and smoking histories, COPD, or asthma when it is otherwise unclear which part of R is due to tracheal obstruction and which part arises from more peripheral airway obstruction. At the time our patient was seen, IOS was not available at our facility. In view of the patient’s stridor and dyspnea, we chose to forgo additional testing and proceeded instead with rigid bronchoscopic dilation.

Comorbidities

The patient had uncontrolled “asthma” on inhaled corticosteroids and rescue albuterol, which had offered only partial relief. No other comorbidities interfered with general anesthesia or bronchoscopy.

Support System

The patient was married, and her husband was very involved in her care. In fact, he had expressed his concerns that the research project into which we intended to enroll his wife was just meant to advance the careers of treating team members and had nothing to do with his wife’s care. Indeed, it is not uncommon to encounter such opinions; surveys show that about one third of respondents among cancer trial participants, for example, and a quarter of nonparticipants fully or partially agreed with the statement that medical research is performed primarily to promote doctors’ careers.¹⁹

Patient Preferences and Expectations

The patient herself believed that our main motive in conducting medical research was our wish to find new treatments or tests that would help other patients like her; in general, this is the opinion of most participants in both cancer and noncancer trials.²⁰ Data on patients’ attitudes toward research in benign tracheal stenosis are not yet available. However, bioethicists believe that biomedical knowledge is a public good (i.e., it is available to any individual even if that individual does not contribute to it); therefore participation in biomedical research is an important way to support the public good, suggesting, at least from a utilitarian perspective, that we all have the duty to participate in it,*²¹ assuming that the risks of participation are not excessive. Another reason why it might be morally wrong to refuse to participate is that from a principle-based ethic (i.e., beneficence), if a person can prevent something bad or can produce something good, then that person has a duty to perform that action. In fact, failure to participate in research could be considered a form of free-riding, in other words, similar to when a person receives a benefit that others pay for, thereby taking advantage of those who contributed but refusing to share the burden of obtaining it.²¹ Of course, the obligation to participate in research is not absolute or legally mandated, and circumstances or reasons may prevail or diminish the force of duty.^†

The purpose of our research protocol was to explore whether novel acoustic and optical imaging technologies could complement traditional diagnostic studies such as WLB and CT, by identifying in vivo airway wall changes before and after laser-assisted dilation of circumferential tracheal strictures. This study was approved by the Institutional Review Board (IRB). Disclosure of IRB approval may be important in helping patients choose to participate in research, as demonstrated in a study from Denmark showing that most patients stated that research ethics committee oversight had had an impact on their decision to participate.²⁰ The fact that a clinical study has been reviewed and approved by an ethics committee apparently gives patients a sense of security and increases their willingness to participate. Sharing such information with patients is essential during the informed consent process. Well-functioning IRBs ensure that patient risks will not be excessive relative to the benefits of participating in the research endeavor. A patient’s moral obligation to participate may be weaker if potential risks to the individual far outweigh potential expected benefits of the research project.

Indications

Bronchoscopic procedures or open surgery is clinically indicated to restore satisfactory airway lumen patency and improve symptoms.^22,23 This patient had symptomatic stenosis. Improving airway patency to less than 50% narrowing could lead to alleviation of her exertional dyspnea.¹⁷

Contraindications

No absolute contraindications to rigid bronchoscopy were identified. No history of heart or lung disease was elicited that could have potentially increased the risk for hemodynamic instability during general anesthesia. No history of renal or liver disease was revealed that would have increased the risk for bleeding.

Expected Results

The morphology of the stricture was circumferential. In this setting, results from studies show that a second endoscopic intervention is required more often than for patients with strictures due to purely eccentric localized hypertrophic tissue.^24,25 In patients with idiopathic subglottic stenosis, however, the therapeutic success of rigid bronchoscopic interventions is variable. In addition to morphology, a major factor impacting success includes stenoses less than 1.0 cm in length and not associated with significant loss of cartilage. In one series, patients were treated bronchoscopically; 60% of them required a second procedure, with a mean interval between procedures of 9 months.²⁵ Similar results have been reported in several other nonrandomized studies of bronchoscopic treatments.^26–28

Team Experience

In our institution, second to malignant central airway obstruction, benign tracheal strictures are the most common diagnoses prompting interventional bronchoscopic procedures. Our patients are managed within the context of a multidisciplinary airway disease program and are routinely discussed with surgical colleagues before bronchoscopic treatments are initiated.

Therapeutic Alternatives

Simple mechanical dilation (with balloons, bougies, or rigid bronchoscope), CO₂ or potassium-titanyl-phosphate (KTP) laser–assisted mechanical dilations, and surgical interventions such as laryngotracheoplasty were discussed with the patient and her husband. Surgical treatment consisting of cricotracheal resection and primary thyrotracheal anastomosis is considered by some to be the procedure of choice for the treatment of severe idiopathic subglottic stenosis (>70% luminal obstruction).²⁹ Our patient had a lesser degree of obstruction and refused open surgery because the success rate of bronchoscopic interventions was acceptable to her. We did not offer her stent placement because of the known frequency of stent-related complications when stents are placed in the subglottic regions (the proximal end of the stent may induce ulceration and granulation tissue formation with subsequent glottis or subglottic restenosis), and because complications from stent insertion might increase the complexity of a stricture or increase the length of the stenotic segment. These two factors might adversely affect open surgical resection at a later date.

Cost-Effectiveness

To our knowledge, no randomized studies have evaluated outcomes or the cost-effectiveness of endoscopic versus open surgical interventions for this disease.

Informed Consent

After they had been advised of all therapeutic alternatives, the patient and her husband elected to proceed with rigid bronchoscopy under general anesthesia. A separate informed consent was obtained for the research protocol.

Anesthesia and Perioperative Care

In general, when the patient is taken to the operating room, the treating team verifies the patient’s name using at least two identifiers (in addition to reading the patient’s name bracelet) and the written history and physical and the informed consent form, thereby ensuring that the patient, the surgeon, and the procedure being performed are correctly identified. The patient or caretaker is asked to confirm the type of procedure being performed. The individual responsible for this communication correlates site marking (when necessary), consent, printed operating room schedule, and history and physical with the patient’s name. If any part is not matching, the process is stopped until it is validated.* In our institution, the person performing the procedure must document in the medical record the appropriate procedure before the procedure takes place.

Many institutions also apply the so-called time-out or preprocedure pause process. This is performed immediately before the procedure for the purpose of preventing medical error by conducting a final verification of correct patient, procedure, and site (when applicable).*