A breathless patient with a suspected underlying neurologic disorder is alarming and requires a swift response. The patient feels that respiratory function is changing and, due to the increased work of breathing, may use descriptions such as “chest tightness.”1 Respiratory distress may not be apparent, noticeable only when provoked by change in position or with testing of respiratory mechanics. It may be the first defining sign of neuromuscular disease.2,3,4 In other circumstances patients may present in coma, catching breaths or are even ceasing to breathe.
Neurologic disease can impair respiration at multiple levels (Fig. 1.1). The interconnections between cerebral hemispheres, respiratory centers in the brain stem and axons, motor neurons, phrenic nerves, and the respiratory muscles provide a functional system that moves air in and out of the lungs. If the system fails, hypercapnia results. The alveoli and pulmonary capillaries subsequently permit efficient gas exchange by diffusion through a foil-thin barrier; if that fails, hypoxemia results. Both conditions may occur simultaneously or one disorder may lead to the other when reduced airflow leads to poor alveolar recruitment and collapse.
There are a number of steps one could take to narrow down the diagnostic possibilities. In the initial evaluation, it is logical to ask three major questions (Table 1.1): Does the patient generate breaths? Is air getting to where it needs to go? Is the pulmonary apparatus intact?5
Acute lesions of the hemisphere or brain stem impact on automatic or voluntary respiratory control. The automatic control of the respiratory drive is generated in the primary ventilatory nuclei in the brain stem (Box 1.1). Loss of automatic control (Ondine's curse) has been reported with neuroblastoma and syringobulbia but is extremely rare.
The voluntary control originates in the cortex and connects to spinal cord levels with motor neurons sending connecting fibers to the diaphragm, intercostal muscles, and abdominal muscles. Impaired voluntary breathing involves a gamut of respiratory disorders and is not discussed further. All of these breathing patterns result in hypoxemia and can rarely be observed well because patients have already been placed on a mechanical ventilator. Some of these patterns are apneustic breathing or cluster breathing, both common with acute lesions in the brain stem. Cheyne-Stokes breathing (an oscillating cycle of 2–3 minutes of hyperpnea separated by apnea) is very frequent and could lead to brief periods of oxygen desaturation.
Failure to maintain a patent airway may have originated directly from acute neurologic disease. Breathing may be obstructed at the pharyngeal or laryngeal level due to tongue displacement, vomit, or tooth fragments. Breathing may also be labored from stridor, recognized by a high-pitched noise at inspiration. It is not infrequent after extubation from subglottic edema or traumatic epithelial injury6 but may be due to laryngeal dystonia (see Chapter 5) or vocal cord paralysis. Failure of gas exchange could be due to profound aspiration or, less commonly, neurogenic pulmonary edema. Frothy sputum and tachypnea often accompany hypoxemia and increased alveolar-arterial oxygen gradient.
Figure 1.1 Causes of respiratory failure in neurologic disease at different levels of the nervous system.
Respiratory mechanics can become impaired due to lesions of the lower motor neuron (Box 1.2). However, disorders of the central nervous system may also impact on mechanics of the rib cage. Restricted muscle movements in advanced stages of Parkinson's disease are also reflected by respiratory muscles causing incoordinated respiratory muscle pump function. Acute spinal cord lesions affecting higher cervical regions (C3–C5) result in ventilator dependence. Lesions below the level of C5 spare the nerve connections to the diaphragm, but expiratory effort is markedly reduced due to involvement of the abdominal and intercostal muscles. Placing the patient in a supine position improves expiration due to pressure of abdominal contents to the chest; breathing becomes labored when the patient is placed in a chair. Other lesions of the spinal cord (particularly multiple sclerosis) may cause diaphragmatic dysfunction, particularly when localized in the upper cervical cord.
Table 1.1. Three Major Causes of Respiratory Failure in Acute Neurologic Disease
The phrenic nerve may be injured, and unilateral damage could cause marked breathlessness during any form of exercise and prevent lying flat. Many cases are unexplained or due to neuralgic amyotrophy (associated with intense pain in shoulder muscles), stretch injury, or traumatic brachial plexus injury.8 It is known to occur in compression of the brachial plexus due to tumor (commonly squamous cell lung carcinoma) or aneurysm (thoracic aorta), prior chest surgery or cannulation of subclavian or internal jugular vein, herpes zoster infection, or chiropractic manipulation.9,10,11
The focus of examination in patients with breathlessness due to neuromuscular disease should be on bulbar dysfunction, paradoxical breathing, and impaired coughing. Careful inspection and testing of the oropharyngeal muscles may point to a diagnosis. Longstanding dysfunction, such as in amyotrophic lateral sclerosis, is often evident by the presence of a wrinkled, fasciculating, slowly moving tongue and a hyperactive jaw reflex. In myasthenia gravis, next to ptosis and ophthalmoparesis, muscle weakness is prominent in the masseter muscles, and repetitive forceful biting on a tongue depressor is soon followed by inability to close the teeth. Passage of air through the nose when asked to blow up the cheek against counterpressure by the examiner's thumb and index finger reveals additional oropharyngeal weakness. The assessment of oropharyngeal weakness hints not only of involvement of the respiratory mechanics but also that ineffective swallowing could lead to aspiration. In a study on predictors of respiratory decline in Guillain-Barré syndrome, bulbar dysfunction predicted later requirement of mechanical ventilation.12
The symptoms and signs of acute neuromuscular respiratory failure are often very subtle. An initial cursory observation of breathing, blood gas, and chest X-ray could indicate no danger to the patient at all. However, after striking up a conversation, it becomes obvious that the patient frequently pauses in sentences to take a breath, displays sweat accumulations at the hairline, demonstrates a mild tachycardia, and, when asked, confirms a sense of discomfort and increased work of breathing. The arterial blood gas can be entirely normal because the patient, due to an increase in frequency, is still able to compensate for a threatening hypoxemia. (The typical response in other medical disorders is increased tidal volume, but this is actually reduced because of respiratory muscle fatigue.) The tachypnea may be subtle, and respiratory rate is often increased to 20 breaths per minute and quickly rises. A useful bedside test is to have the patient count to 20 in one breath after maximal inhalation. If the patient can count, advancing one per second, the vital capacity is probably still within normal range. The classic clinical features of inspiratory paradox are characterized by inward movement of the abdomen during inspiration. With normal inspiration, lungs fill with positive pressure after the chest expands during diaphragm contraction, moving the abdominal contents out. When the diaphragm stops contracting, the positive pressure is replaced by a negative pressure, which causes the inward sucking movement of the abdomen. However, inspiratory paradox due to diaphragm weakness reveals itself late in the illness. Particularly when observed in a patient with an acute neuromuscular disorder, it more than likely indicates that an early opportunity for endotracheal intubation has been missed. These patients are on the verge of apnea, often in the middle of the night.13It is important to note here that hypoxemia and hypercarbia are additional late phenomena, even in patients who are marginally compensated. Hypoxemia occurs due to significant shunting associated with collapse of multiple alveoli that are not recruited from breathing.
Box 1.1. The Anatomy of Central Control of Breathing
Two centers in the brain stem generate a respiratory oscillating pattern: the medulla oblongata central pattern generator and the pontine respiratory group. The medulla oblongata center consists of two major respiratory neuron groups with separate tasks. The dorsal respiratory group times inspiration and the respiratory cycle, and the ventral respiratory group is involved with expiration and includes the expiratory neurons of the Bötzinger complex. Within this ventral population is also the nucleus ambiguus, for dilator function of the upper airway during inspiration, and the nucleus paraambigualis, for inspiratory force. When these respiratory neurons in the medulla oblongata fire, specific patterns of the respiratory cycle are identified, suggesting architectural organization. The pontine center has a connecting link to the medulla center and functions as a time-tuning controller (e.g., setting lung volumes). Input to these centers comes from nonchemical reflexes (pharyngeal and pulmonary receptors and vagus nerve) and chemoreceptors (hydrogen ion). Stimulation is by decreased partial pressure of oxygen (PaO2) (not content such as anemia) and increased partial pressure of arterial carbon dioxide (PaCO2). The cortex can override these control centers, allowing speech, singing, coughing, and breath holding.
Box 1.2. The Anatomy of Pulmonary Mechanics
The activation of muscles in the upper airway, particularly the pharyngeal constrictor muscles and genioglossus, maintains a patent pharynx. The diaphragm is the major contributor to the respiratory pump. Using the abdomen as a bearing, its descent displaces the abdominal contents in a caudal and outward direction. The intercostal muscles run obliquely caudal and backward from the rib above to the rib below and thus provide, with contraction, an additional function. The muscles involved with respiration are not all active, and the inspiratory and expiratory muscles cycle during quiet breathing, the diaphragm contracts synchronously with the intercostal and scalene muscles (inspiration), and abdominal muscles barely assist with passive recoil of the rib cage (expiration).
The diaphragm controls most of the inspiration, and dyspnea is expected with its dysfunction. When it contracts, the rib cage lifts due to its cephalocaudal fiber orientation. When it is not sufficient to lever it, accessory muscles such as sternocleidomastoid, pectoralis, trapezius, and latissimus dorsi, which harness die rib cage, are recruited. Expiration is assisted by contraction of the abdominal muscles. Coughing requires closure of the glottis and contraction of the diaphragm and abdominal muscles. The abdominal muscles can become severely affected in any neuromuscular disorder, reducing the effectiveness of coughing and atelectasis. Conversely, a patient with a forceful cough rarely has significant neuromuscular respiratory failure. Drugs (particularly opioids) and sleep can have an additional detrimental effect on respiratory drive and load.7
Bedside Respiratory and Laboratory Equipment
Pulmonary function tests provide quite useful values and are easy to obtain using non-electrical bedside devices. Commonly used peak flow devices (e.g., for asthma) are unreliable because expiratory peak flow rates can be normal. In neuromuscular respiratory failure, the airway is patent and lung recoil is actually increased. The simplest tests are assessment of vital capacity (VC), maximal inspiratory pressure (PImax), and maximal expiratory pressure (PEmax). The patient's position when these values are obtained is rather critical because clinically relevant diaphragmatic fatigue may become obvious in the supine position.
The technique of obtaining respiratory muscle function values is important, and scuba diving mouthpieces may reduce leakage, particularly when bilateral facial palsy is present. After the patient is connected to this apparatus, a nose clip is placed and the airway is occluded by blocking a port in the valve or by closing a shutter. VC is the volume of gas measured from a slow, complete forced expiration after maximal inspiration. VC can be reduced by additional airway and pulmonary disorders, certainly in patients with prior restricted pulmonary disease. PImax is recorded when a patient forcefully inspires against an occluded device. Typically, PImax is measured near residual volume at the end of maximal expiration and has a negative value as a result of the inspiratory effort in the presence of an occluded airway. PEmax measured near total lung capacity is the maximal pressure that can be generated by the patient making a forceful expiratory effort in an occluded airway. The manometer is able to record from 10 to 200 cm H2O. The coaching of the patient is very important. Many patients have a tendency to produce a Valsalva maneuver, after which the required pressure is not generated and leads to falsely low values. Lack of understanding by the patient on how to perform this test remains a major problem in obtaining these spirometric values.
Normal adults can generate at least –60 cm H2O of inspiratory pressures, and these are decreased in patients who have weakness of the mechanical function of the rib cage. PImax is largely a function of the abdominal and accessory muscles of respiration and some elastic recoil of the lungs. However, PImax can also be decreased in patients with hyperinflation disorders, such as emphysema. This condition in advanced stages makes the diaphragm flat due to trapped gas in the lungs.
There is little, if any, evidence of its usefulness in clinical assessment except in Guillain-Barré syndrome. A retrospective analysis of 114 patients with Guillain-Barré syndrome noted possible critical values of vital capacity of less than 20 mL/kg, PImax less than 30 cm H2O, PEmax less than 40 cm H2O (the so-called 20/30/40 rule) but also any reduction in vital capacity of more than 30% from baseline.12
Another study found VC 60% of the predictive value already a warning sign.14
An important correlation, or lack thereof, is reduction of respiratory muscle strength with PaO2 or increased PaO2. Only when pulmonary function tests are markedly reduced is some rise in PaCO2 expected; however, low-flow (0.5–2 L/min) O2 administration may worsen hypercarbia substantially.15 In these patients, measures to increase alveolar ventilation do not exist. In those patients with longstanding neuromuscular respiratory dysfunction and carbon dioxide retention, there is a dependence on a “hypoxemia drive.” The additional administration of oxygen may cause apnea and hypercarbic coma.
The chest X-ray remains important in assessing pulmonary abnormalities. In neurogenic pulmonary edema it may show hazy opacities and airspace shadowing indistinguishable from cardiogenic pulmonary edema or aspiration pneumonitis (Fig. 1.2). Both acute lung injury and cardiac dysfunction may be present as a result of an adrenergic surge in acute hemispheric lesions or subarachnoid hemorrhage.16 Suppression of cough reflex by sedatives or antiepileptic agents (e.g., barbiturates) may cause mucus plugging of a main bronchus (Fig. 1.3). The chest X-ray may also show indirect signs of phrenic nerve injury (Fig. 1.4). Phrenic nerve conduction tests may document absent responses.8
Figure 1.2 Serial chest X-ray in a patient with subarachnoid hemorrhage showing acute development of pulmonary edema. Left, Chest X-ray on admission. Middle, Acute development of pulmonary infiltrates and enlargement of the heart shadow, indicating pulmonary edema due to cardiac dysfunction. Right, After improvement in cardiac function with use of inotropes, infiltrates remain, suggesting dual injury to lungs and heart.16
Figure 1.3 Acute bronchial occlusion from mucus plug (left), with reexpansion of the lung after bronchoscopic removal (right).
Line of Action
Upper airway obstruction should be relieved, and in stridor laryngoscopy should be performed to view vocal cord erosion or epithelial damage. Elective endotracheal intubation must be performed in any patient with persistent decrease in responsiveness and marginal oxygenation. Patients who have a mild tachycardia, display evidence of hypoxemia on a pulse oximeter, have evidence of increased work of breathing with change in posture,17 or display the appearance of sweat beads need to be intubated preemptively.
In comatose patients, obstruction of the airway occurs for several reasons. First, muscles of the floor of the mouth and tongue become reduced in tone, and this changes the anatomic relationships. The tongue is repositioned to the back wall of the oropharynx and obstructs the airway. This position is even more exaggerated when the head is flexed. Therefore, with a simple technique the airway can be reopened. This so-called head-tilt/chin-lift (Fig. 1.5) tilts the head backward to what is often called the “sniffing position.” In this position, the trachea and pharynx angulation is minimal, allowing for air transport. Also, the index and middle fingers of the examiner's hand lift the mandible and bring the tongue forward.
Figure 1.4 Serial chest X-rays showing development of hemidiaphragm elevation due to phrenic nerve injury on the right. (Left, normal; right, abnormal.)
Another technique is the so-called jaw-thrust/head-tilt. The examiner places the ring, middle, and little fingers underneath the patient's jaw and lifts the chin forward. The examiner's index finger and thumb are free to fit a mask snugly to the face, with the other hand free to operate a resuscitation bag.
When the airway appears blocked by foreign material or dentures, this technique is modified by placing the thumb in the mouth, grasping the chin, and pulling it upward, leaving the other hand to clear any obstructing material from the airway (Fig. 1.5).
An oropharyngeal airway should be placed and is essential in patients who recently had a seizure because it prevents further tongue biting. The placement of this oral airway device is simple. The mouth is opened, a wooden tongue depressor is placed at the base of the tongue, and downward pressure is applied to displace the tongue from the posterior pharyngeal wall. The oropharyngeal tube is then placed close to the posterior wall of the oropharynx and is moved toward the tongue until the teeth are at the bite-block section. Alternatively, the jaw is thrust forward and the device is placed concave toward the palate and then rotated.18 Dental injury, most commonly in patients who have significant dental or periodontal disease, rarely occurs.
Jaw thrust and mask ventilation securely maintain an open airway but must be followed by endotracheal intubation done by an experienced physician. Endotracheal intubation may be complicated in a traumatized patient with possible cervical spine injury. The ideal solution in these patients is to use fiberoptic bronchoscopy because with this procedure the risk of further neck trauma from neck movement is very low. Immediate endotracheal intubation is required in patients with penetrating neck trauma or significant intraoral bleeding. Temporarily, a cricothyrotomy can be made. A 14-gauge needle is inserted through the cricothyroid membrane, followed by insertion of a cannula. (The cricothyroid membrane is located just under the thyroid.) A formal tracheostomy should follow because ventilation through this small, highly flow-resistant tube is compromised.
Hypoxemia is often encountered, and oxygen administration has a high priority in patients with impaired consciousness. Nasal prongs are inefficient because they provide only 30% oxygen concentrations and often dislodge. Nasopharyngeal catheters provide 60% oxygen concentrations (but only when the tip of the catheter is visible above the soft palate or face mask) and are a better alternative. Resuscitation bags are an optimal source of oxygen, and they can deliver a fraction of inspiratory O2 (FiO2) above 0.9 when the oxygen flow in the bag is 10 mL/minute. Oxygenation should be monitored with a pulse oximeter (O2 saturation should exceed 90%) or measurement of an arterial blood gas sample (PaO2 >100 mm Hg).
Figure 1.5 Techniques of airway management. A, Tongue jaw lift/finger sweep. B, Head tilt/chin lift. C, Jaw thrust/mask ventilation. From Wijdicks EFM, Borel CO.18 By permission of Mayo Foundation.
Figure 1.6 Critical steps in imminent neuromuscular respiratory failure. VC, vital capacity; PImax, maximal inspiratory pressure; PEmax, maximal expiratory pressure; PaO2/PaCO2, partial pressure of arterial oxygen/carbon dioxide; BIPAP, bilevel positive airway pressure ventilation.
The indications for intubation in patients with acute neuromuscular failure are shown in Figure 1.6. Noninvasive bilevel positive airway pressure (BiPAP) ventilation in acute myasthenic crises, amyotrophic lateral sclerosis, and likely other chronic neuromuscular disorders should be tried first and may prevent mechanical ventilator dependence, tracheostomy, and volume-controlled, ventilator-associated pulmonary injury. BiPAP may avert intubation in acute myasthenic crises and amyotrophic lateral sclerosis.19,20,21,22 However, presence of hypercapnia predicts failure of BiPAP, and volume-controlled mechanical ventilation is preferred.21 Our initial experience in Guillain-Barré syndrome is disappointing and we have seen respiratory distress persisting or emerging suddenly with BiPAP. Respiratory failure is commonly anticipated in patients with chronic neuromuscular disorders. This option ideally is addressed before intubation is undertaken.23 In some patients (albeit very few), marked atelectasis, intervening pneumonia, or aspiration may have substantially contributed to respiratory failure. Thus, it should be emphasized that ventilatory dependence is not an incontrovertible outcome, and a period of rest with ventilatory support may lead to marked improvement.
After endotracheal intubation, virtually all patients are well served with an initial ventilator order that includes intermittent mandatory ventilation mode. The positive pressure breaths that are delivered by the mechanical ventilator are triggered by the patient, who has to generate only small pressure differences. The patient is able to breathe in between ventilator breaths and has entirely unsupported breaths. This ventilator order is particularly useful in neurologic patients because it allows for spontaneous breathing, and it can deliver hyperventilation if needed. A typical ventilator order for neurologically stable patients is a synchronized intermittent mandatory ventilation mode with an FiO2 of 0.4–1.0, respiratory rate at 8–12 breaths/minute, tidal volume of 10–15 mL/kg, positive end-expiratory pressure of 2–5 cm H2O, and an inspiration/expiration ratio of 1–3.24,25
1. Sivak ED, Shefner JM, Sexton J: Neuromuscular disease and hypoventilation. Curr Opin Pulm Med 5:355, 1999.
2. Braun NMT, Arora NS, Rochester DF: Respiratory muscle and pulmonary function in polymyositis and other proximal myopathies. Thorax 38:616, 1983.
3. De Carvalho M, Matias T, Coelho F, et al.: Motor neuron disease presenting with respiratory failure. J Neurol Sci 139:117, 1996.
4. Kelly BJ, Luce JM: The diagnosis and management of neuromuscular diseases causing respiratory failure. Chest 99:1485, 1991.
5. Wijdicks EFM: Short of breath, short of air, short of mechanics. Pract Neurol 2:208, 2002.
6. Jaber S, Chanques G, Matecki S, et al: Post-extubation stridor in intensive care unit patients: risk factors evaluation and importance of the cuff-leak test. Intensive Care Med 29:69, 2003.
7. Ragette R, Mellies U, Schwake C, et al.: Patterns and predictors of sleep disordered breathing in primary myopathies. Thorax 57:724, 2002.
8. Chen YZ, Zu JG, Shen LJ, et al.: Phrenic nerve conduction study in patients with traumatic brachial plexus palsy. Muscle Nerve 24:1388, 2001.
9. Pandit A, Kalra S, Woodcock A: An unusual cause of bilateral diaphragmatic paralysis, Thorax 47:201, 1992.
10. Soler JJ, Perpi[not tilde]a M, Alfaro A: Hemidiaphragmatic paralysis caused by cervical herpes zoster. Respiration 63:403,1996.
11. Piehler JM, Pairolero PC, Gracey DR, et al.: Unexplained diaphragmatic paralysis: harbinger of malignant disease? J Thorac Cardiovasc Surg 84:861, 1982.
12. Lawn ND, Fletcher DD, Henderson RD, et al.: Anticipating mechanical ventilation in Guillain-Barré syndrome. Arch Neurol 58:893, 2001.
13. Wijdicks EFM, Henderson RD, McClelland RL: Emergency intubation for respiratory failure in Guillain-Barré syndrome. Arch Neurol 60:947, 2003.
14. Sharshar T, Cheveret S, Bourdain F, et al.: Early predictors of mechanical ventilation in Guillain-Barré syndrome. Crit Care Med 31:278, 2003.
15. Gay PC, Edmonds JC: Severe hypercapnia after low-flow oxygen therapy in patients with neuromuscular disease and diaphragmatic dysfunction. Mayo Clin Proc 70:327, 1995.
16. De Chazal IR, Liopyris P, Wijdicks EFM: Cardiogenic shock and acute lung injury after aneurysmal subarachnoid hemorrhage (submitted for publication).
17. Allen SM, Hunt B, Green M: Fall in vital capacity with posture. Br J Dis Chest 79:267, 1985.
18. Wijdicks EFM, Borel CD: Respiratory management in acute neurologic illness. Neurology 50:11, 1998.
19. Lyall RA, Donaldson N, Fleming T, et al.: A prospective study of quality of life in ALS patients treated with non-invasive ventilation. Neurology 57:153, 2001.
20. Miller RG, Rosenberg JA, Gelinas DF, et al.: Practice parameter: the care of the patient with amyotrophic lateral sclerosis (an evidence-based review): report of the Quality Standards Subcommittee of the American Academy of Neurology: ALS Practice Parameter Task Force. Neurology 52:1311, 1999.
21. Rabinstein AA, Wijdicks EFM: BiPAP in acute respiratory failure due to myasthenic crisis may prevent intubation. Neurology 59:1647, 2002.
22. Sivak ED, Shefner JM, Mitsumoto H, et al.: The use of non-invasive positive pressure ventilation (NIPPV) in ALS patients. A need for improved determination of intervention timing. Amyotroph Lateral Scler Other Motor Neuron Disord 2:139, 2001.
23. Rumbak MJ, Walker RM: Should patients with neuromuscular disease be denied the choice of the treatment of mechanical ventilation? Chest 119:683, 2001.
24. Slutsky AS: Mechanical ventilation. American College of Chest Physicians' Consensus Conference. Chest 104:1833, 1993.
25. Wijdicks EFM, Borel CO: Respiratory management in acute neurologic illness. Neurology 50:11, 1998.