Interventional Procedure

Inoperable central airway stenosis due to a malignant tumor is a relatively common condition and may be life threatening. Because of the poor prognosis, palliative methods are needed to maintain airway patency. In patients with severe malignant airway stenosis, interventional bronchoscopy is considered as a method of maintaining airway patency [1].

Flow limitation during forced expiration is affected by the relationship between transmural pressure (Ptm) and the cross-sectional area (A) of the airway. The wave speed is dependent on the stiffness of the airway wall, i.e., dPtm/dA and on the cross-sectional airway itself [2, 3]. The flow-limiting segment (FLS) occurs originally where the cross-sectional area of the airway is the narrowest. On the basis of wave-speed concepts of maximal expiratory flow limitation, stenting at the FLS improved expiratory flow limitation by increasing the cross-sectional area, supporting the weakened airway wall and relieving dyspnea [4, 5].

Assessment of Flow–Volume Curve

The location of the FLS is assessed using flow–volume curves. Analysis of the flow–volume curve can be used to define the nature of the stenosis and provide reliable information on the efficacy of stenting [5,6,7,8,9,10]. In patients with tracheal stenosis, the flow–volume curve shows a marked reduction of the expiratory flow (fixed narrowing patterns) with a plateau. In patients with bronchial stenosis, the flow–volume curve shows decreased flow with expiratory choking (initial transient peak flow followed by acute flow deterioration and consecutive low flow, and dynamic collapse patterns). In patients with carinal stenosis, the flow–volume curve shows a descending expiratory limb with a plateau and choking (combined fixed and dynamic patterns). In patients with extensive stenosis from the trachea and carina, extending to the bronchi due to tumor and/or mediastinal lymphadenopathy, the flow–volume curve shows severe reduction of the expiratory flow (complex patterns containing elements of all the former).

Dyspnea

The degree of dyspnea depends on the degree of airway obstruction and becomes severe when well over 70% of the tracheal lumen is obstructed [11]. In cases with 50% tracheal obstruction, the highest velocities are in the jet, which are generated by glottic constriction. In cases with over 70% tracheal obstruction, peak velocities are generated at the stenosis and exceed velocities in the glottic area. Pressure differences changed dramatically from 70% tracheal obstruction.

The relation between the baseline degree of tracheal obstruction and the changes in MMRC (ΔMMRC) is shown in Table 7.1. Any patient with an improvement in the MMRC scale of 2 or more was considered to be a clinical responder. The clinical responder rate was 84.6% for obstructions above 80% and 58.8% for obstructions between 50% and 80%. Preoperation measures by the baseline degree of tracheal obstruction could be used to predict the post-operation impact on dyspnea [12].

Table 7.1 Relation between the baseline degree of tracheal obstruction and the change in MMRC after interventional bronchoscopy
Full size table

Assessment of Lateral Airway Pressure

Analysis of the flow–volume curve could be used in defining the nature of the stenosis. However, flow–volume curves cannot identify the precise location of the lesion where airway resistance increases, nor can it immediately define the outcome of stenting.

With the use of airway catheters in dogs [13,14,15] and in human subjects [16,17,18], the FLS could be located by measuring lateral airway pressure (Plat) during induced flow limitation generated by either an increase in pleural pressure or a decrease in downstream pressure. Healthy subjects have relatively uniform pressure drop down the bronchial tree during expiration. In patients with airway stenosis, the major pressure drop occurs across the stenosis. By measuring Plat on each side of the stenosis, we could detect the pressure difference between two sites (proximal and distal) of the stenotic segment [12].

After intubation, a double lumen airway catheter was inserted into the trachea during bronchoscopy. Plat was measured simultaneously at two points during spontaneous breathing with light general anesthesia before and after intervention. Plat at the two points was plotted on an oscilloscope (pressure–pressure [P–P] curve). The angle of the P–P curve was defined as the angle between the peak inspiratory and expiratory pressure points and the baseline of the angle. If the cross-sectional area (CSA) was small, the angle was close to 0°; however, after intervention, the CSA significantly increased and the angle was close to 45°.

In healthy subjects, no pressure difference between the carina and trachea was observed (0.10 ± 0.22 cm H2O) during tidal breathing (Fig. 7.1a). The P–P curves were linear, and the angle of the P–P curve was close to 45° (44.6 ± 0.98) (Fig. 7.1b).

Fig. 7.1
Two line graphs, A and B, illustrate the typical patterns of lateral airway pressure measurements. Graph A has fluctuating lines for the carina and trachea. Graph B displays a linear curve that passes through the origin.

Typical patterns of lateral airway pressure (Plat) measurements during tidal breathing in a healthy subject. Plat is measured simultaneously at two points (upper trachea and carina). There are no pressure differences between the carina and upper trachea (a) Lateral airway pressure/time curve. (b) Lateral airway pressure/pressure at carina curve. (Blue: carina; Red: upper trachea.) The angle of pressure–pressure (P–P) curve is defined as the angle between peak inspiratory and expiratory pressure points and the baseline of the angle. The P–P curves are linear, and the angle of P–P curve is close to 45° (Fig. 7.2b)

Full size image

In patients with tracheal obstruction, dyspnea scale, pressure difference, and the angle changed significantly beyond 50% obstruction (Fig. 7.2a, b). After stenting, the pressure difference disappeared, and the angle was close to 45°. The degree of tracheal obstruction was significantly correlated with the pressure difference and the angle (r = 0.83, p < 0.0001 and r = −0.84, p < 0.0001, respectively) [12].

Fig. 7.2
Two scatterplots, A and B, display increasing and decreasing trends for fixed pre, fixed post, variable pre, and variable post data, respectively.

Scatter plot of pressure difference and the angle of the pressure–pressure (P–P) curve versus the degree of tracheal obstruction. Blue diamonds show before intervention and red squares indicate after intervention in cases with fixed stenosis. Green triangles show before intervention and purple X’s indicate after intervention in cases with variable stenosis. Dotted line shows the threshold for 50% tracheal obstruction. The pressure difference (a) and the angle of P–P curves (b) are significantly correlated with the degree of tracheal obstruction. The pressure difference increased significantly above 50% obstruction (a). When the cross-sectional area was small, the angle of the P–P curve was close to 0°. After interventional bronchoscopy, the cross-sectional area increased and the angle of the P–P curve was close to 45° (b)

Full size image

This approach identified a need for additional treatment during interventional bronchoscopy. In a patient with fixed intrathoracic stenosis due to tracheal tuberculosis, CT showed a tracheal stenosis at the middle trachea (Fig. 7.3a). Before treatment, a considerable pressure difference between the upper trachea and carina was noted (Fig. 7.3d), and the angle of the P–P curve was 0.3° (Fig. 7.3i). The flow–volume curve shows marked reduction of the expiratory and inspiratory flow (Fig. 7.3g). After balloon dilation, bronchoscopic imaging revealed greater patency for the trachea (Fig. 7.3b). However, the pressure difference only decreased from 36.6 cm H2O to 20.1 cm H2O (Fig. 7.3e), and the angle of the P–P curve only increased from 0.3° to 5.0° (Fig. 7.3i). Subsequently, a silicone Y stent was implanted from the upper trachea to the both main stem bronchus. After stenting (Fig. 7.3c), pressure differences disappeared (Fig. 7.3f) and the angle of the P–P curve increased from 5.0° to 35.8° (Fig. 7.3i). The MMRC scale decreased from 2 to 0, and flow–volume curve returned to a near normal pattern (Fig. 7.3h). Measuring Plat could estimate the need for additional procedures better than bronchoscopy alone. The direct measurement of the pressure difference and the angle of pressure–pressure curve is a new assessment modality for the success of interventional bronchoscopy.

Fig. 7.3
A set of 3 images and 6 graphs. a. A C T image of the lungs with the narrowed trachea. B. An endoscopic view of the trachea. C. A C T image of the lungs with a Y-shaped stent in the trachea. D to F. 3 line graphs have fluctuating lines for the carina and upper trachea. G and H. 2 graphs plot flow versus volume. I. The graph illustrates three P-P curves.

Lateral airway pressure (Plat) measurements during interventional bronchoscopy with balloon dilation and silicone Y stent implantation in fixed intrathoracic stenosis due to tracheal tuberculosis (before treatment: panels a, d, and g; after balloon dilation: panels b and e; after stenting: panels c, f and h). Plat was measured simultaneously at two points (upper trachea and carina). Blue line shows Plat at carina and the red line indicates Plat at upper trachea (d–f). After each treatment, the angle of P–P curve showed a stepwise improvement over the interventional procedures (i). See text for further explanation

Full size image

Analysis of Pressure–Pressure Curve

Central airway stenosis can be divided into four major types: fixed, variable, extrathoracic, and intrathoracic stenosis. In fixed stenosis, the CSA at the site of the lesion does not change during the respiratory cycle, and the P–P curve was linear. In variable stenosis, the configuration of the stenotic lesion changes between phases of respiration. Airway narrowing occurs during expiration in intrathoracic stenosis, whereas airway narrowing occurs during inspiration in extrathoracic stenosis. In variable extrathoracic stenosis, the angle of the P–P curve during inspiration is smaller than during expiration, and in variable intrathoracic stenosis, the angle of the P–P curve during expiration is smaller than during inspiration.

Conclusions

Placement of the stent at the flow-limiting segment (FLS) provided the greatest functional benefit to patients with central airway stenosis [4, 5]. Although bronchoscopic imaging showed that tracheal patency was restored after procedures, the angle of P–P curve did not always improve. It is difficult to estimate the outcome of interventional procedures by bronchoscopy alone. When the location of the FLS is assessed using flow–volume curves, the pressure difference and the angle of pressure–pressure curve are able to immediately estimate the outcomes of interventional bronchoscopy in real time.