9: Pulmonary Disease

The principal physiologic role of the lungs is to make oxygen available to tissues for metabolism and to remove the main byproduct of that metabolism, carbon dioxide. The lungs perform this function by moving inspired air into close proximity to the pulmonary capillary bed to enable gas exchange by simple diffusion. This is accomplished at a minimal workload, is regulated efficiently over a wide range of metabolic demand, and takes place with close matching of ventilation to lung perfusion. The extensive surface area of the respiratory system must also be protected from a broad variety of infectious or noxious environmental insults.

Humans possess a complex and efficient respiratory system that satisfies these diverse requirements. When injury to components of the respiratory system occurs, the integrated function of the whole is disrupted. The consequences can be profound. Airway injury or dysfunction results in obstructive lung diseases, including bronchitis and asthma, whereas parenchymal lung injury can produce restrictive lung disease or pulmonary vascular disease. To understand the clinical presentations of lung disease, it is necessary first to understand the anatomic and functional organization of the lungs that determines normal function.

Anatomy

The mature respiratory system consists of visceral pleura-covered lungs contained by the chest wall and diaphragm, the latter serving under normal conditions as the principal bellows muscle for ventilation. The lungs are divided into lobes, each demarcated by intervening visceral pleura. Each lung possesses an upper and lower lobe; the middle lobe and lingula are the third lobes in the right and left lungs, respectively. At end expiration, most of the volume of the lungs is air (Table 9–1), whereas almost half of the mass of the lungs is accounted for by blood volume. It is a testament to the delicate structure of the gas-exchanging region of the lungs that alveolar tissue has a total weight of only 250 g but a total surface area of 75 m².

Connective tissue fibers and surfactant serve to maintain the anatomic integrity of this large and complex surface area. The connective tissue fibers are highly organized collagen and elastic structures that radiate into the lungs. These fibers divide segments, invest airways and vessels, and support alveolar walls with a delicate, elastic fibrous network. The multidirectional elastic support provided by this network allows the lung, from alveoli to conducting airways, to support itself and retain airway patency despite large changes in volume.

Surfactant is a complex material produced by type II alveolar cells and composed of multiple phospholipids and specific associated proteins. The physiologic function of surfactant is to enhance the anatomic stability of the lungs. The presence of surfactant covering the alveolar epithelial surface reduces surface tension, allowing expansion of alveoli with a transpulmonary distending pressure of less than 5 cm H₂O. In the absence of this surface-active layer, increasing surface tension associated with a reduction of alveolar volume during expiration would collapse alveoli. The distending pressure required to reexpand these alveoli would be greater than normal ventilatory effort could produce.

Airway & Epithelial Anatomy

Further anatomic division of the lungs is based primarily on the separation of the tracheobronchial tree into conducting airways, which provide for movement of air from the external environment to areas of gas exchange, and terminal respiratory units, or acini, the airways and associated alveolar structures participating directly in gas exchange (Figure 9–1). The proximal conducting airways are lined by ciliated pseudostratified columnar epithelial cells, are supported by a cartilaginous skeleton in their walls, and contain secretory glands in the epithelial wall. The ciliated epithelium has a uniform orientation of cilia that beat in unison toward the pharynx. This ciliary action, together with the mucus layer produced by submucosal mucous secretory glands, provides a mechanism for the continuous transport of contaminating or excess material out of the lungs. Circumferential airway smooth muscle is also present but, as with secretory glands, is reduced and then lost as the airways branch farther into the lung and diminish in caliber. The smallest conducting airways are nonrespiratory bronchioles. They are characterized by a loss of smooth muscle and cartilage but retention of a cuboidal epithelium that may be ciliated and is not a site of gas exchange. The lobes of the lung are divided into less distinct lobules, defined as collections of terminal respiratory units incompletely bounded by connective tissue septa. Terminal respiratory units are the final physiologic and anatomic unit of the lung, with walls of thin alveolar epithelial cells that provide gas exchange with the alveolar-capillary network.

The principal site of resistance to airflow in the lungs is in medium-sized bronchi (Figure 9–2). This seems counterintuitive because one would expect smaller caliber airways to be the major site of resistance. Repetitive branching of the small airways leads to a profound increase in cross-sectional area that does not contribute significantly to airway resistance in healthy individuals (Figure 9–3). Under pathologic conditions such as asthma, in which smaller bronchi and bronchioles become narrowed, airway resistance can increase dramatically.

The pulmonary arteries are found in close association with the branching bronchial tree in the lungs (Figure 9–4). Both arterial blood flow and bronchial airflow are actively regulated by changing vessel or airway caliber. The anatomic relationship between arteries and bronchi provides an ideal setting for the continuous matching of ventilation and perfusion to different lung segments.

Vascular & Lymphatic Anatomy

The pulmonary vascular system includes two distinct circuits that distribute blood through the lungs, the pulmonary and bronchial circulations. The right ventricle pumps its entire output of mixed venous blood through pulmonary arteries toward alveolar capillaries. Pulmonary arteries and arterioles are smooth muscle–invested vessels located adjacent to bronchi within the pulmonary bronchovascular bundle. Pulmonary arterioles are very sensitive to alveolar PO₂, with a prominent vasoconstrictor response to hypoxia. Hypoxic pulmonary vasoconstriction allows matching of alveolar perfusion to ventilation (see below). Pulmonary veins arise from alveolar capillaries to form vessels that traverse the intralobular septa to return oxygenated blood to the left atrium.

Bronchial arteries that arise from the aorta and intercostal arteries deliver oxygenated blood at systemic pressures to nearly all of the intrapulmonary structures proximal to the terminal bronchioles, including the bronchial tree, hilar structures, pulmonary arteries and veins, pulmonary nervous system and lymphatics, connective tissue septa, and visceral pleura. Most lung tumors receive their blood supply from the bronchial circulation. There are abundant bronchopulmonary anastomoses at the arteriolar and capillary levels that are silent in health but may enlarge in disease to contribute to hemoptysis. Drainage of the bronchial circulation occurs both to the right atrium via the azygos vein and to the left atrium via the pulmonary veins. The latter represents an anatomic shunt of deoxygenated blood, typically representing less than 5% of cardiac output.

Pulmonary lymphatics arise in connective tissue spaces beneath the visceral pleura and in deep plexuses at the junction of the terminal bronchioles and alveoli. Lymphatics do not enter the alveolar peri-interstitial space (Figure 9–4). As a result, fluid in the alveolar interstitium must move to the region of terminal bronchioles to gain access to draining lymphatics. Lymphatic ducts travel principally in the peribronchovascular sheath back to hilar and mediastinal lymph nodes before entering the left thoracic duct or right lymphatic duct. Lymphatic drainage of the pleural space occurs through plexuses investing the costal, diaphragmatic, and mediastinal parietal pleura that are anatomically separate from pulmonary lymphatics.

Pulmonary Nervous System

The lungs are richly innervated with neural fibers from parasympathetic (vagal), sympathetic, and the so-called nonadrenergic, noncholinergic (NANC) systems. Efferent fibers include the following: (1) parasympathetic fibers, with muscarinic cholinergic efferents that mediate bronchoconstriction, pulmonary vasodilation, and mucous gland secretion; (2) sympathetic fibers, whose stimulation produces bronchial smooth muscle relaxation, pulmonary vasoconstriction, and inhibition of secretory gland activity; and (3) the NANC system, with multiple transmitters implicated, including adenosine triphosphate (ATP), nitric oxide (NO), and peptide neurotransmitters such as substance P and vasoactive intestinal peptide (VIP). The NANC system participates in inhibitory events, including bronchodilation, and may function as the predominant reciprocal balance to the excitatory cholinergic system.

Pulmonary afferents consist principally of the vagal sensory fibers. These include the following:

1. Fibers from bronchopulmonary stretch receptors, located in the trachea and proximal bronchi. Stimulation of these fibers by lung inflation results in bronchodilation and an increased heart rate.

2. Fibers from irritant receptors, which are also found in proximal airways. Stimulation of these fibers by diverse nonspecific stimuli elicits efferent responses, including cough, bronchoconstriction, and mucus secretion.

3. C fibers, or fibers from juxtacapillary (J) receptors, are unmyelinated fibers ending in lung parenchyma and bronchial walls and respond to mechanical and chemical stimuli. The reflex responses associated with stimulation of C fibers include a rapid shallow breathing pattern, mucus secretion, cough, and heart rate slowing with inspiration.

Immune Structure & Function

Of all the body’s organs, the lungs have a unique exposure to environmental insults. Nonexertional ventilation in an adult totals about 7500 L of air per day, an amount that is increased substantially with activity. This exposure to an open, nonsterile environment imposes an ongoing risk of toxic, infectious, and inflammatory insults. Furthermore, the pulmonary circulation contains the only capillary bed through which the entire circulating blood volume must flow in each cardiac cycle. As a consequence, the lung is an obligatory vascular sieve and functions as a principal site of defense against hematogenous spread of infection or other noxious influences. Protection of the lungs from environmental and infectious injury involves a set of complex responses capable of providing timely and successful defense against attack via the airways or the vascular bed. As outlined in Table 9–2, it is convenient for discussion purposes to separate these responses into two major categories—nonspecific physical and chemical protections and specific immune structures and actions—all functioning to prevent injury to or microbial invasion of the very large epithelial and vascular area of the lung.

Physiology

At rest, the lungs take 4 L/min of air and 5 L/min of blood, direct them within 0.2 μm of each other, and then return both to their respective pools. With maximal exercise, flow may increase to 100 L/min of ventilation and 25 L/min of cardiac output. The lungs thereby perform their primary physiologic function of making oxygen available to the tissues for metabolism and removing the major byproduct of that metabolism, carbon dioxide. The lungs perform this task largely free of conscious control while maintaining PaCO₂ within 5% tolerance. It is a magnificent feat of evolutionary plumbing and neurochemical control.

Static Properties: Compliance & Elastic Recoil

The lung maintains its extremely thin parenchyma over an enormous surface area by means of an intricate supporting architecture of collagen and elastin fibers. Anatomically, as well as physiologically and functionally, the lung is an elastic organ.

The lungs inflate and deflate in response to changes in volume of the semirigid thoracic cage in which they are suspended. An analogy would be to inflate a blacksmith’s bellows by pulling the handles apart, thus increasing the volume of the bellows, lowering pressure, and causing inflow of air. Air enters the lungs when pressure in the pleural space is reduced by the expansion of the chest wall. The volume of air entering the lungs depends on the change in pleural pressure and the compliance of the respiratory system. Compliance is an intrinsic elastic property that relates a change in volume to a change in pressure. The compliance of both the chest wall and the lungs contribute to the compliance of the respiratory system (Figure 9–5). The compliance of the chest wall does not change significantly with thoracic volume, at least within the physiologic range. The compliance of the lungs varies inversely with lung volume. At functional residual capacity (FRC), the lungs are normally very compliant, approximately 200 mL/cm H₂O. Thus, a reduction of only 5 cm H₂O pressure in the pleural space will draw a breath of 1 L.

The tendency of a deformable body to return to its baseline shape is its elastic recoil. The elastic recoil of the chest wall is determined by the shape and structure of the thoracic cage. Lung elastic recoil is determined by two factors, tissue elasticity and the forces needed to change the shape of the air-liquid interface of the alveolus (Figure 9–6). Expanding the lungs requires overcoming local surface forces that are directly proportionate to alveolar surface tension. Surface tension is a physical property that reflects the greater attraction between molecules of a liquid than between molecules of that liquid and adjacent gas. At the air-liquid interface of the lung, molecules of water at the interface are more strongly attracted to each other than they are to the air above. This creates a net force drawing water molecules together in the plane of the interface. If the interface is stretched over a curved surface, that force acts to collapse the curve. The law of Laplace quantifies this force: The pressure needed to keep open the curve (in this case represented by a sphere) is directly proportionate to the surface tension at the interface and inversely proportionate to the radius of the sphere (Figure 9–7).

Surfactant is a mixture of phospholipid (predominantly dipalmitoylphosphatidylcholine [DPPC]) and specific surfactant proteins. These hydrophobic molecules displace water molecules from the air-liquid interface, thereby reducing surface tension. This reduction has three physiologic implications. First, it reduces the elastic recoil pressure of the lungs, thereby reducing the pressure needed to inflate them. This results in reduced work of breathing. Second, it allows surface forces to vary with alveolar surface area, thereby promoting alveolar stability and protecting against atelectasis (Figure 9–7). Third, it limits the reduction of hydrostatic pressure in the pericapillary interstitium caused by surface tension. This reduces the forces promoting transudation of fluid and the tendency to accumulate interstitial edema.

Pathologic states may result from changes in lung elastic recoil related to an increase in compliance (emphysema), a decrease in compliance (pulmonary fibrosis), or a disruption of surfactant with an increase in surface forces (infant respiratory distress syndrome [IRDS]) (Figure 9–8).

Dynamic Properties: Flow & Resistance

Inflation of the lungs must overcome three opposing forces: elastic recoil, including surface forces; inertia of the respiratory system; and resistance to airflow. Since inertia is negligible, the work of breathing can be divided into work to overcome elastic forces and work to overcome flow resistance.

Increased elastic forces predominate in two common disorders, diffuse parenchymal fibrosis and obesity. The reduction in lung compliance in fibrotic lung disease, and in chest wall and respiratory system compliance in obesity, increases the work of breathing. Obese subjects also experience increased airflow resistance, largely though not entirely because of their tendency to breathe at lower lung volumes.

Flow resistance depends on the nature of the flow. Under conditions of laminar or streamlined flow, resistance is described by the Poiseuille equation: Resistance is directly proportionate to the length of the airway and the viscosity of the gas, and inversely proportionate to the fourth power of the radius. A reduction by one half of airway radius leads to a 16-fold increase in airway resistance. Airway caliber is, therefore, the principal determinant of airway resistance under laminar flow conditions. Under conditions of turbulent flow, the driving pressure needed to achieve a given flow rate is proportionate to the square of the flow rate. Turbulent flow is also dependent on gas density and not on gas viscosity.

Most of the resistance to normal breathing arises in medium-sized bronchi and not in smaller bronchioles (Figure 9–2). There are three main reasons for this counterintuitive finding. First, airflow in the normal lung is not laminar but turbulent, at least from the mouth to the small peripheral airways. Thus, where flow is highest (in segmental and subsegmental bronchi), resistance is dependent chiefly on flow rates. Second, in small peripheral airways, where airway caliber is the principal determinant of resistance, repetitive branching creates a very large number of small airways arranged in parallel. Their resistance is reciprocally additive, making their contribution to total airway resistance minor under normal conditions. Third, there is a transition to laminar flow approaching the terminal bronchioles as a consequence of increased cross-sectional area and decreased flow rates (Figure 9–3). In the respiratory bronchioles and alveoli, bulk flow of gas ceases and gas movement occurs by diffusion.

Airway resistance is determined by several factors. Many disease states affect bronchial smooth muscle tone and cause bronchoconstriction, producing an abnormal narrowing of the airways. Airways may also be narrowed by hypertrophy (chronic bronchitis) or infiltration (sarcoidosis) of the airway mucosa. Physiologically, the radial traction of the lung interstitium supports the airways and increases their caliber as lung volume increases. Conversely, as lung volume decreases, airway caliber also decreases and resistance to airflow increases. Patients with airflow obstruction often breathe at large lung volumes because higher volumes tend to increase elastic lung recoil, maximize airway caliber, and minimize flow resistance.

Analysis in terms of laminar and turbulent flow assumes that airways are rigid tubes. In fact, they are highly compressible. The compressibility of the airways underlies the important phenomenon of effort-independent flow; airflow rates during expiration can be increased with effort only up to a certain point. Beyond that point, further increases in effort do not increase flow rates. The explanation for this phenomenon relies on the concept of an equal pressure point. Pleural pressure is generally negative (sub-atmospheric) throughout quiet breathing. Peribronchiolar pressure, the pressure surrounding small, noncartilaginous conducting airways, is closely related to pleural pressure. Hence, during quiet breathing, conducting airways are surrounded by negative pressure that helps to keep them open. Pleural and peribronchiolar pressure becomes positive during forced expiration, subjecting distensible conducting airways to positive pressure. The equal pressure point occurs where the surrounding peribronchiolar pressure equals or exceeds pressure inside the airway, causing dynamic compression of the airways, which leads to instability and potential airway collapse (Figure 9–9).

The equal pressure point is not an anatomic site but a functional result that helps to clarify different mechanisms of airflow obstruction. Because the driving pressure of expiratory airflow is principally lung elastic recoil pressure, a loss of lung elasticity that reduces recoil pressure without changing pleural or peribronchiolar pressure will lead to dynamic compression at higher lung volumes. The resultant air trapping contributes to symptomatic dyspnea in patients with obstructive lung disease. Patients with emphysema lose lung elastic recoil and may have severely impaired expiratory flow even with airways of normal caliber. The presence of airway disease will increase the drop in driving pressure along the airways and may generate an equal pressure point at even higher lung volumes. Conversely, an increase in recoil pressure will oppose dynamic compression. Patients with pulmonary fibrosis may have abnormally high flow rates despite severely reduced lung volumes.

The Work of Breathing

A constant minute ventilation can be achieved through multiple combinations of respiratory rate and tidal volume. The two components of the work of breathing—elastic forces and resistance to airflow—are affected in opposite ways by changes in frequency and depth of breathing. Elastic resistance is minimized by frequent, shallow breaths; resistive forces are minimized by fewer, larger tidal volume breaths. Figure 9–10 shows how these two components can be summed to provide a total work of breathing for different frequencies at a fixed minute ventilation. The set point for basal respiration is that point at which the total work of breathing is minimized. In normal humans, this occurs at a frequency of approximately 15 breaths/min. In different diseases, this pattern is altered to compensate for the underlying physiologic abnormality.

The amount of energy needed to maintain the respiratory muscles during quiet breathing is small, approximately 2% of basal oxygen consumption. In patients with lung disease, the energy requirements are greater at rest and increase dramatically with exercise. Patients with emphysema may not be able to increase their ventilation by more than a factor of 2 because the oxygen cost of breathing exceeds the additional oxygen made available to the body.

Oxygen Transport

Oxygen is poorly soluble in blood. At a temperature of 37°C and an oxygen partial pressure of 100 mm Hg (PaO₂ = 100), total oxygen dissolved in 100 mL of whole blood is approximately 0.3 mL. Because basal oxygen consumption in the average adult human is approximately 250 mL/min, dissolved oxygen content would be inadequate to meet metabolic demands. Instead, the high oxygen needs of complex internal organs are met by a soluble protein that binds oxygen rapidly, reversibly, and with a high storage capacity, namely, hemoglobin.

Hemoglobin is a complex tetramer of two alpha and two beta polypeptide chains, each of which contains a heme group with an iron atom in the ferrous form (Fe²⁺) at its center capable of binding molecular oxygen (O₂). Each molecule of hemoglobin can bind four oxygen molecules. Under physiologic conditions, 1 g of fully saturated hemoglobin can carry approximately 1.34 mL of oxygen. Therefore, 100 mL of blood containing 15 g/dL of saturated hemoglobin contains 20.1 mL of O₂, nearly 70 times the amount in solution. The conventional way to represent oxygen bound to hemoglobin is the hemoglobin saturation (SO₂), the ratio of oxygen bound to hemoglobin divided by the total oxygen binding capacity, typically expressed as a percentage. Note that SO₂ alone does not determine oxygen content. Blood oxygen content is the sum of two terms, dissolved oxygen and oxygen bound to hemoglobin. Dissolved oxygen is a linear function of the oxygen partial pressure (PO₂) and solubility, while oxygen bound to hemoglobin is the product of three terms, oxygen-carrying capacity, hemoglobin concentration, and hemoglobin saturation (SO₂):

CO₂ = (0.003 × PO₂) + (1.34 × [Hemoglobin] × SO₂)

This equation explains why oxygen content and tissue oxygen delivery may be low despite 100% SO₂ if hemoglobin concentration is markedly reduced.

Because of its physical chemistry, hemoglobin saturation has a complex relationship with the partial pressure of oxygen. Interactions among the four polypeptide chains in the heme molecule increase overall affinity for oxygen as each oxygen binding site is filled. If we graph PO₂ against SO₂ to represent the oxyhemoglobin dissociation curve, we see that the relationship is not linear but S-shaped or sigmoidal (Figure 9–11). The curve is very steep in the physiologic range, between 10 and 70 mm Hg PO₂, at which point it flattens out. This relationship explains the suitability of hemoglobin for its primary physiologic role, the reversible binding to oxygen with uptake in the lungs and release in the tissues. Above 70 mm Hg, PO₂ may vary widely with illness or altitude with minimal effect on oxygen content. From 70 to 40 mm Hg, a fall in PO₂ is associated with a proportionally larger increase in the release of oxygen while retaining a relatively high end-capillary PO₂ to promote oxygen diffusion into tissues. Below 40 mm Hg, small changes in PO₂ continue to release oxygen to tissues, down to very low PO₂ levels encountered in some capillary beds.

Distribution of Ventilation & Perfusion

Inhaled air and pulmonary arterial blood flow are not distributed equally to all lung regions. In healthy individuals, heterogeneous distribution is due principally to two factors: the effects of gravity and the fractal geometry of repetitive branching of airways and vessels.

Pleural pressure varies from the top to the bottom of the lung by approximately 0.25 cm H₂O/cm. It is more negative at the apex and more positive at the base. The effect is shifted to an anteroposterior distribution in the supine position and is greatly diminished (although not abolished) at zero gravity. Regional ventilation is dependent on regional pleural pressure (Figure 9–12). More negative pleural pressure at the lung apex causes greater expansion of apical alveoli. Because lung compliance is higher at lower lung volumes, ventilation is preferentially distributed to the lower lobes at FRC.

Pulmonary blood flow is a low-pressure system that functions in a gravitational field across 30 vertical centimeters. In the upright position, there is a nearly linear increase in blood flow from the top to the bottom of the lung. Within any horizontal (isogravitational) plane, however, there is significant heterogeneity of blood flow because of the fractal geometry of repetitive branching of vessels, resulting in heterogeneous resistance. Details of distribution are portrayed in Figure 9–13.

One additional factor that regulates blood flow is hypoxic pulmonary vasoconstriction. The smooth muscle cells of pulmonary arterioles are sensitive to alveolar PO₂ (much more so than to arterial PO₂). As alveolar PO₂ falls, there is arteriolar constriction, an increase in local resistance to flow, and redistribution of flow to regions of higher alveolar PO₂. When regionalized, this is an effective mechanism to diminish local blood flow without a significant increase in mean pulmonary arterial pressure. When it affects more than 20% of the pulmonary circulation, such as in the setting of global alveolar hypoxia, widespread pulmonary vasoconstriction increases mean pulmonary arterial pressure and may result in pulmonary hypertension.

Matching of Ventilation to Perfusion

The functional role of the lungs is to place ambient air in close proximity to circulating blood to permit gas exchange by simple diffusion. To accomplish this, air and blood flow must be directed to the same place at the same time. Optimum functioning of the respiratory system requires that ventilation be matched to perfusion.

In the normal individual, typical resting alveolar ventilation is approximately 4 L/min, whereas pulmonary artery blood flow is 5 L/min. This yields an overall ratio of ventilation to perfusion of 0.8. As noted above, ventilation and perfusion are both preferentially distributed to dependent regions at rest, although the increase in gravity-dependent flow is more marked with perfusion than with ventilation. Hence, the ratio of ventilation to perfusion is highest at the apex and lowest at the base (Figure 9–14).

Regional alterations in this overall distribution of ventilation and perfusion are referred to as V̇/Q̇ mismatch and are an extremely important phenomenon that underlies the functional impairment of many disease states. A distribution may favor high V̇/Q̇ ratios, with the limiting case being alveolar dead space (ventilation without perfusion, or V̇/Q̇ = ∞) or it may favor low V̇/Q̇ ratios, with the limiting case being a shunt (perfusion without ventilation, or V̇/Q̇ = 0). These two types of V̇/Q̇ mismatch affect respiratory function very differently.

In the normal individual, approximately one third of resting minute ventilation fills the main conducting airways. This is the anatomic dead space; it represents ventilation to areas that do not participate in gas exchange. If participating gas-exchanging regions of the lung are ventilated but not perfused, as may occur in pulmonary embolism, pulmonary vascular disease, or emphysema, these regions will also fail to function in gas exchange. Distributions shifted toward high V̇/Q̇ ratios are referred to as alveolar dead space or wasted ventilation (Figure 9–15, lower panel). Functionally, a shift toward high V̇/Q̇ ratios means that more work of breathing supports ventilation that does not participate in gas exchange, reducing the overall efficiency of ventilation.

In the absence of respiratory compensation, the primary effect of a shift toward high V̇/Q̇ ratios will be a rise in arterial PCO₂; PaO₂ may fall slightly as well. Because the respiratory control center is exquisitely sensitive to small changes in PaCO₂, however, the most common integrated response to increasing wasted ventilation is to increase total minute ventilation, thereby maintaining PaCO₂ nearly constant. PaO₂ remains normal or may be reduced if the fraction of wasted ventilation is large. The A-a ΔPO₂ is increased (discussed below). This adaptive response may be done unconsciously, but it presents a clinical problem when the individual can no longer sustain an increased minute ventilation, as in the patient with advanced emphysema.

A shift toward low V̇/Q̇ ratios occurs when regional ventilation is reduced or eliminated but perfusion persists, as might happen with atelectatic lung or in areas of lung consolidation where alveoli are filled with fluid or infected debris (Figure 9–15, mid panel). A shunt is the limiting case of a low V̇/Q̇ area where ventilation is absent and the ratio goes to zero. Pulmonary arterial (mixed venous) blood will then pass to the systemic arterial circulation without coming into contact with alveolar gas. The primary physiologic effect of such a right-to-left shunt is to reduce arterial PO₂.

Ventilation/perfusion mismatching commonly occurs between the limiting cases of true shunts and alveolar dead space. The effect on arterial blood gases of shifts in the distribution of V̇/Q̇ ratios can be predicted from the discussion of the limiting cases (Figure 9–16). At the top of Figure 9–16 is a respiratory unit where on one side (B) ventilation has been reduced but perfusion maintained. This defines an area of low V̇/Q̇ ratio. The effect on lung function can be understood by dividing it into an area with a normal V̇/Q̇ ratio (A) and an area of shunted blood (C). The physiologic effect of low V̇/Q̇ areas is similar to the effect of shunts: hypoxemia without hypercapnia. The difference between low V̇/Q̇ areas and true shunts can also be seen in this schematic. Shunted blood comes into no contact with inspired air; therefore, no amount of additional oxygen supplied to inspired air will reverse the fall in systemic arterial PO₂. A low V̇/Q̇ area does come in contact with inspired air and can be reversed with increased inspired oxygen.

At the bottom of Figure 9–16 is a respiratory unit where on one side blood flow has been decreased (B) but ventilation maintained. This defines an area of high V̇/Q̇ ratio. The effect on lung function can be understood by dividing the unit into an area with a normal V̇/Q̇ ratio (A) and an area of dead space or wasted ventilation (C). The physiologic effect of high V̇/Q̇ ratios is to increase PCO₂, typically leading to increased respiration to return PaCO₂ to normal.

Hyperventilation of unaffected lung regions can compensate for the increase in PCO₂ from alveolar dead space but cannot compensate for the decrease in PO₂ from areas of shunt. The reason is due to the different ways that oxygen and carbon dioxide are carried in blood, and the different relationships between the content and partial pressure of these gases. Because PCO₂ and CO₂ content are linearly related within the normal physiologic range, increased ventilation to one respiratory unit will reduce the PCO₂ and the CO₂ content of blood leaving that unit. Overall CO₂ content will be the mean of affected and unaffected units. Because PCO₂ is proportionate to the CO₂ content, the reduced CO₂ content of hyperventilated lung regions compensates for elevated content from areas of wasted ventilation. The PCO₂ and CO₂ content of the mixture track together.

Hyperventilation or increasing inspired oxygen to unaffected lung regions does not compensate for the decrease in PO₂ from areas of true shunt. The O₂ content of blood is not linearly related to PO₂ (Figure 9–11). The sigmoidal shape of the oxyhemoglobin dissociation curve indicates that hemoglobin is nearly maximally saturated at a PO₂ of 60. Increasing PO₂ from 60 to 600 increases partial pressure 10-fold but O₂ content by only 10%. Increasing ventilation or increasing alveolar PO₂ to unaffected respiratory units can increase end-capillary PO₂ but will not change the O₂ content of blood leaving those units. Overall O₂ content will be the mean of normal blood oxygen content and shunted, desaturated blood. The reduced oxygen content of the mixture tends to lie on the steep portion of the oxyhemoglobin dissociation curve, with the result that modest decreases in oxygen content lead to large decreases in the PO₂.

Arterial blood gases detect major disturbances in respiratory function. One attempt to assess more subtle abnormalities of gas exchange is to calculate the difference between the alveolar and arterial PO₂. This is referred to as the A-a ΔPO₂ or A-a DO₂. The alveolar-capillary membrane permits full equilibration of alveolar and end-capillary oxygen tension under normal V̇/Q̇ matching. There is nonetheless a small A-a ΔPO₂ in normal individuals as a result of right-to-left shunting through the bronchial veins and the thebesian veins of the left heart. This accounts for approximately 2–5% of resting cardiac output and leads to an A-a ΔPO₂ of 5–8 mm Hg in healthy young adults breathing ambient air at sea level. Increasing the fractional inspired concentration of oxygen (FiO₂) increases this value: A normal A-a ΔPO₂ breathing 100% oxygen is approximately 100 mm Hg. Normal values increase with age, presumably as a result of closure of dependent airways with consequent shift toward low V̇/Q̇ ratios. Further increases in the A-a ΔPO₂ reflect areas of low V̇/Q̇ ratio, including shunting.

Control of Breathing

The lungs inflate and deflate passively in response to changes in pleural pressure. Therefore, control over respiration lies in control of the striated muscles—chiefly the diaphragm but also the intercostals and abdominal wall—that change pleural pressure.

These muscles are under both automatic and voluntary control. The rhythm of spontaneous breathing originates in the brainstem, specifically in several groups of interconnected neurons in the medulla. Research into the generation of the respiratory rhythm has identified that it originates in the neurons in the pre-Bötzinger complex. Respiratory neurons are either inspiratory or expiratory and may fire early, late, or in an accelerating fashion during the respiratory cycle. Their integrated output is an efferent signal via the phrenic nerve (diaphragm) and spinal nerves (intercostals and abdominal wall) to generate rhythmic contraction and relaxation of the respiratory musculature. The result is spontaneous breathing without conscious input. However, by attending to breathing, the reader may alter this pattern. Eating, speaking, singing, swimming, and defecating all rely on voluntary control over automatic breathing.

Sensory Input

The frequency, depth, and timing of spontaneous breathing are modified by information provided to the respiratory center from both chemical and mechanical sensors (Figure 9–17).

There are chemoreceptors in the peripheral vasculature and in the brainstem. The peripheral chemoreceptors are the carotid bodies, located at the bifurcation of the common carotid arteries and the aortic bodies near the arch of the aorta. The carotid bodies are particularly important in humans. They function as sensors of arterial oxygenation. There is a graded increase in firing of the carotid body in response to a fall in the PaO₂. This response is most marked below 60 mm Hg. An increase in the PaCO₂ or a fall in arterial pH potentiates the response of the carotid body to decreases in the PaO₂.

In humans, the carotid bodies are solely responsible for the increased ventilation seen in response to hypoxia. Bilateral carotid body resection, which has been performed to treat disabling dyspnea and may happen as an unintended consequence of carotid thromboendarterectomy, results in a complete loss of this hypoxic ventilatory drive while leaving intact the response to changes in PaCO₂. Central chemoreceptors mediate the response to changes in PaCO₂. There is growing evidence that these chemoreceptors are widely dispersed throughout the brainstem. They are separate from the neurons that generate the respiratory rhythm. The increased ventilatory response to elevation in PaCO₂ is mediated through changes in chemoreceptor pH. The blood-brain barrier permits free diffusion of CO₂ but not hydrogen ions. CO₂ is hydrated to carbonic acid, which ionizes and lowers brain pH. Central chemoreceptors probably respond to these changes in intracellular hydrogen ion concentration.

There are a variety of pulmonary stretch receptors located in airway smooth muscle and mucosa whose afferent fibers are carried in the vagus nerve. They discharge in response to lung distention. Increasing lung volume decreases the rate of respiration by increasing expiratory time. This is known as the Hering-Breuer reflex. There are unmyelinated C fibers located near the pulmonary capillaries (hence juxtacapillary [J] receptors). These fibers are quiet during normal breathing but can be directly stimulated by intravenous administration of irritant chemicals such as capsaicin. They appear to stimulate the increased respiratory drive in interstitial edema and pulmonary fibrosis. Skeletal movement transmitted by proprioceptors in joints, muscles, and tendons causes an increase in respiration and may have some role in the increased ventilation of exercise. Finally, there are muscle spindle receptors in the diaphragm and intercostals that provide feedback on muscle force. They may be involved in the sensation of dyspnea when the work of breathing is disproportionate to ventilation.

Integrated Responses

Under normal conditions in healthy adults, the hydrogen ion concentration in the region of the central chemoreceptors determines the drive to breathe. Changes in chemoreceptor pH are largely determined by the PaCO₂. The PaO₂ is not an important part of the baseline respiratory drive under normal conditions.

Breathing is stimulated by a fall in the PaO₂, a rise in the PaCO₂, or an increase in the hydrogen ion concentration of arterial blood (fall in arterial pH).

Ventilation increases approximately 2–3 L/min for every 1 mm Hg rise in PaCO₂. This response (Figure 9–18) occurs first through sensitization of the carotid body receptor. The carotid body will increase its firing in response to an increased PaCO₂ even in the absence of changes in the PaO₂. This accounts for approximately 15% of the ventilatory response to hypercapnia. The majority of the response is mediated through pH changes in the region of the central chemoreceptors. Changes in arterial pH are additive to changes in PaCO₂. CO₂ response curves under conditions of metabolic acidosis have an identical slope but are shifted to the left. The ventilatory response to an increased PaCO₂ falls with age, sleep, and aerobic conditioning and with increased work of breathing.

The individual response to hypoxemia is extremely variable. Normally, there is little increase in ventilation until the PaO₂ falls below 50–60 mm Hg. At this point, there is a rapid increase in ventilation that reaches its maximum at approximately 32 mm Hg. Below this level, further decreases in PaO₂ lead to depression of ventilation. The response to hypoxia is affected by the PaCO₂. An increase in the alveolar PCO₂ will shift the isocapnic O₂ response curve upward and to the right (Figure 9–19).

A fall in arterial hydrogen ion concentration increases minute ventilation. This response results chiefly from stimulation of the carotid bodies and is independent of changes in PaCO₂. There is a response to severe metabolic acidosis in the absence of carotid bodies. It is assumed that this response is mediated by central chemoreceptors; it may represent breakdown of the blood-brain barrier.

Special Situations

1. Chronic hypercapnia—In patients with chronic hypercapnia, brain pH is returned toward normal by compensatory changes in serum and tissue bicarbonate levels. As a result, central chemoreceptors become less sensitive to further changes in PaCO₂. In this instance, a patient’s basal minute ventilation may depend on tonic stimuli from the carotid bodies. If such a patient were given high concentrations of inspired oxygen, carotid body output may be depressed, leading to a fall in minute ventilation. The change in minute ventilation does not fully account for the hypercapnia in response to supplemental oxygen, suggesting that ablation of hypoxic pulmonary vasoconstriction also plays a role.

2. Chronic hypoxia—Long-term residence at high altitude—or sleep apnea with repeated episodes of severe oxygen desaturation—may blunt the hypoxic ventilatory response. In such individuals, the development of lung disease and hypercapnia may attenuate all endogenous stimuli to breathing. This pattern is seen in patients with obesity-hypoventilation syndrome.

3. Exercise—Exercise may increase minute ventilation up to 25 times the resting level. Strenuous but submaximal exercise in a healthy individual typically causes no change or only a slight rise in PaO₂ as a result of increased pulmonary blood flow and better matching of ventilation and perfusion, with no change or a slight fall in PaCO₂. Changes in arterial oxygenation are, therefore, not a factor behind the increased ventilatory response to exercise. The reason for the increased ventilatory response is not known with certainty. Two contributing factors are the increased production of carbon dioxide and increased afferent discharge from joint and muscle proprioceptors.

Obstructive Lung Diseases: Asthma & Chronic Obstructive Pulmonary Disease (COPD)

The fundamental physiologic problem in obstructive diseases is increased resistance to expiratory airflow as a result of caliber reduction of conducting airways. This increased resistance can be caused by processes (1) within the lumen, (2) in the airway wall, or (3) in the supporting structures surrounding the airway. Examples of luminal obstruction include the increased secretions seen in asthma and chronic bronchitis. Airway wall thickening and airway narrowing can result from the inflammation seen in both asthma and chronic bronchitis or from the bronchial smooth muscle contraction in asthma. Emphysema is the classic example of obstruction caused by loss of surrounding supporting structure, with expiratory airway collapse resulting from the destruction of lung elastic tissue. Although the causes and clinical presentations of these diseases are distinct, the common elements of their physiology are instructive.

Asthma

Clinical Presentation

Asthma is a clinical syndrome with multiple phenotypes. This diversity reflects complex interactions between genetic predisposition and environmental exposure and suggests heterogeneity in underlying pathophysiology.

Asthma is a disease of airway inflammation and airflow obstruction characterized by intermittent symptoms, including wheezing, chest tightness, shortness of breath (dyspnea), and cough together with demonstrable bronchial hyperresponsiveness. Exposure to defined allergens or to various nonspecific stimuli initiates a cascade of cellular activation events in the airways, resulting in both acute and chronic inflammatory processes mediated by a complex and integrated assortment of locally released cytokines and other mediators. Release of mediators can alter airway smooth muscle tone and responsiveness, produce mucus hypersecretion, and damage airway epithelium. These pathologic events result in chronically abnormal airway architecture and function.

Inherent in the definition of asthma is the possibility of considerable variation in the magnitude and manifestations of the disease within and between individuals over time. For example, whereas many asthmatic patients have infrequent and mild symptoms, others may have persistent or prolonged symptoms of great severity. Similarly, initiating or exacerbating stimuli may be quite different between individual patients.

Epidemiology and Risk Factors

Asthma is a common chronic pulmonary disease, affecting as many as one third of adolescents in Australia and New Zealand. Overall U.S. prevalence was 8.5% in 2011, with higher rates in males who are younger than 18 years (10.2%) and in females who are older than 18 years (10.0%). Each year, approximately 500,000 hospital admissions and 4500 deaths in the United States are attributed to asthma. Prevalence, hospitalizations, and fatal asthma have all increased in the United States over the past 30 years. Mortality rates reached a plateau in the late 1990s and have declined slightly over the past decade. Hospitalization rates have been highest among blacks and children, and death rates are consistently highest among blacks aged 15–24 years.

The strongest identifiable predisposing factor for the development of asthma is atopy, or the production of immunoglobulin E (IgE) antibodies in response to exposure to allergens. Fifty-six percent of asthma cases were attributable to atopy in the National Health and Nutrition Examination Survey III (NHANES III) of 12,106 Americans done between 1988 and 1994. Exposure of sensitive patients to inhaled allergens increases airway inflammation, airway hyperresponsiveness, and asthma symptoms. Symptoms may develop immediately (immediate asthmatic response) or 4–6 hours after allergen exposure (late asthmatic response). Common allergens are listed in Table 9–3. In multiple studies, obesity is associated with an increased prevalence of asthma.

Indoor exposure to house dust mite or cockroach antigens is a strong risk factor for asthma. At the same time, it is well established that children raised on farms have lower prevalences of atopy and asthma, and the diversity of microbial exposure in childhood was shown in one study to be inversely related to risk of asthma. These observations, and the increased incidence of atopy, allergy, and autoimmune diseases in the developed world, have led researchers to pursue the underlying causes of atopy itself. One theory is that early childhood exposure to infections and/or organic antigens may fundamentally alter adaptive immunity. Some exposures may promote a T_H1 phenotype (differentiation of CD4+ T-helper cells toward a T_H1 response characterized by production of interferon γ), whereas the absence of these exposures may promote a T_H2 phenotype (characterized by a primary cytokine response including interleukin-4 [IL-4], IL-5, IL-13, and tumor necrosis factor that together are associated with atopy, allergic diseases, and asthma). The complexity of the immune response and its interactions with the human microbiome preclude any firm conclusions at this time, but this is a rapidly expanding area of research that promises to reshape our basic understanding of the etiology of atopy and asthma.

Pathogenesis

The fundamental abnormality in asthma is increased reactivity of airways to stimuli. As outlined in Table 9–3, there are many known provocative agents for asthma. These can be broadly categorized as follows: (1) physiologic or pharmacologic mediators of asthmatic airway responses, (2) allergens that can induce airway inflammation and reactivity in sensitized individuals, and (3) exogenous physicochemical agents or stimuli that produce airway hyperreactivity. Some of these provocative agents will produce responses in asthmatics only (eg, exercise, adenosine), whereas others produce characteristically magnified responses in asthmatics that can be used to distinguish them from non-asthmatics under controlled testing conditions (eg, histamine, methacholine; see later discussion). There is no single mechanism that serves to explain the occurrence of asthma in all individuals. There are, however, common events that characterize the pathologic processes that produce asthma. It is important to recognize the central role of airway inflammation in the evolution of asthma.

The earliest events in asthmatic airway responses are the activation of local inflammatory cells, principally mast cells and eosinophils. This can occur by specific IgE-dependent mechanisms or indirectly via other processes (eg, osmotic stimuli or chemical irritant exposure). Acute-acting mediators, including leukotrienes, prostaglandins, and histamine, rapidly induce smooth muscle contraction, mucus hypersecretion, and vasodilation with endothelial leakage and local edema formation. Epithelial cells appear also to be involved in this process, releasing leukotrienes and prostaglandins as well as inflammatory cytokines on activation. Some of these preformed and rapidly acting mediators possess chemotactic activity, recruiting additional inflammatory cells such as eosinophils and neutrophils to airway mucosa.

A critical process that accompanies these acute events is the recruitment, multiplication, and activation of immune inflammatory cells through the actions of a network of locally released cytokines and chemokines. Cytokines and chemokines participate in a complex and prolonged series of events that result in perpetuation of local airway inflammation and hyperresponsiveness (Table 9–4). These events include promoting growth of mast cells and eosinophils, influx and proliferation of T lymphocytes, and differentiation of B lymphocytes to IgE- and IgA-producing plasma cells. An important component of this process now appears to be the differentiation and activation of helper T lymphocytes of the T_H2 phenotype. These T_H2 lymphocytes, through their production of cytokines, including IL-3, IL-4, IL-5, IL-6, IL-9, IL-10, and IL-13, promote activation of mast cells, eosinophils, and other effector cells and drive IgE production by B cells, all of which are pathologic components of the asthma phenotype. Thus, through their specific mediators, these multiple cells participate in the many proinflammatory processes that are active in the airways of asthmatics. Among these are injury to epithelial cells and denudation of the airway, greater exposure of afferent sensory nerves, and consequent neurally mediated smooth muscle hyperresponsiveness; the upregulation of IgE-mediated mast cell and eosinophil activation and mediator release, including acute and long-acting mediators; and submucosal gland hypersecretion with increased mucus volume. Concurrently, the production of growth factors such as TGF-β, TGF-α, and fibroblast growth factor (FGF) by epithelial cells as well as macrophages and other inflammatory cells drives the process of tissue remodeling and submucosal airway fibrosis. This submucosal fibrosis can result in the fixed airway obstruction that may accompany the chronic airway inflammation in asthma.

Pathology

The histopathologic features of asthma reflect the cellular processes at play. Airway mucosa is thickened, edematous, and infiltrated with inflammatory cells, principally lymphocytes, eosinophils, and mast cells. Hypertrophied and contracted airway smooth muscle is seen. Bronchial and bronchiolar epithelial cells are frequently damaged, in part by eosinophil products such as major basic protein and eosinophil chemotactic protein, which are cytotoxic for epithelium. Epithelial injury and death leave portions of the airway lumen denuded, exposing autonomic and probably noncholinergic, nonadrenergic afferents that can mediate airway hyperreactivity. Secretory gland hyperplasia and mucus hypersecretion are seen, with mucus plugging of airways a prominent finding in severe asthma. Even in mildly involved asthmatic airways, inflammatory cells are found in increased numbers in the mucosa and submucosa, and subepithelial myofibroblasts are noted to proliferate and produce increased interstitial collagen; this may explain the component of relatively fixed airway obstruction seen in some asthmatics. The pathologic findings seen in severe fatal asthma parallel the pathologic events described previously but reflect the greater magnitude of the insult. More severe airway epithelial injury and loss are noted, often with severe and complete obstruction of the airway lumen by mucus plugs.

Pathophysiology

Local cellular events in the airways have important effects on lung function. As a consequence of airway inflammation and smooth muscle hyperresponsiveness, the airways are narrowed, resulting in an increase in airway resistance (recall that Raw ∝ 1/radius⁴). The small-caliber peripheral airways do not contribute significantly to airflow resistance in healthy individuals, but as these airways narrow in patients with asthma, they contribute substantially to airflow obstruction. Mucus hypersecretion and additional bronchoconstrictor stimuli may exacerbate obstructive lung physiology. Bronchial neural function also appears to play a role in the evolution of asthma, although this is probably of secondary importance. Cough and reflex bronchoconstriction mediated by vagal efferents follows stimulation of bronchial irritant receptors. Peptide neurotransmitters may also play a role. The proinflammatory neuropeptide substance P can be released from unmyelinated afferent fibers in the airways and can induce smooth muscle contraction and mediator release from mast cells. VIP is the peptide neurotransmitter of some airway nonadrenergic, noncholinergic neurons and functions as a bronchodilator; interruption of its action by cleavage of VIP can promote bronchoconstriction.

Airway obstruction occurs diffusely, although not homogeneously, throughout the lungs. As a result, ventilation of respiratory units becomes nonuniform and the matching of ventilation to perfusion is altered. Areas of both abnormally low and abnormally high V̇/Q̇ ratios exist, with the low V̇/Q̇ ratio regions contributing to hypoxemia. Pure shunt is unusual in asthma even though mucus plugging is a common finding, particularly in severe, fatal asthma. Arterial CO₂ tension is usually normal to low, given the increased ventilation seen with asthma exacerbations. Even mild hypercapnia should be viewed as an ominous sign during a severe asthma attack, indicating progressive airway obstruction, muscle fatigue, and falling alveolar ventilation.

Clinical Manifestations

The manifestations of asthma are readily explained by the presence of airway inflammation and obstruction.

Symptoms and Signs—

The variability of symptoms and signs is an indication of the tremendous range of disease severity, from mild and intermittent disease to chronic, severe, and sometimes fatal asthma.

1. Dyspnea and chest tightness—The sensations of dyspnea and chest tightness result from several concerted physiologic changes. The greater muscular effort required to overcome increased airway resistance is detected by spindle stretch receptors, principally of intercostal muscles and the chest wall. Hyperinflation from airway obstruction results in thoracic distention. Lung compliance falls, and the work of breathing increases, also detected by chest wall sensory nerves and manifested as chest tightness and dyspnea. As obstruction worsens, increased V̇/Q̇ mismatching produces hypoxemia. Rising arterial CO₂ tension and, later, evolving arterial hypoxemia (each alone or together as synergistic stimuli) will stimulate respiratory drive through peripheral and central chemoreceptors. This stimulus in the setting of respiratory muscle fatigue produces progressive dyspnea.

2. Wheezing—Smooth muscle contraction, together with mucus hypersecretion and retention, results in airway caliber reduction and prolonged turbulent airflow, producing auscultatory and audible wheezing. The intensity of wheezing does not correlate well with the severity of airway narrowing; as an example, with extreme airway obstruction, airflow may be so reduced that wheezing is barely detectable if at all.

3. Cough—Cough results from the combination of airway narrowing, mucus hypersecretion, and the neural afferent hyperresponsiveness seen with airway inflammation. It can also be a consequence of nonspecific inflammation after superimposed infections, particularly viral, in asthmatic patients. By virtue of the compressive narrowing and high velocity of airflow in central airways, cough provides sufficient shear and propulsive force to clear collected mucus and retained particles from narrowed airways.

4. Tachypnea and tachycardia—Tachypnea and tachycardia may be absent in mild disease but are virtually universal in acute exacerbations.

5. Pulsus paradoxus—Pulsus paradoxus is a fall of more than 10 mm Hg in systolic arterial pressure during inspiration. It appears to occur as a consequence of lung hyperinflation, with compromise of left ventricular filling together with augmented venous return to the right ventricle during vigorous inspiration in severe obstruction. With increased right ventricular end-diastolic volume during inspiration, the intraventricular septum is moved to the left, compromising left ventricular filling and output. The consequence of this decreased output is a decrease in systolic pressure during inspiration, or pulsus paradoxus.

6. Hypoxemia—Airway narrowing reduces ventilation to affected lung units, causing V̇/Q̇ mismatching with a shift toward low V̇/Q̇ ratios, resulting in an increase in the A-a ΔPO₂ and frank hypoxemia in severe cases. True shunt is unusual except in very severe asthma.

7. Hypercapnia and respiratory acidosis—In mild to moderate asthma, ventilation is normal or increased, and the arterial PCO₂ is either normal or decreased. In severe attacks, airway obstruction may worsen and respiratory muscle fatigue supervenes, with the evolution of alveolar hypoventilation, hypercapnia and respiratory acidosis. Note that this progression can occur despite continued tachypnea. An increased respiratory rate does not reverse alveolar hypoventilation because tidal volumes are reduced secondary to dynamic hyperinflation.

8. Obstructive defects by pulmonary function testing—Patients with mild asthma may have entirely normal pulmonary function between exacerbations. During active asthma attacks, all indices of expiratory airflow are reduced, including FEV₁, FEV₁/FVC (FEV₁%), and peak expiratory flow rate (Figure 9–20). FVC is often also reduced as a result of premature airway closure before full expiration. Administration of a bronchodilator improves airflow obstruction. A consequence of airflow obstruction is incomplete emptying of lung units at end expiration resulting in acute and chronic hyperinflation; total lung capacity (TLC), FRC, and residual volume (RV) can be increased. Pulmonary diffusing capacity for carbon monoxide (DLCO) is often increased as a consequence of the increased lung (and lung capillary blood) volume.

9. Bronchial hyperresponsiveness—Bronchial hyperresponsiveness is defined as either (1) a 12% or greater increase in the FEV₁ in response to an inhaled bronchodilator or (2) a 20% or greater decrease in FEV₁ in response to a provoking factor that, at the same intensity, causes less than a 5% change in a healthy individual. Methacholine and histamine are agents for which standardized provocation testing has been established. Such testing reveals nonspecific hyperresponsiveness in virtually all asthmatics, including those with mild disease and normal spirometry finding. Other agents that have been used to establish specific exposure sensitivities include sulfur dioxide and toluene diisocyanate.

COPD: Chronic Bronchitis & Emphysema

“Chronic obstructive pulmonary disease” is an intentionally imprecise term used to denote a process characterized by the presence of chronic bronchitis or emphysema that may lead to the development of fixed airway obstruction. Although chronic bronchitis and emphysema are often regarded as independent processes, they share some common etiologic factors and treatment strategies and are frequently encountered together in the same patient. It is for the purpose of including both under the same broad category that the definition remains imprecise: It reflects what we currently know about the causes and treatment of these diseases.

Clinical Presentation

Chronic Bronchitis—

Chronic bronchitis is defined by a clinical history of productive cough for 3 months of the year for 2 consecutive years. Dyspnea and airway obstruction, often with an element of reversibility, are intermittently to continuously present. Cigarette smoking is by far the leading cause, although other inhaled irritants may produce the same process. Although the predominant pathologic events are inflammation in larger airways, accompanied by mucosal thickening and mucus hypersecretion, it is the inflammation in smaller bronchioles that is the principal site of increased airflow obstruction.

Emphysema—

Pulmonary emphysema is a condition marked by irreversible enlargement of the airspaces distal to the terminal bronchioles, accompanied by destruction of their walls, most often without obvious fibrosis. In contrast to chronic bronchitis, the primary pathologic defect in emphysema is not in the airways but rather in the respiratory unit walls, where the loss of elastic tissue results in a loss of appropriate recoil tension to support distal airways during expiration. Progressive dyspnea and nonreversible obstruction accompany the airspace destruction without mucus hypersecretion and productive cough. Furthermore, the loss of alveolar surface area and the accompanying capillary bed for gas exchange contribute to the progressive hypoxia and dyspnea. Pathologic and etiologic distinctions can be made among various patterns of emphysema, but the clinical presentations of all are similar.

Etiology and Epidemiology

Epidemiologic data generally consider both chronic bronchitis and emphysema together under the rubric of COPD. COPD affects nearly 15 million persons in the United States; chronic bronchitis is the diagnosis in approximately two thirds of cases and emphysema in the remainder. COPD is the third leading cause of death in the United States, with 138,080 deaths in 2011. The incidence, prevalence, and mortality rates of COPD increase with age and are higher in men, whites, and persons of lower socioeconomic status. Cigarette smoking remains the principal cause of disease in up to 90% of patients with chronic bronchitis and emphysema. COPD is probably significantly under-diagnosed; although only 15–20% of smokers develop severe airflow obstruction, there is a dose-dependent relationship between tobacco smoke exposure and loss of lung function. Population-based studies suggest that chronic dust (including silica and cotton) and chemical fume exposure are significant contributing risk factors for COPD. In the developing world, indoor exposure to smoke from burning biofuels is a major cause of COPD.

The most important identified genetic risk factor for the evolution of COPD is deficiency of α₁-protease (α₁-antitrypsin) inhibitor. Reduced circulating and tissue levels can lead to early onset of severe emphysema. Alpha₁-protease inhibitor is capable of inhibiting several types of proteases, including neutrophil elastase, which is implicated in the genesis of emphysema (see Pathophysiology section below). Autosomal dominant mutations, especially in northern Europeans, produce abnormally low serum and tissue levels of this inhibitor, altering the balance of connective tissue synthesis and proteolysis. A homozygous mutation (the ZZ genotype) results in inhibitor levels 10–15% of normal. The risk of emphysema, particularly in smokers who carry this mutation, is dramatically increased.

Chronic Bronchitis—

A number of pathologic airway changes are seen in chronic bronchitis, although none is uniquely characteristic of this disease. The clinical features of chronic bronchitis can be attributed to chronic airway injury and narrowing. The principal pathologic features are inflammation of airways, particularly small airways, and hypertrophy of large airway mucous glands, with increased mucus secretion and accompanying mucus obstruction of airways (Figure 9–21). The airway mucosa is variably infiltrated with inflammatory cells, including polymorphonuclear leukocytes and lymphocytes. Mucosal inflammation can substantially narrow the bronchial lumen. As a consequence of the chronic inflammation, the normal ciliated pseudostratified columnar epithelium is frequently replaced by patchy squamous metaplasia. In the absence of normal ciliated bronchial epithelium, mucociliary clearance function is severely diminished or completely abolished. Hypertrophy and hyperplasia of submucosal glands are prominent features, with the glands often making up more than 50% of the bronchial wall thickness. Mucus hypersecretion accompanies mucous gland hyperplasia, contributing to luminal narrowing. Bronchial smooth muscle hypertrophy is common, and hyperresponsiveness to nonspecific bronchoconstrictor stimuli (including histamine and methacholine) can be seen. Bronchioles are often infiltrated with inflammatory cells and are distorted, with associated peribronchial fibrosis. Mucus impaction and luminal obstruction of smaller airways are often seen. In the absence of any superimposed process, such as pneumonia, the gas-exchanging lung parenchyma, composed of terminal respiratory units, is largely undamaged. The result of these combined changes is chronic airway obstruction and impaired clearance of airway secretions.

The nonuniform airway obstruction of chronic bronchitis has substantial effects on ventilation and gas exchange. Airway narrowing that leads to prolonged expiratory time produces hyperinflation. Ventilation/perfusion relationships are altered with increased areas of low V̇/Q̇ ratios. These low V̇/Q̇ mismatches are largely responsible for the more significant resting hypoxemia seen in chronic bronchitis, compared with that seen in emphysema. True shunt (perfusion with no ventilation) is unusual in chronic bronchitis.

Emphysema—

The principal pathologic event in emphysema is thought to be a continuing destructive process resulting from an imbalance of local oxidant injury and proteolytic (particularly elastolytic) activity caused by a deficiency of protease inhibitors (Figure 9–22). Oxidants, whether endogenous (superoxide anion) or exogenous (eg, cigarette smoke), can inhibit the normal protective function of protease inhibitors, allowing progressive tissue destruction.

In contrast to chronic bronchitis, which is a disease of the airways, emphysema is a disease of the surrounding lung parenchyma. The physiologic consequences result from three important changes: (1) destruction of terminal respiratory units, (2) loss of alveolar-capillary bed, and (3) loss of the supporting structures of the lung, including elastic connective tissue. The loss of elastic connective tissue reduces normal support of noncartilaginous airways, leading to a lung with diminished elastic recoil and increased compliance. Premature expiratory collapse of airways ensues, with characteristic obstructive symptoms and physiologic findings.

The pathologic picture of emphysema is one of progressive destruction of terminal respiratory units or lung parenchyma distal to terminal bronchioles. Airway inflammatory changes are minimal if present, although some mucous gland hyperplasia can be seen in large conducting airways. The interstitium of respiratory units harbors some inflammatory cells, but the chief finding is a loss of alveolar walls and enlargement of airspaces. Alveolar capillaries are also lost, which can result in decreased diffusing capacity and progressive hypoxemia, particularly with exercise.

Alveolar destruction is not uniform in all cases of emphysema. Anatomic variants have been described on the basis of the pattern of destruction of the terminal respiratory unit (or acinus, as it is also known). In centriacinar emphysema, destruction is focused in the center of the terminal respiratory unit, with the respiratory bronchioles and alveolar ducts relatively spared. This pattern is most frequently associated with prolonged smoking. Panacinar emphysema involves destruction of the terminal respiratory unit globally, with diffuse airspace distention. This pattern is typically, although not uniquely, seen in α₁-protease inhibitor deficiency. It is important to note that the distinction between these two patterns is largely pathologic; there is no significant difference in the clinical presentation. An additional emphysema pattern of clinical importance is bullous emphysema. Bullae are large confluent airspaces formed by greater local destruction or progressive distention of lung units. They are important because of the compressive effect they can have on surrounding lung and the large physiologic dead space associated with these structures.

Clinical Manifestations

Chronic Bronchitis—

The clinical manifestations of chronic bronchitis are principally the result of the obstructive and inflammatory airway process.

1. Cough with sputum production—Cough is productive of thick, often purulent sputum owing to the ongoing local inflammation and the high likelihood of bacterial colonization and infection. Sputum viscosity is increased largely as a result of the presence of free DNA (of high molecular weight and highly viscous) from lysed cells. With increased inflammation and mucosal injury, hemoptysis can occur but is usually scant. Cough, which is very effective in clearing normal airways, is much less effective because of the narrow airway caliber and the greater volume and viscosity of secretions.

2. Wheezing—Persistent airway narrowing and mucus obstruction can produce localized or more diffuse wheezing. This may be responsive to bronchodilators, representing a reversible component to the obstruction.

3. Inspiratory and expiratory coarse crackles—Increased mucus production, together with defective mucociliary escalator function, leaves excessive secretions in the airways, despite increased coughing. These are heard prominently in larger airways during tidal breathing or with cough.

4. Cardiac examination—Tachycardia is common, especially with exacerbations of bronchitis or with hypoxemia. If hypoxemia is significant and chronic, pulmonary hypertension can result; cardiac examination reveals a prominent pulmonary valve closing sound (P₂) or elevated jugular venous pressure and peripheral edema.

5. Imaging—Typical chest radiographic findings include increased lung volumes with relatively depressed diaphragms consistent with hyperinflation. Prominent parallel linear densities (“tram track lines”) of thickened bronchial walls are common. Cardiac size may be increased, suggesting right heart volume overload. Prominent pulmonary arteries are common and are associated with pulmonary hypertension.

6. Pulmonary function tests—Diffuse airway obstruction is demonstrated on pulmonary function testing as a global reduction in expiratory flows and volumes. FEV₁, FVC, and the FEV₁/FVC (FEV₁%) ratio are all reduced. The expiratory flow-volume curve shows substantial limitation of flow (Figure 9–20). Some patients may respond to bronchodilators. Measurement of lung volumes reveals an increase in the RV and FRC, reflecting air trapped in the lung as a result of diffuse airway obstruction and early airway closure at higher lung volumes. DLCO is typically normal, reflecting a preserved alveolar-capillary bed.

7. Arterial blood gases—Ventilation/perfusion mismatching is common in chronic bronchitis. The A-a ΔPO₂ is increased and hypoxemia is common mainly because of significant areas of low V̇/Q̇ ratios (physiologic shunt); hypoxemia at rest tends to be more profound than in emphysema. With increasing obstruction, increasing PCO₂ (hypercapnia) and respiratory acidosis, with compensatory metabolic alkalosis, are seen.

8. Polycythemia—Chronic hypoxemia is associated with a variable erythropoietin-mediated increase in hematocrit. With more severe and prolonged hypoxia, the hematocrit may increase to well over 50%.

Emphysema—

Emphysema presents as a noninflammatory disease manifested by dyspnea, progressive nonreversible airway obstruction, and abnormalities of gas exchange, particularly with exercise.

1. Breath sounds—Breath sounds in emphysema are typically decreased in intensity, reflecting decreased airflow, prolonged expiratory time, and prominent lung hyperinflation. Wheezes, when present, are of diminished intensity. Airway sounds, including crackles and rhonchi, are unusual in the absence of superimposed processes such as infection or pulmonary edema.

2. Cardiac examination—Tachycardia may be present as in chronic bronchitis, especially with exacerbations or hypoxemia. Pulmonary hypertension is a common consequence of pulmonary vascular obliteration. Cardiac examination may reveal prominent pulmonary valve closure (increased P₂, pulmonary component of the second heart sound). Elevated jugular venous pressure and peripheral edema are less common findings than in chronic bronchitis.

3. Imaging—Hyperinflation is common, with flattened hemidiaphragms and an increased anteroposterior chest diameter. Parenchymal destruction produces attenuated lung peripheral vascular markings, often with proximal pulmonary artery dilation as a result of secondary pulmonary hypertension. Cystic or bullous changes may also be seen.

4. Pulmonary function tests—Lung parenchymal destruction and loss of lung elastic recoil are the fundamental causes of pulmonary function abnormalities. The loss of elastic support in lung tissue surrounding the airways results in increased dynamic compression of airways (Figure 9–9), especially during forced expiration. Premature airway collapse reduces all flows including FEV₁, FVC, and the FEV₁/FVC (FEV₁% ratio). As with chronic bronchitis and asthma, the expiratory flow-volume curve shows substantial limitation in flow (Figure 9–20). Expiratory time prolongation, early airway closure caused by loss of elastic recoil, and consequent air trapping increase the RV and FRC. TLC is increased, although often a substantial amount of this increase comes from gas trapped in poorly or noncommunicating lung units, including bullae. The DLCO is generally decreased in proportion to the extent of emphysema, reflecting the progressive loss of alveoli and their capillary beds.

5. Arterial blood gases—Emphysema is a disease of alveolar wall destruction. The loss of alveolar capillaries creates V̇/Q̇ mismatches with areas of high ventilation relative to perfusion. Typically, patients with emphysema adapt to high V̇/Q̇ ratios by increasing their minute ventilation. They may maintain nearly normal PO₂ and PCO₂ levels despite advanced disease. Examination of arterial blood gases invariably reveals an increase in the A-a ΔPO₂. With greater disease severity and further loss of capillary perfusion, the DLCO falls, leading to exercise-related and, ultimately, resting arterial hemoglobin desaturation. Hypercapnia, respiratory acidosis, and a compensatory metabolic alkalosis are common in severe disease.

6. Polycythemia—As in chronic bronchitis, chronic hypoxemia is frequently associated with an elevated hematocrit.

Restrictive Lung Disease: Idiopathic Pulmonary Fibrosis

Interstitial lung disease, or diffuse parenchymal lung disease, is a descriptive term that encompasses over 180 different disorders. These multiple disorders are grouped together because of shared pathologic, physiologic, clinical, and radiographic features. The most common feature of diffuse parenchymal lung disease is the infiltration of the lung by inflammatory cells and fluid, leading to scarring, fibrosis, and capillary obliteration (Figure 9–23). Diffuse lung fibrosis leads to increased lung elastic recoil, decreased lung compliance and lung volumes, and worsening V̇/Q̇ mismatch, leading to impairment in gas exchange, a pattern we know as restrictive lung disease.

Diffuse parenchymal lung disease is often referred to as interstitial lung disease, but the modifier “interstitial” is an incomplete characterization of the pathologic process. The lung interstitium formally refers to the region of the alveolar wall exclusive of and separating the basement membranes of alveolar epithelial and pulmonary capillary endothelial cells. In the normal lung, this interstitium is a potential space that may contain a few mesenchymal cells (eg, fibroblasts), extracellular matrix molecules (eg, collagen, elastin, and proteoglycans), and tissue leukocytes, including mast cells and lymphocytes. Under pathologic conditions, not only is the interstitium affected, but all elements of the alveolar wall may be involved, including alveolar epithelial and capillary endothelial cells. In addition, the alveolar space is often affected including loss of alveolar capillaries. This extensive disruption of normal lung structure by interstitial processes profoundly influences lung function.

The most commonly identified causes of interstitial lung disease are related to occupational and environmental exposures, especially to organic and inorganic dusts. There are also interstitial lung diseases of unknown etiology such as idiopathic pulmonary fibrosis (IPF). The physiologic consequences seen in IPF are typical of other causes of diffuse parenchymal lung disease, particularly in their advanced stages. For that reason, the remainder of this section focuses on IPF.

Clinical Presentation

IPF, previously known as interstitial pulmonary fibrosis or cryptogenic fibrosing alveolitis, is marked by chronic inflammation of alveolar walls, resulting in diffuse and progressive fibrosis and destruction of normal lung architecture. This process produces not only a restrictive defect, with altered ventilation and increased work of breathing, but destructive and obliterative vascular injury that can severely impair normal pulmonary perfusion and gas exchange.

The usual presentation of IPF is with the insidious onset, over months to years, of progressively severe dyspnea, accompanied by a dry and persistent cough. Fever and chest pain are generally absent. With disease progression, dyspnea may occur even at rest. Digital cyanosis, clubbing, and pulmonary hypertension are common in later stages of the disease.

The diagnosis of IPF may be made by high-resolution chest CT in the setting of an appropriate clinical history and pulmonary function tests or by surgical lung biopsy. Typical CT findings are described below. The histopathologic correlate to IPF is usual interstitial pneumonia, or UIP, a temporally and spatially heterogeneous pattern of mature collagen deposition and alveolar wall destruction with scattered clusters of active fibrosis called fibroblastic foci containing both fibroblasts and myofibroblasts. Inflammatory cells are present but usually sparse.

Etiology and Epidemiology

Compared with COPD, IPF is an uncommon disorder with an estimated prevalence of 14 per 100,000 people in the United States. It typically presents in the fifth to seventh decades of life, with a male predominance of 1.5:1.0. By definition, the term “idiopathic” refers to cases with no known causative agent. Major risk factors include tobacco smoke exposure and environmental exposures to organic and inorganic dust. Several systemic diseases including scleroderma, sarcoidosis, and hypersensitivity pneumonitis can, in advanced stages, produce clinical, imaging, and histopathologic findings indistinguishable from IPF. Therefore, one should investigate the occupational history and potential for connective tissue disorders when evaluating a patient with diffuse parenchymal lung disease, because early findings may alter the evaluation or treatment options. An inherited form of pulmonary fibrosis has been described, but typical cases do not appear to have a genetic basis. Associations with aging and other disorders including emphysema, gastroesophageal reflux disease, and obesity remain to be defined.

The natural history of IPF is unremitting progression: Median survival is approximately 3 years from diagnosis. The clinical course is nonetheless very heterogeneous and may reflect separate phenotypes rather than variable progression. Risk factors for accelerated progression include older age at diagnosis (>70 years), cumulative tobacco smoke exposure, and severity of disease by symptoms (dyspnea score) or standardized assessment (extent of radiographic disease, severity of pulmonary restriction on pulmonary function tests, presence of pulmonary hypertension).

Pathophysiology

Diffuse parenchymal lung disease encompasses many disorders with different precipitating events and possibly different cellular and molecular mechanisms. We can describe a common series of cellular events that mediate and regulate lung inflammatory processes and fibrotic responses (Table 9–5). These events include (1) initial tissue injury; (2) vascular injury and endothelial cell activation, with increased permeability, exudation of plasma proteins into the extravascular space, and variable thrombosis and thrombolysis; (3) alveolar epithelial cell injury and activation, with loss of barrier integrity and release of proinflammatory mediators; (4) increased leukocyte adherence to activated endothelium, with transit of activated leukocytes into the interstitium; and (5) continued injury and repair processes characterized by alterations in cell populations and increased matrix production.

The challenge in describing the pathophysiology of IPF and its histopathologic correlate UIP is that neither the primary injury nor the mechanisms that perpetuate an unregulated fibrotic response is known. Indeed, current research on IPF focuses on ways that it may be distinct from other causes of diffuse parenchymal lung disease since IPF is characterized by progressive fibrosis in the setting of minimal inflammation. Exciting work is being done on alveolar epithelial cell injury and apoptosis leading to abnormal activation of adjacent epithelial cells, alterations in surfactant protein folding, deregulation of embryological pathways in the response to injury, and accelerated shortening of telomeres.

A unique feature of the histologic findings in IPF is that the process of lung injury and scarring is neither uniform nor synchronous: areas of intense injury and fibrosis are often intermixed with relatively spared lung. In the early stages of disease, patchy type II alveolar epithelial hyperplasia accompanies infiltration of alveolar structures by leukocytes. Damage to type II alveolar epithelial cells alters the production and turnover of surfactant, with an increase in alveolar surface tension in affected lung units. This is followed by increasing tissue leukocytosis, fibroblast proliferation, and scar formation. Lymphocytes—predominantly T cells—and mast cells are found in increased numbers in alveolar interstitium and submucosal regions. Collagen and elastin deposition are markedly increased. Later in the course of the disease, there is progressive alveolar destruction, with large areas of fibrosis and residual airspaces lined by cuboidal epithelium; these appear on radiographs as honeycombing. Along with alveolar destruction, the accompanying vascular bed is obliterated, also in a patchy pattern. The result is an altered physiology that includes increased elastic recoil and poor lung compliance, altered gas exchange, and pulmonary vascular abnormalities. The pathophysiology of interstitial lung diseases is summarized in Table 9–6.

Clinical Manifestations

Symptoms and Signs

1. Cough—An intermittent, irritating, nonproductive cough is often the first symptom of IPF. It may be refractory to antitussive therapy. The mechanism is likely multifactorial, with fibrotic damage to terminal respiratory units causing bronchial and bronchiolar distortion, which leads to alterations in both stimulatory and inhibitory nerve fibers involved in cough reflexes. Although epithelial cells may be injured, mucus hypersecretion and a productive cough are not typically seen in early disease.

2. Dyspnea and tachypnea—Multiple factors contribute to dyspnea in patients with IPF. Fibrosis of lung parenchyma decreases lung compliance; in combination with alterations in surfactant turnover, the distending pressure required to inflate the lungs increases, as does the work of breathing. Increased stimuli from C fibers in fibrotic alveolar walls or stretch receptors in the chest wall may sense the increased force necessary to inflate less compliant lungs. Tachypnea results from increased lung sensory receptor stimuli and the attempt to maintain a normal alveolar minute ventilation (and hence normal PaCO₂) as lung volumes decrease. A rapid, shallow breathing pattern also reduces ventilatory work in the face of increased lung elastic recoil. The diminished capillary bed and thickened alveolar-capillary membrane contribute to hypoxemia with exercise. In advanced disease, altered gas exchange with severe V̇/Q̇ mismatching can produce hypoxemia at rest.

3. Inspiratory crackles—Diffuse fine, dry inspiratory crackles are common and reflect the successive opening on inspiration of respiratory units that are collapsed owing to the fibrosis and the loss of normal surfactant.

4. Digital clubbing—Clubbing of the fingers and toes is a common finding, but the cause is unknown. There is no established link with any specific physiologic variable, including hypoxemia.

5. Cardiac examination—As with hypoxemia from other causes, cardiac examination can reveal evidence of pulmonary hypertension with prominent pulmonary valve closure sound (P₂). This can be accompanied by right heart overload or decompensation, with elevated jugular venous pressure, the murmur of tricuspid regurgitation, or a right-sided third heart sound (S₃).

Imaging

Characteristic chest radiograph findings include reduced lung volumes with increased reticular opacities prominent in the lung periphery and loss of definition of vascular structures, diaphragms, and cardiac border. Fibrosis surrounding expanded small airspaces is seen as honeycombing. With pulmonary hypertension, central pulmonary arteries may be enlarged. Typical chest CT findings include diffuse septal thickening, increased subpleural reticular opacities, traction bronchiectasis with pleural scalloping, and subpleural clustered small (3–10 mm) cystic airspaces (honeycombing). Ground glass opacities are usually absent.

Pulmonary Function Tests

Lung fibrosis typically produces a restrictive ventilatory pattern, with reductions in TLC, FEV₁, and FVC, while maintaining a preserved or even increased ratio of FEV₁/FVC (FEV₁%) (Figure 9–20). The increased elastic recoil produces normal to increased expiratory flow rates when adjusted for lung volume. The DLCO in lung fibrosis is progressively reduced as a function of the fibrotic obliteration of lung capillaries.

Arterial Blood Gases

Hypoxemia is common in advanced IPF. It results from the patchy nature of fibrosis, causing extreme variability in regional compliance and ventilation leading to prominent V̇/Q̇ mismatching shifted toward low V̇/Q̇ ratios. Cardiac output tends to be low, which reduces mixed venous PO₂ (P_mvO₂). Diffusion impairment increases with the severity of fibrosis but rarely contributes to resting hypoxemia. Diffusion impairment is a significant contributor to exercise-induced desaturation when the combination of a low P_mvO₂ and reduced capillary transit time limits oxygen loading of hemoglobin (Figure 9–24). Arterial PCO₂ is typically low because of increased ventilation due to hypoxia and the irritant stimuli of lung fibrosis. Only in the later stages of disease, when increased lung elastic recoil and work of breathing prevent appropriate ventilation, does the PaCO₂ rise above normal. Hypercapnia is a grave sign, implying an inability to maintain adequate alveolar ventilation as a result of excess work of breathing.

Pulmonary Edema

Clinical Presentation

Pulmonary edema is the accumulation of excess fluid in the extravascular compartment of the lungs, principally in the interstitium and alveolar spaces. This accumulation may occur slowly, as in a patient with occult renal failure, or emergently, as in left ventricular failure after an acute myocardial infarction. Pulmonary edema most commonly presents with dyspnea. Dyspnea is breathing perceived by a patient as both uncomfortable or anxiety-provoking and disproportionate to the level of activity. Dyspnea from pulmonary edema may be present only with exertion, or the patient may experience dyspnea at rest. In severe cases, pulmonary edema may be accompanied by edema fluid in the sputum and can cause acute respiratory failure.

Etiology

Pulmonary edema is a common problem associated with a variety of medical conditions (Table 9–6). In light of these multiple causes, it is helpful to think about pulmonary edema in terms of underlying physiologic principles.

Pathophysiology

All blood vessels leak; under normal conditions, fluid moves between blood vessels and the spaces around them while protein flux is minimal. In the adult human, the pulmonary capillaries are the major site of fluid flux from the pulmonary vasculature. The Starling equation—(Jv ≈ K ([Pc − Pi] − σ[πc − πi])—describes the movement of fluid between the pulmonary capillaries and the pulmonary extravascular compartment. The flux of fluid across a semipermeable membrane (Jv) is related to the inherent permeability of the membrane to fluid (K = fluid filtration coefficient of the capillary endothelium) and macromolecules (σ = protein reflection coefficient of the capillary endothelium) as well as the hydrostatic and colloid oncotic pressure gradients across the membrane (Pc = capillary hydrostatic pressure, Pi = interstitial hydrostatic pressure, and πc and πi are the capillary and interstitial colloid oncotic pressures, respectively).

Extravasation of fluid from the capillaries into the surrounding interstitial spaces is limited by tight junctions between pulmonary capillary endothelial cells. The pulmonary alveolar spaces are normally protected from extravascular fluid in the interstitial space by three mechanisms: (i) an alveolar epithelial cell barrier that is nearly impermeable to the passage of protein, (ii) pericapillary lymphatics that normally clear fluid before it can accumulate sufficiently to overwhelm the alveolar epithelial barrier, and (iii) active transport of sodium from the alveolus that regulates the amount of fluid in the alveolar space. Under normal conditions:

1. Pulmonary capillary hydrostatic pressure exceeds interstitial hydrostatic pressure, and therefore, hydrostatic forces favor fluid movement out of the capillaries into the interstitial space.

2. Pulmonary capillary colloid oncotic pressure exceeds interstitial colloid oncotic pressure, favoring fluid movement out of the interstitial space into the capillaries. In addition, the interstitial colloid oncotic pressure exceeds the alveolar fluid colloid oncotic pressure, favoring fluid movement out of the alveolar space into the interstitium.

3. The effect of hydrostatic forces is greater than that of colloid oncotic forces, and thus, there is a tendency for net fluid movement out of the capillaries into the interstitial spaces. The net rate of fluid movement out of the pulmonary capillaries under normal conditions is approximately 15–20 mL/h, representing less than 0.01% of pulmonary blood flow.

4. Fluid in the lung interstitial space is removed by pericapillary lymphatics that do not enter the alveolar wall and are termed “juxta-alveolar.” The pericapillary interstitium is contiguous with the perivascular and peribronchial interstitium. Interstitial pressure in these more central areas is negative relative to the pericapillary interstitium, so fluid tracks centrally away from the airspaces. In effect, the perivascular and peribronchiolar interstitium acts as a sump for fluid, and it can accommodate approximately 500 mL with a negligible increase in interstitial hydrostatic pressure. Because this fluid is protein-depleted relative to blood, osmotic forces favor resorption from the interstitium into blood vessels adjacent to these central areas. This is the major site of resorption of fluid from the perivascular and peribronchiolar interstitium. Edema fluid may track further into the mediastinum where it is taken up by mediastinal lymphatics. The perivascular and peribronchiolar interstitium is also contiguous with the interlobular septa and the visceral pleura. In some patients, a significant amount of fluid transits out through the visceral pleural into the pleural space where there is high capacity resorption through pores on the parietal pleura into parietal pleural lymphatics.

The rate of fluid resorption by the lymphatic system is usually sufficient to prevent accumulation of fluid in the interstitium and the alveolar spaces. Pulmonary edema occurs when fluid exits the pulmonary vascular space, exceeds the capacity for fluid clearance, and accumulates in the extravascular spaces of the lung. At some undefined critical level after the perivascular and peribronchiolar interstitium has been filled, increased interstitial hydrostatic pressure causes edema fluid to enter the alveolar space (Figure 9–25). The pathway into the alveolar space remains unknown but is thought to occur by bulk flow. Pulmonary edema may occur under a number of different conditions:

1. The hydrostatic pressure gradient increases (elevated pulmonary capillary hydrostatic pressure). Pulmonary edema that occurs in this setting is termed cardiogenic or hydrostatic pulmonary edema. This is primarily a mechanical process resulting in an ultrafiltrate of plasma. Edema fluid in this setting has a relatively low protein content, generally less than 60% of a patient’s plasma protein content. In normal individuals, pulmonary capillary pressure (ie, pulmonary capillary wedge pressure) must exceed approximately 20 mm Hg before the fluid leaving the vascular space exceeds the rate of resorption, leading to accumulation of interstitial and ultimately alveolar fluid that we describe as pulmonary edema. Cardiogenic or hydrostatic pulmonary edema classically results from elevated pulmonary venous and left atrial pressures due to left ventricular systolic or diastolic failure, mitral stenosis, or mitral regurgitation.

2. Vascular endothelial cell and/or alveolar epithelial cell permeability increases. Pulmonary edema that occurs in this setting is termed non-cardiogenic or permeability pulmonary edema. The permeability of the endothelial or epithelial barrier may increase as a result of cellular injury. This is primarily an inflammatory process that typically results in dysfunction of both the endothelial and epithelial barriers. In this setting, both fluid and protein permeability increase, although there may be little change in hydrostatic pressure. Under conditions of increased membrane permeability, edema fluid has protein content similar to intravascular fluid, generally at least 70% of the plasma protein content. The acute respiratory distress syndrome (ARDS) epitomizes this type of pulmonary edema (detailed discussion below).

3. The oncotic pressure gradient decreases (low plasma colloid oncotic pressure). Edema fluid in this setting has a relatively low protein content. Hypoalbuminemia due to prolonged illness or nephrotic syndrome can cause this type of pulmonary edema.

4. Lymphatic drainage is impaired. This form of pulmonary edema is rare but may be seen with physical obstruction of the lymphatic system from malignancy (lymphoma) or infection (histoplasmosis, tuberculosis), from obliteration of lymphatics due to radiation therapy for breast or lung cancer, or from idiopathic causes (yellow nail syndrome).

Hydrostatic and permeability pulmonary edema are not mutually exclusive; indeed, they are closely linked. Pulmonary edema occurs when the hydrostatic pressure is excessive for a given capillary permeability and for a given rate of clearance of interstitial fluid. For instance, in the presence of damaged capillary endothelium, small increases in an otherwise normal hydrostatic pressure gradient may cause large increases in edema formation. Similarly, if the alveolar epithelial barrier is damaged, even the baseline flux of fluid across an intact capillary endothelium may cause alveolar filling.

The pathophysiology of increased permeability pulmonary edema (ARDS) is complex and may result from multiple different insults. Alveolar fluid accumulates as a result of loss of integrity of the alveolar epithelial barrier, allowing solutes and large molecules such as albumin to enter the alveolar space. This loss of integrity may result from direct injury to the alveolar epithelium by inhaled toxins or pulmonary infection, or they may occur after primary injury to the pulmonary capillary endothelium by circulating toxins, as in sepsis or pancreatitis, followed by secondary inflammatory injury to the alveolar epithelial barrier. This is in contrast to cardiogenic pulmonary edema, in which both the alveolar epithelium and the capillary endothelium are usually intact. The potential causes of pulmonary edema leading to ARDS include a diverse group of clinical entities (see Table 9–7). Many diverse problems are grouped together in the syndrome called ARDS because they share injury to the alveolar epithelium and impaired pulmonary surfactant, which results in characteristic changes in pulmonary mechanics and lung function.

With inhalation injury, such as that produced by mustard gas during World War I, there is direct chemical injury to the alveolar epithelium that disrupts this normally tight cellular barrier. The presence of high-protein fluid in the alveolus, particularly the presence of fibrinogen and fibrin degradation products, inactivates pulmonary surfactant, causing large increases in surface tension. This results in a fall in pulmonary compliance and alveolar instability, leading to areas of atelectasis. Increased surface tension decreases the interstitial hydrostatic pressure and favors further fluid movement into the alveolus. A damaged surfactant monolayer may increase susceptibility to infection as well.

Circulating factors may act directly on the capillary endothelium or may affect it through various immunologic mediators. A common clinical scenario is gram-negative bacteremia. Bacterial endotoxin does not cause endothelial damage directly; it causes neutrophils and macrophages to adhere to endothelial surfaces and to release a variety of inflammatory mediators such as leukotrienes, thromboxanes, and prostaglandins as well as oxygen radicals that cause oxidant injury. Both macrophages and neutrophils may release proteolytic enzymes that cause further damage. Alveolar macrophages may also be stimulated. Vasoactive substances may cause intense pulmonary vasoconstriction, leading to capillary failure.

The histopathology of increased permeability pulmonary edema reflects these changes. The lungs appear grossly edematous and heavy. The surface appears violaceous, and hemorrhagic fluid exudes from the cut pleural surface. Microscopically, there is cellular infiltration of the interalveolar septa and the interstitium by inflammatory cells and erythrocytes. Type I pneumocytes are damaged, leaving a denuded alveolar barrier. Sheets of pink material composed of plasma proteins, fibrin, and coagulated cellular debris line the denuded basement membrane. These are called hyaline membranes. The inflammatory injury progresses to fibrosis in some cases, although complete recovery with regeneration of the alveolar epithelium from the type II pneumocytes may also occur.

Clinical Manifestations

Cardiogenic and noncardiogenic pulmonary edema both result in increased extravascular lung water, and both may result in respiratory failure. Given the differences in pathophysiology, it is not surprising that the clinical manifestations are very different in the two syndromes.

Increased Hydrostatic Pulmonary Edema (Cardiogenic Pulmonary Edema)

Early increases in pulmonary venous pressure may be asymptomatic. The patient may notice only mild exertional dyspnea or a nonproductive cough stimulated by activation of irritant receptors coupled with C fibers. Orthopnea and paroxysmal nocturnal dyspnea occur when recumbency causes redistribution of blood or edema fluid, usually pooled in the lower extremities, into the venous circulation, thereby increasing thoracic blood volume and pulmonary venous pressures.

Clinical signs begin with the accumulation of interstitial fluid. Cardiac examination may reveal a third heart sound, but there is a paucity of lung findings in purely interstitial edema. The earliest sign is frequently a chest radiograph showing an increase in the caliber of the upper lobe vessels (“pulmonary vascular redistribution”) and fluid accumulating in the perivascular and peribronchial spaces (“cuffing”). It may also show Kerley B lines, which represent fluid in the interlobular septa. Pulmonary compliance falls, and the patient begins to breathe more rapidly and shallowly to adapt to increased elastic work of breathing. As alveolar flooding begins, there are further decreases in lung volume and pulmonary compliance. With some alveoli filled with fluid, there is an increase in the fraction of the lung that is perfused but poorly ventilated. This shift toward low V̇/Q̇ ratios causes an increase in A-a ΔPO₂, if not frank hypoxemia. Supplemental oxygen corrects the hypoxemia. The PaCO₂ is normal or low, reflecting the increased drive to breathe. The patient may become sweaty and cyanotic. The sputum may show edema fluid that is pink from capillary hemorrhage from high pulmonary venous pressures. Auscultation reveals inspiratory crackles chiefly at the bases, where the hydrostatic pressure is greatest, but potentially throughout both lungs. Rhonchi and wheezing (“cardiac asthma”) may occur. The radiograph shows bilateral perihilar ground glass opacities representative of areas of interstitial and alveolar edema.

Increased Permeability Pulmonary Edema (Noncardiogenic Pulmonary Edema)

The most common form of increased-permeability pulmonary edema is ARDS. According to the consensus Berlin Definition, ARDS is characterized by acute onset (<7 days) of bilateral radiographic pulmonary infiltrates and respiratory failure not fully explained by heart failure or volume overload, with associated impairment in oxygenation defined as a PaO₂/FiO₂ ratio of 300 or less. The severity of ARDS is defined by the severity of impairment in oxygenation. Mild, moderate, and severe ARDS are defined by PaO₂/FiO₂ ratios between 200–300 mm Hg, 100–200 mm Hg, and less than 100 mm Hg, respectively. ARDS is the final common pathway of a number of different serious medical conditions, all of which lead to increased pulmonary capillary leak. The range of clinical presentations includes all the diagnoses in the adult ICU, including sepsis, pneumonia, pancreatitis, aspiration of gastric contents, shock, lung contusion, nonthoracic trauma, toxic inhalation, near-drowning, and multiple blood transfusions. About one third of ARDS patients initially have sepsis syndrome.

Although the mechanism of injury varies, damage to capillary endothelial cells and alveolar epithelial cells is common to ARDS regardless of cause. After the initial insult (eg, an episode of high-grade bacteremia), there is generally a period of stability, reflecting the time it takes for proinflammatory mediators released from stimulated inflammatory cells to cause damage. Damage to endothelial and epithelial cells causes increased vascular permeability and reduced surfactant production and activity. These abnormalities lead to interstitial and alveolar pulmonary edema, alveolar collapse, a significant increase in surface forces, markedly reduced pulmonary compliance, and hypoxemia. For the first 24–48 hours after the insult, the patient may experience increased work of breathing, manifested by dyspnea and tachypnea but without abnormalities in the chest radiograph. At this early stage, the increased A-a ΔPO₂ reflects alveolar edema with V̇/Q̇ mismatching shifted to low V̇/Q̇ ratios that can be corrected by increased FiO₂ and increased minute ventilation. The clinical picture may improve, or there may be a further fall in compliance and disruption of pulmonary capillaries, leading to areas of true shunting and refractory hypoxemia. The combination of increased work of breathing and progressive hypoxemia usually requires mechanical ventilation. Because the underlying process is heterogeneous, with normal-appearing lung adjacent to atelectatic or consolidated lung, ventilating patients at typical tidal volumes may overdistend normal alveoli, reduce blood flow to areas of adequate ventilation, and precipitate further lung injury (“volu-trauma”).

Hypoxemia can be profound, typically followed days later by hypercapnia due to increasing dead space ventilation. Radiographically, there may be patchy alveolar opacities or “whiteout” of the lungs, representing diffuse confluent alveolar filling. Pathologically, diffuse alveolar damage (DAD) is seen, characterized by inflammatory cells and the formation of hyaline membranes. Mortality is 30–40%. Most patients die from some complication of their presenting illness, not from refractory hypoxemia. Of those who survive, a majority will recover near-normal lung function, but their recovery may be prolonged to 6–12 months. A significant number will develop new reactive airway disease or pulmonary fibrosis.

Pulmonary Embolism

Clinical Presentation

The English word “embolus” derives from a Greek word meaning “plug” or “stopper.” A pulmonary embolus consists of material that gains access to the venous system and then to the pulmonary circulation. Eventually, it reaches a vessel whose caliber is too small to permit free passage where it forms a plug, occluding the lumen and obstructing perfusion. There are many types of pulmonary emboli (Table 9–8). The most common is pulmonary thromboembolism, which occurs when venous thrombi, chiefly from the lower extremities, migrate to the pulmonary circulation. Note that it is a normal function of the pulmonary microcirculation to prevent embolic material from entering the systemic arterial system. The lungs possess both excess functional capacity and a redundant vascular supply, allowing them to filter a significant amount of thrombi and platelet aggregates with minimal impact on lung function or hemodynamics. However, large thromboemboli, or a large enough accumulation of smaller ones, can cause substantial impairment of cardiac and respiratory function and death.

Pulmonary thromboemboli are common and cause significant morbidity. They are found at autopsy in 25–50% of hospitalized patients and are considered a major contributing cause of death in a third of those. However, the diagnosis is made antemortem in only 10–20% of cases.

Etiology & Epidemiology

Pulmonary embolism (PE) and deep venous thrombosis represent a continuum of a single disease that has been coined venous thromboembolic disease, or VTE. Thromboemboli almost never originate in the pulmonary circulation; they arrive there by traveling through the venous circulation.

More than 95% of pulmonary thromboemboli arise from thrombi in the deep veins of the lower extremity: the popliteal, femoral, and iliac veins. Venous thrombosis below the popliteal veins or occurring in the superficial veins of the leg is clinically common but not a risk factor for pulmonary thromboembolism because thrombi in these locations rarely migrate to the pulmonary circulation without first extending above the knee. Because fewer than 20% of calf thrombi will extend into the popliteal veins, isolated calf thrombi may be observed with serial tests to exclude extension into the deep system and do not necessarily require anticoagulation. Venous thromboses occasionally occur in the upper extremities or in the right side of the heart; this happens most commonly in the presence of intravenous catheters or cardiac pacing wires and may be of increasing clinical importance as the use of long-term intravenous catheters increases.

Risk factors for pulmonary thromboembolism are, therefore, the risk factors for the development of venous thrombosis in the deep veins of the legs (deep venous thrombosis) (Table 9–9). The German pathologist Rudolf Virchow stated these risk factors in 1856: venous stasis, injury to the vascular wall, and increased activation of the clotting system. His observations remain valid today.

The most prevalent risk factor in hospitalized patients is stasis from immobilization, especially in those undergoing surgical procedures. The incidence of calf vein thrombosis in patients who do not receive heparin prophylaxis after total knee replacement is reported to be as high as 84%; it is more than 50% after hip surgery or prostatectomy. The risk of fatal pulmonary thromboembolism in these patients may be as high as 5%. Physicians caring for these patients must, therefore, be aware of the magnitude of the risk and institute appropriate prophylactic therapy (Tables 9–9 and 9–10)

Malignancy and tissue damage at surgery are the two most common causes of increased activation of the coagulation system. Abnormalities in the vessel wall contribute little to venous as opposed to arterial thrombosis. However, prior thrombosis can damage venous valves and lead to venous incompetence, which promotes stasis.

Advances now permit identification of genetic disorders in up to one third of unselected patients with venous thrombosis and in more than half of patients with familial thrombosis (Table 9–9). It is now clear that these genetic variants may interact with other factors (eg, oral contraceptive use, dietary deficiencies) to increase thrombosis risk.

Pathophysiology

Venous thrombi are composed of a friable mass of fibrin, with many erythrocytes and a few leukocytes and platelets randomly enmeshed in the matrix. When a venous thrombus travels to the pulmonary circulation, it causes a broad array of pathophysiologic changes (Table 9–11).

Hemodynamic Changes

Every patient with a pulmonary embolus has some degree of mechanical obstruction. The effect of mechanical obstruction depends on the proportion of the pulmonary circulation obstructed, on neurohumoral reflexes stimulated by the thrombus, and on the presence or absence of preexisting cardiopulmonary disease. Patients without preexisting cardiopulmonary disease can accommodate occlusion of up to roughly one third of the pulmonary circulation with a negligible increase in pulmonary vascular resistance and pulmonary arterial pressure. The pulmonary circulation adapts to increased flow through a reduced vascular bed by recruitment of underperfused capillaries (Figure 9–13) and vascular dilation from increased flow. These adaptive mechanisms fail with increased occlusion of the pulmonary circulation by emboli, at which point pulmonary vascular resistance and pulmonary arterial pressure increase. In patients with preexisting cardiopulmonary disease, increases in pulmonary artery pressures have not been shown to correlate with extent of embolization. The likely explanation is that normal adaptive mechanisms are ineffective in patients with preexisting pulmonary hypertension, making them susceptible to significant instability with any subsequent impairment of the pulmonary vasculature.

Large emboli that do not completely occlude vessels, particularly in patients with compromised cardiac function, may cause an acute increase in pulmonary vascular resistance. This leads to acute right ventricular strain and can lead to a fatal fall in cardiac output. The most devastating and feared complication of acute pulmonary thromboembolism is sudden occlusion of the pulmonary outflow tract (“saddle embolus”), reducing cardiac output to zero and causing immediate cardiovascular collapse and death. Such dramatic presentations occur in less than 5% of cases and are essentially untreatable. They serve to highlight the importance of primary prevention of venous thrombosis.

Changes in Ventilation/Perfusion Relationships

Pulmonary thromboembolism reduces or eliminates perfusion distal to the site of the occlusion. The immediate effect is increased V̇/Q̇ mismatching, with a shift in the proportion of lung segments with high V̇/Q̇ ratios (alveolar dead space or wasted ventilation). A shift toward high V̇/Q̇ ratios impairs the excretion of carbon dioxide with minimal effect on oxygenation. The patient compensates for this increase in wasted ventilation by increasing total minute ventilation. After several hours, local hypoperfusion interferes with production of surfactant by alveolar type II cells. Surfactant is subsequently depleted, resulting in alveolar edema, alveolar collapse, and areas of atelectasis creating lung units with little or no ventilation. Depending on the level of perfusion to these segments, there will be an increase in lung units with low V̇/Q̇ ratios including some areas of true shunting, both of which contribute to an increased A-a ΔPO₂ and arterial hypoxemia.

Hypoxemia

Mild to moderate hypoxemia with a low PaCO₂ is the most common finding in acute pulmonary thromboembolism. Mild hypoxemia may be obscured by the tendency to rely on oximetry alone, because more than half of patients will have oxygen saturations (SaO₂) above 90% (Figure 9–26). Historically, the A-a ΔPO₂ was thought to be a more sensitive indicator of PE because it compensates for the presence of hypocapnia and the amount of inspired FiO₂. However, the Prospective Investigation of Pulmonary Embolism Diagnosis II (PIOPED II) study called this thinking into question. An A-a ΔPO₂ less than 20, which is normal or near normal depending on patient age, was found in one third of patients with an acute PE identified by CT scanning (Figure 9–26).

No one mechanism fully accounts for hypoxemia in acute PE. At least five mechanisms have been suggested:

1. Loss of surfactant, resulting in atelectasis and localized pulmonary edema over the first 24 hours after pulmonary vascular obstruction. When these areas are reperfused, atelectatic lung units represent low V̇/Q̇ areas causing hypoxemia.

2. Enhanced perfusion of poorly ventilated or nonventilated lung zones. Perfusion is normally reduced to hypoventilated lung regions through hypoxic pulmonary vasoconstriction. However, if pulmonary artery pressure increases after thromboembolism, perfusion may increase in areas of vasoconstriction, resulting in shifts toward low V̇/Q̇ areas causing hypoxemia.

3. True right-to-left shunts. Such shunts have been described in a small percentage of patients with hypoxemia in the setting of acute pulmonary thromboembolism. It has been proposed that these shunts result from the opening of a foramen ovale or from pulmonary artery to pulmonary venous shunting, but their exact location is unknown.

4. Low mixed venous PO₂. In some patients, with preexisting impaired cardiac function or with large emboli that cause acute right ventricular strain, cardiac output may fall, with a resultant fall in mixed venous oxygen concentration. This is an important cause of hypoxemia in seriously ill patients.

5. Decreased pulmonary capillary surface area resulting in decreased lung diffusion capacity.

Bronchoconstriction

Reflex bronchoconstriction causes wheezing and increased work of breathing in some patients.

Pulmonary Infarction

Obstruction of small pulmonary arterial branches that act as end arteries leads to pulmonary infarction in about 10% of cases. It is generally associated with some concomitant abnormality of the bronchial circulation such as that seen in patients with left ventricular failure and chronically elevated left atrial pressures.

Clinical Manifestations

Symptoms and Signs

The classic triad of a sudden onset of dyspnea, pleuritic chest pain, and hemoptysis occurs in a minority of cases. In a large study of patients with PE, dyspnea was present in 73% of cases and pleuritic chest pain was present 44% of the time. Dyspnea probably results from reflex bronchoconstriction as well as increased pulmonary artery pressure, loss of pulmonary compliance, and stimulation of C fibers. In patients with large emboli, acute right heart strain may contribute to dyspnea. Pleuritic chest pain is more common than pulmonary infarction; one group has suggested that the pain is caused by areas of pulmonary hemorrhage. Hemoptysis is seen with pulmonary infarction but may also result from transmission of systemic arterial pressures to the microvasculature via bronchopulmonary anastomoses, with subsequent capillary disruption. It may reflect hemorrhagic pulmonary edema from surfactant depletion or neutrophil-associated capillary injury. Syncope may signal a massive embolus.

The most compelling physical finding is not in the chest but the leg: a swollen, tender, warm and reddened calf that provides evidence for deep venous thrombosis. The absence of such evidence does not exclude the diagnosis, because the clinical examination is insensitive, and the absence of signs may indicate that the entire thrombus has embolized. Auscultatory chest findings are common but nonspecific. Atelectasis may lead to inspiratory crackles; infarction may cause a focal pleural friction rub; and the release of mediators may cause bronchoconstriction and wheezing. In large embolization, one may find signs of acute right ventricular strain such as a right ventricular lift and accentuation of the pulmonary component of the second heart sound.

Electrocardiography

The electrocardiogram is abnormal in 70% of patients with acute PE. However, the most common abnormalities are sinus tachycardia and nonspecific ST and T wave changes, each seen in approximately 40% of patients. The classic finding of an acute right ventricular strain pattern on ECG—a deep S wave in lead I and both a Q wave and an inverted T wave in lead III (S₁Q₃T₃)—was observed in 11% of patients in the Urokinase Pulmonary Embolism Trial.

Laboratory Findings

An increase in the A-a ΔPO₂ is seen in more than two thirds of cases, and hypoxemia is a common yet nonspecific finding. Measurement of the degradation product of cross-linked fibrin, D-dimers, can be used to exclude the diagnosis of acute PE in symptomatic outpatients deemed to have a low pretest probability of PE based on clinical criteria. Depending on the specific assay and patient population, the D-dimer has a high sensitivity (85–99%) and moderate to high specificity (40–93%). Most studies suggest that D-dimer cannot be used to exclude PE in a patient with an intermediate or a high pretest probability for PE.

Brain natriuretic peptide (BNP), an indicator of ventricular stretch, and cardiac troponins, which indicate cardiac myocyte injury, are commonly measured in patients with PE. Because of low sensitivity and specificity, these markers cannot be used to diagnose PE. However, an elevation of BNP or troponins in the setting of known PE has been shown to correlate with the presence of right ventricular overload and greater risk of adverse outcomes, including respiratory failure and death.

Chest Radiography

The chest radiograph was normal in only 12% of patients with confirmed pulmonary thromboembolism in the PIOPED study. The most common findings were atelectasis, parenchymal opacities, and small pleural effusions. However, the prevalence of these findings was the same in hospitalized patients without suspected pulmonary thromboembolism. Local oligemia (Westermark sign) or pleura-based areas of increased opacity that represent intraparenchymal hemorrhage (Hampton hump) are rare. The chest radiograph is necessary to exclude other common lung diseases and to permit interpretation of the ventilation/perfusion scan, but it does not itself establish the diagnosis. Paradoxically, it may be most helpful when normal in the setting of acute severe hypoxemia.

Ventilation/Perfusion Scanning

A perfusion scan is obtained by injecting microaggregated radiolabeled albumin with a particle size of 50–100 μm into the venous system and allowing the particles to embolize to the pulmonary capillary bed (approximate diameter 10 μm). The substance is labeled with a gamma-emitting isotope of technetium (Tc-99m pertechnetate) that permits imaging of the distribution of pulmonary blood flow. A ventilation scan is performed by having the patient breathe xenon (Xe-133) or a radioactive aerosol and doing sequential scans during inhalation and exhalation. A normal perfusion scan excludes clinically significant pulmonary thromboembolism. A segmental or larger perfusion defect in a radiographically normal area that shows normal ventilation is diagnostic. This is referred to as a “mismatched” defect and is highly specific (97%) for pulmonary thromboembolism.

Only a minority of ventilation/perfusion scans reveal clearly diagnostic findings, however. The PIOPED study demonstrated that nondiagnostic ventilation/perfusion scans can stratify a patient’s risk of pulmonary thromboembolism. Furthermore, within the categories of high-, medium-, and low-probability studies, the clinician’s pretest assessment of the probability of pulmonary thromboembolism can further stratify patients.

Computed Tomography and Pulmonary Angiography

Computed tomography scanning with intravenous contrast (CT pulmonary angiography) has widely supplanted ventilation/perfusion scanning as the initial test of choice to diagnose PE. The diagnostic strength of this imaging modality lies in its high negative predictive value and its ability to identify other conditions that cause dyspnea and chest pain (eg, pneumonia, aortic dissection). Multiple trials have shown a high sensitivity and specificity of this imaging technique, although the diagnostic test characteristics are dependent on patient selection, expertise of the technician performing the contrast injection, and the experience of the interpreting radiologist. The PIOPED II trial evaluated CT angiography for the diagnosis of PE and found a sensitivity of 83% and specificity 96% (Table 9–12). Several other studies indicate that the risk of PE after a negative CT scan in patients with a low or intermediate clinical probability of PE is less than 2%. Consistent with the first PIOPED trial comparing ventilation/perfusion scanning and traditional pulmonary angiography, pretest probability based on clinical risk scores must be taken into account when interpreting CT pulmonary angiography. If the results are discordant, further testing, such as ventilation/perfusion scanning or lower extremity Doppler ultrasonography, must be considered.

Pulmonary angiography is a safe but invasive procedure with well-defined morbidity and mortality data. Minor complications occur in approximately 5% of patients. Most are allergic contrast reactions or transient kidney injury or are related to percutaneous catheter insertion; cardiac perforation and arrhythmias are reported but rare. Among the PIOPED I patients who underwent angiography, five deaths (0.7%) were directly related to the procedure. Pulmonary angiography remains the reference standard for the diagnosis of PE, but its role compared with CT angiography is a subject of ongoing debate. There is general agreement that angiography is indicated when the diagnosis is in doubt but there is a high clinical pretest probability of PE, or when the diagnosis of PE must be established with certainty, as when anticoagulation is contraindicated or placement of an inferior vena cava filter is contemplated. An intraluminal filling defect in more than one projection establishes a definitive diagnosis. Secondary findings highly suggestive of PE include abrupt arterial cutoff, asymmetry of blood flow—especially segmental oligemia—or a prolonged arterial phase with slow filling. Pulmonary angiography was performed in 755 patients in the PIOPED I study. A definitive diagnosis was established in 97%; in 3%, the studies were nondiagnostic. Four patients (0.8%) with negative angiograms subsequently had pulmonary thromboemboli at autopsy. Serial angiography has demonstrated minimal resolution of thrombus prior to day 7 following presentation. Thus, negative angiography within 7 days of presentation excludes the diagnosis.

Resolution

The variability among patients is so great that generalizations are difficult. The largest number of patients monitored serially with quantitative assessments was in the Urokinase Pulmonary Embolism Trial. In that study, serial perfusion scans showed resolution of 35–56% of perfusion defects at 9–14 days. More recent studies, some involving quantitative angiography, have tended to support the time course of these findings.

In a few patients, pulmonary emboli do not resolve completely but become organized and incorporated into the pulmonary arterial wall as an epithelialized fibrous mass, producing what is termed chronic pulmonary thromboembolism. This entity presents with stenosis of the central pulmonary arteries with associated pulmonary hypertension and right ventricular failure (cor pulmonale). Treatment is surgical.

Yeong Kwok, MD

Case 43

Asthma

A 25-year-old previously well woman presents to your office with complaints of episodic shortness of breath and chest tightness. She has had the symptoms on and off for about 2 years but states that they have worsened lately, occurring two or three times a month. She notes that the symptoms are worse during the spring months. She has no exercise-induced or nocturnal symptoms. The family history is notable for a father with asthma. She is single and works as a secretary in a high-tech firm. She lives with a roommate, who moved in approximately 2 months ago. The roommate has a cat. The patient smokes occasionally when out with friends, drinks socially, and has no history of drug use. Examination is notable for mild end-expiratory wheezing. The history and physical examination are consistent with a diagnosis of asthma. Pulmonary function tests are ordered to confirm the diagnosis.

Questions

A. What are the three categories of provocative agents that can trigger asthma? What are some possible triggers in this patient?

The fundamental abnormality in asthma is increased reactivity of airways to stimuli. Asthma can be induced by many provocative agents. These can be broadly categorized as (1) physiologic or pharmacologic mediators of asthmatic airway responses, (2) allergens that can induce airway inflammation and reactivity in sensitized individuals, and (3) exogenous physicochemical agents or stimuli that produce airway hyperreactivity. This patient’s history (seasonal predilection) is most consistent with allergen-induced asthma. The worsening symptoms in the last few months may be due to an allergic reaction to the roommate’s cat.

B. Describe the early events responsible for the pathogenesis of asthma. How does this result in chronic airway inflammation and airway hyperresponsiveness?

The earliest events in asthma are the activation of local inflammatory cells, primarily mast cells and eosinophils, by the provocative agents described previously. This can occur by specific IgE-dependent mechanisms or indirectly by chemical irritant exposure or osmotic stimuli. Acute-acting mediators, including leukotrienes, prostaglandins, and histamine, induce smooth muscle contraction, mucus hypersecretion, and vasodilation with endothelial leakage and local edema formation. Epithelial cells also participate, releasing leukotrienes, prostaglandins, and inflammatory cytokines. Additional inflammatory cells, including neutrophils and eosinophils, are recruited to the airway mucosa. In addition, the cell cytokines released promote growth of mast cells and eosinophils, the influx and proliferation of T cells, and the differentiation of B lymphocytes into IgE- and IgA-producing plasma cells. Ultimately, this ongoing inflammation results in injury to epithelial cells, denudation of the airway, greater exposure of afferent sensory nerves, and subsequent smooth muscle hyperresponsiveness, chronic inflammation, submucosal gland hypersecretion, and increased mucus volume.

C. What pathogenetic mechanisms are responsible for this patient’s symptoms of wheezing, shortness of breath, and chest tightness?

Wheezing is caused by a combination of smooth muscle contraction and mucus hypersecretion and retention, resulting in airway caliber reduction and prolonged turbulent airflow. The sensations of shortness of breath and chest tightness are also the result of a number of concerted changes. These include the detection by spindle cell stretch receptors of the greater muscular effort required to overcome the increased airway resistance as well as detection of thoracic distention resulting from chest hyperinflation, decreased lung compliance, and increased work of breathing. These are sensed by the chest wall nerves and manifested as chest tightness and shortness of breath. As obstruction worsens, hypoxemia and CO₂ retention occur, further stimulating respiratory drive through peripheral and central chemoreceptors. This stimulus in the setting of respiratory muscle fatigue produces progressive dyspnea.

D. What might you expect the results of her pulmonary function tests to be? Why?

This patient’s symptoms are relatively mild, occurring only intermittently. In between exacerbations, her pulmonary function tests may be normal. During an attack, all indices of expiratory airflow may be reduced, including FEV₁, FEV₁/FVC, and peak expiratory flow rate. FVC may also be reduced as a result of premature airway closure. Total lung capacity, functional residual capacity, and residual volume may be increased as a consequence of airflow obstruction and incomplete emptying of lung units. DLCO may be increased because of increased lung and capillary blood volume.

Case 44

Chronic Obstructive Pulmonary Disease (COPD)

A 67-year-old man presents to your office with worsening cough, sputum production, and shortness of breath. He has been a cigarette smoker for the past 50 years, smoking approximately 1 pack a day. He has a chronic AM cough productive of some yellow sputum but generally feels okay during the day. He was in his usual state of health until two weeks ago when he developed a cold. Since then, he has had a hacking cough and increased thick sputum production. He also has had difficulty walking more than a block without stopping due to shortness of breath. Physical examination reveals prolonged expiration, audible wheezing, and diffuse rhonchi throughout both lung fields. Chest x-ray shows hyperinflation of both lungs with a flattened diaphragm.

Questions

A. What are the two major clinical syndromes that are classified as chronic obstructive pulmonary disease? How do they differ?

Chronic obstructive pulmonary disease (COPD) is an intentionally imprecise term used to denote a process characterized by the presence of chronic bronchitis or emphysema that may lead to the development of fixed airway obstruction. Chronic bronchitis and emphysema are frequently encountered together in the same patient.

Chronic bronchitis is defined by a clinical history of productive cough for 3 months of the year for 2 consecutive years. Dyspnea and airway obstruction, often with an element of reversibility, are intermittently to continuously present. Cigarette smoking is by far the leading cause of this disease, although other inhaled irritants may induce the same process. While the predominant pathologic events are inflammation in larger airways, accompanied by mucosal thickening and mucus hypersecretion, it is the inflammation in smaller bronchioles that is the principal site of increased airflow obstruction.

Pulmonary emphysema is a condition marked by irreversible enlargement of the airspaces distal to the terminal bronchioles, accompanied by destruction of their walls, most often without obvious fibrosis. In contrast to chronic bronchitis, the primary pathologic defect in emphysema is not in the airways but rather in the respiratory unit walls, where the loss of elastic tissue results in a loss of appropriate recoil tension to support distal airways during expiration. Progressive dyspnea and nonreversible obstruction accompany the airspace destruction without mucus hypersecretion and productive cough. Furthermore, the loss of alveolar surface area and the accompanying capillary bed for gas exchange contribute to the progressive hypoxia and dyspnea.

B. Of the two syndromes above, which is predominant in this patient? What is the epidemiology and predisposing factors for this condition?

The chronic productive cough and thick sputum production present in this patient are characteristic of chronic bronchitis. Cigarette smoking remains the principal cause of disease in up to 90% of patients with chronic bronchitis and emphysema. COPD is probably significantly underdiagnosed; although only 15–20% of smokers develop severe airflow obstruction, there is a dose-dependent relationship between tobacco smoke exposure and loss of lung function. Population-based studies suggest that chronic dust (including silica and cotton) or chemical fume exposure is a significant contributing risk factor for COPD. In the developing world, indoor exposure to smoke from burning biofuels is a major cause of COPD. The most important identified genetic risk factor for the evolution of COPD is deficiency of α₁-protease (α₁-antitrypsin) inhibitor. Reduced circulating and tissue levels can lead to early onset of severe emphysema, not chronic bronchitis.

C. What might the pulmonary function tests show in this patient?

Diffuse airway obstruction is demonstrated on pulmonary function testing as a global reduction in expiratory flows and volumes. FEV₁, FVC, and the FEV₁/FVC (FEV₁%) ratio are all reduced. The expiratory flow-volume curve shows substantial limitation of flow. Some patients may respond to bronchodilators. Measurement of lung volumes reveals an increase in the RV and FRC, reflecting air trapped in the lung as a result of diffuse airway obstruction and early airway closure at higher lung volumes. This is characterized by flattened diaphragm on chest x-ray. DLCO is typically normal, reflecting a preserved alveolar capillary bed.

D. How do arterial blood gases differ in chronic bronchitis and emphysema?

Ventilation/perfusion mismatching is common in chronic bronchitis. The A-a ΔPO₂ is increased and hypoxemia is common mainly because of significant areas of low V̇/Q̇ ratios (physiologic shunt); hypoxemia at rest tends to be more profound than in emphysema. There is also V̇/Q̇ mismatch in emphysema, but typically, patients with emphysema adapt to high V̇/Q̇ ratios by increasing their minute ventilation. They may maintain nearly normal PO₂ and PCO₂ levels despite advanced disease. However, with greater disease severity and further loss of capillary perfusion, the DLCO falls, leading to exercise-related and, ultimately, resting arterial hemoglobin desaturation. In both conditions, increasing PCO₂ (hypercapnia) and respiratory acidosis, with compensatory metabolic alkalosis, are seen in severe disease.

Case 45

Idiopathic Pulmonary Fibrosis

A 68-year-old man presents to the clinic with a complaint of shortness of breath. He states that he has become progressively more short of breath for the last 2 months, such that he is now short of breath with walking one block. In addition, he has noted a nonproductive cough. He denies fever, chills, night sweats, chest pain, orthopnea, or paroxysmal nocturnal dyspnea. He has noted no lower extremity edema. The medical history is unremarkable. Physical examination is remarkable for a respiratory rate of 19/min and fine dry inspiratory crackles heard throughout both lung fields. Digital clubbing is present. A diagnosis of idiopathic pulmonary fibrosis is made.

Questions

A. What are the cellular events involved in lung injury and fibrosis in idiopathic pulmonary fibrosis?

Diffuse parenchymal lung disease encompasses many disorders with different precipitating events and possibly different cellular and molecular mechanisms. A common series of cellular events mediate and regulate lung inflammatory processes and fibrotic responses (Table 9–5). These events include (1) initial tissue injury; (2) vascular injury and endothelial cell activation, with increased permeability, exudation of plasma proteins into the extravascular space, and variable thrombosis and thrombolysis; (3) alveolar epithelial cell injury and activation, with loss of barrier integrity and release of proinflammatory mediators; (4) increased leukocyte adherence to activated endothelium, with transit of activated leukocytes into the interstitium; and (5) continued injury and repair processes characterized by alterations in cell populations and increased matrix production.

B. What pathophysiologic mechanisms are responsible for this patient’s symptoms of dyspnea and cough? What pathogenetic mechanisms are responsible for his physical findings of tachypnea, inspiratory crackles, and digital clubbing?

An intermittent, irritating, non-productive cough is often the first symptom of idiopathic pulmonary fibrosis (IPF). It may be refractory to antitussive therapy. The mechanism is likely multifactorial with fibrotic damage to terminal respiratory units causing bronchial and bronchiolar distortion, leading to alterations in both stimulatory and inhibitory nerve fibers involved in cough reflexes.

Multiple factors contribute to dyspnea in patients with IPF. Fibrosis of lung parenchyma decreases lung compliance; in combination with alterations in surfactant turnover, the distending pressure required to inflate the lungs increases, as does the work of breathing. Increased stimuli from C fibers in fibrotic alveolar walls or stretch receptors in the chest wall may sense the increased force necessary to inflate less compliant lungs.

Tachypnea results from increased lung sensory receptor stimuli and the attempt to maintain a normal alveolar minute ventilation (and hence normal PaCO₂) as lung volumes decrease. A rapid, shallow breathing pattern also reduces ventilatory work in the face of increased lung elastic recoil. The diminished capillary bed and thickened alveolar-capillary membrane contribute to hypoxemia with exercise. In advanced disease, altered gas exchange with severe V̇/Q̇ mismatching can produce hypoxemia at rest.

The diffuse inspiratory crackles reflect the successive opening on inspiration of respiratory units that are collapsed owing to the fibrosis and the loss of normal surfactant. The cause of digital clubbing is not known.

C. What might you expect the chest x-ray film to show? The pulmonary function tests?

Characteristic chest radiograph findings include reduced lung volumes with increased reticular opacities that are prominent in the lung periphery and cause loss of definition of vascular structures, diaphragms, and cardiac border. Fibrosis surrounding expanded small airspaces is seen as honeycombing. With pulmonary hypertension, central pulmonary arteries may be enlarged. Pulmonary fibrosis produces a restrictive pattern on pulmonary function tests. This is manifested as reductions in TLC, FEV₁, and FVC, with preservation of or increases in FEV₁/FVC and expiratory flow rates. DLCO decreases progressively as fibrosis continues and lung capillaries are obliterated.

Case 46

Pulmonary Edema

A 72-year-old man presents to the emergency department complaining of severe shortness of breath. He has long-standing poorly controlled hypertension and history of coronary artery disease and two myocardial infarctions. About 1 week before admission, he had an episode of substernal chest pain lasting approximately 30 minutes. Since then he has noted progressive shortness of breath to the point that he is now dyspneic on minimal exertion such as walking across the room. He notes a new onset of shortness of breath while lying down. He is only comfortable when propped up by three pillows. He is occasionally awakened from sleep acutely short of breath. On examination he is afebrile, with a blood pressure of 160/100 mm Hg, heart rate of 108/min, respiratory rate of 22/min, and oxygen saturation of 88% on room air. He is pale, cool, and diaphoretic. Jugular venous pressure is 10 cm H₂O. Chest auscultation reveals rales in both lungs to the mid lung fields. Cardiac examination reveals tachycardia, with an audible S₃ and S₄. No murmurs or rubs are heard. Extremities are without edema. The ECG shows left ventricular hypertrophy and Q waves in the anterior and lateral leads, consistent with this patient’s history of hypertension and myocardial infarction. Chest x-ray film reveals bilateral fluffy infiltrates consistent with pulmonary edema. He is admitted to the ICU with a diagnosis of heart failure and possible myocardial infarction.

Questions

A. What are the four factors that account for almost all cases of pulmonary edema? Which are probably responsible for this patient’s pulmonary edema?

The four factors that account for almost all cases of pulmonary edema are (1) an increase in the hydrostatic pressure gradient (cardiogenic pulmonary edema); (2) an increase in vascular endothelial cell and/or alveolar epithelial cell permeability (noncardiogenic pulmonary edema); (3) a decrease in the oncotic pressure gradient (usually due to low protein content of the plasma); and (4) impaired lymphatic drainage either from physical lymphatic obstruction or from lymphatic obliteration that can occur in the setting of radiation treatment. This patient’s history of prior myocardial infarction, long-standing history of hypertension, and possible recent ischemia make it likely that he has cardiogenic pulmonary edema.

B. How does poor cardiac function cause pulmonary edema?

Cardiogenic or hydrostatic pulmonary edema classically results from elevated pulmonary venous and left atrial pressures due to left ventricular systolic or diastolic failure, mitral stenosis, or mitral regurgitation. This is primarily a mechanical process resulting in an ultrafiltrate of plasma. Edema fluid in this setting has a relatively low protein content, generally less than 60% of a patient’s plasma protein content. In normal individuals, pulmonary capillary pressure (ie, pulmonary capillary wedge pressure) must exceed approximately 20 mm Hg before the fluid leaving the vascular space exceeds the rate of resorption, leading to accumulation of interstitial (Figure 9–26) and ultimately alveolar fluid that we describe as pulmonary edema.

Case 47

Pulmonary Embolism

A 57-year-old man undergoes total knee replacement for severe degenerative joint disease. Four days after surgery, he develops an acute onset of shortness of breath and right-sided pleuritic chest pain. He is now in moderate respiratory distress with a respiratory rate of 28/min, heart rate of 120 bpm, and blood pressure of 110/70 mm Hg. Oxygen saturation is 90% on room air. Lung examination is normal. Cardiac examination reveals tachycardia but is otherwise unremarkable. The right lower extremity is postsurgical, healing well, with 2+ pitting edema, calf tenderness, erythema, and warmth; the left leg is normal. He has a positive Homan sign on the right. Acute pulmonary embolism is suspected.

Questions

A. Where did the pulmonary embolism probably arise from?

Thromboemboli almost never originate in the pulmonary circulation. More than 95% of pulmonary thromboemboli arise from the deep veins of the lower extremity: the popliteal, femoral, and iliac veins. The findings of right lower extremity warmth, erythema, and swelling in this patient support the view that this is very likely the site of origin of thromboembolism. It is important to note, however, that the absence of such lower extremity findings does not exclude the diagnosis of thrombus from the lower extremity, because findings are insensitive.

B. What are this patient’s risk factors for thromboembolism?

This patient has multiple risk factors for pulmonary embolism, and he was at high risk for such an event. He is older than 40 years, was anesthetized for more than 30 minutes for his total knee replacement, and underwent orthopedic surgery (risk imposed by immobilization). His risk for calf vein thrombosis is as high as 84%, and the risk of fatal pulmonary embolism is approximately 5%. All such patients should receive prophylactic therapy with anticoagulants postoperatively.

C. What are the hemodynamic changes seen in acute pulmonary embolism?

All patients with pulmonary emboli have some degree of mechanical obstruction. The effect depends on the proportion of the pulmonary circulation obstructed (how large the pulmonary embolus is), neurohumoral reflexes stimulated by the thrombus, and the severity of preexisting cardiopulmonary disease. As the degree of obstruction of pulmonary circulation increases, pulmonary artery pressures rise, ultimately leading to right ventricular strain. In severe pulmonary embolism, occlusion of the pulmonary outflow tract may occur, severely reducing cardiac output and causing cardiovascular collapse and death.

D. What changes might be expected in ventilation/perfusion relationships? What might you expect this patient’s A-a ΔPO₂ to be?

Pulmonary embolism decreases or eliminates perfusion distal to the site of the occlusion. The immediate effect is increased V̇/Q̇ mismatching, with a shift in the proportion of lung segments with high V̇/Q̇ ratios (alveolar dead space or wasted ventilation). A shift toward high V̇/Q̇ ratios impairs the excretion of carbon dioxide with minimal effect on oxygenation. The patient compensates for this increase in wasted ventilation by increasing total minute ventilation. After several hours, local hypoperfusion reduces production of surfactant by type II alveolar cells, resulting in edema, alveolar collapse, and atelectasis, creating lung units with little or no ventilation. Depending on the level of perfusion to these segments, there will be an increase in lung units with low V̇/Q̇ ratios including some areas of true shunting, both of which contribute to an increased A-a ΔPO₂ and arterial hypoxemia.

Case 48

Acute Respiratory Distress Syndrome (ARDS)

A 46-year-old man presents to the hospital with a 5-day history of worsening cough, high fever, and shortness of breath. On physical examination, he is noted to be tachypneic (respiratory rate of 30 breaths/min), hypoxic with a low oxygen saturation (89%), and febrile (39°C). Chest x-ray film reveals infiltrates in both lower lobes. A complete blood count reveals a high white blood cell count. He is admitted to the hospital. Despite treatment with oxygen and antibiotics, he becomes more hypoxic and requires endotracheal intubation and mechanical ventilation. Blood cultures grow Streptococcus pneumoniae. Despite mechanical ventilation using high oxygen concentrations, his arterial blood oxygen level remains low. His chest x-ray film shows progression of infiltrates throughout both lung fields. He is diagnosed with acute respiratory distress syndrome (ARDS).

Questions

A. What are the main pathophysiologic factors in ARDS that cause accumulation of extravascular fluid in the lungs?

The pathophysiology of increased permeability pulmonary edema (ARDS) is complex and may result from multiple different insults. Alveolar fluid accumulates as a result of loss of integrity of the alveolar epithelial barrier, allowing solutes and large molecules such as albumin to enter the alveolar space. This loss of integrity may result from direct injury to the alveolar epithelium by inhaled toxins or pulmonary infection, or they may occur after primary injury to the pulmonary capillary endothelium by circulating toxins as in sepsis or pancreatitis, followed by secondary inflammatory injury to the alveolar epithelial barrier. The presence of high-protein fluid in the alveolus, particularly the presence of fibrinogen and fibrin-degradation products, inactivates pulmonary surfactant, causing large increases in surface tension. This results in a fall in pulmonary compliance and alveolar instability, leading to areas of atelectasis. Increased surface tension decreases the interstitial hydrostatic pressure and favors further fluid movement into the alveolus. A damaged surfactant monolayer may increase susceptibility to infection as well.

B. What are the common causes of ARDS?

ARDS is the final common pathway of a number of different serious medical conditions, all of which lead to increased pulmonary capillary leak. The range of clinical presentations includes all of the diagnoses in the adult ICU, including sepsis, pneumonia, pancreatitis, aspiration of gastric contents, shock, lung contusion, nonthoracic trauma, toxic inhalation, near-drowning, and multiple blood transfusions. About one third of ARDS patients initially have sepsis syndrome.

C. What accounts for the severe hypoxia often found in ARDS, despite the use of mechanical ventilation and high concentrations of oxygen?

The severe hypoxia found in ARDS is due to several factors. Damage to endothelial and epithelial cells causes increased vascular permeability and reduced surfactant production and activity. These abnormalities lead to interstitial and alveolar pulmonary edema, alveolar collapse, a significant increase in surface forces, markedly reduced pulmonary compliance, and hypoxemia. As the process worsens, there may be a further fall in compliance and disruption of pulmonary capillaries, leading to areas of true shunting and refractory hypoxemia. The combination of increased work of breathing and progressive hypoxemia usually requires mechanical ventilation. Since the underlying process is heterogeneous, with normal-appearing lung adjacent to atelectatic or consolidated lung, ventilating patients at typical tidal volumes may overdistend normal alveoli, reduce blood flow to areas of adequate ventilation, and precipitate further lung injury (“volu-trauma”). Hypoxemia can be profound, typically followed days later by hypercapnia due to increasing dead space ventilation.