Tracheobronchial Structure

The trachea extends from the cricoid cartilage at the level of C6 to the carina at approximately the level of T4/5, at which point it branches into the right and left mainstem bronchi. The trachea is approximately 12 cm in women and 14 cm in men, and ranges from a mean of 19 ± 1.5 mm in diameter in women to 22 ± 1.5 mm in men. The right mainstem bronchus is shorter, wider, and more vertical than the left mainstem bronchus. Its length is approximately 1.4 to 1.8 cm compared with 4.4 to 4.9 cm on the left; thus the right lung is generally more susceptible to aspiration.

There are three right-sided lobes: the upper, middle, and lower lobes. On the left, there are two lobes, the upper and lower lobes, with the lingula being a “tongue-like” extension of the left upper lobe. The lobes are covered with visceral pleura and are separated by folds of pleura known as the horizontal and oblique fissures on the right, and by the oblique fissure on the left. Each lobe has a lobar, or secondary bronchus, as well as lobar blood supply. The lobes are further divided into bronchopulmonary segments, which again have dedicated named bronchi and blood supply. The right and left lungs each have 10 segments, although nomenclature for these segments may vary slightly in the literature (Fig. 5.1).^2,3

The primary, or mainstem, bronchi branch from the trachea at the carina, then further branch into the lobar bronchi, and the segmental bronchi. These airways contain rings of cartilage that help keep them open. Once the airways branch into bronchioles, the walls are made of smooth muscle, and at the level of the respiratory bronchioles, the walls are made of only epithelium. This marks the transition from the conducting airways, encompassing the oropharynx and nasopharynx through bronchioles, to the respiratory airways, which participate in gas exchange and include the respiratory bronchioles and the alveoli.² On average, the lungs have 23 generations of dichotomous branching, as shown in Fig. 5.2.⁴

Respiratory Airways

The respiratory bronchioles end in the acinus, which is the functional unit of the lung, and consists of the alveolar ducts and alveolar sacs. The acinus appears as a cluster of grapes, each of which may contain up to 2000 alveoli. In total, the human lung consists of approximately 300 million alveoli creating a surface area for gas exchange of 40 to 80 m.^2,4 The alveoli contain alveolar type I cells, alveolar type II cells, and alveolar macrophages. Type I cells are squamous cells, comprising the very thin alveolar walls to facilitate gas exchange, and they cover up to 95% of the surface area of the alveoli. Type II cells are cuboidal and contain lamellar bodies, which secrete pulmonary surfactant, a film of phospholipids which reduces alveolar surface tension to facilitate lung expansion. Alveolar macrophages are scavengers that engulf foreign particles throughout the lungs, playing a key immunologic role.^2,3

Transpulmonary Pressure

The transpulmonary pressure, P_L, represents the distending pressure across the lungs, which is the difference between the airway pressure and the pleural pressure. It is an essential concept because it represents the pressure that effectively promotes air flow and distends the lungs. At end-inspiration, if respiratory flow equals zero and there is a clear plateau of the airway pressure, end-inspiratory P_L = P_plateau – end-inspiratory P_pleural represents the pressure acting across the alveolar units. At end exhalation, that pressure would be represented, in the absence of auto-PEEP, by end-expiratory P_L = PEEP – end-expiratory P_pleural. The importance of the concept of P_L becomes clear as the inspiratory P_L is the pressure which provides tidal ventilation at the same time that could produce stretch lung injury; and the end-expiratory P_L is that required to prevent lung collapse at end-exhalation by being kept positive.

Unfortunately, the assessment of P_L is not usual in clinical practice because measurement of pleural pressure is not simple, in contrast to measurement of airway pressure. Yet, pleural pressure may be estimated using an esophageal balloon, which is either standalone or attached to an esophagogastric tube and positioned in the lowest third of the esophagus^23–25 (Fig. 5.6). These esophagogastric tubes are generally 100 cm long, and the average depth for a correctly positioned balloon is approximately 35 to 45 cm. The catheter is connected to a pressure transducer via a three-way stopcock, and the balloon is inflated to a standard volume depending on the device, which is necessary to prevent artifact because of passive recoil of the balloon walls. Further, the balloon should be long enough to avoid regional variability, providing an average measurement of esophageal pressure to estimate pleural pressure. Although use of an esophageal balloon represents an improvement in the accuracy of pulmonary pressure monitoring, it is not without its limitations. Artifact may arise from mediastinal weight, gravitational gradients in pleural pressure, and airway closure at end exhalation.²³ Furthermore, there is considerable controversy in the literature on whether the esophageal pressure measurements should be directly taken as absolute estimates of pleural pressures for PEEP adjustments,²⁶ or whether changes in esophageal pressures should be used to compute chest wall and lung elastances, which are then used to set the level of the administered PEEP.²⁷ Although a ventilator strategy guided by the transpulmonary pressure measured with esophageal balloons improved oxygenation and compliance in a small group of acute respiratory distress syndrome (ARDS) patients,²⁸ a subsequent large trial on patients with moderate to severe ARDS resulted in no significant difference in death and days free from mechanical ventilation between patients receiving an esophageal balloon-guided PEEP and those receiving empirical high PEEP-FiO₂ settings.²⁹ Those findings did not support esophageal balloon-guided PEEP titration in ARDS. In critically ill patients with class 3 obesity (body mass index [BMI] ≥40 kg/m²) and ARDS, a retrospective analysis compared a standard protocol of ventilator settings determined by the ARDS net table for lower PEEP/higher FiO₂ with a “lung rescue” management strategy with settings determined by an individualized protocol based on lung recruitment maneuvers, esophageal manometry, and hemodynamic monitoring.²⁹ Patients receiving the standard protocol had almost double the 28-day mortality compared with the lung rescue cohort (31% vs. 16%, P = .012; hazard ratio [HR], 0.32; 95% confidence interval [CI] 95%, 0.13–0.78).

In robotic-assisted laparoscopic surgery, esophageal manometry can help delineate the portion of increased driving pressure, which is applied to the lungs versus that applied to the chest wall; the increase in elastance associated with pneumoperitoneum is substantially greater for the chest wall (E_CW) than for the lungs (E_LS).³⁰ This may improve the ability to maintain open lung by appropriately titrating PEEP in this and other perioperative conditions with high susceptibility to lung derecruitment.

Driving Pressure

Driving pressure (ΔP) is the difference between the plateau pressure (P_plat) and PEEP (Fig. 5.7). The driving pressure can also be expressed as the V_T normalized by respiratory system compliance (C_rs)—that is, the ΔP is proportional to the change in volume of the lung in a breath (V_T) divided by its initial volume, to the extent that C_rs changes in proportion to lung volume. Change in volume divided by initial volume is, by definition, the volumetric strain. Thus physiologically, the reason why ΔP may be important is because it provides a rough measure of the global volumetric strain of the lungs.^31–33

Recent studies suggest that such physiologic considerations may be very important in the clinical setting. Driving pressures have been shown to be better associated with relevant clinical outcomes in ARDS^32,34 and surgical patients^33,35,36 than V_Ts. Indeed, a direct association of increased intraoperative ΔP with increased risk of significant postoperative pulmonary complications has been shown in patients undergoing noncardiothoracic surgery, in the absence of a corresponding association with V_T^33,35, in addition the relative risk of in-hospital death in relation to plateau pressure and driving pressure is depicted in Fig. 5.8. Such a relationship was also found in ARDS patients, with a greater association between increased ΔP and mortality compared with that of V_T³² (Fig. 5.9). This finding was reinforced in a secondary analysis of observational data on patients with moderate to severe ARDS ventilated with standard lung-protective ventilation.³⁷ Variation in V_T or PEEP had no impact on mortality. In contrast, above a plateau pressure of 29 cm H₂O or a driving pressure of 19 cm H₂O, there was an ordinal increment in risk of death.³⁷

In thoracic anesthesia, comparison of conventional lung-protective ventilation (one-lung ventilation [OLV] with V_T = 6 mL/kg ideal body weight, PEEP 5 cm H₂O, and recruitment maneuvers) with ΔP-guided ventilation (same VTs and recruitment, individualized PEEP to produce the lowest ΔP during OLV) showed a smaller incidence of pneumonia or ARDS in the ΔP-guided group.³⁸ A retrospective analysis of data of patients undergoing OLV for thoracic surgery (hospital electronic medical records and the Society of Thoracic Surgeons database) indicated that ΔP, when modeled together with V_T, predicted major morbidity, whereas V_T alone inversely correlated with respiratory complications.³⁹ This implied that low V_T by itself was not associated with less postoperative pulmonary complications, whereas low V_T taken together with strategies to minimize ΔP may be.³⁹

Elastance and Compliance

The lungs and the chest wall have elastic properties that tend to restore them to their combined equilibrium configuration when they are displaced from baseline by spontaneous or mechanical ventilation. Thus if the pleural space is open at FRC, the lungs would recoil inward toward collapse, while the chest wall would recoil outward, toward expansion. Elastance is the change in distending pressure for a given change in volume (E = ΔP/ΔV). The elastance of the total respiratory system (E_RS) is the sum of the lung (E_L) and the chest wall (E_CW) elastances:

ERS=EL+ECW

Compliance is the reciprocal of the elastance (C = ΔV/ ΔP); that is, the change in volume for a given change in distending pressure. Compliance tends to be more frequently used ­clinically. Thus the earlier equation can be alternatively expressed as:

1CRS=1CL+1CCW

where C_RS, C_L, and C_CW are the compliances of the respiratory system, lung, and chest wall, respectively.

During mechanical ventilation with volume control, a peak inspiratory pressure (PIP) is usually seen, which is followed by a plateau pressure (P_plat) if an inspiratory pause (i.e., a period of no flow at end of inspiration) is provided (see Fig. 5.7). The difference between peak and plateau pressures is caused by dynamic factors, particularly airway resistance. Dynamic compliance (C_dyn) is defined as the compliance calculated using the PIP:

Cdyn=VTPIP−PEEP

where PEEP is the positive end-expiratory pressure (see Fig. 5.7). C_dyn, thus, conflates the distensibility of the lungs themselves with the resistance of the airways, in addition to the properties of the chest wall, to the overall quantification of compliance. Static compliance (C_stat) aims to eliminate the dynamic airflow-dependent component from the computation of compliance by using P_plat:

CStat=VTPPlat− PEEP

Lung compliance represents the distensibility of the lung parenchyma; it is altered in various disease states. Compliance is decreased with consolidation, atelectasis, pulmonary edema, interstitial disease or fibrosis, pneumothorax, surgical resection, large effusion, and mainstem intubation. It is increased in emphysema because of loss of elastic tissue. In addition, lung compliance depends upon the lung volume in a nonlinear manner. The highest lung compliance is observed when lungs are at moderate volumes, and lowest at the extremes of low or high lung volumes. Such relationships are expressed in the pressure-volume curve of the lung (Fig. 5.10). That curve also shows that the pressure-volume relationship and the compliance for the same volume (or pressure) on inhalation differs from that on exhalation, a phenomenon known as hysteresis.

To compute the compliance of the lungs, the contribution of the chest wall needs to be subtracted. Accordingly, lung compliance (C_L = ΔV_LΔP_L) can be estimated as:

CL=  VTPawEI−PesEI−PawEE−PesEE

,where P_awEI is the airway pressure at end inhalation, P_esEI is the esophageal pressure at end inhalation, P_awEE is the airway pressure at end exhalation, and P_esEE is the esophageal pressure at end exhalation, measured using esophageal manometry, as explained earlier.

Chest wall compliance reflects the distensibility of the muscles, bones, and soft tissue that comprise the chest wall. Chest wall compliance is reduced with increased abdominal pressure (e.g., insufflation, abdominal compartment syndrome), chest wall edema, thoracic deformity, muscle tone, extensive thoracic/abdominal scar, supine positioning, and obesity. Chest wall compliance is increased with flail chest, muscle relaxation, and upright positioning. Using estimates of pleural pressure, the compliance of the chest wall (C_CW) can be calculated as:

CCW=VT/PesEI−OesEE

with variables that have been defined earlier.

Resistance

Resistance is the opposition to fluid flow because of friction and may be calculated as the ratio of the driving pressure to air flow (R_aw = ΔP/V.). Airway resistance is determined by the dimensions of the airway, the viscosity of the gas, and whether the flow is laminar or turbulent, which in turn is predicted by the Reynold’s number, Re:

Re=ρuLμ

where ρ is the fluid/gas density (kg/m³), u is the velocity (m/s), L is the length (m), and μ is the viscosity (kg/ms). Laminar flow dominates at Re less than 2300, while turbulent flow dominates at Re greater than 2900.

Laminar flow through a long cylindric pipe is governed by Poiseuille’s Law, which is appropriately applied to conducting airways. Resistance is relatively low in laminar flow, and is calculated as follows:

R=8Luπr4

where L is the length, u is the velocity, and r is the radius. Thus the airway radius is especially important in determining the resistance; if the radius is halved, the resistance increases by a factor of 16. Pathologic states, such as tracheal stenosis and bronchospasm, raise the resistance and increase the work of breathing. Reducing the density of the gas mixture (e.g., through mixing with helium, “Heliox”) can improve resistance and lead to decreased work of breathing.

Because of the dependence on radius, an individual small airway has a higher resistance than a large airway. However, because the airways branch so prolifically, the overall resistance depends on the number of parallel pathways present, and it is the medium-sized airways that have the overall greatest resistance (see Fig. 5.2).

Stress and Strain

Stress is defined as force divided by the area over which it is applied.³¹ Lung stress can be estimated through the transpulmonary pressure (P_L). This will be larger at end-inspiration and lower at end-exhalation, in the absence of patient effort. Strain is the change in dimension relative to the material’s initial dimension. Based on this definition, lung volumetric strain can be calculated as the ratio of the change in lung volume (dV) to the resting lung volume (V₀), that is, dV/V_0.⁴⁰ Assuming a linear relationship, lung stress and strain are related to one another by the specific lung elastance, K.

stress PL=K⋅strain dVV0

.

Experimental studies indicate that when stress and strain are outside the normal physiologic limits, the lung releases inflammatory cytokines and mediators, which indicate, produce, and exacerbate lung injury. Very large V_Ts conducive to large strains, either through spontaneous or mechanical ventilation, can produce lung injury.^41,42 When mechanical ventilation is applied for more than 48 hours in initially normal animal lungs, injury occurs when lung expansion reaches approximately the TLC, that is when strain would be (TLC – FRC)/FRC.^40,43,44 In experiments representing usual clinical conditions, those large values are not present.⁴⁵ This suggests an increased susceptibility of lung tissue to strain in the presence of an additional injurious trigger (i.e., systemic inflammation).^46,47 A large animal study modeled moderate–severe ARDS (respiratory system compliance = 14.0 ± 4.9 mL/cm H₂O; PaO₂/FiO₂ = 92 ± 59) with surfactant depletion and ventilator-induced lung injury (VILI) and quantified volumetric strain using CT to support a relationship between strain and local inflammation assessed with positron emission tomography (PET) in those extreme lung injury cases.⁴⁸

Stress is approximated by the transpulmonary pressure distending the airspaces, but the regional stress of the lung will vary because of gravity and other factors producing heterogeneous lung expansion. This will produce a distribution of stress throughout the lungs. For example, in uniformly expanded lungs, alveolar pressure is the same across the whole lung, but when the lung is not uniformly expanded (such as when air spaces are isolated from one another because of obstruction or closure), alveolar pressures will not be uniform throughout.⁴⁹ The presence of such heterogeneity leads to regional stress values substantially larger than those present at the global level.⁴⁵ This means that even when global transpulmonary pressures are low, local stresses can be high, even to the limit of being injurious.^45,50