Chapter 6: Integrated Respiratory System Compliance and the Work of Breathing

The tissue elastic and surface tension recoil forces that provide inherent compliance to normal lung tissue were reviewed in Chaps. 4 and 5. For excised lungs, these forces can be examined as if they operated independently of constraints placed on lung expansion by bones and muscles of the thorax. However, the chest wall and airways strongly influence lung distension and the work of breathing (Fig. 6.1). Incorporating these factors into a broad model of ventilation will facilitate understanding of the major categories of lung disorders, the obstructive diseases and restrictive diseases.

The chest wall and diaphragm normally oppose the recoil forces of lung tissue elasticity and surface tension. Thus, the negative P_IP at functional residual capacity (FRC) (see Fig. 4.6) represents the equilibrium when these musculoskeletal elements are bowed inwardly by lung recoil. Since this balance is not intuitive, consider a bilateral pneumothorax when P_IP = P_B (Fig. 6.2). When P_IP increases from −5 to 0 cm H₂O, healthy lungs recoil toward their minimal volume, while the chest walls spring outward and the diaphragm drops toward the abdominal viscera. Indeed, the volume of the open chest when P_IP = P_B is about 80% of a subject’s normal total lung capacity (TLC), implying that musculoskeletal elements will oppose all forces attempting to make the chest either larger or smaller than about 80% of TLC.

Thus, an in situ compliance curve for the entire lung-thorax system is the algebraic sum of the compliance curves for each component. Such a composite compliance curve can be obtained by having a subject perform a relaxation pressure-volume maneuver (Fig. 6.3) using a specialized apparatus for whole-body plethysmography (Chap. 16). While simultaneously monitoring P_IT in the airway and intraesophageal pressure to approximate P_IP, subjects inspire or expire to a series of lung volumes between RV and TLC, and then are locked momentarily at that volume by a valve so they can completely relax their chest wall. Under these conditions, the maneuver begins and the subject’s FRC is empirically defined as that volume when P_IT = P_B = 0 at the mouthpiece. This FRC also is the volume where inwardly directed lung recoil forces are equal and opposite to outwardly directed forces exerted by the chest wall plus diaphragm.

When subjects who are at FRC expire toward RV, their P_IT is less than P_B and a sensation of suction is felt as the thoracic structures are increasingly deviated inward from their preferred resting positions seen with a pneumothorax. Conversely, when the subject inspires to lung volumes above FRC and toward TLC, P_IT exceeds P_B and a sensation of chest compression develops as lung elastic and tension recoil forces increase. Indeed, the subject’s sensation of chest tightness will maximize at lung volumes >80% of TLC (ie, the size of the thoracic space after a pneumothorax) since the chest wall and diaphragm will also then be exerting inwardly directed positive pressures that augment the positive pressure created by lung recoil.

From this explanation, it becomes apparent that if a subject breathes at lung volumes between FRC and about 80% of TLC, each inhalation to inflate the lungs is achieved in part by utilizing potential energy stored in the chest wall and diaphragm as they were pulled inward during the preceding exhalation. By the same argument, every exhalation uses potential energy stored in the inflated lungs to expel the tidal breath, while simultaneously storing potential energy in musculo-skeletal elements being bowed inward. Thus, it is not accurate to consider normal inhalation to be entirely "active" or normal exhalation to be entirely "passive". Rather, normal breathing is carefully controlled to oscillate around the midrange of available lung volumes so that inhalation consumes less energy than one might think, but exhalation also costs energy.

To this point, models of static lung and chest wall compliance have been used to estimate resistances to ventilation that are generated by surface and tissue recoil of lung tissue, and by thoracic musculoskeletal elements. However, additional resistance to ventilation also occurs in the airways during flow, only to disappear under no-flow conditions such as FRC. Such dynamic airway resistance is nevertheless a powerful determinant of the pattern, energetic cost, and overall efficiency of ventilation. Thus, dynamic airway compliance measurements are made to directly determine in-line resistances due to the airways. Such measurements have shown that most dynamic airway resistance resides in the upper airways (generations 1-7) because their flow rates tend to be high, their end-to-end pressure changes are large, and their aggregate cross-sectional areas are small. By the same argument, the smaller peripheral airways usually contribute less resistance despite having smaller individual diameters, due to their shorter lengths and their exponentially larger cross-sectional areas (Fig. 6.4).

To appreciate the role that airway dimensions play in creating resistance to ventilation and thereby add to the work of breathing, recall Ohm’s law:

where I = current, ΔV = voltage potential, and R = resistance

In airways and other tubes, this becomes a generalized flux equation:

where ΔP = pressure differential, either end-to-end or end-to-side, and

R = resistance due to airway dimensions and gas viscosity, and thus:

where r = airway radius, η = viscosity, and l = airway segment length

Substituting terms gives the well-known Poiseuille equation:

Considering the dependence of resistance on radius to the fourth power, the importance of even a small change in airway radius is apparent. It will be shown in later chapters that many respiratory diseases pivot on this aspect of airway resistance, versus a lesser dependence on airway length or viscosity.

Applying the Poiseuille equation to human ventilation presumes an idealized laminar flow pattern, but evidence is not persuasive as to how often and how distally such flow exists in the airways (Fig. 6.5). Most bronchial airways are too wide, too frequently bifurcated, or too short before branching to promote consistent laminar flow. Rather, it seems likely transitional flow patterns predominate at rest, that deteriorate to turbulent patterns during exercise and other events associated with hyperventilation.

Clinically, airway resistance is calculated using the pressure gradient from trachea to alveolus divided by flow rate; the subject’s tracheal pressure and flow rate are estimated at the mouth by anemometer, and alveolar pressure by esophageal transducer or plethysmography (Chap. 16). When the in vivo volume/pressure changes are plotted for such a dynamic compliance maneuver, the area between the inflation and deflation curves representing hysteresis is larger than for excised lungs of the same size (see Fig. 5.2). Most importantly, this increased area represents additional nonrecoverable work during both inflation and deflation to overcome airway resistance.

In summary, the total work of breathing in a healthy adult can be partitioned into energy that must be applied to overcome:

• Static lung resistance caused by tissue elasticity and surface tension forces that must be overcome regardless of dynamic resistance, normally 65%-70% but higher when pathologies involving surfactant or the respiratory parenchyma are present;

• Dynamic airway resistance that exists only when gas is moving, normally 25%-30% and mostly in the upper 10 generations, but much higher in patients with obstructive disorders; and

• Tissue viscous drag due to physical abrasion and friction between pleural surfaces of the lung and chest, normally 5% but can become fatally high if adhesions, congenital anomalies, or effusions are present.

The rationale for devoting such attention to the work of breathing and determination of lung compliance is that most respiratory diseases arise in patients who present with impaired ventilation characterized as an obstructive airway disease or a restrictive lung disease. In general, a disease is obstructive if there is an increased resistance to airflow, most evident during expiration but also detectable upon inspiration. Specific examples include asthma, bronchitis, and emphysema. Patients with restrictive lung diseases do not necessarily show airflow obstruction, but their ventilation is impaired due to restrictions in the maximal size or movement of the lungs or chest wall (or both). Restrictive chest wall diseases include kyphoscoliosis, neuromuscular disorders, and spinal trauma. Restrictive lung diseases include interstitial fibrosis, sarcoidosis, scarring after pneumonia or tuberculosis, bronchopulmonary dysplasia, and alveolar edema due to congestive heart failure. Patients can show features of both obstructive and restrictive lung diseases.

In Chap. 4, procedures to measure forced vital capacity (FVC) and forced expiratory volume (FEV) were introduced, in which patients on a spirometer or flowmeter inhale to TLC and then exhale as quickly as possible as volume and flow rate are recorded (Fig. 6.6). Exhaled volumes are reported as FVC (= TLC − RV) in liters, and as percentages of the actual FVC that are expired within the first second (FEV₁), first two seconds (FEV₂), and first three seconds (FEV₃) of the maneuver. For a healthy adult with FVC = 6 L, an FEV₁ = 5 L or 83% FVC would be normal; an FEV₂ of ~95% and FEV₃ of ~99% would also be expected.

In patients with obstructive airway diseases (Fig. 6.6), both FVC and FEV₁ are lower than predicted based on age, gender, and size. However, the patient’s TLC is often larger than predicted because low expiratory flow rates lead to peripheral gas trapping that increases RV. As RV expands with disease severity, such patients will be forced to use their principal ventilatory muscles at distinct mechanical disadvantages. Indeed, the resulting hyperinflation of the chest as RV and FRC increase eventually flattens the diaphragm, so that its contraction fails to increase intra-thoracic volume. In contrast, a patient with restrictive lung disease shows lower FVC and TLC than predicted by age, gender, and size, due to physical limits imposed on lung and/or chest volumes (Fig. 6.6). Of note, given the smaller lung or thoracic volumes in such patients, the FEV₁ may be normal or even increased as a percent of FVC unless airway obstruction is also present.

Flow rate (in L/sec) is measured at the mouth continuously during an FVC test to obtain an expiratory flow-volume loop, with airflow plotted versus lung volume from TLC to RV on the abscissa (Fig. 6.7). The expiratory flow rate in healthy subjects quickly reaches a maximum that corresponds to the steepest portion of a conventional spirometer tracing (Fig. 6.6). Despite continued maximal effort by the subject, airways with high flow rates begin to collapse because pressures within parenchyma surrounding the airways soon exceed airway pressures themselves. This effect can be explained by principles developed by the mathematician and physician Bernoulli. The eventual decline in flow rate as a forceful exhalation proceeds is termed effort-independent: no matter how forcefully the subject attempts to exhale, their volume-dependent airway compression during dynamic conditions limits maximal flow rate. In fact, even if subjects exert a submaximal effort initially and then a maximal effort later, the effort-independent portion of their flow-volume loop is not displaced upward or to the right.

When compared to healthy subjects, individuals with obstructive or restrictive disease patterns show distinctive shapes for their flow-volume loops (Fig. 6.8). Patients with obstructive airway disease often begin an FEV maneuver at an abnormally higher TLC due to hyperinflation, and end the maneuver at a substantially increased RV. Such patients also take longer to achieve maximal flow rates, which are lower than expected, and their effort-independent region may be distinctly concave as they approach RV. Patients with restrictive lung or chest disease have a low or normal RV, depending on the nature of their disorder. By definition, such patients have abnormally low total lung capacities. However, their peak flow rates often resemble the effort-independent portion of normal subjects, if airway caliber is not a prominent feature of their disease.

When patients are being mechanically ventilated, it is common to monitor their inspiratory and expiratory flow-volume loops at the bedside. These data permit the clinical team to assess ongoing changes in dynamic airway resistance and in lung and chest wall compliances as these are altered by disease course or by interventions during hospitalization. This topic is discussed further in Chaps. 28 and 30.

It has been assumed to this point that the lungs are uniformly ventilated, and that their compliance shows no regional heterogeneity. However, lungs are air-filled organs suspended in a partial vacuum within a body cavity susceptible to the effects of gravity. Lungs also have progressively smaller airways and blood vessels, moving distally from the trachea and pulmonary artery. These anatomical and physical conditions create a situation that is far from ideal. Indeed, when an upright subject breathes air containing trace amounts of ¹³³Xenon [which, like helium (He), is poorly soluble in fluid] while the chest is scanned with appropriate detectors, it is found that more inhaled ¹³³Xe goes to the lung base than to its apex. This result is not due simply to the greater volume of the lung bases. Rather, the inhaled tracer is always observed to distribute preferentially to dependent lung zones, defined as those regions lowest in the gravitational field. So when subjects inhale ¹³³Xe while standing on their head or lying supine, the tracer preferentially enters apical alveoli or those nearest the spine, respectively.

These nonintuitive results occur because at any instant, P_IP is not uniform for the entire lung but varies from the gravitational top of the lungs to their bottom within the thoracic space. When upright, P_IP is most negative at the thoracic apex, due to the hanging weight of the lungs as they maintain contact with adjacent apical pleural surfaces. In the same upright position, P_IP at the lung base is less negative or may even be positive, because here the lungs pull less on basal pleural membranes, or may even push against them. Such gradients in P_IP exist for a subject in any assumed body position and persist throughout a ventilation cycle. Thus, individual alveoli in more dependent regions are smaller at any lung volume, due to the less negative P_IP they are experiencing versus alveoli above them that are more distended due to the more negative P_IP adjacent to them in the same gravitational field.

At the same time, all alveoli in the lung still experience nearly identical levels of P_IT and P_A throughout a ventilation cycle since they are connected via common airways from primary bronchi to the most distal airspaces (Fig. 6.9). Consequently, all alveoli experience the same ΔP_IT and ΔP_A during inhalation, but those in the more dependent zones will inflate more easily because they are nearer to their own RV and thus more compliant (see Fig. 4.6). At the same instant, alveoli in gravitationally higher regions inflate less since they are already closer to their own TLC, and thus less compliant. Analogous gravitational effects that act upon the pulmonary arteries and the patency of alveolar capillaries will be discussed in Chap. 7.

There are significant clinical consequences to these regional differences in ventilation. In a subject at FRC (Fig. 6.9), basal alveoli are more easily ventilated because they are on the steepest portion of their individual compliance curves. However, if the same lungs are studied at a low lung volume like RV instead, perhaps in a comatose patient, the P_IP at the lung base is positive, and little or no alveolar expansion occurs during a typical tidal breath of only 0.5-0.7 L. Such situations also develop rather commonly in sedated patients, whose spontaneous ventilation may not prevent collapse of their dependent alveoli, particularly when an otherwise desirable increase in F_Io₂ accelerates this effect by absorption atelectasis (Chap. 30). In an awake and healthy subject, alveolar collapse is prevented or reversed by periodic sighs, with a sigh operationally defined as three times a subject’s normal tidal volume (V_T).