The Controller: Directing the Orchestra

Central Control

The Automatic Centers ↑

The neural structures responsible for the automatic control of breathing appear to be in the medulla. Two aggregates of neurons, termed the dorsal respiratory group (DRG) and the ventrolateral group (VRG), have both inspiratory and expiratory neurons. In addition, the DRG seems to play an important role in processing information from receptors in the lungs, chest wall, and chemoreceptors that modulate breathing. Neural activity from the DRG is important in activation of the diaphragm, and the VRG, in addition to having a role in determining the rhythm of breathing, regulates the changes in diameter of the upper airway that occur with breathing by stimulating muscles to expand the upper airway during inspiration.

There are also neurons in the pons (the pontine respiratory group [PRG]) that may contribute to the transitions or switching from inspiration to expiration. Damage to the respiratory neurons in the pons leads to an increase in inspiratory time, a decrease in respiratory frequency, and an increase in the tidal volume.

Within the medulla, there appear to be inspiratory neurons that have a pacemaker function, that is, they have intrinsic properties that give them rhythmic activity. Similar to the cardiac pacemaker, they fire at a particular rate that can be modified by other factors, such as disturbances in gas exchange or stimulation of pulmonary receptors, which are discussed at greater length later in this chapter. Information from pulmonary receptors is transmitted to and processed in the DRG. This feedback from the lungs to the medulla is believed to affect the respiratory pattern.

Some of the neurons in the medulla fire during inspiration; others seem to have a role in the transition from inspiration to expiration; and others appear to fire during expiration, primarily serving an inhibitory role on the diaphragm. Neurons in the DRG increase their rate of firing during inspiration. Others in the VRG seem to increase activity during expiration. Taken together, all of these respiratory neurons responsible for our automatic rhythmic breathing are termed the central pattern generator (CPG).

If you examine the activity of the inspiratory neurons during a respiratory cycle, you see three phases of activity (Fig. 6-2). During the inspiratory phase of the respiratory cycle, the activity of the inspiratory neurons rapidly increases. This phase appears to be terminated or switched off by the action of inhibitory neurons. At the beginning of expiration, however, there may be another burst of inspiratory activity that serves to slow or put a brake on expiratory flow. In asthma, this inspiratory activity during expiration appears to be accentuated and may contribute to the hyperinflation seen in individuals with that disease.

Volitional Breathing ↑

If someone asks you to hold your breath, you can for a short time. If someone asks you to breathe at 30 breaths a minute, you can respond appropriately. If someone asks you to take a breath with a tidal volume of 2 L, you inhale deeply. Clearly, you can send messages from your motor cortex to your ventilatory muscles. These neural impulses descend in the spinal cord in corticospinal tracts that are separate from the tracts containing axons originating in the automatic centers. The nerves that innervate the diaphragm exit the spinal cord in the medcervical level (C3–C5). Thus, people who sustain spinal cord injuries below this level in the cervical cord may be severely impaired with quadriplegia, but they can still inspire normally. The intercostal muscles are innervated by nerves exiting the spinal cord along the thoracic column, so they may not function, depending on the level of a spinal cord injury. But as long as the diaphragmatic innervation is intact, inspiration can still take place. Ventilation, therefore, is much more severely compromised by a cervical spinal cord lesion, which prevents activation of both the diaphragm and intercostal muscles, than by a thoracic spinal cord injury.

When you are awake, there may be elements of volitional control of breathing, even when you are not consciously thinking about taking a breath. When asleep, a person’s P_aCO₂ tends to be higher than when awake, for example. An unusual medical problem, called congenital central hypoventilation syndrome (CCHS), offers another example of “volitional” control in the absence of actual thought directed to breathing. CCHS is characterized by the absence of a functioning central pattern generator. In this rare disorder (fewer than 200 cases are reported in the medical literature), infants stop breathing when they go to sleep, and they are identified shortly after birth when they suffer a respiratory arrest during sleep. These individuals must be maintained on ventilators at night throughout their lives. When awake, however, they breathe quite normally, although their respiratory pattern is a bit more irregular than those without this syndrome. The presumption is that the reticular activating system may have connections to the respiratory control areas in the brain that help to regulate breathing when we are awake.

The concept of volitional control of breathing also includes the effects of discomfort and anxiety on our breathing. When experiencing pain or shortness of breath, most people increase their respiratory rate, and total ventilation increases. The pattern of breathing may also reflect attempts to reduce the discomfort associated with ventilation. Patients with significantly reduced respiratory system compliance, for example, tend to breathe with a rapid, shallow pattern. Because the system is stiff, it requires less work to breathe in this way than to exert the large intrathoracic pressures necessary to distend the system. For patients with increased airway resistance, on the other hand, the high flow required for rapid, shallow breathing requires considerable work. These patients tend to adopt a slower breathing pattern with large tidal volumes. Anxious patients typically increase their respiratory rates.

Sources of Information That Modulate the Control of Breathing ↑

The central control centers for the respiratory system do not act in isolation. The brain constantly receives information from the upper airways, lungs, and chest wall that tell it how the ventilatory pump is responding to the messages exiting the controller. Most of the receptors in the upper airways, lungs, and chest wall are mechanoreceptors, so called because they are activated by mechanical distortion of their local environment. In addition, the status of the gas exchanger is also followed closely, and the controller reacts to decreases in the P_aO₂ and increases in P_aCO₂. Information arising from the periphery (the so-called afferent neural pathways) travels to the brain. For example, sensory input from the upper airways, lungs, and peripheral chemoreceptors travels up the ninth and tenth cranial nerves to the region in the medulla where the DRG is located. We will now examine some of the key sources of afferent information that affect the activity of the central controller.

Upper Airway Receptors

Some receptors in the airways appear to sense and monitor flow. These receptors likely respond, however, to changes in temperature resulting from the flow of air past them rather than directly to flow itself. Stimulation of flow receptors appears to inhibit the central controller.

Contained within the walls of the pharynx are receptors that appear to be activated in association with swallowing. Respiratory activity ceases during swallowing as the epiglottis covers the larynx. From an evolutionary standpoint, this sequence of events minimizes the risk of aspiration of food and liquid into the lungs.

Pulmonary Receptors

The lungs contain two types of stretch receptors, both of which are myelinated. In addition, C fiber endings may monitor changes in the pulmonary circulation and interstitial space and transmit information to the brain via unmyelinated fibers in the vagus nerve.

Stretch receptors in the lungs are categorized based on the speed with which the rate of firing of the receptor changes as the local environment, specifically, changes in lung volume, in which the receptor is located is altered. Slowly adapting stretch receptors (SARs) continue to fire at a fairly constant rate after being stimulated by a stretch, even after they have reached a new constant length (i.e., they adapt slowly to the new lung volume). In contrast, rapidly adapting stretch receptors (RARs) change their firing rate quickly upon reaching a new level in the environment (Fig. 6-3).

Use Animated Figure 6-3 to observe how the SARs and the RARs respond to a change in lung volume. Listen as you watch the neural impulses to get a feel for how quickly each type of receptor adapts to the new conditions.

Animated Figure 6-3 (Work in progress)

It is believed that the SARs are located among smooth muscle cells within the intra- and extrathoracic airways. When these airway receptors are stimulated by lung inflation, the expiratory phase of respiration is prolonged. These receptors may also play a role in the early termination of inspiration when tidal volume increases. Deflation of the lung causes a decrease in the basal firing rate of the SARs, which appears to lead to an increase in the respiratory rate.

The RARs are thought to be located within the airways in association with airway epithelial cells, with the majority appearing to be near the region of the carina and in the larger bronchi. The RARs can be stimulated both by chemical (e.g., cigarette smoke, histamine, prostaglandins) and mechanical stimuli. An older name for some of these nerves is irritant receptors, which reflects their activation in the presence of chemical stimuli that are perceived as irritating to the lungs. Activation of the more centrally located RARs may lead to cough, bronchospasm, and increased mucus production, and stimulation of the receptors located more deeply within the lungs can lead to hyperpnea. Lung deflation activates these receptors and can contribute to an increase in respiratory rate as well as the periodic large breaths that we take (i.e., sighs).

The unmyelinated fibers, termed C fibers, in the lungs carry information from a variety of receptors whose function is not totally understood. A group of C fibers that arise from deep within the lungs are believed to carry information from J receptors. These receptors are located near (or in juxtaposition with, hence the J in J receptors) to the pulmonary capillaries. Researchers believe that these receptors may be activated by increases in pulmonary capillary pressures or the accumulation of interstitial fluid, as is seen in the setting of congestive heart failure (CHF). C fibers may also arise from receptors in bronchi. Both chemical and mechanical factors may stimulate receptors that are served by C fibers, some of which may play a role in bronchoconstriction. The exact role of C fibers in the modulation of respiratory control is uncertain, but it is possible that they contribute to the tachypnea seen in patients with CHF and other conditions that are associated with acute changes in pulmonary capillary pressure (Table 6-1).

Chest Wall Receptors

The primary receptors in the chest wall that have a role in monitoring respiration are the muscle spindles and Golgi tendon organs. Their function appears to pertain primarily to alerting the controller to the fact that the physiology of the ventilatory pump has changed, that is, that airway resistance has increased or respiratory system compliance has decreased. Such changes are generally characterized as an increased "load" on the respiratory system. In other words, for a given efferent, or outgoing neural discharge from the brain to the muscles, the respiratory system is not responding appropriately. Having received afferent, or incoming signals from these peripheral receptors to the brain, the typical response of the controller to an increased load is to increase the efferent activity to the muscles in an effort to "compensate" for the load.

Muscle spindles are located in the skeletal muscles. They are found in intercostal muscles and, to a lesser degree, in the diaphragm. The muscle spindle responds to mechanical stimuli (contraction of the muscle) and behaves like a slowly adapting receptor. The spindle receptor is attached to spindle muscle fibers that are innervated by gamma fibers. The spindle muscles are arranged in parallel with the main contracting fibers of the muscle. When a message is sent from the brain to the ventilatory muscle to contract, a message sent over alpha motor neurons, a simultaneous message is sent to the spindle muscle over the gamma fibers. When the spindle muscle contracts, the spindle receptor is stretched and activated. If the main muscle, which is simultaneously receiving a message to contract, also contracts and shortens, the effect on the spindle is to reduce its tension to baseline, and there will be no change in output from the spindle receptor (Fig. 6-4).

Thus, the spindle activity provides feedback on the response of the chest wall to the command to the ventilatory muscles to contract. If there is no load on the respiratory system, the ventilatory muscles will shorten significantly, the spindle will no longer be stretched, and the firing frequency will be diminished. If there is a high mechanical load on the ventilatory pump (imagine the extreme case in which a person has aspirated a piece of food and the trachea is nearly completely occluded), the ventilatory muscles will contract but will shorten minimally because little air can enter the lung because of the region of high resistance in the trachea. The spindle muscles, however, will shorten, the spindle receptor will be activated and will remain activated, and the receptor will send afferent messages to the brain to further increase the contraction of the ventilatory muscles. As a result, load compensation will have occurred, that is, the controller will have made an adjustment to the presence of a severe resistive load on the ventilatory pump by instructing the muscles to contract more forcefully.

Use Animated Figure 6-4 to explore the role of muscle spindles in load compensation. Observe the activity of the spindle receptors when they respond to different loading conditions. Note that for conditions of high load with poor muscle shortening (as with the aspirated bolus of food blocking airflow, for example), the feedback from the muscle spindle leads to a further increase in the contraction force of the muscle.

Animated Figure 6-4 (Work in progress)

Golgi tendon organs are located at the point of insertion of the muscle fiber into the tendon. They appear to monitor tension on the muscles, and it is thought that when tension is very high, messages from these receptors inhibit further contraction. This mechanism may serve to protect muscles from injury when the load on the system is very high.

Chemoreceptors

The controller monitors changes in P_aO₂, P_aCO₂, and arterial pH via specialized neural structures called chemoreceptors. The peripheral chemoreceptors, located in the carotid bodies and the aortic arch, respond to changes in P_aO₂, P_aCO₂, and arterial pH. The central chemoreceptors, located in the medulla, respond to alterations in PCO₂ and pH in the cerebrospinal fluid (CSF), which reflects the variables in the arterial blood.

The carotid bodies are found at the bifurcation of the common carotid artery into the internal and external carotid arteries. Because the chemoreceptors receive more blood than they need to meet their local metabolic needs, the P_aO₂ within the chemoreceptor reflects the delivery and consumption of oxygen by the rest of the body. This allows the chemoreceptor to sense and respond to the requirements of the body as a whole. The capillaries that perfuse the chemoreceptors are intimately associated with specialized nerve endings. Information from these nerve endings is carried to the brain via the ninth cranial nerve. It is not entirely clear how the carotid bodies sense hypoxemia, but it is clear that the stimulus for increased ventilation is P_aO₂, not the oxygen content of the blood. At normal levels of P_aO₂, some neural activity arises from the carotid bodies. At hyperoxic (above normal) levels, this activity is reduced but does not cease. As P_aO₂ decreases below 60 mm Hg, the rate of firing rapidly increases. Also, in most mammals, some tissues in the aortic body, located in the arch of the aorta, appear to be sensitive to hypoxemia. Removal of the carotid bodies in humans, however, abolishes the ventilatory response to hypoxemia, which suggests that the carotid bodies play little role in ventilatory control in people.

The activity of the peripheral chemoreceptors also increases with high levels of P_aCO₂ (leading to increased ventilation) and reduced levels of arterial pH (leading to increased ventilation). Because elevations of carbon dioxide in the blood are also associated with a decrease in pH (more in Chapter 7), it is not immediately evident whether P_aCO₂ or pH is the stimulus under conditions of acute hypercapnia. Experiments in which the two variables are modified independently, however, suggest that the carotid body can respond to either stimulus. Although responsive to changes in both oxygen and carbon dioxide levels, the chemoreceptor is much more sensitive to acute hypercapnia than to hypoxemia. (Note: we may refer to the peripheral and central chemoreceptors in the singular, as above, even though each is composed of many nerve endings).

The central chemoreceptor appears to be a less discrete anatomic site than the peripheral chemoreceptors. Nerves that respond to change in PCO₂ (by increasing ventilation in response to increased PCO₂) and pH (by increasing ventilation in response to a decreased pH) appear to be located in both ventral and dorsal regions of the medulla. The ventral regions, in particular, appear to be near the surface of the brainstem in close proximity to CSF. This location may facilitate the ability of the central chemoreceptor to monitor changes in PCO₂ and pH. Changes in arterial PCO₂ and pH alter the levels of carbon dioxide and protons in the CSF, although, as we will discuss shortly, at different rates. As with the peripheral chemoreceptor, under experimental conditions, either an elevated PCO₂ or a reduced pH can independently cause an increase in respiratory-related neural activity from these regions. Our experience with mixed acid-base disorders, however, suggests that pH may be a more important factor in the regulation of breathing.

In assessing the response of the controller to changes in P_aO₂, P_aCO₂ and arterial pH, one must also examine the interactions of the peripheral and central chemoreceptors. Because of the blood–brain barrier, the neurons in each receptor may have somewhat different local environments at a given point in time. This can cause the chemoreceptors to seem to be out of phase with each other. In fact, it serves to smooth out the response to an acute change in gas exchange or the acid–base status of the body.

The blood–brain barrier has a differential permeability to ions such as H⁺ (low permeability) and lipid-soluble molecules such as carbon dioxide (high permeability). If one were to infuse an acid into the blood, the peripheral chemoreceptor would respond by increasing ventilation before the local environment in the fluid bathing the medulla reflected the acid pH in the blood. As ventilation increases by virtue of the stimulation of the peripheral chemoreceptor, the P_aCO₂ decreases, which results in the diffusion of carbon dioxide from the fluid surrounding the brain back into the blood. The environment of the central chemoreceptor would rapidly reflect the lower PCO₂, but only later reflect the elevated H⁺ concentration of the blood (because of the extra time needed for the H⁺ ions to cross the blood–brain barrier, as already mentioned). The activity of the central chemoreceptor would decrease in the short term, which would attenuate the body’s total response to the acid challenge. Alternatively, if one infused a buffer into the blood, such as sodium bicarbonate, and the arterial pH level increased, the activity of the peripheral chemoreceptor would decrease, ventilation would decrease, and the P_aCO₂ level would increase. Carbon dioxide would then diffuse across the blood–brain barrier and increase the PCO₂ level in the brain. Again, because the equilibration of H⁺ across the blood–brain barrier occurs more slowly, in the short term, the activity of the central chemoreceptor would increase (Fig. 6-5).

Use Animated Figure 6-5 to acutely change the pH and PCO₂ of the blood, and observe the changes in the fluid bathing the brain, as well as the effects on the activity of the peripheral and central chemoreceptors. In particular, notice how the central chemoreceptors can serve to attenuate the short-term ventilatory response to an acid or base load.

Animated Figure 6-5 (Work in progress)

The responses described thus far reflect the sequence of events that occurs as the result of acute changes in carbon dioxide or pH. Both the blood and the brain have mechanisms to restore pH toward normal levels when acute disruptions occur. Thus, when P_aCO₂ is elevated chronically, as might occur in a patient with severe COPD, the activity of the peripheral and central chemoreceptors decrease within a few days as pH is normalized. At extremely high levels of carbon dioxide (PCO₂ > 80–100 mm Hg), an anesthetic effect may be produced, and ventilation decreases rather than increases (Table 6-2).

The Ventilatory Response to Hypoxemia ↑

One can characterize the ventilatory response of an individual to acute derangements in P_aO₂ and P_aCO₂ with a laboratory evaluation in which the person breathes a mixture of gases that alter the partial pressure of the gas in the blood in a progressive manner. The ventilation is measured continuously and plotted as a function of the partial pressure of oxygen or carbon dioxide (note: in healthy lungs, the alveolar and arterial PO₂ levels will be quite close, but in diseased lungs, there may be a range of alveolar PO₂ values, but there can be only one arterial PO₂; the ventilatory response is dependent on the P_aO₂ level and is most accurately represented as a reflection of arterial PO₂).

In keeping with the activity of the peripheral chemoreceptor, as described previously, ventilation changes little as the P_aO₂ decreases from 95 to 60 mm Hg. At that point, the ventilation starts to increase (Fig. 6-6).

At moderate degrees of hypoxemia—P_aO₂ between 45 and 60 mm Hg, for example— the ventilation is only elevated to approximately twice the normal level. It is only after the P_aO₂ decreases below 40 mm Hg that ventilation increases sharply. When acute hypercapnia is present simultaneously with acute hypoxemia (this occurs typically when a patient has disease of the ventilatory pump or gas exchanger of such severity that the increased stimulation of the controller from the hypoxemia alone is insufficient to lead to hypocapnia), there is a synergistic effect, and ventilation is substantially elevated. The acute hypercapnia produces further stimulation of the peripheral and central chemoreceptors. This is in contrast to the counterbalancing effect that acute hypocapnia, which normally accompanies the hyperventilation associated with hypoxemia, would have on the ventilatory response to hypoxemia.

The Ventilatory Response To Hypercapnia ↑

In contrast to the ventilatory response to hypoxemia, ventilation increases linearly with acute increases in P_aCO₂. The normal range of the response is between 2 and 5 L/min increase in ventilation for each 1-mm Hg increase in P_aCO₂ (Fig. 6-7).

The slope of the ventilatory response appears to be genetically determined. Family members tend to be similar in their location within the normal range, that is, they all tend to have relatively brisk or blunted responses.

The ventilatory response to acute hypercapnia is more pronounced than the response to hypoxemia. Whereas a decline in the P_aO₂ from 90 to 45 mm Hg may cause the ventilation to double from 5 to 10 L/min, an increase in the P_aCO₂ from 40 to 50 mm Hg may cause ventilation to increase from 5 to 40 L/min. The reason for this observation is not known, but with our knowledge of respiratory system physiology, we can speculate about a few possibilities. First, with acute hypoxemia, the increase in ventilation leads to hypocapnia and a blunting of the activity of the central chemoreceptor. In contrast, both the peripheral and central chemoreceptors increase their activity with acute hypercapnia. Second, acute hypercapnia results in the accumulation of acid in the blood, which further stimulates the peripheral and central chemoreceptors.

The magnitude of the ventilatory response to hypercapnia can be affected by a number of other factors. When one is asleep, for example, the curve is shifted to the right (less of an effect on ventilation). Similarly, drugs that depress the central nervous system such as narcotics and anesthetics depress the response.