Neuromuscular respiratory failure
The first priority in evaluating the patient with rapidly progressive weakness is to determine whether they will require immediate intubation. Warning signs of impending respiratory disaster include tachypnea, punctuated speech, and accessory respiratory muscle use. Check inspiratory muscle strength at the bedside by asking the patient to count to 30 as quickly as possible. The ability to do so in a single breath suggests sufficient diaphragmatic strength to maintain adequate gas exchange. While performing the initial bedside assessment, check for a strong, forceful cough, which indicates the ability to “clear the airway.”
Unlike most cardiopulmonary sources of respiratory distress, early neuromuscular respiratory failure is usually not accompanied by a decline in oxygen saturation. Similarly, typical arterial blood gas abnormalities that reflect hypoventilation (low PO2 and high PCO2) do not appear early enough to predict the need for intubation. Rather, bedside spirometry is the most important quantitative measurement of neuromuscular respiratory failure. The two most important spirometry values are the negative inspiratory force (NIF) and forced vital capacity (FVC). Values that should prompt you to consider intubation or at least monitoring in an intensive care unit setting are a NIF <40 cmH2O and a FVC <1 liter. Use clinical judgment in interpreting these numbers, as poor patient effort and bulbar weakness (resulting in inability to form an adequate seal around the spirometer) lead to false-positive results.
Initial pattern of weakness
All severe generalized weakness ultimately produces a flaccid, intubated patient. It is the initial pattern of weakness, therefore, that helps to determine the etiology. Severe weakness beginning in the extraocular muscles and descending over a period of hours suggests botulism. Fluctuating extraocular weakness with ptosis for several weeks or more prior to the onset of generalized weakness is most consistent with myasthenia gravis. Weakness that begins in the bulbar muscles, descends rapidly, and is accompanied by a pharyngeal exudate is classic for diphtheria. Limb weakness sparing the face is most consistent with pathology at the level of the cervical spine. Weakness that begins in the legs and ascends rapidly over several days to a few weeks is the classic (but not exclusive) presentation of Guillain–Barré syndrome (GBS; acute inflammatory demyelinating polyradiculoneuropathy). Generalized weakness that develops over seconds is most consistent with infarction or hemorrhage of the ventral pons.
Although many patients with rapidly progressive weakness are intubated and incapable of cooperating with the neurological examination, it is essential to perform as careful an assessment as the circumstances will allow. Mental status should be tested in as much detail as possible in order to exclude coma and severe encephalopathy masquerading as weakness (see Chapters 1 and 2). Important components of cranial nerve examination include pupillary reactions and eye movements. Absent pupillary reactions are characteristic of but not universal in botulism: more than half of patients with botulism actually have normal, reactive pupils.1 While diplopia and ptosis are common to many causes of rapidly progressive weakness, fluctuating rather than fixed extraocular muscle weakness strongly suggests myasthenia gravis. Preserved vertical eye movements in a tetraplegic, seemingly unresponsive patient point to a locked-in state caused by pontine hemorrhage or infarction. Deep tendon reflex testing assumes utmost importance in patients with rapidly progressive weakness. Hyporeflexia or areflexia are prerequisites for the diagnosis of GBS. While spasticity and hyperreflexia might be expected in CNS disorders, patients with spinal cord lesions sufficient to produce spinal shock are flaccid and areflexic. Sensory examination is often unhelpful in patients with rapidly progressive weakness, as cooperation is usually limited. In a patient who is sufficiently awake, decreased sensation points to a peripheral nerve or spinal cord lesion, whereas preserved sensation makes the muscle or neuromuscular junction more likely localizations.
If there are signs of a specific etiology of rapidly progressive weakness, investigations should be tailored to confirm that diagnosis. In many cases, however, a broad spectrum of studies is required, as the diagnosis is not apparent from the clinical history and examination. Although this “shotgun” approach may appear inelegant, the gravity of rapidly progressive weakness and limitations in the examination necessitate a thorough battery of tests in many cases. The most important diagnostic tools are neuroimaging studies of the brain and spine, nerve conduction studies (NCS) and electromyography (EMG), and lumbar puncture.
In most cases, rapidly progressive weakness results from injury to the peripheral nervous system rather than the CNS. In some cases, however, ischemic, inflammatory, or space-occupying lesions of the CNS may produce acute-onset weakness. MRI of the brain and entire spine with and without contrast should be performed if CNS injury is suspected or if evaluation of the peripheral nervous system is unrevealing.
Electromyography and nerve conduction studies
These are the most useful diagnostic studies for localizing the source of rapidly progressive weakness. Nerve conduction studies involve stimulating and recording from a select group of sensory and motor nerves, thereby allowing measurement of the amplitude and velocity of peripheral nerve responses. Needle EMG allows recording of spontaneous activity from muscles and motor unit analysis, and is helpful in patients with muscle, nerve, and nerve root disease. Repetitive nerve stimulation and single-fiber EMG are special studies that assess the neuromuscular junction.
Neurophysiological testing has several important limitations when used for evaluation of patients with severe generalized weakness. First, a fair amount of cooperation is required for needle EMG, and single-fiber EMG is essentially impossible in severely weak patients. Secondly, many patients with severe weakness are intubated and have a variety of intravenous lines and other catheters that prevent adequate exposure for electrode placement. Finally, electrical interference from intensive care unit equipment creates excessive electrical noise.
The following is a brief summary of the most important patterns of EMG and NCS abnormalities in patients with rapidly progressive weakness.
The neurophysiological hallmark of GBS is acquired demyelination: NCS show prolonged distal latencies, markedly slowed conduction velocities (<75% of normal values), abnormal temporal dispersion, conduction block, and prolonged late responses (see Chapter 10, Figure 10.4). As discussed below, many of the findings of primary demyelination are not present in early GBS. Needle EMG showing decreased motor unit recruitment may be the only electrophysiological abnormality in the first week of the disease.
Axonal neuropathies are less likely to lead to rapidly progressive weakness than are demyelinating ones. Neurophysiological characteristics of axonal neuropathy include low response amplitudes and mildly reduced (never below 75% of normal) conduction velocities. Electromyography shows decreased motor unit recruitment and fibrillation potentials. Rapidly progressive weakness produced by an axonal polyneuropathy should prompt investigation for heavy metal intoxication, porphyria, and vasculitis. Axonal variants of GBS are less common outside China.
Presynaptic neuromuscular junction dysfunction
Botulism is the most important presynaptic neuromuscular junction disorder leading to rapidly progressive weakness. The hallmark of presynaptic neuromuscular junction dysfunction is that exercise or fast repetitive nerve stimulation results in an increase in motor amplitudes (see Chapter 10, Figure 10.3).
Postsynaptic neuromuscular junction dysfunction
Myasthenia gravis and other postsynaptic neuromuscular junction transmission disorders are accompanied by abnormal decremental responses with repetitive nerve stimulation (Chapter 10, Figure 10.2). If it is technically possible in the intensive care unit, perform single-fiber EMG to investigate for increased jitter and blocking.
Muscle diseases uncommonly cause rapidly progressive weakness, but may cause failure to wean from the ventilator (see below). The EMG characteristics of myopathy are short-duration, small-amplitude, polyphasic motor units with early recruitment. Motor NCS may show decreased response amplitudes, while sensory NCS are normal.
The characteristic CSF finding in GBS is albuminocytological dissociation in which elevated protein is accompanied by a normal cell count. This abnormality, however, is present in only 70% of patients within 1 week of symptom onset.2 The spinal fluid in acute disseminated encephalomyelitis and other inflammatory myelopathies shows increased white blood cell counts and protein. Viral antibodies are present in patients with poliomyelitis caused by West Nile virus. CSF protein may be slightly elevated in patients with botulism.
Blood, urine, and stool examination
Abnormalities in blood, urine, and stool may help to reach a diagnosis in some patients with rapidly progressive weakness. Patients with lead, arsenic, or thallium poisoning will have elevated levels of the relevant heavy metal. Antibodies to West Nile virus are seen in some patients with poliomyelitis. Antibodies to the acetylcholine receptor are present in the majority of patients with myasthenic crisis. If tested early enough, botulinum toxin is detectable in the stool or serum of patients with botulism.1 Elevated levels of 24-hour urine porphobilinogen excretion are diagnostic of porphyria.
Causes of acute paralysis
Guillain–Barré syndrome (GBS), or acute inflammatory demyelinating polyradiculoneuropathy (AIDP), is the most common source of acute-onset generalized paralysis in the industrialized world. As its name suggests, it is an autoimmune demyelinating disorder of the peripheral nerves and roots. The classical clinical history begins with a prodrome of gastrointestinal or respiratory illness or vaccination, which is followed several weeks later by numbness and tingling in the extremities and weakness that begins in the feet and spreads upwards to the legs, arms, face, and respiratory muscles. Many cases of GBS, however, do not conform to this pattern, and weakness may also begin in the proximal muscles or even occasionally in the bulbar muscles. Back pain is often exquisite and may lead to exhaustive but fruitless investigation for structural spinal disease. The symptoms of GBS become maximal within 2–4 weeks of onset. Untreated, most patients develop rapidly progressive flaccid weakness, which often requires intubation. Occasional patients may develop only mild distal weakness and paresthesias.
The two most important physical examination findings are symmetric weakness and decreased reflexes. While there may be some side-to-side asymmetry in patients in the early stages of GBS, the left and right sides should be symmetrically weak as the disorder progresses. Consider a diagnosis other than GBS in the absence of hyporeflexia or areflexia. Bulbar weakness may be present in patients with more advanced weakness. Sensory loss is less severe than motor dysfunction, with large-fiber modalities such as vibration and proprioception typically more affected than small-fiber modalities such as pinprick and temperature.
In many cases, the diagnosis of GBS is obvious from the clinical history and examination. Lumbar puncture and EMG play a confirmatory role, but may be especially important in patients with very early disease in whom the syndrome is incomplete. Albuminocytological dissociation of the cerebrospinal fluid in which the protein is high and the white blood cell count is low (<10 cells/mm3) is present in approximately 70% of patients with symptom duration of 1 week or less.2 If >10 cells/mm3 are present, consider HIV seroconversion as an alternative explanation.3 Electromyography findings diagnostic of GBS (discussed above) are present in only 50% of patients within the first 5 days of symptoms.2
Plasmapheresis and intravenous immunoglobulin (IVIg) are the mainstays of the immunomodulatory treatment of GBS. Plasmapheresis is performed as a series of five exchanges every other day. Complications of plasmapheresis include those related to central line insertion and fluctuations in blood pressure due to large-volume fluid shifts. The IVIg is administered at a dose of 2 g/kg, usually divided over 2 days. Important side effects of IVIg include headache, aseptic meningitis, kidney failure, and hypercoaguability. Both plasmapheresis and IVIg reduce the time required to regain the ability the walk, and should be started as quickly as possible.4,5 Although neither treatment is superior to the other, I choose IVIg more often, as it can be started more quickly, has fewer serious associated side effects, and, in at least one relatively large study, showed a trend towards producing faster improvement than plasmapheresis.6
Supportive care is also important in patients with GBS. Patients with airway compromise, cardiac arrhythmias, and blood pressure fluctuations should be treated in the intensive care unit. Patients with GBS often require narcotics (including patient-controlled anesthesia in some cases) and agents for neuropathic pain in order to obtain adequate pain control.
As might be guessed, the ultimate prognosis of GBS is correlated with the maximal severity of clinical deficits. Electrophysiologically, finding reduced compound muscle action potential (CMAP) amplitudes on NCS predicts slower and incomplete recovery. Some patients who are diagnosed and treated at an early enough stage may be able to walk out of the hospital unassisted. Recovery for patients with severe disease, however, usually takes 3–6 months or longer. Because improvement may be slow, there is often a temptation to treat patients with another course of IVIg or plasmapheresis, or to switch between treatments. There is no evidence, however, to support either of these approaches.
A myasthenic crisis characterized by respiratory muscle paralysis may occur at any stage of the disorder, even as its initial presentation. It may be differentiated from the other causes of rapid-onset weakness by the preceding weakness of extraocular and bulbar muscles. Precipitants of myasthenic crisis include infection and medications that are known to exacerbate myasthenia, especially aminoglycosides. High-dose corticosteroids prescribed for myasthenia gravis may paradoxically worsen symptoms within 1 or 2 weeks of initiation, often leading to crisis. The mechanism for this worsening is unclear, but corticosteroids are usually titrated slowly upwards to prevent this from happening.
Myasthenic crisis should be differentiated from the much less common cholinergic crisis caused by acetylcholinesterase inhibitors. Telltale signs of cholinergic crisis include excessive oral and nasal secretions, fasciculations, and gastrointestinal cramps. Unlike myasthenic crisis, cholinergic crisis resolves within several hours of stopping acetycholinesterase inhibitors. It is best, however, to anticipate the need to intubate all myasthenics with respiratory muscle weakness, rather than to assume that they will improve spontaneously.
Treating myasthenic crisis must obviously begin with pulmonary function testing and intubation as necessary. Superimposed respiratory, gastrointestinal, and urinary tract infections must be addressed. In rare cases, treating such an infection may resolve the crisis. Almost all patients, however, will require a series of five plasma exchanges or IVIg given at a dose of 2 mg/kg divided over 5 days. A response is typically noted within 1–3 weeks, if it does occur. I prefer to treat myasthenics in crisis with plasmapheresis based on personal experience and verified by the findings of Gajdos and colleagues, which suggested that patients who underwent plasmapheresis improved more quickly than those who received IVIg.7 Larger doses of acetylcholinesterase inhibitors are not helpful in treating myasthenic crisis.
Because the benefits of IVIg and plasmapheresis are transient, disease-modifying regimens need to be augmented in patients with myasthenic crisis. Increase steroids to their goal doses while patients receive IVIg or plasmapheresis. In patients who are already taking steroids, start a steroid-sparing agent such as mycophenolate mofetil (500 mg bid to start, increase to 1000 mg bid in 1 week) or azathioprine (50 mg qd to start, increase by 50 mg per dose each week to a goal of 1–2 mg/kg). Be aware, however, that these agents will not provide any symptomatic benefit for several months.
Large brainstem lesions, typically in the ventral pons, may produce severe, instantaneous weakness. The classic example is the locked-in state caused by basilar artery thrombosis or pontine hemorrhage. This state is characterized by complete paralysis of the face and limbs with sparing of vertical eye movements and blinking. Other causes of brainstem catastrophe include inflammatory, neoplastic, infectious mass lesions, and central pontine myelinolysis caused by overly rapid correction of hyponatremia. Patients with suspected brainstem catastrophe should undergo MRI with diffusion-weighted imaging and contrast, and treatment should be directed at the underlying cause.
Less common causes of acute paralysis
Food-borne botulism is an uncommon form of acute-onset paralysis caused by ingestion of botulinum toxin, an inhibitor of presynaptic acetylcholine release.1 Gastrointestinal distress usually begins 12–36 hours after toxin ingestion. Diplopia, ptosis, dysarthria, and
Table 12.1 Uncommon neuropathies causing rapidly progressive weakness
dysphagia are the first neurological symptoms. Large unreactive pupils are present in approximately 50% of patients with botulism, distinguishing the disorder from other causes of acute paralysis. Weakness of the arms, respiratory apparatus, and legs follows within several hours to a few days. The toxin may be identified from the serum or stool if checked within 48–72 hours of symptom onset. Nerve conduction study findings of botulism include small motor amplitudes that increase in size with sustained exercise or rapid (50 Hz) repetitive nerve stimulation. If botulism is strongly suspected, then initiate treatment with antitoxin. Although this intervention will not reverse active symptoms, it may prevent new ones from developing. Recovery from botulism takes many months, but is usually complete if the comorbidities of chronic ventilation can be avoided.
Several uncommon polyneuropathies may produce an acute paralysis that mimics GBS. Because these neuropathies are all rare, they will only be mentioned briefly (Table 12.1).
While it is uncommon in the industrialized world, poliomyelitis due to poliovirus still affects patients in the developing world. After a viral prodrome, patients with poliomyelitis develop painful monoparesis followed by the rapid onset of flaccid paralysis. During the acute phase, the spinal fluid of patients with poliomyelitis shows an increased neutrophil count, which distinguishes it from GBS. West Nile virus is an important cause of poliomyelitis, and may be diagnosed by finding antibodies in the serum or spinal fluid.8
Spinal cord insults
Lesions of the high cervical spinal cord may produce acute paralysis of all four limbs while sparing the face. Trauma is the most important cause of severe cervical myelopathy. Other important spinal processes that cause rapidly progressive weakness include herniated intervertebral discs, transverse myelitis, and spinal cord stroke (Chapters 17 and 22).
Difficulty weaning from the ventilator
Difficulty weaning from the ventilator is usually due to cardiopulmonary disease.9 In some cases, intensivists are not able to find a cardiopulmonary explanation for failure to wean, and neurologists are consulted. Begin the diagnostic process by addressing possible CNS processes such as coma and encephalopathy. Next, exclude disorders acquired prior to intubation such as GBS and myasthenia gravis. Finally, investigate for neuromuscular disorders acquired in the intensive care unit including critical illness polyneuropathy, critical illness myopathy, and prolonged neuromuscular junction blockade.
Critical illness polyneuropathy
Critical illness polyneuropathy (CIP) occurs in the setting of severe sepsis and multiorgan failure, usually in patients with intensive care unit stays of at least a week in duration.10 Clinically, CIP is characterized by flaccid areflexic weakness and sensory loss. Nerve conduction studies show evidence for axonal polyneuropathy in the form of low sensory and motor response amplitudes. Electromyography shows fibrillations and positive sharp waves, which reflect denervation. Unfortunately, there is no specific treatment for CIP, and patients who have the condition usually need several months (or longer) to allow axonal regrowth and clinical recovery.
Critical illness myopathy
Critical illness myopathy (CIM) also occurs in the setting of severe sepsis. Use of corticosteroids and neuromuscular junction blocking agents are additional risk factors for CIM.11 Patients with CIM are usually weak and areflexic, much like those with CIP. Unlike patients with CIP, the sensory examination (if assessable) should be normal in patients with CIM. Creatine kinase levels may be normal, modestly elevated, or markedly elevated. Nerve conduction studies show decreased motor response amplitudes with normal sensory response amplitudes. Like CIP, needle EMG shows signs of denervation including fibrillation potentials and positive sharp waves. If motor units can be activated, they are small, polyphasic, and show early recruitment. It is often difficult to distinguish between CIM and CIP by clinical examination and with standard neurophysiological assessment, and many patients have both conditions simultaneously. The best way to distinguish between the two is by finding electrical inexcitability upon direct muscle stimulation in CIM.12 The distinction between CIP and CIM is not necessarily important, as neither has a specific treatment, and both require weeks to months of supportive care before a substantial improvement occurs.
Prolonged neuromuscular junction blockade
Neuromuscular junction blocking agents are used to paralyze patients for surgery or to maintain an airway in patients with cardiopulmonary disease who are resisting the ventilator. Unfortunately, weakness may persist for several days after these are withdrawn, especially in patients with hepatic or renal dysfunction.13 Although this process is localized to the neuromuscular junction, creatine kinase levels are often elevated. Although abnormal repetitive nerve stimulation may help to diagnose prolonged neuromuscular junction blockade, interference from electrical equipment in the intensive care unit generally makes this technique unfeasible. Other than discontinuing the responsible agents and waiting for recovery, there is no specific treatment for prolonged neuromuscular junction blockade. While most patients recover in a few days, some have prolonged deficits that may overlap clinically with CIM.
1. Hughes JM, Blumenthal JR, Merson MH, et al. Botulinum stool and serum testing – clinical features of types A and B food-borne botulism. Ann Intern Med 1981;95:442–445.
2. Gordon PH, Wilbourn AJ. Early electrodiagnostic findings in Guillain–Barré syndrome. Arch Neurol 2001;58:913–917.
3. Cornblath DR, McArthur JC, Kennedy PGE, Witte AS, Griffin JW. Inflammatory demyelinating peripheral neuropathies associated with human T-cell lymphotropic virus type III infection. Ann Neurol1987;21:32–40.
4. Hughes RAC, Swan AV, van Doorn PA. Intravenous immunoglobulin for Guillain–Barré syndrome. Cochrane Database Syst Rev 2006;(1):CD002063.
5. Raphael JC, Chevret S, Hughes RAC, Annane D. Plasma exchange for Guillain–Barré syndrome. Cochrane Database Syst Rev 2002;(2):CD001798.
6. van der Meche FG, Schmitz PI. A randomized trial comparing intravenous immune globulin and plasma exchange in Guillain–Barré syndrome. Dutch Guillain–Barré Study Group. N Engl J Med1992;326:1123–1129.
7. Gajdos P, Chevret S, Clair B, Tranchant C, Chastang C. Clinical trial of plasma exchange and high dose intravenous immunoglobulin in myasthenia gravis. Ann Neurol 1997;41:789–796.
8. Petersen LR, Marfin AA. West Nile virus: a primer for the clinician. Ann Intern Med 2002;137:173–179.
9. De Jonghe B, Sharshar T, Lefaucheur JP, et al. Paresis acquired in the intensive care unit. A prospective multicenter study. JAMA 2002;288:2859–2867.
10. Bolton CF, Gilbert JJ, Hahn AF, Sibbald WJ. Polyneuropathy in critically ill patients. J Neurol Neurosurg Psychiatry 1984;47:1223–1231.
11. Lacomis D, Giuliani MJ, Van Cott A, Kramer DJ. Acute myopathy of intensive care: clinical, electromyographic, and pathological aspects. Ann Neurol 1996;40:645–654.
12. Rich MM, Bird SJ, Raps EC, McCluskey LF, Teener JW. Direct muscle stimulation in acute quadriplegic myopathy. Muscle Nerve 1997;20:665–673.
13. Gooch JL. AAEM Case Report #29: prolonged paralysis after neuromuscular blockade. Muscle Nerve 1995;18:937–942.